Applied Evidence

CASE 

Mr. A is a 30-year-old man with neurofibromatosis and myelopathy with associated quadriplegia, complicated by dysphasia and chronic hypercapnic respiratory failure requiring a tracheostomy. He is cared for at home by his very competent mother but requires regular visits with his medical providers for assistance with his complex care needs. Due to logistical challenges, he had been receiving regular home visits even before the ­COVID-19 pandemic.

After estimating the risk of exposure to the patient, Mr. A’s family and his physician’s office staff scheduled a home visit. Before the appointment, the doctor conducted a virtual visit with the patient and family members to screen for COVID-19 infection, which proved negative. The doctor arranged a visit to coincide with Mr. A’s regular appointment with the home health nurse. He invited the patient’s social worker to attend, as well.

The providers donned masks, face shields, and gloves before entering the home. Mr. A’s temperature was checked and was normal. The team completed a physical exam, assessed the patient’s current needs, and refilled prescriptions. The doctor, nurse, and social worker met afterward in the family’s driveway to coordinate plans for the patient’s future care.

This encounter allowed a vulnerable patient with special needs to have access to care while reducing his risk of undesirable exposure. Also, his health care team’s provision of care in the home setting reduced Mr. A’s anxiety and that of his family members.

Home visits have long been an integral part of what it means to be a family physician. In 1930, roughly 40% of all patient-physician encounters in the United States occurred in patients’ homes. By 1980, this number had dropped to < 1%.¹ Still, a 1994 survey of American doctors in 3 primary care specialties revealed that 63% of family physicians, more than the other 2 specialties, still made house calls.² A 2016 analysis of Medicare claims data showed that between 2006 and 2011, only 5% of American doctors overall made house calls on Medicare recipients, but interestingly, the total number of home visits was increasing.³

This resurgence of interest in home health care is due in part to the increasing number of homebound patients in America, which exceeds the number of those in nursing homes.⁴ Further, a growing body of evidence indicates that home visits improve patient outcomes. And finally, many family physicians whose work lives have been centered around a busy office or hospital practice have found satisfaction in once again seeing patients in their own homes.

The COVID-19 pandemic has of course presented unique challenges—and opportunities, too—for home visits, which we discuss at the end of the article.

In the elderly, home visits have reduced functional decline, nursing home admissions, and mortality by 25% to 33%.

Why aren’t more of us making home visits?

For most of us, the decision not to make home visits is simply a matter of time and money. Although Medicare reimbursement for a home visit is typically about 150% that of a comparable office visit,⁵ it’s difficult, if not impossible, to make 2 home visits in the time you could see 3 patients in the office. So, economically it’s a net loss. Furthermore, we tend to feel less comfortable in our patients’ homes than in our offices. We have less control outside our own environment, and what happens away from our office is often less predictable—sometimes to the point that we may be concerned for our safety.

Continue to: So why make home visits at all?

So why make home visits at all?

First and foremost, home visits improve patient outcomes. This is most evident in our more vulnerable patients: newborns and the elderly, those who have been recently hospitalized, and those at risk because of their particular home situation. Multiple studies have shown that, for elders, home visits reduce functional decline, nursing home admissions, and mortality by around 25% to 33%.^6-8 For those at risk of abuse, a recent systematic review showed that home visits reduce intimate partner violence and child abuse.⁹ Another systematic review demonstrated that patients with diabetes who received home visits vs usual care were more likely to show improvements in quality of life.¹⁰ These patients were also more likely to have lower HbA1c levels and lower systolic blood pressure readings.¹⁰ A few caveats apply to these studies:

• all of them targeted “vulnerable” patients
• most studies enlisted interdisciplinary teams and had regular team meetings
• most findings reached significance only after multiple home visits.

A further reason for choosing to become involved in home care is that it builds relationships, understanding, and empathy with our patients. “There is deep symbolism in the home visit.... It says, ‘I care enough about you to leave my power base … to come and see you on your own ground.’”¹¹ And this benefit is 2-way; we also grow to understand and appreciate our patients better, especially if they are different from us culturally or socioeconomically.

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention. For the patient, home visits foster a strong sense of trust with the individual doctor and our health delivery network, and they decrease the need to seek emergency services. Finally, it has been demonstrated that provider satisfaction improves when home visits are incorporated into the work week.¹²

What is the role of community health workers in home-based care?

Community health workers (CHWs), defined as “frontline public health workers who are trusted members of and/or have an unusually close understanding of the community they serve,”¹³ can be an integral part of the home-based care team. Although CHWs have variable amounts of formal training, they have a unique perspective on local health beliefs and practices, which can assist the home-care team in providing culturally competent health care services and reduce health care costs.

In a study of children with asthma in Seattle, Washington, patients were randomized to a group that had 4 home visits by CHWs and a group that received usual care. The group that received home visits demonstrated more asthma symptom–free days, improved quality-of-life scores, and fewer urgent care visits.¹⁴ Furthermore, the intervention was estimated to save approximately $1300 per patient, resulting in a return on investment of 190%. Similarly, in a study comparing inappropriate emergency department (ED) visits between children who received CHW visits and those who did not, patients in the intervention group were significantly less likely to visit the ED for ambulatory complaints (18.2% vs 35.1%; P = .004).¹⁵

Continue to: What is the role of social workersin home-based care?

What is the role of social workersin home-based care?

Social workers can help meet the complex medical and biopsychosocial needs of the homebound population.¹⁶ A study by Cohen et al based in Israel concluded that homebound participants had a significantly higher risk for mortality, higher rates of depression, and difficulty completing instrumental activities of daily living when compared with their non-homebound counterparts.¹⁷

The Mount Sinai (New York) Visiting Doctors Program (MSVD) is a home-based care team that uses social workers to meet the needs of their complex patients.¹⁸ The social workers in the MSVD program provide direct counseling, make referrals to government and community resources, and monitor caregiver burden. Using a combination of measurement tools to assess caregiver burden, Ornstein et al demonstrated that the MSVD program led to a decrease in unmet needs and in caregiver burden.^19,20 Caregiver burnout can be assessed using the Caregiver Burden Inventory, a validated 24-item questionnaire.²¹

What electronic tools are availableto monitor patients at home?

Although expensive in terms of both dollars and personnel time, telemonitoring allows home care providers to receive real-time, updated information regarding their patients.

Chronic obstructive pulmonary disease (COPD). One systematic review showed that although telemonitoring of patients with COPD improved quality of life and decreased COPD exacerbations, it did not reduce the risk of hospitalization and, therefore, did not reduce health care costs.²² Telemonitoring in COPD can include transmission of data about spirometry parameters, weight, temperature, blood pressure, sputum color, and 6-minute walk distance.^23,24

Congestive heart failure (CHF). A 2010 Cochrane review found that telemonitoring of patients with CHF reduced all-cause mortality (risk ratio [RR] = 0.66; P < .0001).²⁵ The Telemedical Interventional Management in Heart Failure II (TIM-HF2) trial,conducted from 2013 to 2017, compared usual care for CHF patients with care incorporating daily transmission of body weight, blood pressure, heart rate, electrocardiogram tracings, pulse oximetry, and self-rated health status.²⁶ This study showed that the average number of days lost per year due to hospital admission was less in the telemonitoring group than in the usual care group (17.8 days vs. 24.2 days; P = .046). All-cause mortality was also reduced in the telemonitoring group (hazard ratio = 0.70; P = .028).

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention.

Continue to: What role do “home hospitals” play?

What role do “home hospitals” play?

Home hospitals provide acute or subacute treatment in a patient’s home for a condition that would normally require hospitalization.²⁷ In a meta-analysis of 61 studies evaluating the effectiveness of home hospitals, this option was more likely to reduce mortality (odds ratio [OR] = 0.81; P = .008) and to reduce readmission rates (OR = 0.75; P = .02).²⁸ In a study of 455 older adults, Leff et al found that hospital-at-home was associated with a shorter length of stay (3.2 vs. 4.9 days; P = .004) and that the mean cost was lower for hospital-at-home vs traditional hospital care.²⁹

However, a 2016 Cochrane review of 16 randomized controlled trials comparing hospital-at-home with traditional hospital care showed that while care in a hospital-­at-home may decrease formal costs, if costs for caregivers are taken into account, any difference in cost may disappear.³⁰

Although the evidence for cost saving is variable, hospital-at-home admission has been shown to reduce the likelihood of living in a residential care facility at 6 months (RR = 0.35; P < .0001).³⁰ Further, the same Cochrane review showed that admission avoidance may increase patient satisfaction with the care provided.³⁰

Finally, a recent randomized trial in a Boston-area hospital system showed that patients cared for in hospital-at-home were significantly less likely to be readmitted within 30 days and that adjusted cost was about two-thirds the cost of traditional hospital care.³¹

What is the physician’s rolein home health care?

While home health care is a team effort, the physician has several crucial roles. First, he or she must make the determination that home care is appropriate and feasible for a particular patient. Appropriate, meaning there is evidence that this patient is likely to benefit from home care. Feasible, meaning there are resources available in the community and family to safely care for the patient at home. “Often a house call will serve as the first step in developing a home-based-management plan.”³²

Patients with diabetes receiving home care are more likely to have improved quality of life, lower HbA1c levels, and lower systolic BP readings.

Continue to: Second, the physician serves...

Second, the physician serves an important role in directing and coordinating the team of professionals involved. This primarily means helping the team to communicate with one another. Before home visits begin, the physician’s office should reach out not only to the patient and family, but also to any other health care personnel involved in the patient’s home care. Otherwise, many of the health care providers involved will never have face-to-face interaction with the physician. Creation of the coordinated health team minimizes duplication and miscommunication; it also builds a valuable bond.

How does one go about making a home visit?

Scheduling. What often works best in a busy practice is to schedule home visits for the end of the workday or to devote an entire afternoon to making home visits to several patients in one locale. Also important is scheduling times, if possible, when important family members or other caregivers are at home or when other members of the home care team can accompany you.

What to bring along. Carry a “home visit bag” that includes equipment you’re likely to need and that is not available away from your office. A minimally equipped visit bag would include different-sized blood pressure cuffs, a glucometer, a pulse oximeter, thermometers, and patient education materials. Other suggested contents are listed in TABLE 1. Due to the COVID-19 pandemic, providers should also carry adequate personal protective equipment (PPE), including an N-95 mask.

Dos and don’ts. Take a few minutes when you first arrive to simply visit with the patient. Sit down and introduce yourself and any members of the home care team that the patient has not met. Take an interim history. While you’re doing this, be observant: Is the home neat or cluttered? Is the indoor temperature comfortable? Are there fall hazards? Is there a smell of cigarette smoke? Are there any indoor combustion sources (eg, wood stove or kerosene heater)? Ask questions such as: Who lives here with you? Can you show me where you keep your medicines? (If the patient keeps insulin or any other medicines in the refrigerator, ask to see it. Note any apparent food scarcity.)

During your exam, pay particular attention to whether vital signs are appreciably different than those measured in the office or hospital. Pay special attention to the patient’s functional abilities. “A subtle, but critical distinction between medical management in the home and medical management in the hospital, clinic, or office is the emphasis on the patient’s functional abilities, family assistance, and environmental factors.”³³

Observe the patient’s use of any home technology, if possible; this can be as simple as home oxygenation or as complex as home hemodialysis. Assess for any apparent caregiver stress. Finally, don’t neglect to offer appropriate emotional and spiritual support to the patient and family and to schedule the next follow-up visit before you leave.

Continue to:  Documentation and reimbursement.

Not all assessments will be able to be done at each visit but seeing them listed in the template can be helpful. Billing follows the same principles as for office visits and has similar requirements for documentation. Codes for the most common types of home visits are listed in TABLE 3.

Where can I get help?

Graduates of family medicine residency programs are required to receive training in home visits by the Accreditation Council for Graduate Medical Education (ACGME). Current ACGME program requirements stipulate that “residents must demonstrate competence to independently diagnose, manage, and integrate the care of patients of all ages in various outpatient settings, including the FMP [family medicine practice] site and home environment,” and “residents must be primarily responsible for a panel of continuity patients, integrating each patient’s care across all settings, including the home ...” [emphasis added].³⁴

For those already in practice, one of the hardest parts of doing home visits is feeling alone, especially if few other providers in your community engage in home care. As you run into questions and challenges with incorporating home care of patients into your practice, one excellent resource is the American Academy of Home Care Medicine (www.aahcm.org/). Founded in 1988 and headquartered in Chicago, it not only provides numerous helpful resources, but serves as a networking tool for physicians involved in home care.

Using a home visit template can help with documentation and reimbursement.

This unprecedented pandemichas allowed home visits to shine

As depicted in our opening patient case, patients who have high-risk conditions and those who are older than 65 years of age may be cared for more appropriately in a home visit rather than having them come to the office. Home visits may also be a way for providers to “lay eyes” on patients who do not have technology available to participate in virtual visits.

Before performing a home visit, inquire as to whether the patient has symptoms of COVID-19. Adequate PPE should be donned at all times and social distancing should be practiced when appropriate. With adequate PPE, home visits may also allow providers to care for low-risk patients known to have ­COVID-19 and thereby minimize risks to staff and other patients in the office. JFP

CORRESPONDENCE
Curt Elliott, MD, Prisma Health USC Family Medicine Center, 3209 Colonial Drive, Columbia, SC 29203; [email protected].

CASE 

Mr. A is a 30-year-old man with neurofibromatosis and myelopathy with associated quadriplegia, complicated by dysphasia and chronic hypercapnic respiratory failure requiring a tracheostomy. He is cared for at home by his very competent mother but requires regular visits with his medical providers for assistance with his complex care needs. Due to logistical challenges, he had been receiving regular home visits even before the ­COVID-19 pandemic.

After estimating the risk of exposure to the patient, Mr. A’s family and his physician’s office staff scheduled a home visit. Before the appointment, the doctor conducted a virtual visit with the patient and family members to screen for COVID-19 infection, which proved negative. The doctor arranged a visit to coincide with Mr. A’s regular appointment with the home health nurse. He invited the patient’s social worker to attend, as well.

The providers donned masks, face shields, and gloves before entering the home. Mr. A’s temperature was checked and was normal. The team completed a physical exam, assessed the patient’s current needs, and refilled prescriptions. The doctor, nurse, and social worker met afterward in the family’s driveway to coordinate plans for the patient’s future care.

This encounter allowed a vulnerable patient with special needs to have access to care while reducing his risk of undesirable exposure. Also, his health care team’s provision of care in the home setting reduced Mr. A’s anxiety and that of his family members.

Home visits have long been an integral part of what it means to be a family physician. In 1930, roughly 40% of all patient-physician encounters in the United States occurred in patients’ homes. By 1980, this number had dropped to < 1%.¹ Still, a 1994 survey of American doctors in 3 primary care specialties revealed that 63% of family physicians, more than the other 2 specialties, still made house calls.² A 2016 analysis of Medicare claims data showed that between 2006 and 2011, only 5% of American doctors overall made house calls on Medicare recipients, but interestingly, the total number of home visits was increasing.³

This resurgence of interest in home health care is due in part to the increasing number of homebound patients in America, which exceeds the number of those in nursing homes.⁴ Further, a growing body of evidence indicates that home visits improve patient outcomes. And finally, many family physicians whose work lives have been centered around a busy office or hospital practice have found satisfaction in once again seeing patients in their own homes.

The COVID-19 pandemic has of course presented unique challenges—and opportunities, too—for home visits, which we discuss at the end of the article.

In the elderly, home visits have reduced functional decline, nursing home admissions, and mortality by 25% to 33%.

Why aren’t more of us making home visits?

For most of us, the decision not to make home visits is simply a matter of time and money. Although Medicare reimbursement for a home visit is typically about 150% that of a comparable office visit,⁵ it’s difficult, if not impossible, to make 2 home visits in the time you could see 3 patients in the office. So, economically it’s a net loss. Furthermore, we tend to feel less comfortable in our patients’ homes than in our offices. We have less control outside our own environment, and what happens away from our office is often less predictable—sometimes to the point that we may be concerned for our safety.

Continue to: So why make home visits at all?

So why make home visits at all?

First and foremost, home visits improve patient outcomes. This is most evident in our more vulnerable patients: newborns and the elderly, those who have been recently hospitalized, and those at risk because of their particular home situation. Multiple studies have shown that, for elders, home visits reduce functional decline, nursing home admissions, and mortality by around 25% to 33%.^6-8 For those at risk of abuse, a recent systematic review showed that home visits reduce intimate partner violence and child abuse.⁹ Another systematic review demonstrated that patients with diabetes who received home visits vs usual care were more likely to show improvements in quality of life.¹⁰ These patients were also more likely to have lower HbA1c levels and lower systolic blood pressure readings.¹⁰ A few caveats apply to these studies:

• all of them targeted “vulnerable” patients
• most studies enlisted interdisciplinary teams and had regular team meetings
• most findings reached significance only after multiple home visits.

A further reason for choosing to become involved in home care is that it builds relationships, understanding, and empathy with our patients. “There is deep symbolism in the home visit.... It says, ‘I care enough about you to leave my power base … to come and see you on your own ground.’”¹¹ And this benefit is 2-way; we also grow to understand and appreciate our patients better, especially if they are different from us culturally or socioeconomically.

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention. For the patient, home visits foster a strong sense of trust with the individual doctor and our health delivery network, and they decrease the need to seek emergency services. Finally, it has been demonstrated that provider satisfaction improves when home visits are incorporated into the work week.¹²

What is the role of community health workers in home-based care?

Community health workers (CHWs), defined as “frontline public health workers who are trusted members of and/or have an unusually close understanding of the community they serve,”¹³ can be an integral part of the home-based care team. Although CHWs have variable amounts of formal training, they have a unique perspective on local health beliefs and practices, which can assist the home-care team in providing culturally competent health care services and reduce health care costs.

In a study of children with asthma in Seattle, Washington, patients were randomized to a group that had 4 home visits by CHWs and a group that received usual care. The group that received home visits demonstrated more asthma symptom–free days, improved quality-of-life scores, and fewer urgent care visits.¹⁴ Furthermore, the intervention was estimated to save approximately $1300 per patient, resulting in a return on investment of 190%. Similarly, in a study comparing inappropriate emergency department (ED) visits between children who received CHW visits and those who did not, patients in the intervention group were significantly less likely to visit the ED for ambulatory complaints (18.2% vs 35.1%; P = .004).¹⁵

Continue to: What is the role of social workersin home-based care?

What is the role of social workersin home-based care?

Social workers can help meet the complex medical and biopsychosocial needs of the homebound population.¹⁶ A study by Cohen et al based in Israel concluded that homebound participants had a significantly higher risk for mortality, higher rates of depression, and difficulty completing instrumental activities of daily living when compared with their non-homebound counterparts.¹⁷

The Mount Sinai (New York) Visiting Doctors Program (MSVD) is a home-based care team that uses social workers to meet the needs of their complex patients.¹⁸ The social workers in the MSVD program provide direct counseling, make referrals to government and community resources, and monitor caregiver burden. Using a combination of measurement tools to assess caregiver burden, Ornstein et al demonstrated that the MSVD program led to a decrease in unmet needs and in caregiver burden.^19,20 Caregiver burnout can be assessed using the Caregiver Burden Inventory, a validated 24-item questionnaire.²¹

What electronic tools are availableto monitor patients at home?

Although expensive in terms of both dollars and personnel time, telemonitoring allows home care providers to receive real-time, updated information regarding their patients.

Chronic obstructive pulmonary disease (COPD). One systematic review showed that although telemonitoring of patients with COPD improved quality of life and decreased COPD exacerbations, it did not reduce the risk of hospitalization and, therefore, did not reduce health care costs.²² Telemonitoring in COPD can include transmission of data about spirometry parameters, weight, temperature, blood pressure, sputum color, and 6-minute walk distance.^23,24

Congestive heart failure (CHF). A 2010 Cochrane review found that telemonitoring of patients with CHF reduced all-cause mortality (risk ratio [RR] = 0.66; P < .0001).²⁵ The Telemedical Interventional Management in Heart Failure II (TIM-HF2) trial,conducted from 2013 to 2017, compared usual care for CHF patients with care incorporating daily transmission of body weight, blood pressure, heart rate, electrocardiogram tracings, pulse oximetry, and self-rated health status.²⁶ This study showed that the average number of days lost per year due to hospital admission was less in the telemonitoring group than in the usual care group (17.8 days vs. 24.2 days; P = .046). All-cause mortality was also reduced in the telemonitoring group (hazard ratio = 0.70; P = .028).

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention.

Continue to: What role do “home hospitals” play?

What role do “home hospitals” play?

Home hospitals provide acute or subacute treatment in a patient’s home for a condition that would normally require hospitalization.²⁷ In a meta-analysis of 61 studies evaluating the effectiveness of home hospitals, this option was more likely to reduce mortality (odds ratio [OR] = 0.81; P = .008) and to reduce readmission rates (OR = 0.75; P = .02).²⁸ In a study of 455 older adults, Leff et al found that hospital-at-home was associated with a shorter length of stay (3.2 vs. 4.9 days; P = .004) and that the mean cost was lower for hospital-at-home vs traditional hospital care.²⁹

However, a 2016 Cochrane review of 16 randomized controlled trials comparing hospital-at-home with traditional hospital care showed that while care in a hospital-­at-home may decrease formal costs, if costs for caregivers are taken into account, any difference in cost may disappear.³⁰

Although the evidence for cost saving is variable, hospital-at-home admission has been shown to reduce the likelihood of living in a residential care facility at 6 months (RR = 0.35; P < .0001).³⁰ Further, the same Cochrane review showed that admission avoidance may increase patient satisfaction with the care provided.³⁰

Finally, a recent randomized trial in a Boston-area hospital system showed that patients cared for in hospital-at-home were significantly less likely to be readmitted within 30 days and that adjusted cost was about two-thirds the cost of traditional hospital care.³¹

What is the physician’s rolein home health care?

While home health care is a team effort, the physician has several crucial roles. First, he or she must make the determination that home care is appropriate and feasible for a particular patient. Appropriate, meaning there is evidence that this patient is likely to benefit from home care. Feasible, meaning there are resources available in the community and family to safely care for the patient at home. “Often a house call will serve as the first step in developing a home-based-management plan.”³²

Patients with diabetes receiving home care are more likely to have improved quality of life, lower HbA1c levels, and lower systolic BP readings.

Continue to: Second, the physician serves...

Second, the physician serves an important role in directing and coordinating the team of professionals involved. This primarily means helping the team to communicate with one another. Before home visits begin, the physician’s office should reach out not only to the patient and family, but also to any other health care personnel involved in the patient’s home care. Otherwise, many of the health care providers involved will never have face-to-face interaction with the physician. Creation of the coordinated health team minimizes duplication and miscommunication; it also builds a valuable bond.

How does one go about making a home visit?

Scheduling. What often works best in a busy practice is to schedule home visits for the end of the workday or to devote an entire afternoon to making home visits to several patients in one locale. Also important is scheduling times, if possible, when important family members or other caregivers are at home or when other members of the home care team can accompany you.

What to bring along. Carry a “home visit bag” that includes equipment you’re likely to need and that is not available away from your office. A minimally equipped visit bag would include different-sized blood pressure cuffs, a glucometer, a pulse oximeter, thermometers, and patient education materials. Other suggested contents are listed in TABLE 1. Due to the COVID-19 pandemic, providers should also carry adequate personal protective equipment (PPE), including an N-95 mask.

Dos and don’ts. Take a few minutes when you first arrive to simply visit with the patient. Sit down and introduce yourself and any members of the home care team that the patient has not met. Take an interim history. While you’re doing this, be observant: Is the home neat or cluttered? Is the indoor temperature comfortable? Are there fall hazards? Is there a smell of cigarette smoke? Are there any indoor combustion sources (eg, wood stove or kerosene heater)? Ask questions such as: Who lives here with you? Can you show me where you keep your medicines? (If the patient keeps insulin or any other medicines in the refrigerator, ask to see it. Note any apparent food scarcity.)

During your exam, pay particular attention to whether vital signs are appreciably different than those measured in the office or hospital. Pay special attention to the patient’s functional abilities. “A subtle, but critical distinction between medical management in the home and medical management in the hospital, clinic, or office is the emphasis on the patient’s functional abilities, family assistance, and environmental factors.”³³

Observe the patient’s use of any home technology, if possible; this can be as simple as home oxygenation or as complex as home hemodialysis. Assess for any apparent caregiver stress. Finally, don’t neglect to offer appropriate emotional and spiritual support to the patient and family and to schedule the next follow-up visit before you leave.

Continue to:  Documentation and reimbursement.

Not all assessments will be able to be done at each visit but seeing them listed in the template can be helpful. Billing follows the same principles as for office visits and has similar requirements for documentation. Codes for the most common types of home visits are listed in TABLE 3.

Where can I get help?

Graduates of family medicine residency programs are required to receive training in home visits by the Accreditation Council for Graduate Medical Education (ACGME). Current ACGME program requirements stipulate that “residents must demonstrate competence to independently diagnose, manage, and integrate the care of patients of all ages in various outpatient settings, including the FMP [family medicine practice] site and home environment,” and “residents must be primarily responsible for a panel of continuity patients, integrating each patient’s care across all settings, including the home ...” [emphasis added].³⁴

For those already in practice, one of the hardest parts of doing home visits is feeling alone, especially if few other providers in your community engage in home care. As you run into questions and challenges with incorporating home care of patients into your practice, one excellent resource is the American Academy of Home Care Medicine (www.aahcm.org/). Founded in 1988 and headquartered in Chicago, it not only provides numerous helpful resources, but serves as a networking tool for physicians involved in home care.

Using a home visit template can help with documentation and reimbursement.

This unprecedented pandemichas allowed home visits to shine

As depicted in our opening patient case, patients who have high-risk conditions and those who are older than 65 years of age may be cared for more appropriately in a home visit rather than having them come to the office. Home visits may also be a way for providers to “lay eyes” on patients who do not have technology available to participate in virtual visits.

Before performing a home visit, inquire as to whether the patient has symptoms of COVID-19. Adequate PPE should be donned at all times and social distancing should be practiced when appropriate. With adequate PPE, home visits may also allow providers to care for low-risk patients known to have ­COVID-19 and thereby minimize risks to staff and other patients in the office. JFP

CORRESPONDENCE
Curt Elliott, MD, Prisma Health USC Family Medicine Center, 3209 Colonial Drive, Columbia, SC 29203; [email protected].

CASE 

Mr. A is a 30-year-old man with neurofibromatosis and myelopathy with associated quadriplegia, complicated by dysphasia and chronic hypercapnic respiratory failure requiring a tracheostomy. He is cared for at home by his very competent mother but requires regular visits with his medical providers for assistance with his complex care needs. Due to logistical challenges, he had been receiving regular home visits even before the ­COVID-19 pandemic.

After estimating the risk of exposure to the patient, Mr. A’s family and his physician’s office staff scheduled a home visit. Before the appointment, the doctor conducted a virtual visit with the patient and family members to screen for COVID-19 infection, which proved negative. The doctor arranged a visit to coincide with Mr. A’s regular appointment with the home health nurse. He invited the patient’s social worker to attend, as well.

The providers donned masks, face shields, and gloves before entering the home. Mr. A’s temperature was checked and was normal. The team completed a physical exam, assessed the patient’s current needs, and refilled prescriptions. The doctor, nurse, and social worker met afterward in the family’s driveway to coordinate plans for the patient’s future care.

This encounter allowed a vulnerable patient with special needs to have access to care while reducing his risk of undesirable exposure. Also, his health care team’s provision of care in the home setting reduced Mr. A’s anxiety and that of his family members.

Home visits have long been an integral part of what it means to be a family physician. In 1930, roughly 40% of all patient-physician encounters in the United States occurred in patients’ homes. By 1980, this number had dropped to < 1%.¹ Still, a 1994 survey of American doctors in 3 primary care specialties revealed that 63% of family physicians, more than the other 2 specialties, still made house calls.² A 2016 analysis of Medicare claims data showed that between 2006 and 2011, only 5% of American doctors overall made house calls on Medicare recipients, but interestingly, the total number of home visits was increasing.³

This resurgence of interest in home health care is due in part to the increasing number of homebound patients in America, which exceeds the number of those in nursing homes.⁴ Further, a growing body of evidence indicates that home visits improve patient outcomes. And finally, many family physicians whose work lives have been centered around a busy office or hospital practice have found satisfaction in once again seeing patients in their own homes.

The COVID-19 pandemic has of course presented unique challenges—and opportunities, too—for home visits, which we discuss at the end of the article.

In the elderly, home visits have reduced functional decline, nursing home admissions, and mortality by 25% to 33%.

Why aren’t more of us making home visits?

For most of us, the decision not to make home visits is simply a matter of time and money. Although Medicare reimbursement for a home visit is typically about 150% that of a comparable office visit,⁵ it’s difficult, if not impossible, to make 2 home visits in the time you could see 3 patients in the office. So, economically it’s a net loss. Furthermore, we tend to feel less comfortable in our patients’ homes than in our offices. We have less control outside our own environment, and what happens away from our office is often less predictable—sometimes to the point that we may be concerned for our safety.

Continue to: So why make home visits at all?

So why make home visits at all?

First and foremost, home visits improve patient outcomes. This is most evident in our more vulnerable patients: newborns and the elderly, those who have been recently hospitalized, and those at risk because of their particular home situation. Multiple studies have shown that, for elders, home visits reduce functional decline, nursing home admissions, and mortality by around 25% to 33%.^6-8 For those at risk of abuse, a recent systematic review showed that home visits reduce intimate partner violence and child abuse.⁹ Another systematic review demonstrated that patients with diabetes who received home visits vs usual care were more likely to show improvements in quality of life.¹⁰ These patients were also more likely to have lower HbA1c levels and lower systolic blood pressure readings.¹⁰ A few caveats apply to these studies:

• all of them targeted “vulnerable” patients
• most studies enlisted interdisciplinary teams and had regular team meetings
• most findings reached significance only after multiple home visits.

A further reason for choosing to become involved in home care is that it builds relationships, understanding, and empathy with our patients. “There is deep symbolism in the home visit.... It says, ‘I care enough about you to leave my power base … to come and see you on your own ground.’”¹¹ And this benefit is 2-way; we also grow to understand and appreciate our patients better, especially if they are different from us culturally or socioeconomically.

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention. For the patient, home visits foster a strong sense of trust with the individual doctor and our health delivery network, and they decrease the need to seek emergency services. Finally, it has been demonstrated that provider satisfaction improves when home visits are incorporated into the work week.¹²

What is the role of community health workers in home-based care?

Community health workers (CHWs), defined as “frontline public health workers who are trusted members of and/or have an unusually close understanding of the community they serve,”¹³ can be an integral part of the home-based care team. Although CHWs have variable amounts of formal training, they have a unique perspective on local health beliefs and practices, which can assist the home-care team in providing culturally competent health care services and reduce health care costs.

In a study of children with asthma in Seattle, Washington, patients were randomized to a group that had 4 home visits by CHWs and a group that received usual care. The group that received home visits demonstrated more asthma symptom–free days, improved quality-of-life scores, and fewer urgent care visits.¹⁴ Furthermore, the intervention was estimated to save approximately $1300 per patient, resulting in a return on investment of 190%. Similarly, in a study comparing inappropriate emergency department (ED) visits between children who received CHW visits and those who did not, patients in the intervention group were significantly less likely to visit the ED for ambulatory complaints (18.2% vs 35.1%; P = .004).¹⁵

Continue to: What is the role of social workersin home-based care?

What is the role of social workersin home-based care?

Social workers can help meet the complex medical and biopsychosocial needs of the homebound population.¹⁶ A study by Cohen et al based in Israel concluded that homebound participants had a significantly higher risk for mortality, higher rates of depression, and difficulty completing instrumental activities of daily living when compared with their non-homebound counterparts.¹⁷

The Mount Sinai (New York) Visiting Doctors Program (MSVD) is a home-based care team that uses social workers to meet the needs of their complex patients.¹⁸ The social workers in the MSVD program provide direct counseling, make referrals to government and community resources, and monitor caregiver burden. Using a combination of measurement tools to assess caregiver burden, Ornstein et al demonstrated that the MSVD program led to a decrease in unmet needs and in caregiver burden.^19,20 Caregiver burnout can be assessed using the Caregiver Burden Inventory, a validated 24-item questionnaire.²¹

What electronic tools are availableto monitor patients at home?

Although expensive in terms of both dollars and personnel time, telemonitoring allows home care providers to receive real-time, updated information regarding their patients.

Chronic obstructive pulmonary disease (COPD). One systematic review showed that although telemonitoring of patients with COPD improved quality of life and decreased COPD exacerbations, it did not reduce the risk of hospitalization and, therefore, did not reduce health care costs.²² Telemonitoring in COPD can include transmission of data about spirometry parameters, weight, temperature, blood pressure, sputum color, and 6-minute walk distance.^23,24

Congestive heart failure (CHF). A 2010 Cochrane review found that telemonitoring of patients with CHF reduced all-cause mortality (risk ratio [RR] = 0.66; P < .0001).²⁵ The Telemedical Interventional Management in Heart Failure II (TIM-HF2) trial,conducted from 2013 to 2017, compared usual care for CHF patients with care incorporating daily transmission of body weight, blood pressure, heart rate, electrocardiogram tracings, pulse oximetry, and self-rated health status.²⁶ This study showed that the average number of days lost per year due to hospital admission was less in the telemonitoring group than in the usual care group (17.8 days vs. 24.2 days; P = .046). All-cause mortality was also reduced in the telemonitoring group (hazard ratio = 0.70; P = .028).

Home visits allow the medical team to see challenges the patient has grown accustomed to, and perhaps ones that the patient has deemed too insignificant to mention.

Continue to: What role do “home hospitals” play?

What role do “home hospitals” play?

Home hospitals provide acute or subacute treatment in a patient’s home for a condition that would normally require hospitalization.²⁷ In a meta-analysis of 61 studies evaluating the effectiveness of home hospitals, this option was more likely to reduce mortality (odds ratio [OR] = 0.81; P = .008) and to reduce readmission rates (OR = 0.75; P = .02).²⁸ In a study of 455 older adults, Leff et al found that hospital-at-home was associated with a shorter length of stay (3.2 vs. 4.9 days; P = .004) and that the mean cost was lower for hospital-at-home vs traditional hospital care.²⁹

However, a 2016 Cochrane review of 16 randomized controlled trials comparing hospital-at-home with traditional hospital care showed that while care in a hospital-­at-home may decrease formal costs, if costs for caregivers are taken into account, any difference in cost may disappear.³⁰

Although the evidence for cost saving is variable, hospital-at-home admission has been shown to reduce the likelihood of living in a residential care facility at 6 months (RR = 0.35; P < .0001).³⁰ Further, the same Cochrane review showed that admission avoidance may increase patient satisfaction with the care provided.³⁰

Finally, a recent randomized trial in a Boston-area hospital system showed that patients cared for in hospital-at-home were significantly less likely to be readmitted within 30 days and that adjusted cost was about two-thirds the cost of traditional hospital care.³¹

What is the physician’s rolein home health care?

While home health care is a team effort, the physician has several crucial roles. First, he or she must make the determination that home care is appropriate and feasible for a particular patient. Appropriate, meaning there is evidence that this patient is likely to benefit from home care. Feasible, meaning there are resources available in the community and family to safely care for the patient at home. “Often a house call will serve as the first step in developing a home-based-management plan.”³²

Patients with diabetes receiving home care are more likely to have improved quality of life, lower HbA1c levels, and lower systolic BP readings.

Continue to: Second, the physician serves...

Second, the physician serves an important role in directing and coordinating the team of professionals involved. This primarily means helping the team to communicate with one another. Before home visits begin, the physician’s office should reach out not only to the patient and family, but also to any other health care personnel involved in the patient’s home care. Otherwise, many of the health care providers involved will never have face-to-face interaction with the physician. Creation of the coordinated health team minimizes duplication and miscommunication; it also builds a valuable bond.

How does one go about making a home visit?

Scheduling. What often works best in a busy practice is to schedule home visits for the end of the workday or to devote an entire afternoon to making home visits to several patients in one locale. Also important is scheduling times, if possible, when important family members or other caregivers are at home or when other members of the home care team can accompany you.

What to bring along. Carry a “home visit bag” that includes equipment you’re likely to need and that is not available away from your office. A minimally equipped visit bag would include different-sized blood pressure cuffs, a glucometer, a pulse oximeter, thermometers, and patient education materials. Other suggested contents are listed in TABLE 1. Due to the COVID-19 pandemic, providers should also carry adequate personal protective equipment (PPE), including an N-95 mask.

Dos and don’ts. Take a few minutes when you first arrive to simply visit with the patient. Sit down and introduce yourself and any members of the home care team that the patient has not met. Take an interim history. While you’re doing this, be observant: Is the home neat or cluttered? Is the indoor temperature comfortable? Are there fall hazards? Is there a smell of cigarette smoke? Are there any indoor combustion sources (eg, wood stove or kerosene heater)? Ask questions such as: Who lives here with you? Can you show me where you keep your medicines? (If the patient keeps insulin or any other medicines in the refrigerator, ask to see it. Note any apparent food scarcity.)

During your exam, pay particular attention to whether vital signs are appreciably different than those measured in the office or hospital. Pay special attention to the patient’s functional abilities. “A subtle, but critical distinction between medical management in the home and medical management in the hospital, clinic, or office is the emphasis on the patient’s functional abilities, family assistance, and environmental factors.”³³

Observe the patient’s use of any home technology, if possible; this can be as simple as home oxygenation or as complex as home hemodialysis. Assess for any apparent caregiver stress. Finally, don’t neglect to offer appropriate emotional and spiritual support to the patient and family and to schedule the next follow-up visit before you leave.

Continue to:  Documentation and reimbursement.

Not all assessments will be able to be done at each visit but seeing them listed in the template can be helpful. Billing follows the same principles as for office visits and has similar requirements for documentation. Codes for the most common types of home visits are listed in TABLE 3.

Where can I get help?

Graduates of family medicine residency programs are required to receive training in home visits by the Accreditation Council for Graduate Medical Education (ACGME). Current ACGME program requirements stipulate that “residents must demonstrate competence to independently diagnose, manage, and integrate the care of patients of all ages in various outpatient settings, including the FMP [family medicine practice] site and home environment,” and “residents must be primarily responsible for a panel of continuity patients, integrating each patient’s care across all settings, including the home ...” [emphasis added].³⁴

For those already in practice, one of the hardest parts of doing home visits is feeling alone, especially if few other providers in your community engage in home care. As you run into questions and challenges with incorporating home care of patients into your practice, one excellent resource is the American Academy of Home Care Medicine (www.aahcm.org/). Founded in 1988 and headquartered in Chicago, it not only provides numerous helpful resources, but serves as a networking tool for physicians involved in home care.

Using a home visit template can help with documentation and reimbursement.

This unprecedented pandemichas allowed home visits to shine

As depicted in our opening patient case, patients who have high-risk conditions and those who are older than 65 years of age may be cared for more appropriately in a home visit rather than having them come to the office. Home visits may also be a way for providers to “lay eyes” on patients who do not have technology available to participate in virtual visits.

Before performing a home visit, inquire as to whether the patient has symptoms of COVID-19. Adequate PPE should be donned at all times and social distancing should be practiced when appropriate. With adequate PPE, home visits may also allow providers to care for low-risk patients known to have ­COVID-19 and thereby minimize risks to staff and other patients in the office. JFP

CORRESPONDENCE
Curt Elliott, MD, Prisma Health USC Family Medicine Center, 3209 Colonial Drive, Columbia, SC 29203; [email protected].

Mention the word “botulinum toxin” and one’s mind is likely to go to the big business of cosmetic procedures. Among the 15.7 minimally invasive cosmetic procedures performed in 2017, botulinum toxin type A ­(BoNT-A) made up the largest share, with 7.23 million procedures.¹ However, botulinum toxin—which was first recognized for the ability to paralyze muscles through decreased release of acetylcholine—also has many pain-related and noncosmetic uses; some are approved by the US Food and Drug Administration (FDA) and others are off-label (see TABLE 1^2-31). This review provides an evidence-based look at these uses, from those that have good evidence to support them—including chronic migraine and overactive bladder—to those that have limited (or no) evidence to support them—such as chronic pelvic pain and cluster headache.

BoNT-A is 1 of 7 recognized serotypes derived from Clostridium botulinum.

But before we get into the evidence behind specific uses for botulinum toxin, let’s review the available options and the potential risks they pose.

Many options

Although botulinum toxin is produced by Clostridium botulinum, the synthetic process to produce pharmaceuticals is patented and branded. BoNT-A is 1 of 7 recognized serotypes derived from C botulinum; some examples of BoNT-A include onabotulinumtoxinA, abobotulinumtoxinA, and incobotulinumtoxinA. Clinically, the differences are minor, but they do allow for use of other brands if a patient becomes intolerant to the selected therapy. Treatment doses and costs for each brand vary.

Training. Primary care providers can obtain didactic training from pharmaceutical companies as well as skills training through workshops on botulinum toxin. Credentialed providers can perform some procedures in the primary care setting (TABLE 2).

Adverse effects also vary depending on the formulation and the sites injected. Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints. However, systemic events such as anaphylaxis and antibody development can occur. Depending on the formulation injected, the molecule can migrate and cause weakness in adjacent muscles, leading to undesired effects. Compensatory muscles can become strained, resulting in pain. Serious complications such as pneumonia and death have occurred with injection of botulinum toxin in or around the neck.

A note about pain management. In addition to muscle relaxation, analgesic properties are among the identified benefits of BoNT-A injections.^32,33 BoNT-A suppresses the release of norepinephrine, substance P, and glutamate, which reduces pain sensitization.³² However, the extent of ongoing research involving BoNT-A uses in pain management exceeds the scope of this article. Some pain-related indications will be discussed, but the focus will be on other noncosmetic uses.

Headache disorders

Chronic migraine affects 1.3% to 2.2% of the population and is defined as headaches occurring ≥ 15 days (≥ 8 migrainous days) per month.² To qualify for BoNT-A treatment, patients must have tried 2 prophylactic medications that failed to provide relief, and their headaches must last at least 4 hours. Injections every 12 weeks with 5 U in each of 31 prescribed sites is effective, as shown in the PREEMPT 2 study² with external verification.³ The 24-week, double-blind, placebo-­controlled study showed that BoNT-A treatment reduced headache days by 9 days (P < .001) and migraine days by 8.7 days (P < .001)² and, at 108 weeks, injections reduced headache days by 10.7 days (P < .0001).^4,5

Continue to: Episodic migraine, tension headache, and cluster headaches

Episodic migraine, tension headache, and cluster headaches. There is no significant BoNT-A-related pain reduction in episodic migraine (n = 1838; 0.05 headaches/mo; 95% CI, –0.26 to –0.36) or tension headaches (n = 675; –1.43 headaches/mo; 95% CI, –3.13 to –0.27).^5,6 For cluster headaches, a single prospective study with low enrollment showed no consistent benefit,⁷ while a pilot study showed some improvement, with reduction of attacks by 50% in half of subjects.⁸

Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints.

Occipital neuralgia and trigeminal neuralgia entail paroxysmal, brief, shock-like pain without associated deficits affecting the respective nerve distributions. Multiple prospective and double-blind placebo-­controlled studies with relatively low enrollment show consistent improvement in pain intensity, number of pain-free days, analgesic consumption, and headache frequency with BoNT-A added to nerve blocks.⁶

ENT disorders

Tinnitus by involuntary palatal tremor causes a discontinuous clicking noise. Palatal tremor can be treated with BoNT-A 15 U to tensor veli palatini and levator veli muscles to provide temporary relief for 2 to 6 months.⁹

Spasmodic dysphonia and voice tremor are the result of laryngeal hyperkinesis, and BoNT-A has been deemed the gold standard of treatment. BoNT-A is administered via bilateral injection of the thyroarytenoid muscles for patients with adductor-type spasmodic dysphonia and of the posterior cricoarytenoid muscles for those with the abductor type. A series of 1300 patients (predominantly with the adductor type) treated with BoNT-A showed a 100% improvement in symptoms for 6 to 15 weeks. Patients with abductor-type spasmodic dysphonia were found to have 89% improvement in Voice Related Quality of Life Index score.¹⁰

Secretory disorders

Primary axillary hyperhidrosis (PAH) is an idiopathic excessive production of sweat occurring for at least 6 months, typically with onset before age 25 years. PAH can cause significant psychosocial and physical impairment. Current treatments include topical aluminum chloride, systemic anticholinergics, and thoracic sympathectomy, which can provide temporary relief but are not well tolerated.

Continue to: BoNT-A treatment is efficacious...

Evaluation of BoNT-A as an adjunctive therapy in cerebral palsy has been extensive and conflicting.

BoNT-A treatment is efficacious, safe, and improves quality of life for PAH patients. A 52-week, multicenter, double-blind, randomized, placebo-controlled study showed significant reductions in symptom severity, decreased sweating at rest by gravimetric testing, and improvements in self-reported quality of life.¹¹ A 10-year retrospective study in patients ages 12 years and older showed a 75% to 100% improvement in hyperhidrosis, with a median treatment effect duration of 7 months.¹²

Sialorrhea, or hypersalivation, is typically associated with neurological conditions such as cerebral palsy, amyotrophic lateral sclerosis, Parkinson disease, and posttraumatic brain injuries. It typically is treated with anticholinergic drugs, surgery, and irradiation of salivary glands, which can have significant adverse effects and complications. In a randomized blinded study, BoNT-A injections in the parotid and submandibular glands resulted in a dramatic reduction of sialorrhea and were safe and well tolerated.¹³

Gastric disorders

Achalasia is a syndrome of aperistalsis and incomplete lower esophageal sphincter (LES) relaxation with a “bird beak” appearance on barium swallow. Patients who meet diagnostic criteria are treated with pneumatic dilation or myotomy; however, some patients demonstrate symptoms of achalasia but don’t meet the diagnostic criteria. In these patients, BoNT-A injection in the LES provides symptomatic relief. In a case series, LES BoNT-A injections 20 U were used as a decision tool in whether to proceed with definitive treatment.¹⁴

Gastroparesis is a disorder of impaired gastric motility without mechanical obstruction. Pyloric sphincter BoNT-A injections are useful in refractory patients. Multiple prospective, noncontrolled (4), retrospective (3), and randomized placebo-controlled (2), studies with limited enrollment showed benefit for 37.5% to 100% of patients receiving ­BoNT-A injections of 80 to 200 U.¹⁵

Musculoskeletal disorders

Cervical dystonia (CD) entails involuntary contractions of the neck and upper shoulder musculature, causing abnormal neck, shoulder, and head posturing. BoNT-A is first-line treatment for CD.⁵ BoNT-A is more efficacious than trihexyphenidyl based on multiple large, high-quality studies.¹⁶

Continue to: Chronic low back pain

Chronic low back pain (CLBP) is defined as back pain persisting ≥ 12 weeks. More than 80% of adults have had at least 1 episode of back pain in their lifetime. A 14-month open-label, pilot study evaluating the short- and long-term effects of paraspinal muscle ­BoNT-A injections for refractory CLBP showed reduced pain intensity, reduced number of pain days, and functional improvements.¹⁷

Myofascial pain syndrome (MPS) consists of myofascial trigger points (palpable, tender nodules that produce pain) with multiple pathophysiological etiologies that include dysfunctional acetylcholine activity, which releases nociceptive neurotransmitters. Studies have yielded inconsistent effects of BoNT-A on MPS.¹⁸

Spastic disorders

Cerebral palsy (CP) involves altered muscle tone, posture, and movement secondary to central motor dysfunction with spasticity. Evaluation of BoNT-A as an adjunctive therapy in CP has been extensive and conflicting. A prospective cohort study evaluating gastrocsoleus BoNT-A injections along with gait analysis in 37 children with CP showed no significant improvements.³⁰ In 60 children with CP who received BoNT-A injections, there was improvement in muscle tone and range of motion, while gait improved in patients up to (but not after) age 7 years.¹⁹ A multicenter Dutch study of 65 children compared BoNT-A injections in addition to a comprehensive rehabilitation program vs rehabilitation alone, with no difference identified.²⁰

Neonatal brachial plexus palsy (NBPP) is damage to the brachial plexus as a result of trauma during the perinatal period. It is typically self-resolving but can cause residual functional impairment. Surgery is recommended for serious injuries or if functional recovery is not achieved within 9 months. Off-label use of BoNT-A has been shown to be effective in relieving muscle contractures and imbalance, but data are limited and there have only been small studies performed.²¹ A retrospective cohort study of 59 patients with NBPP who received BoNT-A injections showed improved range of motion and function of the affected extremity. Moreover, surgical intervention was deferred, modified, or averted in patients who were under consideration for more invasive treatment.²¹

Post-stroke spasticity can be temporarily relieved with the use of BoNT-A injections. Several studies have examined the effect of BoNT-A coupled with rehabilitation programs vs injections alone in the treatment of post-stroke spasticity. Devier et al found that improvements in spasticity scores did not differ between groups; however, implementing rehabilitation after BoNT-A injections was associated with improved function compared to injection alone.³¹ A 2018 randomized, double-blind, placebo-controlled trial demonstrated improvements in both treatment groups: those who received ­BoNT-A plus targeted rehab regimen and those who received saline injection plus rehab.²² In this case, it appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.⁵

Continue to: Hemifacial spasm

Hemifacial spasm is an involuntary, brief, irregular unilateral (sometimes bilateral) spasm of the face in the distribution of the facial nerve. Injections with BoNT-A have been deemed effective by the American Academy of Neurology.²³ A 16-year retrospective study examined the efficacy and adverse effects of BoNT-A in the treatment of hemifacial spasm in 113 patients with a mean age of 63.1 years; it demonstrated high efficacy and mild temporary adverse effects.²⁴ The duration of improvement averaged 16 weeks; pretarsal injections had better results than preseptal injections; and there were no differences between the commercial brands.

Blepharospasm is a focal dystonia marked by excessive blinking and involuntary eye closures due to overexcitability of orbicularis oculi and periocular muscles, and BoNT-A is the treatment of choice.^5,25 A retrospective review of 19 patients with blepharospasm who were treated with BoNT-A for more than 5 years found that BoNT-A is a stable and effective treatment with an adverse event rate of 4%. Additionally, there were no differences found in clinical efficacy between the 4 BoNT-A brands on the market.²⁵

It appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.

Laryngeal tics can cause significant psychosocial distress for patients. This condition is characterized by involuntary, recurrent rhythmic sounds that are often preceded by premonitory urges that are relieved by the behavior. An open-label, uncontrolled, confirmatory study with 30 subjects showed that bilateral vocal cord BoNT-A injections resulted in 93% improvement in vocal tics.²⁶ A subsequent study highlighted case histories of 2 patients with laryngeal tics who received thyroarytenoid muscle BoNT-A injections and had marked reduction in symptoms and premonitory sensations.²⁷ Although these small studies have suggested possible effectiveness of BoNT-A for laryngeal tics, there is no high-quality evidence.

Urologic disorders

Overactive, idiopathic overactive, or neurogenic bladder causes increased urinary frequency, urgency, and nocturia without infectious etiology; they can be a result of neurologic dysregulation, detrusor overactivity, or idiopathic causes. Intravesical BoNT-A injection of 100 to 300 U has been found effective for symptoms refractory to anticholinergic and lifestyle therapy, with increased cystometric capacity (229.1 to 427 mL, P < .00001), decreased maximum detrusor pressure (60.7 to 26.1 cm H2O, P < .00001), and resolution of urgency in 87% of patients (P < .001).²⁸

Interstitial cystitis, also known as painful bladder syndrome, is characterized by reduced bladder emptying, urethral pressure, and residual urine pressure, with symptoms of increased urinary frequency without infection. Intravesicular BoNT-A injections have not consistently been effective in treatment of this condition.²⁸

Continue to: Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome

Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome involve urethral sphincter spasticity with difficulty passing urine and possibly retention. Urethral sphincter injections of 100 U ­BoNT-A improved flow rates and decreased residual volume. A randomized, double-blinded, ­placebo-controlled study showed a significantly improved International Prostate Symptom Score (IPSS), quality of life index, maximum flow rate, voided volume, and decreased detrusor voiding pressure at 1 month.²⁹

Intravesicular BoNT-A injections have not consistently been effective in the treatment of interstitial cystitis.

Benign prostatic hypertrophy (BPH) is a very common condition leading to outlet obstruction. The mainstays of treatment are 5-α-reductase inhibitors, α-adrenergic blockers, and surgical removal. Intraprostatic BoNT-A injections of 100 to 200 U were initially promising, and subsequent randomized, double-blind, placebo-controlled studies demonstrated patients with moderate-to-severe symptoms (IPSS ≥ 19) had improved IPSS, maximum flow rate, and post-void residual volume compared to placebo.²⁹

Gynecologic disorders

Vaginismus is the involuntary, recurrent, or persistent contraction of the perineal muscles surrounding the outer third of the vagina; it is classified by 4 progressively more severe degrees of intensity. Levator ani, bulbospongiosus, bulbocavernosus, pubococcygeus, and/or puborectalis muscle BoNT-A injections have shown benefits in decreasing resistance to vaginal exams (95.8%) and the ability to achieve satisfactory sexual intercourse after first injection (75%-100%). Effects were transient for up to 15.4% of patients requiring repeat injections.²⁸

Vulvodynia is vulvar pain and orgasmic difficulties and has been treated with bulbospongiosus muscle BoNT-A injections in retrospective studies. A single randomized, double-blinded, placebo-controlled study showed significantly improved pain scores after 1 to 2 injection series.²⁸

Chronic pelvic pain is a syndrome of somatic functional or regional pain, which can be caused by the spasm of the pelvic musculature with or without trigger points. Patients with pain refractory to treatment have been treated with levator ani injections. A retrospective cohort study found 79.3% of patients experienced pain relief and 20.7% reported improved symptoms. In a double-blind, randomized, placebo-controlled trial, pelvic floor muscles were injected with 80 U BoNT or saline, and symptoms were evaluated along with vaginal manometry. BoNT was associated with a reduction in some pain but not as much as placebo, while vaginal pressures decreased more with BoNT than with placebo.²⁸

CORRESPONDENCE
Blake Busey, DO, FAAFP, Texas Tech University of Health Sciences El Paso–Transmountain, 2000B Transmountain Road, Suite B400, El Paso, TX 79911; [email protected]

Mention the word “botulinum toxin” and one’s mind is likely to go to the big business of cosmetic procedures. Among the 15.7 minimally invasive cosmetic procedures performed in 2017, botulinum toxin type A ­(BoNT-A) made up the largest share, with 7.23 million procedures.¹ However, botulinum toxin—which was first recognized for the ability to paralyze muscles through decreased release of acetylcholine—also has many pain-related and noncosmetic uses; some are approved by the US Food and Drug Administration (FDA) and others are off-label (see TABLE 1^2-31). This review provides an evidence-based look at these uses, from those that have good evidence to support them—including chronic migraine and overactive bladder—to those that have limited (or no) evidence to support them—such as chronic pelvic pain and cluster headache.

BoNT-A is 1 of 7 recognized serotypes derived from Clostridium botulinum.

But before we get into the evidence behind specific uses for botulinum toxin, let’s review the available options and the potential risks they pose.

Many options

Although botulinum toxin is produced by Clostridium botulinum, the synthetic process to produce pharmaceuticals is patented and branded. BoNT-A is 1 of 7 recognized serotypes derived from C botulinum; some examples of BoNT-A include onabotulinumtoxinA, abobotulinumtoxinA, and incobotulinumtoxinA. Clinically, the differences are minor, but they do allow for use of other brands if a patient becomes intolerant to the selected therapy. Treatment doses and costs for each brand vary.

Training. Primary care providers can obtain didactic training from pharmaceutical companies as well as skills training through workshops on botulinum toxin. Credentialed providers can perform some procedures in the primary care setting (TABLE 2).

Adverse effects also vary depending on the formulation and the sites injected. Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints. However, systemic events such as anaphylaxis and antibody development can occur. Depending on the formulation injected, the molecule can migrate and cause weakness in adjacent muscles, leading to undesired effects. Compensatory muscles can become strained, resulting in pain. Serious complications such as pneumonia and death have occurred with injection of botulinum toxin in or around the neck.

A note about pain management. In addition to muscle relaxation, analgesic properties are among the identified benefits of BoNT-A injections.^32,33 BoNT-A suppresses the release of norepinephrine, substance P, and glutamate, which reduces pain sensitization.³² However, the extent of ongoing research involving BoNT-A uses in pain management exceeds the scope of this article. Some pain-related indications will be discussed, but the focus will be on other noncosmetic uses.

Headache disorders

Chronic migraine affects 1.3% to 2.2% of the population and is defined as headaches occurring ≥ 15 days (≥ 8 migrainous days) per month.² To qualify for BoNT-A treatment, patients must have tried 2 prophylactic medications that failed to provide relief, and their headaches must last at least 4 hours. Injections every 12 weeks with 5 U in each of 31 prescribed sites is effective, as shown in the PREEMPT 2 study² with external verification.³ The 24-week, double-blind, placebo-­controlled study showed that BoNT-A treatment reduced headache days by 9 days (P < .001) and migraine days by 8.7 days (P < .001)² and, at 108 weeks, injections reduced headache days by 10.7 days (P < .0001).^4,5

Continue to: Episodic migraine, tension headache, and cluster headaches

Episodic migraine, tension headache, and cluster headaches. There is no significant BoNT-A-related pain reduction in episodic migraine (n = 1838; 0.05 headaches/mo; 95% CI, –0.26 to –0.36) or tension headaches (n = 675; –1.43 headaches/mo; 95% CI, –3.13 to –0.27).^5,6 For cluster headaches, a single prospective study with low enrollment showed no consistent benefit,⁷ while a pilot study showed some improvement, with reduction of attacks by 50% in half of subjects.⁸

Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints.

Occipital neuralgia and trigeminal neuralgia entail paroxysmal, brief, shock-like pain without associated deficits affecting the respective nerve distributions. Multiple prospective and double-blind placebo-­controlled studies with relatively low enrollment show consistent improvement in pain intensity, number of pain-free days, analgesic consumption, and headache frequency with BoNT-A added to nerve blocks.⁶

ENT disorders

Tinnitus by involuntary palatal tremor causes a discontinuous clicking noise. Palatal tremor can be treated with BoNT-A 15 U to tensor veli palatini and levator veli muscles to provide temporary relief for 2 to 6 months.⁹

Spasmodic dysphonia and voice tremor are the result of laryngeal hyperkinesis, and BoNT-A has been deemed the gold standard of treatment. BoNT-A is administered via bilateral injection of the thyroarytenoid muscles for patients with adductor-type spasmodic dysphonia and of the posterior cricoarytenoid muscles for those with the abductor type. A series of 1300 patients (predominantly with the adductor type) treated with BoNT-A showed a 100% improvement in symptoms for 6 to 15 weeks. Patients with abductor-type spasmodic dysphonia were found to have 89% improvement in Voice Related Quality of Life Index score.¹⁰

Secretory disorders

Primary axillary hyperhidrosis (PAH) is an idiopathic excessive production of sweat occurring for at least 6 months, typically with onset before age 25 years. PAH can cause significant psychosocial and physical impairment. Current treatments include topical aluminum chloride, systemic anticholinergics, and thoracic sympathectomy, which can provide temporary relief but are not well tolerated.

Continue to: BoNT-A treatment is efficacious...

Evaluation of BoNT-A as an adjunctive therapy in cerebral palsy has been extensive and conflicting.

BoNT-A treatment is efficacious, safe, and improves quality of life for PAH patients. A 52-week, multicenter, double-blind, randomized, placebo-controlled study showed significant reductions in symptom severity, decreased sweating at rest by gravimetric testing, and improvements in self-reported quality of life.¹¹ A 10-year retrospective study in patients ages 12 years and older showed a 75% to 100% improvement in hyperhidrosis, with a median treatment effect duration of 7 months.¹²

Sialorrhea, or hypersalivation, is typically associated with neurological conditions such as cerebral palsy, amyotrophic lateral sclerosis, Parkinson disease, and posttraumatic brain injuries. It typically is treated with anticholinergic drugs, surgery, and irradiation of salivary glands, which can have significant adverse effects and complications. In a randomized blinded study, BoNT-A injections in the parotid and submandibular glands resulted in a dramatic reduction of sialorrhea and were safe and well tolerated.¹³

Gastric disorders

Achalasia is a syndrome of aperistalsis and incomplete lower esophageal sphincter (LES) relaxation with a “bird beak” appearance on barium swallow. Patients who meet diagnostic criteria are treated with pneumatic dilation or myotomy; however, some patients demonstrate symptoms of achalasia but don’t meet the diagnostic criteria. In these patients, BoNT-A injection in the LES provides symptomatic relief. In a case series, LES BoNT-A injections 20 U were used as a decision tool in whether to proceed with definitive treatment.¹⁴

Gastroparesis is a disorder of impaired gastric motility without mechanical obstruction. Pyloric sphincter BoNT-A injections are useful in refractory patients. Multiple prospective, noncontrolled (4), retrospective (3), and randomized placebo-controlled (2), studies with limited enrollment showed benefit for 37.5% to 100% of patients receiving ­BoNT-A injections of 80 to 200 U.¹⁵

Musculoskeletal disorders

Cervical dystonia (CD) entails involuntary contractions of the neck and upper shoulder musculature, causing abnormal neck, shoulder, and head posturing. BoNT-A is first-line treatment for CD.⁵ BoNT-A is more efficacious than trihexyphenidyl based on multiple large, high-quality studies.¹⁶

Continue to: Chronic low back pain

Chronic low back pain (CLBP) is defined as back pain persisting ≥ 12 weeks. More than 80% of adults have had at least 1 episode of back pain in their lifetime. A 14-month open-label, pilot study evaluating the short- and long-term effects of paraspinal muscle ­BoNT-A injections for refractory CLBP showed reduced pain intensity, reduced number of pain days, and functional improvements.¹⁷

Myofascial pain syndrome (MPS) consists of myofascial trigger points (palpable, tender nodules that produce pain) with multiple pathophysiological etiologies that include dysfunctional acetylcholine activity, which releases nociceptive neurotransmitters. Studies have yielded inconsistent effects of BoNT-A on MPS.¹⁸

Spastic disorders

Cerebral palsy (CP) involves altered muscle tone, posture, and movement secondary to central motor dysfunction with spasticity. Evaluation of BoNT-A as an adjunctive therapy in CP has been extensive and conflicting. A prospective cohort study evaluating gastrocsoleus BoNT-A injections along with gait analysis in 37 children with CP showed no significant improvements.³⁰ In 60 children with CP who received BoNT-A injections, there was improvement in muscle tone and range of motion, while gait improved in patients up to (but not after) age 7 years.¹⁹ A multicenter Dutch study of 65 children compared BoNT-A injections in addition to a comprehensive rehabilitation program vs rehabilitation alone, with no difference identified.²⁰

Neonatal brachial plexus palsy (NBPP) is damage to the brachial plexus as a result of trauma during the perinatal period. It is typically self-resolving but can cause residual functional impairment. Surgery is recommended for serious injuries or if functional recovery is not achieved within 9 months. Off-label use of BoNT-A has been shown to be effective in relieving muscle contractures and imbalance, but data are limited and there have only been small studies performed.²¹ A retrospective cohort study of 59 patients with NBPP who received BoNT-A injections showed improved range of motion and function of the affected extremity. Moreover, surgical intervention was deferred, modified, or averted in patients who were under consideration for more invasive treatment.²¹

Post-stroke spasticity can be temporarily relieved with the use of BoNT-A injections. Several studies have examined the effect of BoNT-A coupled with rehabilitation programs vs injections alone in the treatment of post-stroke spasticity. Devier et al found that improvements in spasticity scores did not differ between groups; however, implementing rehabilitation after BoNT-A injections was associated with improved function compared to injection alone.³¹ A 2018 randomized, double-blind, placebo-controlled trial demonstrated improvements in both treatment groups: those who received ­BoNT-A plus targeted rehab regimen and those who received saline injection plus rehab.²² In this case, it appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.⁵

Continue to: Hemifacial spasm

Hemifacial spasm is an involuntary, brief, irregular unilateral (sometimes bilateral) spasm of the face in the distribution of the facial nerve. Injections with BoNT-A have been deemed effective by the American Academy of Neurology.²³ A 16-year retrospective study examined the efficacy and adverse effects of BoNT-A in the treatment of hemifacial spasm in 113 patients with a mean age of 63.1 years; it demonstrated high efficacy and mild temporary adverse effects.²⁴ The duration of improvement averaged 16 weeks; pretarsal injections had better results than preseptal injections; and there were no differences between the commercial brands.

Blepharospasm is a focal dystonia marked by excessive blinking and involuntary eye closures due to overexcitability of orbicularis oculi and periocular muscles, and BoNT-A is the treatment of choice.^5,25 A retrospective review of 19 patients with blepharospasm who were treated with BoNT-A for more than 5 years found that BoNT-A is a stable and effective treatment with an adverse event rate of 4%. Additionally, there were no differences found in clinical efficacy between the 4 BoNT-A brands on the market.²⁵

It appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.

Laryngeal tics can cause significant psychosocial distress for patients. This condition is characterized by involuntary, recurrent rhythmic sounds that are often preceded by premonitory urges that are relieved by the behavior. An open-label, uncontrolled, confirmatory study with 30 subjects showed that bilateral vocal cord BoNT-A injections resulted in 93% improvement in vocal tics.²⁶ A subsequent study highlighted case histories of 2 patients with laryngeal tics who received thyroarytenoid muscle BoNT-A injections and had marked reduction in symptoms and premonitory sensations.²⁷ Although these small studies have suggested possible effectiveness of BoNT-A for laryngeal tics, there is no high-quality evidence.

Urologic disorders

Overactive, idiopathic overactive, or neurogenic bladder causes increased urinary frequency, urgency, and nocturia without infectious etiology; they can be a result of neurologic dysregulation, detrusor overactivity, or idiopathic causes. Intravesical BoNT-A injection of 100 to 300 U has been found effective for symptoms refractory to anticholinergic and lifestyle therapy, with increased cystometric capacity (229.1 to 427 mL, P < .00001), decreased maximum detrusor pressure (60.7 to 26.1 cm H2O, P < .00001), and resolution of urgency in 87% of patients (P < .001).²⁸

Interstitial cystitis, also known as painful bladder syndrome, is characterized by reduced bladder emptying, urethral pressure, and residual urine pressure, with symptoms of increased urinary frequency without infection. Intravesicular BoNT-A injections have not consistently been effective in treatment of this condition.²⁸

Continue to: Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome

Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome involve urethral sphincter spasticity with difficulty passing urine and possibly retention. Urethral sphincter injections of 100 U ­BoNT-A improved flow rates and decreased residual volume. A randomized, double-blinded, ­placebo-controlled study showed a significantly improved International Prostate Symptom Score (IPSS), quality of life index, maximum flow rate, voided volume, and decreased detrusor voiding pressure at 1 month.²⁹

Intravesicular BoNT-A injections have not consistently been effective in the treatment of interstitial cystitis.

Benign prostatic hypertrophy (BPH) is a very common condition leading to outlet obstruction. The mainstays of treatment are 5-α-reductase inhibitors, α-adrenergic blockers, and surgical removal. Intraprostatic BoNT-A injections of 100 to 200 U were initially promising, and subsequent randomized, double-blind, placebo-controlled studies demonstrated patients with moderate-to-severe symptoms (IPSS ≥ 19) had improved IPSS, maximum flow rate, and post-void residual volume compared to placebo.²⁹

Gynecologic disorders

Vaginismus is the involuntary, recurrent, or persistent contraction of the perineal muscles surrounding the outer third of the vagina; it is classified by 4 progressively more severe degrees of intensity. Levator ani, bulbospongiosus, bulbocavernosus, pubococcygeus, and/or puborectalis muscle BoNT-A injections have shown benefits in decreasing resistance to vaginal exams (95.8%) and the ability to achieve satisfactory sexual intercourse after first injection (75%-100%). Effects were transient for up to 15.4% of patients requiring repeat injections.²⁸

Vulvodynia is vulvar pain and orgasmic difficulties and has been treated with bulbospongiosus muscle BoNT-A injections in retrospective studies. A single randomized, double-blinded, placebo-controlled study showed significantly improved pain scores after 1 to 2 injection series.²⁸

Chronic pelvic pain is a syndrome of somatic functional or regional pain, which can be caused by the spasm of the pelvic musculature with or without trigger points. Patients with pain refractory to treatment have been treated with levator ani injections. A retrospective cohort study found 79.3% of patients experienced pain relief and 20.7% reported improved symptoms. In a double-blind, randomized, placebo-controlled trial, pelvic floor muscles were injected with 80 U BoNT or saline, and symptoms were evaluated along with vaginal manometry. BoNT was associated with a reduction in some pain but not as much as placebo, while vaginal pressures decreased more with BoNT than with placebo.²⁸

CORRESPONDENCE
Blake Busey, DO, FAAFP, Texas Tech University of Health Sciences El Paso–Transmountain, 2000B Transmountain Road, Suite B400, El Paso, TX 79911; [email protected]

Mention the word “botulinum toxin” and one’s mind is likely to go to the big business of cosmetic procedures. Among the 15.7 minimally invasive cosmetic procedures performed in 2017, botulinum toxin type A ­(BoNT-A) made up the largest share, with 7.23 million procedures.¹ However, botulinum toxin—which was first recognized for the ability to paralyze muscles through decreased release of acetylcholine—also has many pain-related and noncosmetic uses; some are approved by the US Food and Drug Administration (FDA) and others are off-label (see TABLE 1^2-31). This review provides an evidence-based look at these uses, from those that have good evidence to support them—including chronic migraine and overactive bladder—to those that have limited (or no) evidence to support them—such as chronic pelvic pain and cluster headache.

BoNT-A is 1 of 7 recognized serotypes derived from Clostridium botulinum.

But before we get into the evidence behind specific uses for botulinum toxin, let’s review the available options and the potential risks they pose.

Many options

Although botulinum toxin is produced by Clostridium botulinum, the synthetic process to produce pharmaceuticals is patented and branded. BoNT-A is 1 of 7 recognized serotypes derived from C botulinum; some examples of BoNT-A include onabotulinumtoxinA, abobotulinumtoxinA, and incobotulinumtoxinA. Clinically, the differences are minor, but they do allow for use of other brands if a patient becomes intolerant to the selected therapy. Treatment doses and costs for each brand vary.

Training. Primary care providers can obtain didactic training from pharmaceutical companies as well as skills training through workshops on botulinum toxin. Credentialed providers can perform some procedures in the primary care setting (TABLE 2).

Adverse effects also vary depending on the formulation and the sites injected. Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints. However, systemic events such as anaphylaxis and antibody development can occur. Depending on the formulation injected, the molecule can migrate and cause weakness in adjacent muscles, leading to undesired effects. Compensatory muscles can become strained, resulting in pain. Serious complications such as pneumonia and death have occurred with injection of botulinum toxin in or around the neck.

A note about pain management. In addition to muscle relaxation, analgesic properties are among the identified benefits of BoNT-A injections.^32,33 BoNT-A suppresses the release of norepinephrine, substance P, and glutamate, which reduces pain sensitization.³² However, the extent of ongoing research involving BoNT-A uses in pain management exceeds the scope of this article. Some pain-related indications will be discussed, but the focus will be on other noncosmetic uses.

Headache disorders

Chronic migraine affects 1.3% to 2.2% of the population and is defined as headaches occurring ≥ 15 days (≥ 8 migrainous days) per month.² To qualify for BoNT-A treatment, patients must have tried 2 prophylactic medications that failed to provide relief, and their headaches must last at least 4 hours. Injections every 12 weeks with 5 U in each of 31 prescribed sites is effective, as shown in the PREEMPT 2 study² with external verification.³ The 24-week, double-blind, placebo-­controlled study showed that BoNT-A treatment reduced headache days by 9 days (P < .001) and migraine days by 8.7 days (P < .001)² and, at 108 weeks, injections reduced headache days by 10.7 days (P < .0001).^4,5

Continue to: Episodic migraine, tension headache, and cluster headaches

Episodic migraine, tension headache, and cluster headaches. There is no significant BoNT-A-related pain reduction in episodic migraine (n = 1838; 0.05 headaches/mo; 95% CI, –0.26 to –0.36) or tension headaches (n = 675; –1.43 headaches/mo; 95% CI, –3.13 to –0.27).^5,6 For cluster headaches, a single prospective study with low enrollment showed no consistent benefit,⁷ while a pilot study showed some improvement, with reduction of attacks by 50% in half of subjects.⁸

Patients generally tolerate the procedure well, with discomfort from injections and localized bleeding as the major complaints.

Occipital neuralgia and trigeminal neuralgia entail paroxysmal, brief, shock-like pain without associated deficits affecting the respective nerve distributions. Multiple prospective and double-blind placebo-­controlled studies with relatively low enrollment show consistent improvement in pain intensity, number of pain-free days, analgesic consumption, and headache frequency with BoNT-A added to nerve blocks.⁶

ENT disorders

Tinnitus by involuntary palatal tremor causes a discontinuous clicking noise. Palatal tremor can be treated with BoNT-A 15 U to tensor veli palatini and levator veli muscles to provide temporary relief for 2 to 6 months.⁹

Spasmodic dysphonia and voice tremor are the result of laryngeal hyperkinesis, and BoNT-A has been deemed the gold standard of treatment. BoNT-A is administered via bilateral injection of the thyroarytenoid muscles for patients with adductor-type spasmodic dysphonia and of the posterior cricoarytenoid muscles for those with the abductor type. A series of 1300 patients (predominantly with the adductor type) treated with BoNT-A showed a 100% improvement in symptoms for 6 to 15 weeks. Patients with abductor-type spasmodic dysphonia were found to have 89% improvement in Voice Related Quality of Life Index score.¹⁰

Secretory disorders

Primary axillary hyperhidrosis (PAH) is an idiopathic excessive production of sweat occurring for at least 6 months, typically with onset before age 25 years. PAH can cause significant psychosocial and physical impairment. Current treatments include topical aluminum chloride, systemic anticholinergics, and thoracic sympathectomy, which can provide temporary relief but are not well tolerated.

Continue to: BoNT-A treatment is efficacious...

Evaluation of BoNT-A as an adjunctive therapy in cerebral palsy has been extensive and conflicting.

BoNT-A treatment is efficacious, safe, and improves quality of life for PAH patients. A 52-week, multicenter, double-blind, randomized, placebo-controlled study showed significant reductions in symptom severity, decreased sweating at rest by gravimetric testing, and improvements in self-reported quality of life.¹¹ A 10-year retrospective study in patients ages 12 years and older showed a 75% to 100% improvement in hyperhidrosis, with a median treatment effect duration of 7 months.¹²

Sialorrhea, or hypersalivation, is typically associated with neurological conditions such as cerebral palsy, amyotrophic lateral sclerosis, Parkinson disease, and posttraumatic brain injuries. It typically is treated with anticholinergic drugs, surgery, and irradiation of salivary glands, which can have significant adverse effects and complications. In a randomized blinded study, BoNT-A injections in the parotid and submandibular glands resulted in a dramatic reduction of sialorrhea and were safe and well tolerated.¹³

Gastric disorders

Achalasia is a syndrome of aperistalsis and incomplete lower esophageal sphincter (LES) relaxation with a “bird beak” appearance on barium swallow. Patients who meet diagnostic criteria are treated with pneumatic dilation or myotomy; however, some patients demonstrate symptoms of achalasia but don’t meet the diagnostic criteria. In these patients, BoNT-A injection in the LES provides symptomatic relief. In a case series, LES BoNT-A injections 20 U were used as a decision tool in whether to proceed with definitive treatment.¹⁴

Gastroparesis is a disorder of impaired gastric motility without mechanical obstruction. Pyloric sphincter BoNT-A injections are useful in refractory patients. Multiple prospective, noncontrolled (4), retrospective (3), and randomized placebo-controlled (2), studies with limited enrollment showed benefit for 37.5% to 100% of patients receiving ­BoNT-A injections of 80 to 200 U.¹⁵

Musculoskeletal disorders

Cervical dystonia (CD) entails involuntary contractions of the neck and upper shoulder musculature, causing abnormal neck, shoulder, and head posturing. BoNT-A is first-line treatment for CD.⁵ BoNT-A is more efficacious than trihexyphenidyl based on multiple large, high-quality studies.¹⁶

Continue to: Chronic low back pain

Chronic low back pain (CLBP) is defined as back pain persisting ≥ 12 weeks. More than 80% of adults have had at least 1 episode of back pain in their lifetime. A 14-month open-label, pilot study evaluating the short- and long-term effects of paraspinal muscle ­BoNT-A injections for refractory CLBP showed reduced pain intensity, reduced number of pain days, and functional improvements.¹⁷

Myofascial pain syndrome (MPS) consists of myofascial trigger points (palpable, tender nodules that produce pain) with multiple pathophysiological etiologies that include dysfunctional acetylcholine activity, which releases nociceptive neurotransmitters. Studies have yielded inconsistent effects of BoNT-A on MPS.¹⁸

Spastic disorders

Cerebral palsy (CP) involves altered muscle tone, posture, and movement secondary to central motor dysfunction with spasticity. Evaluation of BoNT-A as an adjunctive therapy in CP has been extensive and conflicting. A prospective cohort study evaluating gastrocsoleus BoNT-A injections along with gait analysis in 37 children with CP showed no significant improvements.³⁰ In 60 children with CP who received BoNT-A injections, there was improvement in muscle tone and range of motion, while gait improved in patients up to (but not after) age 7 years.¹⁹ A multicenter Dutch study of 65 children compared BoNT-A injections in addition to a comprehensive rehabilitation program vs rehabilitation alone, with no difference identified.²⁰

Neonatal brachial plexus palsy (NBPP) is damage to the brachial plexus as a result of trauma during the perinatal period. It is typically self-resolving but can cause residual functional impairment. Surgery is recommended for serious injuries or if functional recovery is not achieved within 9 months. Off-label use of BoNT-A has been shown to be effective in relieving muscle contractures and imbalance, but data are limited and there have only been small studies performed.²¹ A retrospective cohort study of 59 patients with NBPP who received BoNT-A injections showed improved range of motion and function of the affected extremity. Moreover, surgical intervention was deferred, modified, or averted in patients who were under consideration for more invasive treatment.²¹

Post-stroke spasticity can be temporarily relieved with the use of BoNT-A injections. Several studies have examined the effect of BoNT-A coupled with rehabilitation programs vs injections alone in the treatment of post-stroke spasticity. Devier et al found that improvements in spasticity scores did not differ between groups; however, implementing rehabilitation after BoNT-A injections was associated with improved function compared to injection alone.³¹ A 2018 randomized, double-blind, placebo-controlled trial demonstrated improvements in both treatment groups: those who received ­BoNT-A plus targeted rehab regimen and those who received saline injection plus rehab.²² In this case, it appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.⁵

Continue to: Hemifacial spasm

Hemifacial spasm is an involuntary, brief, irregular unilateral (sometimes bilateral) spasm of the face in the distribution of the facial nerve. Injections with BoNT-A have been deemed effective by the American Academy of Neurology.²³ A 16-year retrospective study examined the efficacy and adverse effects of BoNT-A in the treatment of hemifacial spasm in 113 patients with a mean age of 63.1 years; it demonstrated high efficacy and mild temporary adverse effects.²⁴ The duration of improvement averaged 16 weeks; pretarsal injections had better results than preseptal injections; and there were no differences between the commercial brands.

Blepharospasm is a focal dystonia marked by excessive blinking and involuntary eye closures due to overexcitability of orbicularis oculi and periocular muscles, and BoNT-A is the treatment of choice.^5,25 A retrospective review of 19 patients with blepharospasm who were treated with BoNT-A for more than 5 years found that BoNT-A is a stable and effective treatment with an adverse event rate of 4%. Additionally, there were no differences found in clinical efficacy between the 4 BoNT-A brands on the market.²⁵

It appears BoNT-A acts as more of an adjunct to physical therapy in the treatment of post-stroke spasticity.

Laryngeal tics can cause significant psychosocial distress for patients. This condition is characterized by involuntary, recurrent rhythmic sounds that are often preceded by premonitory urges that are relieved by the behavior. An open-label, uncontrolled, confirmatory study with 30 subjects showed that bilateral vocal cord BoNT-A injections resulted in 93% improvement in vocal tics.²⁶ A subsequent study highlighted case histories of 2 patients with laryngeal tics who received thyroarytenoid muscle BoNT-A injections and had marked reduction in symptoms and premonitory sensations.²⁷ Although these small studies have suggested possible effectiveness of BoNT-A for laryngeal tics, there is no high-quality evidence.

Urologic disorders

Overactive, idiopathic overactive, or neurogenic bladder causes increased urinary frequency, urgency, and nocturia without infectious etiology; they can be a result of neurologic dysregulation, detrusor overactivity, or idiopathic causes. Intravesical BoNT-A injection of 100 to 300 U has been found effective for symptoms refractory to anticholinergic and lifestyle therapy, with increased cystometric capacity (229.1 to 427 mL, P < .00001), decreased maximum detrusor pressure (60.7 to 26.1 cm H2O, P < .00001), and resolution of urgency in 87% of patients (P < .001).²⁸

Interstitial cystitis, also known as painful bladder syndrome, is characterized by reduced bladder emptying, urethral pressure, and residual urine pressure, with symptoms of increased urinary frequency without infection. Intravesicular BoNT-A injections have not consistently been effective in treatment of this condition.²⁸

Continue to: Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome

Dysfunctional voiding, urethral sphincter overactivity, and Fowler syndrome involve urethral sphincter spasticity with difficulty passing urine and possibly retention. Urethral sphincter injections of 100 U ­BoNT-A improved flow rates and decreased residual volume. A randomized, double-blinded, ­placebo-controlled study showed a significantly improved International Prostate Symptom Score (IPSS), quality of life index, maximum flow rate, voided volume, and decreased detrusor voiding pressure at 1 month.²⁹

Intravesicular BoNT-A injections have not consistently been effective in the treatment of interstitial cystitis.

Benign prostatic hypertrophy (BPH) is a very common condition leading to outlet obstruction. The mainstays of treatment are 5-α-reductase inhibitors, α-adrenergic blockers, and surgical removal. Intraprostatic BoNT-A injections of 100 to 200 U were initially promising, and subsequent randomized, double-blind, placebo-controlled studies demonstrated patients with moderate-to-severe symptoms (IPSS ≥ 19) had improved IPSS, maximum flow rate, and post-void residual volume compared to placebo.²⁹

Gynecologic disorders

Vaginismus is the involuntary, recurrent, or persistent contraction of the perineal muscles surrounding the outer third of the vagina; it is classified by 4 progressively more severe degrees of intensity. Levator ani, bulbospongiosus, bulbocavernosus, pubococcygeus, and/or puborectalis muscle BoNT-A injections have shown benefits in decreasing resistance to vaginal exams (95.8%) and the ability to achieve satisfactory sexual intercourse after first injection (75%-100%). Effects were transient for up to 15.4% of patients requiring repeat injections.²⁸

Vulvodynia is vulvar pain and orgasmic difficulties and has been treated with bulbospongiosus muscle BoNT-A injections in retrospective studies. A single randomized, double-blinded, placebo-controlled study showed significantly improved pain scores after 1 to 2 injection series.²⁸

Chronic pelvic pain is a syndrome of somatic functional or regional pain, which can be caused by the spasm of the pelvic musculature with or without trigger points. Patients with pain refractory to treatment have been treated with levator ani injections. A retrospective cohort study found 79.3% of patients experienced pain relief and 20.7% reported improved symptoms. In a double-blind, randomized, placebo-controlled trial, pelvic floor muscles were injected with 80 U BoNT or saline, and symptoms were evaluated along with vaginal manometry. BoNT was associated with a reduction in some pain but not as much as placebo, while vaginal pressures decreased more with BoNT than with placebo.²⁸

CORRESPONDENCE
Blake Busey, DO, FAAFP, Texas Tech University of Health Sciences El Paso–Transmountain, 2000B Transmountain Road, Suite B400, El Paso, TX 79911; [email protected]

Stroke ranks second behind heart disease as the leading cause of mortality worldwide, accounting for 1 of every 19 deaths,¹ and remains a serious cause of morbidity. Best practices in stroke diagnosis and management can seem elusive to front-line clinicians, for 2 reasons: the rate of proliferation and nuance in stroke medicine and the fact that the typical scope of primary care practice exists apart from much of the diagnostic tools and management schema provided in stroke centers.² In this article, we describe and update the diagnosis of stroke and review imaging modalities, their nuances, and their application in practice.

Diagnosis of acute stroke

Acute stroke is diagnosed upon observation of new neurologic deficits and congruent neuroimaging. Some updated definitions favor a silent form of cerebral ischemia manifested by imaging pathology only; this form is not discussed in this article. Although there are several characteristically distinct stroke syndromes, there is no way to clinically distinguish ischemic pathology from hemorrhagic pathology.

Some common symptoms that should prompt evaluation for stroke are part of the American Stroke Association FAST mnemonic designed to promote public health awareness^3-5:

• face drooping
• arm weakness
• speech difficulty
• time to call 911.

There are several characteristically distinct stroke syndromes, but no way to differentiate ischemic and hemorrhagic pathologies clinically.

Other commonly reported stroke symptoms include unilateral weakness or numbness, confusion, word-finding difficulty, visual problems, difficulty ambulating, dizziness, loss of balance or coordination, and thunderclap headache. A stroke should also be considered in the presence of any new focal neurologic deficit.^3,4

Stroke patients should be triaged by emergency medical services using a stroke screening scale, such as BE-FAST⁵ (a modification of FAST that adds balance and eye assessments); the Los Angeles Prehospital Stroke Screen (LAPSS)^6,7; the Rapid Arterial oCclusion Evaluation (RACE)⁸; and the Cincinnati Prehospital Stroke Severity Scale (CP-SSS)^9,10 (see “Stroke screening scales for early identification and triage"). Studies have not found that any single prehospital stroke scale is superior to the others for reliably predicting large-vessel occlusion; therefore, prehospital assessment is typically based on practice patterns in a given locale.¹¹ A patient (or family member or caregiver) who seeks your care for stroke symptoms should be told to call 911 and get emergency transport to a health care facility that can capably administer intravenous (IV) thrombolysis.^a

First responders should elicit “last-known-normal” time; this critical information can aid in diagnosis and drive therapeutic options, especially if patients are unaccompanied at time of transport to a higher echelon of care. A point-of-care blood glucose test should be performed by emergency medical staff, with dextrose administered for a level < 45 mg/dL. Establishing IV access for fluids, medications, and contrast can be considered if it does not delay transport. A 12-lead electrocardiogram can also be considered, again, as long as it does not delay transport to a facility capable of providing definitive therapy. Notification by emergency services staff before arrival and transport of the patient to such a facility is the essential element of prehospital care, and should be prioritized above ancillary testing beyond the stroke assessment.¹⁴

Guidelines recommend use of the National Institutes of Health Stroke Scale ­(NIHSS; www.stroke.nih.gov/resources/scale.htm) for clinical evaluation upon arrival at the ED.¹⁵ Although no scale has been identified that can reliably predict large-vessel occlusion amenable to endovascular therapy (EVT), no other score has been found to outperform the NIHSS in achieving meaningful patient outcomes.¹⁶ Furthermore, NIHSS has been validated to track clinical changes in response to therapy, is widely utilized, and is free.

Continue to: A criticism of the NIHSS...

A criticism of the NIHSS is its bias toward left-hemispheric ischemic pathology.¹⁷ NIHSS includes 11 questions on a scale of 0 to 42; typically, a score < 4 is associated with a higher chance of a positive clinical outcome.¹⁸ There is no minimum or maximum NIHSS score that precludes treatment with thrombolysis or EVT.

Other commonly used scores in acute stroke include disability assessments. The modified Rankin scale, which is used most often, features a score of 0 (symptom-free) to 6 (death). A modified Rankin scale score of 0 or 1 is considered an indication of a favorable outcome after stroke.¹⁹ Note that these functional scores are not always part of an acute assessment but can be done early in the clinical course to gauge the response to treatment, and are collected for stroke-center certification.

Imaging modalities

Imaging is recommended within 20 minutes of arrival in the ED in a stroke patient who might be a candidate for thrombolysis or thrombectomy.³ There, imaging modalities commonly performed are noncontrast-enhanced head computed tomography (NCHCT); computed tomography (CT) angiography, with or without perfusion; and diffusion-weighted magnetic resonance imaging (MRI).^20,21 In addition, more highly specialized imaging modalities are available for the evaluation of the stroke patient in specific, often limited, circumstances. All these modalities are described below and compared in the TABLE,^20,21 using the ACR Appropriateness Criteria (of the American College of Radiology),²¹ which are guidelines for appropriate imaging of stroke, based on a clinical complaint. Separate recommendations and appraisals are offered by the most recent American Heart Association/American Stroke Association (AHA/ASA) guideline.³

NCHCT. This study should be performed within 20 minutes after arrival at the ED because it provides rapid assessment of intracerebral hemorrhage, can effectively corroborate the diagnosis of some stroke mimickers, and identifies some candidates for EVT or thrombolysis^3,21,22 (typically, the decision to proceed with EVT is based on adjunct imaging studies discussed in a bit). Evaluation for intracerebral hemorrhage is required prior to administering thrombolysis. Ischemic changes can be seen with variable specificity and sensitivity on NCHCT, depending on how much time has passed since the original insult. In all historical trials, CT was the only imaging modality used in the diagnosis of acute ischemic stroke (AIS) that suggested benefit from IV thrombolysis.^23-25

Acute, subacute, and chronic changes can be seen on NCHCT, although the modality has limited sensitivity for identifying AIS (ie, approximately 75% within 6 hours after the original insult):

• Acute findings on NCHCT include intracellular edema, which causes loss of the gray matter–white matter interface and effacement of the cortical sulci. This occurs as a result of increased cellular uptake of water in response to ischemia and cell death, resulting in a decreased density of tissue (hypoattenuation) in affected areas.
• Subacute changes appear in the 2- to 5-day window, including vasogenic edema with greater mass effect, hypoattenuation, and well-defined margins.^3,20,21
• Chronic vascular findings on NCHCT include loss of brain tissue and hypoattenuation.

Continue to: NCHCT is typically performed...

NCHCT is typically performed in advance of other adjunct imaging modalities.^3,20,21 Baseline NCHCT can be performed on patients with advanced kidney disease and those who have an indwelling metallic device.

CT angiography is performed with timed contrast, providing a 3-dimensional representation of the cerebral vasculature; the entire intracranial and extracranial vasculature, including the aortic arch, can be mapped in approximately 60 seconds. CT angiography is sensitive in identifying areas of stenosis > 50% and identifies clinically significant areas of stenosis up to approximately 90% of the time.²⁶ For this reason, it is particularly helpful in identifying candidates for an interventional strategy beyond pharmacotherapeutic thrombolysis. In addition, CT angiography can visualize aneurysmal dilation and dissection, and help with the planning of interventions—specifically, the confident administration of thrombolysis or more specific planning for target lesions and EVT.

Commonly used scoring systems in acute stroke include disability assessments— not always part of the acute assessment but undertaken early in the clinical course to gauge response to treatment.

It also can help identify a host of vascular phenomena, such as arteriovenous malformations, Moyamoya disease (progressive arterial blockage within the basal ganglia and compensatory microvascularization), and some vasculopathies.^20,27 In intracranial hemorrhage, CT with angiography can help evaluate for structural malformations and identify patients at risk of hematoma expansion.²²

CT perfusion. Many stroke centers will perform a CT perfusion study,²⁸ which encompasses as many as 3 different CT sequences:

• NCHCT
• vertex-to-arch angiography with contrast bolus
• administration of contrast and capture of a dynamic sequence through 1 or 2 slabs of tissue, allowing for the generation of maps of cerebral blood flow (CBF), mean transit time (MTT), and cerebral blood volume (CBV) of the entire cerebral vasculature.

The interplay of these 3 sequences drives characterization of lesions (ie, CBF = CBV/MTT). An infarct is characterized by low CBF, low CBV, and elevated MTT. In penumbral tissue, MTT is elevated but CBF is slightly decreased and CBV is normal or increased. Using CT perfusion, areas throughout the ischemic penumbra can be surveyed for favorable interventional characteristics.^20,29

Continue to: A CT perfusion study adds...

A CT perfusion study adds at least 60 seconds to NCHCT. This modality can be useful in planning interventions and for stratifying appropriateness of reperfusion strategies in strokes of unknown duration.^3,30 CT perfusion can be performed on any multidetector CT scan but (1) requires specialized software and expertise to interpret and (2) subjects the patient to a significant radiation dose, which, if incorrectly administered, can be considerably higher than intended.^20,26,27

Diffusion-weighted MRI. This is the most sensitive study for demonstrating early ischemic changes; however, limitations include lack of availability, contraindication in patients with metallic indwelling implants, and duration of the study—although, at some stroke centers, diffusion-weighted MRI can be performed in ≤ 10 minutes.

MRI and NCHCT have comparable sensitivity in detecting intracranial hemorrhage. MRI is likely more sensitive in identifying areas of microhemorrhage: In diffusion-weighted MRI, the sensitivity of stroke detection increases to > 95%.³¹ The modality relies on the comparable movement of water through damaged vs normal neuronal tissue. Diffusion-weighted MRI does not require administration of concomitant contrast, which can be a benefit in patients who are allergic to gadolinium-based contrast agents or have advanced kidney disease that precludes the use of contrast. It typically does not result in adequate characterization of extracranial vasculature.

Other MRI modalities. These MRI extensions include magnetic resonance (MR) perfusion and MR angiography. Whereas diffusion-weighted MRI (discussed above) offers the most rapid and sensitive evaluation for ischemia, fluid-attenuated inversion recovery (FLAIR) imaging has been utilized as a comparator to isolated diffusion-weighted MRI to help determine stroke duration. FLAIR signal positivity typically occurs 6 to 24 hours after the initial insult but is negative in stroke that occurred < 3 hours earlier.³²

MRI is limited, in terms of availability and increased study duration, especially when it comes to timely administration of thrombolysis. A benefit of this modality is less radiation and, as noted, superior sensitivity for ischemia. Diffusion-weighted MRI combined with MR perfusion analysis can help isolate areas of the ischemic penumbra. MR perfusion is performed for a similar reason as CT perfusion, although logistical execution across those modalities is significantly different. Considerations for choosing MR perfusion or CT perfusion should be made on an individual basis and based on available local resources and accepted local practice patterns.²⁶

Continue to: In the subacute setting...

Knowledge of historic details of the event, the patient (eg, known atrial fibrillation), and findings on imaging can facilitate communication between the primary care physician and inpatient teams.

In the subacute setting, MR perfusion and MR angiography of the head and the neck are often performed to identify stenosis, dissection, and more subtle mimickers of cerebrovascular accident not ascertained on initial CT evaluation. These studies are typically performed well outside the window for thrombolysis or intervention.²⁶ No guidelines specifically direct or recommend this practice pattern. The superior sensitivity and cerebral blood flow mapping of MR perfusion and MR angiography might be useful for validating a suspected diagnosis of ischemic stroke and providing phenotypic information about AIS events.

Transcranial Doppler imaging relies on bony windows to assess intracranial vascular flow, velocity, direction, and reactivity. This information can be utilized to diagnose stenosis or occlusion. This modality is principally used to evaluate for stenosis in the anterior circulation (sensitivity, 70%-90%; specificity, 90%-95%).²⁰ Evaluation of the basilar, vertebral, and internal carotid arteries is less accurate (sensitivity, 55%-80%).²⁰ Transcranial Doppler imaging is also used to assess for cerebral vasospasm after subarachnoid hemorrhage, monitor sickle cell disease patients’ overall risk for ischemic stroke, and augment thrombolysis. It is limited by the availability of an expert technician, and therefore is typically reserved for unstable patients or those who cannot receive contrast.²⁰

Carotid duplex ultrasonography. A dynamic study such as duplex ultrasonography can be strongly considered for flow imaging of the extracranial carotids to evaluate for stenosis. Indications for carotid stenting or endarterectomy include 50% to 79% occlusion of the carotid artery on the same side as a recent transient ischemic attack or AIS. Carotid stenosis > 80% warrants consideration for intervention independent of a recent cerebrovascular accident. Interventions are typically performed 2 to 14 days after stroke.³³ Although this study is of limited utility in the hyperacute setting, it is recommended within 24 hours after nondisabling stroke in the carotid territory, when (1) the patient is otherwise a candidate for a surgical or procedural intervention to address the stenosis and (2) none of the aforementioned studies that focus on neck vasculature have been performed.

Conventional (digital subtraction) ­angiography is the gold standard for mapping cerebrovascular disease because it is dynamic and highly accurate. It is, however, typically limited by the number of required personnel, its invasive nature, and the requirement for IV contrast. This study is performed during intra-arterial intervention techniques, including stent retrieval and intra-arterial thrombolysis.²⁶

Impact of imaging on treatment

Imaging helps determine the cause and some characteristics of stroke, both of which can help determine therapy. Strokes can be broadly subcategorized as hemorrhagic or ischemic; recent studies suggest that 87% are ischemic.³⁴ Knowledge of the historic details of the event, the patient (eg, known atrial fibrillation, anticoagulant use, history of falls), and findings on imaging can contribute to determine the cause of AIS, and can facilitate communication and consultation between the primary care physician and inpatient teams.³⁵

Continue to: Best practices for stroke treatment...

Best practices for stroke treatment are based on the cause of the event.³ To identify the likely cause, the aforementioned characteristics are incorporated into one of the scoring systems, which seek to clarify either the cause or the phenotypic appearance of the AIS, which helps direct further testing and treatment. (The ASCOD³⁶ and TOAST³⁷ classification schemes are commonly used phenotypic and causative classifications, respectively.) Several (not all) of the broad phenotypic imaging patterns, with myriad clinical manifestations, are reviewed below. They include:

• Embolic stroke, which, classically, involves end circulation and therefore has cortical involvement. Typically, these originate from the heart or large extracranial arteries, and higher rates of atrial fibrillation and hypercoagulable states are implicated.
• Thrombotic stroke, which, typically, is from large vessels or small vessels, and occurs as a result of atherosclerosis. These strokes are more common at the origins or bifurcations of vessels. Symptoms of thrombotic stroke classically wax and wane slightly more frequently. Lacunar strokes are typically from thrombotic causes, although there are rare episodes of an embolic source contributing to a lacunar stroke syndrome.³⁸

There is evidence for using MRI discrepancies between diffusion-weighted and FLAIR imaging to time AIS findings in so-called wake-up strokes.³⁹ The rationale is that strokes < 4.5 hours old can be identified because they would have abnormal diffusion imaging components but normal findings with FLAIR. When these criteria were utilized in considering whether to treat with thrombolysis, there was a statistically significant improvement in 90-day modified Rankin scale (odds ratio = 1.61; 95% confidence interval, 1.09-2.36), but also an increased probability of death and intracerebral hemorrhage.³⁹

This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome.

A recent multicenter, randomized, open-label trial, with blinded outcomes assessment, showcased the efficacy of thrombectomy as an adjunct when ischemic brain territory was identified without frank infarction, as ascertained by CT perfusion within the anterior circulation. This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome with favorable imaging characteristics (on the modified Rankin scale, an ordinal score of 4 with medical therapy and an ordinal score of 3 with EVT [odds ratio = 2.77; 95% confidence interval, 1.63-4.70]).²⁹ A 2018 multicenter, prospective, randomized trial with blinded assessment of endpoints extended this idea, demonstrating that, when there was mismatch of the clinical deficit (ie, high NIHSS score) and infarct volume (measured on diffusion-weighted MRI or CT perfusion), thrombectomy as late as 24 hours after the patient was last known to be well was beneficial for lesions in the anterior circulation—specifically, the intracranial internal carotid artery or the proximal middle cerebral artery.⁴⁰

^a Whether local emergency departments (EDs) should be bypassed in favor of a specialized stroke center is the subject of debate. The 2019 American Heart Association/American Stroke Association guidelines note the AHA’s Mission: Lifeline Stroke EMS algorithm, which bypasses the nearest ED in feared cases of large-vessel occlusion if travel to a comprehensive stroke center can be accomplished within 30 minutes of arrival at the scene. This is based on expert consensus.^3,12,13

CORRESPONDENCE
Brian Ford, MD, 4301 Jones Bridge Road, Bethesda, MD; [email protected].

Stroke ranks second behind heart disease as the leading cause of mortality worldwide, accounting for 1 of every 19 deaths,¹ and remains a serious cause of morbidity. Best practices in stroke diagnosis and management can seem elusive to front-line clinicians, for 2 reasons: the rate of proliferation and nuance in stroke medicine and the fact that the typical scope of primary care practice exists apart from much of the diagnostic tools and management schema provided in stroke centers.² In this article, we describe and update the diagnosis of stroke and review imaging modalities, their nuances, and their application in practice.

Diagnosis of acute stroke

Acute stroke is diagnosed upon observation of new neurologic deficits and congruent neuroimaging. Some updated definitions favor a silent form of cerebral ischemia manifested by imaging pathology only; this form is not discussed in this article. Although there are several characteristically distinct stroke syndromes, there is no way to clinically distinguish ischemic pathology from hemorrhagic pathology.

Some common symptoms that should prompt evaluation for stroke are part of the American Stroke Association FAST mnemonic designed to promote public health awareness^3-5:

• face drooping
• arm weakness
• speech difficulty
• time to call 911.

There are several characteristically distinct stroke syndromes, but no way to differentiate ischemic and hemorrhagic pathologies clinically.

Other commonly reported stroke symptoms include unilateral weakness or numbness, confusion, word-finding difficulty, visual problems, difficulty ambulating, dizziness, loss of balance or coordination, and thunderclap headache. A stroke should also be considered in the presence of any new focal neurologic deficit.^3,4

Stroke patients should be triaged by emergency medical services using a stroke screening scale, such as BE-FAST⁵ (a modification of FAST that adds balance and eye assessments); the Los Angeles Prehospital Stroke Screen (LAPSS)^6,7; the Rapid Arterial oCclusion Evaluation (RACE)⁸; and the Cincinnati Prehospital Stroke Severity Scale (CP-SSS)^9,10 (see “Stroke screening scales for early identification and triage"). Studies have not found that any single prehospital stroke scale is superior to the others for reliably predicting large-vessel occlusion; therefore, prehospital assessment is typically based on practice patterns in a given locale.¹¹ A patient (or family member or caregiver) who seeks your care for stroke symptoms should be told to call 911 and get emergency transport to a health care facility that can capably administer intravenous (IV) thrombolysis.^a

First responders should elicit “last-known-normal” time; this critical information can aid in diagnosis and drive therapeutic options, especially if patients are unaccompanied at time of transport to a higher echelon of care. A point-of-care blood glucose test should be performed by emergency medical staff, with dextrose administered for a level < 45 mg/dL. Establishing IV access for fluids, medications, and contrast can be considered if it does not delay transport. A 12-lead electrocardiogram can also be considered, again, as long as it does not delay transport to a facility capable of providing definitive therapy. Notification by emergency services staff before arrival and transport of the patient to such a facility is the essential element of prehospital care, and should be prioritized above ancillary testing beyond the stroke assessment.¹⁴

Guidelines recommend use of the National Institutes of Health Stroke Scale ­(NIHSS; www.stroke.nih.gov/resources/scale.htm) for clinical evaluation upon arrival at the ED.¹⁵ Although no scale has been identified that can reliably predict large-vessel occlusion amenable to endovascular therapy (EVT), no other score has been found to outperform the NIHSS in achieving meaningful patient outcomes.¹⁶ Furthermore, NIHSS has been validated to track clinical changes in response to therapy, is widely utilized, and is free.

Continue to: A criticism of the NIHSS...

A criticism of the NIHSS is its bias toward left-hemispheric ischemic pathology.¹⁷ NIHSS includes 11 questions on a scale of 0 to 42; typically, a score < 4 is associated with a higher chance of a positive clinical outcome.¹⁸ There is no minimum or maximum NIHSS score that precludes treatment with thrombolysis or EVT.

Other commonly used scores in acute stroke include disability assessments. The modified Rankin scale, which is used most often, features a score of 0 (symptom-free) to 6 (death). A modified Rankin scale score of 0 or 1 is considered an indication of a favorable outcome after stroke.¹⁹ Note that these functional scores are not always part of an acute assessment but can be done early in the clinical course to gauge the response to treatment, and are collected for stroke-center certification.

Imaging modalities

Imaging is recommended within 20 minutes of arrival in the ED in a stroke patient who might be a candidate for thrombolysis or thrombectomy.³ There, imaging modalities commonly performed are noncontrast-enhanced head computed tomography (NCHCT); computed tomography (CT) angiography, with or without perfusion; and diffusion-weighted magnetic resonance imaging (MRI).^20,21 In addition, more highly specialized imaging modalities are available for the evaluation of the stroke patient in specific, often limited, circumstances. All these modalities are described below and compared in the TABLE,^20,21 using the ACR Appropriateness Criteria (of the American College of Radiology),²¹ which are guidelines for appropriate imaging of stroke, based on a clinical complaint. Separate recommendations and appraisals are offered by the most recent American Heart Association/American Stroke Association (AHA/ASA) guideline.³

NCHCT. This study should be performed within 20 minutes after arrival at the ED because it provides rapid assessment of intracerebral hemorrhage, can effectively corroborate the diagnosis of some stroke mimickers, and identifies some candidates for EVT or thrombolysis^3,21,22 (typically, the decision to proceed with EVT is based on adjunct imaging studies discussed in a bit). Evaluation for intracerebral hemorrhage is required prior to administering thrombolysis. Ischemic changes can be seen with variable specificity and sensitivity on NCHCT, depending on how much time has passed since the original insult. In all historical trials, CT was the only imaging modality used in the diagnosis of acute ischemic stroke (AIS) that suggested benefit from IV thrombolysis.^23-25

Acute, subacute, and chronic changes can be seen on NCHCT, although the modality has limited sensitivity for identifying AIS (ie, approximately 75% within 6 hours after the original insult):

• Acute findings on NCHCT include intracellular edema, which causes loss of the gray matter–white matter interface and effacement of the cortical sulci. This occurs as a result of increased cellular uptake of water in response to ischemia and cell death, resulting in a decreased density of tissue (hypoattenuation) in affected areas.
• Subacute changes appear in the 2- to 5-day window, including vasogenic edema with greater mass effect, hypoattenuation, and well-defined margins.^3,20,21
• Chronic vascular findings on NCHCT include loss of brain tissue and hypoattenuation.

Continue to: NCHCT is typically performed...

NCHCT is typically performed in advance of other adjunct imaging modalities.^3,20,21 Baseline NCHCT can be performed on patients with advanced kidney disease and those who have an indwelling metallic device.

CT angiography is performed with timed contrast, providing a 3-dimensional representation of the cerebral vasculature; the entire intracranial and extracranial vasculature, including the aortic arch, can be mapped in approximately 60 seconds. CT angiography is sensitive in identifying areas of stenosis > 50% and identifies clinically significant areas of stenosis up to approximately 90% of the time.²⁶ For this reason, it is particularly helpful in identifying candidates for an interventional strategy beyond pharmacotherapeutic thrombolysis. In addition, CT angiography can visualize aneurysmal dilation and dissection, and help with the planning of interventions—specifically, the confident administration of thrombolysis or more specific planning for target lesions and EVT.

Commonly used scoring systems in acute stroke include disability assessments— not always part of the acute assessment but undertaken early in the clinical course to gauge response to treatment.

It also can help identify a host of vascular phenomena, such as arteriovenous malformations, Moyamoya disease (progressive arterial blockage within the basal ganglia and compensatory microvascularization), and some vasculopathies.^20,27 In intracranial hemorrhage, CT with angiography can help evaluate for structural malformations and identify patients at risk of hematoma expansion.²²

CT perfusion. Many stroke centers will perform a CT perfusion study,²⁸ which encompasses as many as 3 different CT sequences:

• NCHCT
• vertex-to-arch angiography with contrast bolus
• administration of contrast and capture of a dynamic sequence through 1 or 2 slabs of tissue, allowing for the generation of maps of cerebral blood flow (CBF), mean transit time (MTT), and cerebral blood volume (CBV) of the entire cerebral vasculature.

The interplay of these 3 sequences drives characterization of lesions (ie, CBF = CBV/MTT). An infarct is characterized by low CBF, low CBV, and elevated MTT. In penumbral tissue, MTT is elevated but CBF is slightly decreased and CBV is normal or increased. Using CT perfusion, areas throughout the ischemic penumbra can be surveyed for favorable interventional characteristics.^20,29

Continue to: A CT perfusion study adds...

A CT perfusion study adds at least 60 seconds to NCHCT. This modality can be useful in planning interventions and for stratifying appropriateness of reperfusion strategies in strokes of unknown duration.^3,30 CT perfusion can be performed on any multidetector CT scan but (1) requires specialized software and expertise to interpret and (2) subjects the patient to a significant radiation dose, which, if incorrectly administered, can be considerably higher than intended.^20,26,27

Diffusion-weighted MRI. This is the most sensitive study for demonstrating early ischemic changes; however, limitations include lack of availability, contraindication in patients with metallic indwelling implants, and duration of the study—although, at some stroke centers, diffusion-weighted MRI can be performed in ≤ 10 minutes.

MRI and NCHCT have comparable sensitivity in detecting intracranial hemorrhage. MRI is likely more sensitive in identifying areas of microhemorrhage: In diffusion-weighted MRI, the sensitivity of stroke detection increases to > 95%.³¹ The modality relies on the comparable movement of water through damaged vs normal neuronal tissue. Diffusion-weighted MRI does not require administration of concomitant contrast, which can be a benefit in patients who are allergic to gadolinium-based contrast agents or have advanced kidney disease that precludes the use of contrast. It typically does not result in adequate characterization of extracranial vasculature.

Other MRI modalities. These MRI extensions include magnetic resonance (MR) perfusion and MR angiography. Whereas diffusion-weighted MRI (discussed above) offers the most rapid and sensitive evaluation for ischemia, fluid-attenuated inversion recovery (FLAIR) imaging has been utilized as a comparator to isolated diffusion-weighted MRI to help determine stroke duration. FLAIR signal positivity typically occurs 6 to 24 hours after the initial insult but is negative in stroke that occurred < 3 hours earlier.³²

MRI is limited, in terms of availability and increased study duration, especially when it comes to timely administration of thrombolysis. A benefit of this modality is less radiation and, as noted, superior sensitivity for ischemia. Diffusion-weighted MRI combined with MR perfusion analysis can help isolate areas of the ischemic penumbra. MR perfusion is performed for a similar reason as CT perfusion, although logistical execution across those modalities is significantly different. Considerations for choosing MR perfusion or CT perfusion should be made on an individual basis and based on available local resources and accepted local practice patterns.²⁶

Continue to: In the subacute setting...

Knowledge of historic details of the event, the patient (eg, known atrial fibrillation), and findings on imaging can facilitate communication between the primary care physician and inpatient teams.

In the subacute setting, MR perfusion and MR angiography of the head and the neck are often performed to identify stenosis, dissection, and more subtle mimickers of cerebrovascular accident not ascertained on initial CT evaluation. These studies are typically performed well outside the window for thrombolysis or intervention.²⁶ No guidelines specifically direct or recommend this practice pattern. The superior sensitivity and cerebral blood flow mapping of MR perfusion and MR angiography might be useful for validating a suspected diagnosis of ischemic stroke and providing phenotypic information about AIS events.

Transcranial Doppler imaging relies on bony windows to assess intracranial vascular flow, velocity, direction, and reactivity. This information can be utilized to diagnose stenosis or occlusion. This modality is principally used to evaluate for stenosis in the anterior circulation (sensitivity, 70%-90%; specificity, 90%-95%).²⁰ Evaluation of the basilar, vertebral, and internal carotid arteries is less accurate (sensitivity, 55%-80%).²⁰ Transcranial Doppler imaging is also used to assess for cerebral vasospasm after subarachnoid hemorrhage, monitor sickle cell disease patients’ overall risk for ischemic stroke, and augment thrombolysis. It is limited by the availability of an expert technician, and therefore is typically reserved for unstable patients or those who cannot receive contrast.²⁰

Carotid duplex ultrasonography. A dynamic study such as duplex ultrasonography can be strongly considered for flow imaging of the extracranial carotids to evaluate for stenosis. Indications for carotid stenting or endarterectomy include 50% to 79% occlusion of the carotid artery on the same side as a recent transient ischemic attack or AIS. Carotid stenosis > 80% warrants consideration for intervention independent of a recent cerebrovascular accident. Interventions are typically performed 2 to 14 days after stroke.³³ Although this study is of limited utility in the hyperacute setting, it is recommended within 24 hours after nondisabling stroke in the carotid territory, when (1) the patient is otherwise a candidate for a surgical or procedural intervention to address the stenosis and (2) none of the aforementioned studies that focus on neck vasculature have been performed.

Conventional (digital subtraction) ­angiography is the gold standard for mapping cerebrovascular disease because it is dynamic and highly accurate. It is, however, typically limited by the number of required personnel, its invasive nature, and the requirement for IV contrast. This study is performed during intra-arterial intervention techniques, including stent retrieval and intra-arterial thrombolysis.²⁶

Impact of imaging on treatment

Imaging helps determine the cause and some characteristics of stroke, both of which can help determine therapy. Strokes can be broadly subcategorized as hemorrhagic or ischemic; recent studies suggest that 87% are ischemic.³⁴ Knowledge of the historic details of the event, the patient (eg, known atrial fibrillation, anticoagulant use, history of falls), and findings on imaging can contribute to determine the cause of AIS, and can facilitate communication and consultation between the primary care physician and inpatient teams.³⁵

Continue to: Best practices for stroke treatment...

Best practices for stroke treatment are based on the cause of the event.³ To identify the likely cause, the aforementioned characteristics are incorporated into one of the scoring systems, which seek to clarify either the cause or the phenotypic appearance of the AIS, which helps direct further testing and treatment. (The ASCOD³⁶ and TOAST³⁷ classification schemes are commonly used phenotypic and causative classifications, respectively.) Several (not all) of the broad phenotypic imaging patterns, with myriad clinical manifestations, are reviewed below. They include:

• Embolic stroke, which, classically, involves end circulation and therefore has cortical involvement. Typically, these originate from the heart or large extracranial arteries, and higher rates of atrial fibrillation and hypercoagulable states are implicated.
• Thrombotic stroke, which, typically, is from large vessels or small vessels, and occurs as a result of atherosclerosis. These strokes are more common at the origins or bifurcations of vessels. Symptoms of thrombotic stroke classically wax and wane slightly more frequently. Lacunar strokes are typically from thrombotic causes, although there are rare episodes of an embolic source contributing to a lacunar stroke syndrome.³⁸

There is evidence for using MRI discrepancies between diffusion-weighted and FLAIR imaging to time AIS findings in so-called wake-up strokes.³⁹ The rationale is that strokes < 4.5 hours old can be identified because they would have abnormal diffusion imaging components but normal findings with FLAIR. When these criteria were utilized in considering whether to treat with thrombolysis, there was a statistically significant improvement in 90-day modified Rankin scale (odds ratio = 1.61; 95% confidence interval, 1.09-2.36), but also an increased probability of death and intracerebral hemorrhage.³⁹

This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome.

A recent multicenter, randomized, open-label trial, with blinded outcomes assessment, showcased the efficacy of thrombectomy as an adjunct when ischemic brain territory was identified without frank infarction, as ascertained by CT perfusion within the anterior circulation. This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome with favorable imaging characteristics (on the modified Rankin scale, an ordinal score of 4 with medical therapy and an ordinal score of 3 with EVT [odds ratio = 2.77; 95% confidence interval, 1.63-4.70]).²⁹ A 2018 multicenter, prospective, randomized trial with blinded assessment of endpoints extended this idea, demonstrating that, when there was mismatch of the clinical deficit (ie, high NIHSS score) and infarct volume (measured on diffusion-weighted MRI or CT perfusion), thrombectomy as late as 24 hours after the patient was last known to be well was beneficial for lesions in the anterior circulation—specifically, the intracranial internal carotid artery or the proximal middle cerebral artery.⁴⁰

^a Whether local emergency departments (EDs) should be bypassed in favor of a specialized stroke center is the subject of debate. The 2019 American Heart Association/American Stroke Association guidelines note the AHA’s Mission: Lifeline Stroke EMS algorithm, which bypasses the nearest ED in feared cases of large-vessel occlusion if travel to a comprehensive stroke center can be accomplished within 30 minutes of arrival at the scene. This is based on expert consensus.^3,12,13

CORRESPONDENCE
Brian Ford, MD, 4301 Jones Bridge Road, Bethesda, MD; [email protected].

Stroke ranks second behind heart disease as the leading cause of mortality worldwide, accounting for 1 of every 19 deaths,¹ and remains a serious cause of morbidity. Best practices in stroke diagnosis and management can seem elusive to front-line clinicians, for 2 reasons: the rate of proliferation and nuance in stroke medicine and the fact that the typical scope of primary care practice exists apart from much of the diagnostic tools and management schema provided in stroke centers.² In this article, we describe and update the diagnosis of stroke and review imaging modalities, their nuances, and their application in practice.

Diagnosis of acute stroke

Acute stroke is diagnosed upon observation of new neurologic deficits and congruent neuroimaging. Some updated definitions favor a silent form of cerebral ischemia manifested by imaging pathology only; this form is not discussed in this article. Although there are several characteristically distinct stroke syndromes, there is no way to clinically distinguish ischemic pathology from hemorrhagic pathology.

Some common symptoms that should prompt evaluation for stroke are part of the American Stroke Association FAST mnemonic designed to promote public health awareness^3-5:

• face drooping
• arm weakness
• speech difficulty
• time to call 911.

There are several characteristically distinct stroke syndromes, but no way to differentiate ischemic and hemorrhagic pathologies clinically.

Other commonly reported stroke symptoms include unilateral weakness or numbness, confusion, word-finding difficulty, visual problems, difficulty ambulating, dizziness, loss of balance or coordination, and thunderclap headache. A stroke should also be considered in the presence of any new focal neurologic deficit.^3,4

Stroke patients should be triaged by emergency medical services using a stroke screening scale, such as BE-FAST⁵ (a modification of FAST that adds balance and eye assessments); the Los Angeles Prehospital Stroke Screen (LAPSS)^6,7; the Rapid Arterial oCclusion Evaluation (RACE)⁸; and the Cincinnati Prehospital Stroke Severity Scale (CP-SSS)^9,10 (see “Stroke screening scales for early identification and triage"). Studies have not found that any single prehospital stroke scale is superior to the others for reliably predicting large-vessel occlusion; therefore, prehospital assessment is typically based on practice patterns in a given locale.¹¹ A patient (or family member or caregiver) who seeks your care for stroke symptoms should be told to call 911 and get emergency transport to a health care facility that can capably administer intravenous (IV) thrombolysis.^a

First responders should elicit “last-known-normal” time; this critical information can aid in diagnosis and drive therapeutic options, especially if patients are unaccompanied at time of transport to a higher echelon of care. A point-of-care blood glucose test should be performed by emergency medical staff, with dextrose administered for a level < 45 mg/dL. Establishing IV access for fluids, medications, and contrast can be considered if it does not delay transport. A 12-lead electrocardiogram can also be considered, again, as long as it does not delay transport to a facility capable of providing definitive therapy. Notification by emergency services staff before arrival and transport of the patient to such a facility is the essential element of prehospital care, and should be prioritized above ancillary testing beyond the stroke assessment.¹⁴

Guidelines recommend use of the National Institutes of Health Stroke Scale ­(NIHSS; www.stroke.nih.gov/resources/scale.htm) for clinical evaluation upon arrival at the ED.¹⁵ Although no scale has been identified that can reliably predict large-vessel occlusion amenable to endovascular therapy (EVT), no other score has been found to outperform the NIHSS in achieving meaningful patient outcomes.¹⁶ Furthermore, NIHSS has been validated to track clinical changes in response to therapy, is widely utilized, and is free.

Continue to: A criticism of the NIHSS...

A criticism of the NIHSS is its bias toward left-hemispheric ischemic pathology.¹⁷ NIHSS includes 11 questions on a scale of 0 to 42; typically, a score < 4 is associated with a higher chance of a positive clinical outcome.¹⁸ There is no minimum or maximum NIHSS score that precludes treatment with thrombolysis or EVT.

Other commonly used scores in acute stroke include disability assessments. The modified Rankin scale, which is used most often, features a score of 0 (symptom-free) to 6 (death). A modified Rankin scale score of 0 or 1 is considered an indication of a favorable outcome after stroke.¹⁹ Note that these functional scores are not always part of an acute assessment but can be done early in the clinical course to gauge the response to treatment, and are collected for stroke-center certification.

Imaging modalities

Imaging is recommended within 20 minutes of arrival in the ED in a stroke patient who might be a candidate for thrombolysis or thrombectomy.³ There, imaging modalities commonly performed are noncontrast-enhanced head computed tomography (NCHCT); computed tomography (CT) angiography, with or without perfusion; and diffusion-weighted magnetic resonance imaging (MRI).^20,21 In addition, more highly specialized imaging modalities are available for the evaluation of the stroke patient in specific, often limited, circumstances. All these modalities are described below and compared in the TABLE,^20,21 using the ACR Appropriateness Criteria (of the American College of Radiology),²¹ which are guidelines for appropriate imaging of stroke, based on a clinical complaint. Separate recommendations and appraisals are offered by the most recent American Heart Association/American Stroke Association (AHA/ASA) guideline.³

NCHCT. This study should be performed within 20 minutes after arrival at the ED because it provides rapid assessment of intracerebral hemorrhage, can effectively corroborate the diagnosis of some stroke mimickers, and identifies some candidates for EVT or thrombolysis^3,21,22 (typically, the decision to proceed with EVT is based on adjunct imaging studies discussed in a bit). Evaluation for intracerebral hemorrhage is required prior to administering thrombolysis. Ischemic changes can be seen with variable specificity and sensitivity on NCHCT, depending on how much time has passed since the original insult. In all historical trials, CT was the only imaging modality used in the diagnosis of acute ischemic stroke (AIS) that suggested benefit from IV thrombolysis.^23-25

Acute, subacute, and chronic changes can be seen on NCHCT, although the modality has limited sensitivity for identifying AIS (ie, approximately 75% within 6 hours after the original insult):

• Acute findings on NCHCT include intracellular edema, which causes loss of the gray matter–white matter interface and effacement of the cortical sulci. This occurs as a result of increased cellular uptake of water in response to ischemia and cell death, resulting in a decreased density of tissue (hypoattenuation) in affected areas.
• Subacute changes appear in the 2- to 5-day window, including vasogenic edema with greater mass effect, hypoattenuation, and well-defined margins.^3,20,21
• Chronic vascular findings on NCHCT include loss of brain tissue and hypoattenuation.

Continue to: NCHCT is typically performed...

NCHCT is typically performed in advance of other adjunct imaging modalities.^3,20,21 Baseline NCHCT can be performed on patients with advanced kidney disease and those who have an indwelling metallic device.

CT angiography is performed with timed contrast, providing a 3-dimensional representation of the cerebral vasculature; the entire intracranial and extracranial vasculature, including the aortic arch, can be mapped in approximately 60 seconds. CT angiography is sensitive in identifying areas of stenosis > 50% and identifies clinically significant areas of stenosis up to approximately 90% of the time.²⁶ For this reason, it is particularly helpful in identifying candidates for an interventional strategy beyond pharmacotherapeutic thrombolysis. In addition, CT angiography can visualize aneurysmal dilation and dissection, and help with the planning of interventions—specifically, the confident administration of thrombolysis or more specific planning for target lesions and EVT.

Commonly used scoring systems in acute stroke include disability assessments— not always part of the acute assessment but undertaken early in the clinical course to gauge response to treatment.

It also can help identify a host of vascular phenomena, such as arteriovenous malformations, Moyamoya disease (progressive arterial blockage within the basal ganglia and compensatory microvascularization), and some vasculopathies.^20,27 In intracranial hemorrhage, CT with angiography can help evaluate for structural malformations and identify patients at risk of hematoma expansion.²²

CT perfusion. Many stroke centers will perform a CT perfusion study,²⁸ which encompasses as many as 3 different CT sequences:

• NCHCT
• vertex-to-arch angiography with contrast bolus
• administration of contrast and capture of a dynamic sequence through 1 or 2 slabs of tissue, allowing for the generation of maps of cerebral blood flow (CBF), mean transit time (MTT), and cerebral blood volume (CBV) of the entire cerebral vasculature.

The interplay of these 3 sequences drives characterization of lesions (ie, CBF = CBV/MTT). An infarct is characterized by low CBF, low CBV, and elevated MTT. In penumbral tissue, MTT is elevated but CBF is slightly decreased and CBV is normal or increased. Using CT perfusion, areas throughout the ischemic penumbra can be surveyed for favorable interventional characteristics.^20,29

Continue to: A CT perfusion study adds...

A CT perfusion study adds at least 60 seconds to NCHCT. This modality can be useful in planning interventions and for stratifying appropriateness of reperfusion strategies in strokes of unknown duration.^3,30 CT perfusion can be performed on any multidetector CT scan but (1) requires specialized software and expertise to interpret and (2) subjects the patient to a significant radiation dose, which, if incorrectly administered, can be considerably higher than intended.^20,26,27

Diffusion-weighted MRI. This is the most sensitive study for demonstrating early ischemic changes; however, limitations include lack of availability, contraindication in patients with metallic indwelling implants, and duration of the study—although, at some stroke centers, diffusion-weighted MRI can be performed in ≤ 10 minutes.

MRI and NCHCT have comparable sensitivity in detecting intracranial hemorrhage. MRI is likely more sensitive in identifying areas of microhemorrhage: In diffusion-weighted MRI, the sensitivity of stroke detection increases to > 95%.³¹ The modality relies on the comparable movement of water through damaged vs normal neuronal tissue. Diffusion-weighted MRI does not require administration of concomitant contrast, which can be a benefit in patients who are allergic to gadolinium-based contrast agents or have advanced kidney disease that precludes the use of contrast. It typically does not result in adequate characterization of extracranial vasculature.

Other MRI modalities. These MRI extensions include magnetic resonance (MR) perfusion and MR angiography. Whereas diffusion-weighted MRI (discussed above) offers the most rapid and sensitive evaluation for ischemia, fluid-attenuated inversion recovery (FLAIR) imaging has been utilized as a comparator to isolated diffusion-weighted MRI to help determine stroke duration. FLAIR signal positivity typically occurs 6 to 24 hours after the initial insult but is negative in stroke that occurred < 3 hours earlier.³²

MRI is limited, in terms of availability and increased study duration, especially when it comes to timely administration of thrombolysis. A benefit of this modality is less radiation and, as noted, superior sensitivity for ischemia. Diffusion-weighted MRI combined with MR perfusion analysis can help isolate areas of the ischemic penumbra. MR perfusion is performed for a similar reason as CT perfusion, although logistical execution across those modalities is significantly different. Considerations for choosing MR perfusion or CT perfusion should be made on an individual basis and based on available local resources and accepted local practice patterns.²⁶

Continue to: In the subacute setting...

Knowledge of historic details of the event, the patient (eg, known atrial fibrillation), and findings on imaging can facilitate communication between the primary care physician and inpatient teams.

In the subacute setting, MR perfusion and MR angiography of the head and the neck are often performed to identify stenosis, dissection, and more subtle mimickers of cerebrovascular accident not ascertained on initial CT evaluation. These studies are typically performed well outside the window for thrombolysis or intervention.²⁶ No guidelines specifically direct or recommend this practice pattern. The superior sensitivity and cerebral blood flow mapping of MR perfusion and MR angiography might be useful for validating a suspected diagnosis of ischemic stroke and providing phenotypic information about AIS events.

Transcranial Doppler imaging relies on bony windows to assess intracranial vascular flow, velocity, direction, and reactivity. This information can be utilized to diagnose stenosis or occlusion. This modality is principally used to evaluate for stenosis in the anterior circulation (sensitivity, 70%-90%; specificity, 90%-95%).²⁰ Evaluation of the basilar, vertebral, and internal carotid arteries is less accurate (sensitivity, 55%-80%).²⁰ Transcranial Doppler imaging is also used to assess for cerebral vasospasm after subarachnoid hemorrhage, monitor sickle cell disease patients’ overall risk for ischemic stroke, and augment thrombolysis. It is limited by the availability of an expert technician, and therefore is typically reserved for unstable patients or those who cannot receive contrast.²⁰

Carotid duplex ultrasonography. A dynamic study such as duplex ultrasonography can be strongly considered for flow imaging of the extracranial carotids to evaluate for stenosis. Indications for carotid stenting or endarterectomy include 50% to 79% occlusion of the carotid artery on the same side as a recent transient ischemic attack or AIS. Carotid stenosis > 80% warrants consideration for intervention independent of a recent cerebrovascular accident. Interventions are typically performed 2 to 14 days after stroke.³³ Although this study is of limited utility in the hyperacute setting, it is recommended within 24 hours after nondisabling stroke in the carotid territory, when (1) the patient is otherwise a candidate for a surgical or procedural intervention to address the stenosis and (2) none of the aforementioned studies that focus on neck vasculature have been performed.

Conventional (digital subtraction) ­angiography is the gold standard for mapping cerebrovascular disease because it is dynamic and highly accurate. It is, however, typically limited by the number of required personnel, its invasive nature, and the requirement for IV contrast. This study is performed during intra-arterial intervention techniques, including stent retrieval and intra-arterial thrombolysis.²⁶

Impact of imaging on treatment

Imaging helps determine the cause and some characteristics of stroke, both of which can help determine therapy. Strokes can be broadly subcategorized as hemorrhagic or ischemic; recent studies suggest that 87% are ischemic.³⁴ Knowledge of the historic details of the event, the patient (eg, known atrial fibrillation, anticoagulant use, history of falls), and findings on imaging can contribute to determine the cause of AIS, and can facilitate communication and consultation between the primary care physician and inpatient teams.³⁵

Continue to: Best practices for stroke treatment...

Best practices for stroke treatment are based on the cause of the event.³ To identify the likely cause, the aforementioned characteristics are incorporated into one of the scoring systems, which seek to clarify either the cause or the phenotypic appearance of the AIS, which helps direct further testing and treatment. (The ASCOD³⁶ and TOAST³⁷ classification schemes are commonly used phenotypic and causative classifications, respectively.) Several (not all) of the broad phenotypic imaging patterns, with myriad clinical manifestations, are reviewed below. They include:

• Embolic stroke, which, classically, involves end circulation and therefore has cortical involvement. Typically, these originate from the heart or large extracranial arteries, and higher rates of atrial fibrillation and hypercoagulable states are implicated.
• Thrombotic stroke, which, typically, is from large vessels or small vessels, and occurs as a result of atherosclerosis. These strokes are more common at the origins or bifurcations of vessels. Symptoms of thrombotic stroke classically wax and wane slightly more frequently. Lacunar strokes are typically from thrombotic causes, although there are rare episodes of an embolic source contributing to a lacunar stroke syndrome.³⁸

There is evidence for using MRI discrepancies between diffusion-weighted and FLAIR imaging to time AIS findings in so-called wake-up strokes.³⁹ The rationale is that strokes < 4.5 hours old can be identified because they would have abnormal diffusion imaging components but normal findings with FLAIR. When these criteria were utilized in considering whether to treat with thrombolysis, there was a statistically significant improvement in 90-day modified Rankin scale (odds ratio = 1.61; 95% confidence interval, 1.09-2.36), but also an increased probability of death and intracerebral hemorrhage.³⁹

This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome.

A recent multicenter, randomized, open-label trial, with blinded outcomes assessment, showcased the efficacy of thrombectomy as an adjunct when ischemic brain territory was identified without frank infarction, as ascertained by CT perfusion within the anterior circulation. This trial showed that thrombectomy could be performed as long as 16 hours after the patient was last well-appearing and still result in an improved outcome with favorable imaging characteristics (on the modified Rankin scale, an ordinal score of 4 with medical therapy and an ordinal score of 3 with EVT [odds ratio = 2.77; 95% confidence interval, 1.63-4.70]).²⁹ A 2018 multicenter, prospective, randomized trial with blinded assessment of endpoints extended this idea, demonstrating that, when there was mismatch of the clinical deficit (ie, high NIHSS score) and infarct volume (measured on diffusion-weighted MRI or CT perfusion), thrombectomy as late as 24 hours after the patient was last known to be well was beneficial for lesions in the anterior circulation—specifically, the intracranial internal carotid artery or the proximal middle cerebral artery.⁴⁰

^a Whether local emergency departments (EDs) should be bypassed in favor of a specialized stroke center is the subject of debate. The 2019 American Heart Association/American Stroke Association guidelines note the AHA’s Mission: Lifeline Stroke EMS algorithm, which bypasses the nearest ED in feared cases of large-vessel occlusion if travel to a comprehensive stroke center can be accomplished within 30 minutes of arrival at the scene. This is based on expert consensus.^3,12,13

CORRESPONDENCE
Brian Ford, MD, 4301 Jones Bridge Road, Bethesda, MD; [email protected].

Pruritus, defined as a sensation that induces a desire to scratch¹ and classified as acute or chronic (lasting > 6 weeks),² is one of the most common complaints among primary care patients: Approximately 1% of ambulatory visits in the United States are linked to pruritus.³

Chronic pruritus impairs quality of life; its impact has been compared to that of chronic pain.⁴ Treatment should therefore be instituted promptly. Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition, as we describe here. When persistent pruritus is refractory to treatment, systemic causes should be fully explored.

In this article, we discuss the pathogenesis and management of pruritus without skin eruption in the adult nonpregnant patient. We also present practice recommendations to help you determine whether your patient’s pruritus is indicative of a serious systemic condition.

An incomplete understanding of the pathophysiology of pruritus

The pathophysiology of pruritus is not fully understood. It is generally recognized, however, that pruritus starts in the peripheral nerves located in the dermal–epidermal junction of the skin.⁵ The sensation is then transmitted along unmyelinated slow-conducting C fibers to the dorsal horn of the spinal cord.^5,6 There are 2 types of C fibers that transmit the itch impulse⁶: A histamine-dependent type and a non-­histamine-dependent type, which might explain why pruritus can be refractory to antihistamine treatment.⁶

Once the itch impulse has moved from the spinal cord, it travels along the spinothalamic tract up to the contralateral thalamus.¹ From there, the impulse ascends to the cerebral cortex.¹ In the cortex, the impulse triggers multiple areas of the brain, such as those responsible for sensation, motor function, reward, memory, and emotion.⁷

Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition.

Several chemical mediators have been found to be peripheral and central inducers of pruritus: histamine, endogenous opioids, substance P, and serotonin.² There are indications that certain receptors, such as mu-opioid receptors and kappa-opioid receptors, are key contributors to itch as well.²

A diverse etiology

The International Forum for the Study of Itch (IFSI) has established 6 main categories of causes of pruritus(TABLE 1)²:

• dermatologic
• systemic
• neurologic
• psychogenic
• mixed
• other.

Continue to: In this review...

In this review, we focus on the work-up and management of 3 of those categories: systemic, neurologic, and psychogenic causes of pruritus.

Systemic causes

Research has shown that 14% to 24% of patients who seek the care of a dermatologist for chronic itch without skin lesions have a systemic illness.⁸

Renal disease. Approximately 40% of patients with end-stage renal disease who are on hemodialysis or peritoneal dialysis have uremic pruritus.2 The itch is mostly generalized but can be pronounced on the back. For most patients, the itch is worse at night, causing a major impact on quality of life.⁶

Liver disease. In hepatic disease, there is often impairment in the secretion of bile, which can lead to cholestatic pruritus.² This condition commonly affects the hands and feet first; later, it becomes generalized.² Cholestatic pruritus can be elicited by tight-fitting clothing. Relief is not achieved by scratching.⁹ This type of itch effects 70% of patients with primary biliary cirrhosis and 15% of patients with hepatitis C infection.⁹

Hematologic disorders. Pruritus is a hallmark symptom of polycythemia rubra vera. Almost 50% of patients with this disorder report pruritus that occurs after exposure to water⁹; aquagenic pruritus can precede the formal diagnosis of polycythemia rubra vera by years.² It has been speculated that platelet aggregation in this disorder leads to release of serotonin and histamine, which, in turn, causes itch.⁹

Continue to: Endocrine disorders

Endocrine disorders. Approximately 4% to 11% of patients with thyrotoxicosis have pruritus.¹ It has been suggested that vasodilation, increased skin temperature, and a decreased itch threshold from untreated Graves disease might be inciting factors.

Malignancy. In generalized chronic pruritus without a known cause, strongly consider the likelihood of underlying malignancy^8,10; for 10% of these patients, their chronic pruritus is a paraneoplastic sign. Paraneoplastic pruritus is characterized as an itch that predates clinical onset, or occurs early in the course, of a malignancy.⁹ The condition is most strongly linked to cancers of the liver, gallbladder, biliary tract, hematologic system, and skin.¹¹

Palpate the liver, spleen, lymph nodes, and thyroid for organomegaly, which could indicate a serious systemic condition as the cause of pruritus.

Chronic pruritus affects 30% of patients with Hodgkin lymphoma.⁹ General pruritus can precede this diagnosis by months, even years.¹ In Hodgkin lymphoma patients who are in remission, a return of pruritic symptoms can be a harbinger of recurrence.⁹

Neurologic causes

A recent study found that 8% to 15% of patients referred to a dermatology clinic for chronic pruritus without skin eruption had underlying neurologic pathology.¹² Although the specific mechanisms of neuropathic itch are still poorly understood, it has been theorized that the itch emanates from neuronal damage, which can come from peripheral or central nervous system lesions.⁹

Brachioradial pruritus. There are divergent theories about the etiology of brachioradial pruritus. One hypothesis is that the condition is caused by cervical nerve-root impingement at the level of C5-C8 that leads to nerve damage²; another is that chronic exposure to sunlight causes injury to peripheral cutaneous nerves.² Brachioradial pruritus is localized to the dorsolateral forearm; it can also involve the neck, back, shoulder, upper arm, and chest, unilaterally and bilaterally. This pruritus can be intermittent and become worse upon exposure to sunlight.²

Continue to: Notalgia paresthetica

Notalgia paresthetica. This condition might also cause neuropathic pruritus as a consequence of nerve impingement. The itch of notalgia paresthesia is located on the skin, medial to the scapular border on the upper or mid-back.² It has been postulated that the itch is caused by nerve entrapment of the posterior rami of spinal nerves arising from T2-T6.⁹ However, another theory suggests that the itch is caused by damage to peripheral nerves.⁹ The itch of notalgia paresthetica can wax and wane.²

Poststroke pruritus. Brain lesions, most often caused by stroke, can cause neuropathic itch. One of the best-known syndromes related to poststroke itch is Wallenberg syndrome (ischemia from a lateral medullary infarction), which typically presents with itch, thermalgic hypoesthesia of the face, cerebellar dysfunction, nausea, and vomiting.⁷

Shingles. More than one-half of patients who develop postherpetic neuralgia as a consequence of a herpes zoster infection also develop neuropathic pruritus.⁹ It is thought that postherpetic pruritus shares a comparable pathophysiology with postherpetic neuralgia, in which neurons involved in itch stimuli become damaged.⁷

Diabetes mellitus. Pruritus from diabetes can be classified as systemic or neuropathic. Diabetes is one of the most common causes of small-fiber polyneuropathy, which can cause neuropathic pruritus.¹³

Multiple sclerosis. Central nervous system lesions that affect sensory pathways can lead to neuropathic itch in multiple sclerosis. Patients can have severe episodes of generalized pruritus. It has been hypothesized that the neuropathic itch in multiple sclerosis is induced by activation of artificial synapses in demyelinated areas.²

Continue to: Psychogenic pruritus

Chronic pruritus can be a comorbidity of psychiatric illness. A retrospective study found that pruritus occurs in 32% to 42% of psychiatric inpatients.¹⁴ Depression, anxiety, bipolar disorders, obsessive–compulsive disorders, somatoform disorders, psychosis, and substance abuse all have a strong link to psychogenic excoriation.¹⁵ Psychogenic excoriation, which can cause secondary skin lesions, occurs in psychiatric patients who excessively pick and scratch normal skin because they perceive an itch sensation or have a delusion of infestation.² Affected skin can be marked by scattered crusted lesions (FIGURE) anywhere on the body that the patient can reach—most commonly, the extremities.²

Delusion of infestation. Patients with a delusion of infestation have a strong belief that their body is infected by some kind of insect or microorganism.¹⁶ Before a diagnosis of delusion of infestation can be made, other organic causes must be excluded, including withdrawal from such substances as cocaine, amphetamines, and alcohol.¹⁶ Patients with a delusion of infestation can have, and maintain, a symptomatic response with continuing use of an atypical antipsychotic agent, including risperidone and olanzapine.¹⁷

Evaluation and diagnostic work-up

A thorough medical history, review of systems, medication review, social history, and family history are important when evaluating a patient with chronic pruritus.¹⁸ These items can be valuable in formulating a differential diagnosis, even before a physical examination.

Physical examination. The physical exam should include detailed inspection of the entire skin and hair¹⁸; such a comprehensive physical exam can determine whether the source of the itch is cutaneous.⁷ This, in turn, can help further narrow the differential diagnosis. It is crucial that the physical exam include palpation of the liver, spleen, lymph nodes, and thyroid for organomegaly,⁸ which could indicate a serious systemic condition, such as lymphoma.

The ice-pack sign—in which an ice pack applied to the pruritic area provides immediate relief—is considered pathognomonic for brachioradial pruritus.

The ice-pack sign—in which an ice pack is applied to the pruritic area, the patient experiences immediate relief of pruritus, and the itch returns soon after the ice pack is removed—is considered pathognomonic for brachioradial pruritus.¹⁹

Continue to: Chronic pruritus with abnormal findings...

Chronic pruritus with abnormal findings on the physical exam should prompt an initial work-up.¹⁸ Also consider an initial work-up for a patient with chronic pruritus whose symptom has not been relieved with conservative treatment.¹⁸

Laboratory testing. The initial laboratory work-up could include any of the following evaluations: complete blood count, measurement of thyroid-stimulating hormone, comprehensive metabolic panel (liver function, renal function, and the serum glucose level) and the erythrocyte sedimentation rate (TABLE 2).¹⁸ If warranted by the evaluation and physical exam, blood work can also include serologic studies for human immunodeficiency virus infection and ­hepatitis.¹⁷

Imaging. Chest radiography should be performed if there is suspicion of malignancy, such as lymphoma.⁷ Although brachioradial pruritus and notalgia paresthetica have been postulated to be caused by impingement of spinal nerves, obtaining spinal imaging, such as magnetic resonance imaging, as part of the initial work-up is not recommended; because spinal images might not show evidence of spinal disease, obtaining spinal imaging is not a requirement before treating brachioradial pruritus and notalgia paresthetica. Do consider spinal imaging, however, for patients in whom brachioradial pruritus or notalgia paresthetica is suspected and conservative treatment has not produced a response.

Treatment: Nondrug approaches, topicals, systemic agents

Start conservatively. Treatment of pruritus should begin with behavior modification and nonpharmacotherapeutic options (TABLE 3⁸). Educate the patient that scratching might cause secondary skin lesions; empowering them with that knowledge is sometimes enough to help break the scratching cycling—especially if the patient combines behavior modification with proper skin hydration with an emollient. To prevent secondary skin lesions through involuntary scratching, consider recommending that lesions be covered with an occlusive dressing or protective clothing.¹³

Stress has been shown to make chronic itch worse; therefore, stress-reduction activities, such as exercise, meditation, and yoga, might be helpful.20 For patients in whom pruritus has a psychological component, referral to a psychiatrist or psychologist might be therapeutic.

Continue to: When a patient complains...

When a patient complains of severe pruritus at first presentation, consider pharmacotherapy in conjunction with nonpharmacotherapeutic options. Several of the more effective topical therapies for pruritus^a are listed in TABLE 4.²⁰ Well-known systemic agents for this purpose are reviewed below and listed in TABLE 5.⁷

Systemic treatment

Antihistamines. A staple in the treatment of pruritus for many years, antihistamines are not effective for all causes; however, they are effective in treating paraneoplastic pruritus.²⁰ First-generation antihistamines, with their sedating effect, can be useful for patients who experience generalized pruritus at night.²⁰

Anticonvulsants. Gabapentin and pregabalin are analogs of the neurotransmitter gamma-aminobutyric acid.²⁰ This drug class is helpful in neuropathic pruritus specifically caused by impingements, such as brachioradial pruritus and notalgia paresthetica.²⁰ In addition, of all systemic therapies used to treat uremic pruritus, gabapentin has, in clinical trials, most consistently been found effective for uremic pruritus.⁶ (Note: Use renal dosing of gabapentin in patients with renal failure.)

Antidepressants. Selective serotonin reuptake inhibitors (SSRIs; eg, fluvoxamine, paroxetine, and sertraline) might cause itch to subside by increasing the serotonin level, which, in turn, works to decrease inflammatory substances that cause itch.⁷ SSRIs have been used to treat patients with psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.⁷

Start conservatively: Use behavior modification and nonpharmacotherapeutic options for pruritus first.

Tricyclic antidepressants (eg, amitriptyline and doxepin) lessen the itch by antagonizing histamine receptors and through anticholinergic mechanisms. Tricyclics are best used in the treatment of psychogenic and nocturnal itch.⁷

Continue to: Mirtazapine...

Mirtazapine, a tetracyclic antidepressant, works in patients with uremic pruritus, psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.¹

Substance P antagonist. Aprepitant, a neurokinin receptor I antagonist, is a newer agent that inhibits binding of the itch mediator substance P to the neurokinin receptor. The drug has been found helpful in patients with drug-induced, paraneoplastic, and brachioradial pruritus.⁷

Opioid-receptor agents. Naltrexone, as a mu opioid-receptor antagonist, has shown promise as a treatment for uremic pruritus and cholestatic pruritus. Nalfurafine, a kappa opioid-receptor agonist, is emerging as a possible therapy for uremic pruritus.⁷

Bile-acid sequestrants. A few small studies have shown that treatment with a bile-acid sequestrant, such as cholestyramine and ursodiol, induces moderate improvement in symptoms in patients with cholestatic pruritus.²¹

CORRESPONDENCE
Matasha Russell, MD, Department of Family and Community Medicine, The University of Texas Health Science Center at Houston, McGovern Medical School, 6431 Fannin Street, JJL 324, Houston, TX 77030; [email protected].

Pruritus, defined as a sensation that induces a desire to scratch¹ and classified as acute or chronic (lasting > 6 weeks),² is one of the most common complaints among primary care patients: Approximately 1% of ambulatory visits in the United States are linked to pruritus.³

Chronic pruritus impairs quality of life; its impact has been compared to that of chronic pain.⁴ Treatment should therefore be instituted promptly. Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition, as we describe here. When persistent pruritus is refractory to treatment, systemic causes should be fully explored.

In this article, we discuss the pathogenesis and management of pruritus without skin eruption in the adult nonpregnant patient. We also present practice recommendations to help you determine whether your patient’s pruritus is indicative of a serious systemic condition.

An incomplete understanding of the pathophysiology of pruritus

The pathophysiology of pruritus is not fully understood. It is generally recognized, however, that pruritus starts in the peripheral nerves located in the dermal–epidermal junction of the skin.⁵ The sensation is then transmitted along unmyelinated slow-conducting C fibers to the dorsal horn of the spinal cord.^5,6 There are 2 types of C fibers that transmit the itch impulse⁶: A histamine-dependent type and a non-­histamine-dependent type, which might explain why pruritus can be refractory to antihistamine treatment.⁶

Once the itch impulse has moved from the spinal cord, it travels along the spinothalamic tract up to the contralateral thalamus.¹ From there, the impulse ascends to the cerebral cortex.¹ In the cortex, the impulse triggers multiple areas of the brain, such as those responsible for sensation, motor function, reward, memory, and emotion.⁷

Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition.

Several chemical mediators have been found to be peripheral and central inducers of pruritus: histamine, endogenous opioids, substance P, and serotonin.² There are indications that certain receptors, such as mu-opioid receptors and kappa-opioid receptors, are key contributors to itch as well.²

A diverse etiology

The International Forum for the Study of Itch (IFSI) has established 6 main categories of causes of pruritus(TABLE 1)²:

• dermatologic
• systemic
• neurologic
• psychogenic
• mixed
• other.

Continue to: In this review...

In this review, we focus on the work-up and management of 3 of those categories: systemic, neurologic, and psychogenic causes of pruritus.

Systemic causes

Research has shown that 14% to 24% of patients who seek the care of a dermatologist for chronic itch without skin lesions have a systemic illness.⁸

Renal disease. Approximately 40% of patients with end-stage renal disease who are on hemodialysis or peritoneal dialysis have uremic pruritus.2 The itch is mostly generalized but can be pronounced on the back. For most patients, the itch is worse at night, causing a major impact on quality of life.⁶

Liver disease. In hepatic disease, there is often impairment in the secretion of bile, which can lead to cholestatic pruritus.² This condition commonly affects the hands and feet first; later, it becomes generalized.² Cholestatic pruritus can be elicited by tight-fitting clothing. Relief is not achieved by scratching.⁹ This type of itch effects 70% of patients with primary biliary cirrhosis and 15% of patients with hepatitis C infection.⁹

Hematologic disorders. Pruritus is a hallmark symptom of polycythemia rubra vera. Almost 50% of patients with this disorder report pruritus that occurs after exposure to water⁹; aquagenic pruritus can precede the formal diagnosis of polycythemia rubra vera by years.² It has been speculated that platelet aggregation in this disorder leads to release of serotonin and histamine, which, in turn, causes itch.⁹

Continue to: Endocrine disorders

Endocrine disorders. Approximately 4% to 11% of patients with thyrotoxicosis have pruritus.¹ It has been suggested that vasodilation, increased skin temperature, and a decreased itch threshold from untreated Graves disease might be inciting factors.

Malignancy. In generalized chronic pruritus without a known cause, strongly consider the likelihood of underlying malignancy^8,10; for 10% of these patients, their chronic pruritus is a paraneoplastic sign. Paraneoplastic pruritus is characterized as an itch that predates clinical onset, or occurs early in the course, of a malignancy.⁹ The condition is most strongly linked to cancers of the liver, gallbladder, biliary tract, hematologic system, and skin.¹¹

Palpate the liver, spleen, lymph nodes, and thyroid for organomegaly, which could indicate a serious systemic condition as the cause of pruritus.

Chronic pruritus affects 30% of patients with Hodgkin lymphoma.⁹ General pruritus can precede this diagnosis by months, even years.¹ In Hodgkin lymphoma patients who are in remission, a return of pruritic symptoms can be a harbinger of recurrence.⁹

Neurologic causes

A recent study found that 8% to 15% of patients referred to a dermatology clinic for chronic pruritus without skin eruption had underlying neurologic pathology.¹² Although the specific mechanisms of neuropathic itch are still poorly understood, it has been theorized that the itch emanates from neuronal damage, which can come from peripheral or central nervous system lesions.⁹

Brachioradial pruritus. There are divergent theories about the etiology of brachioradial pruritus. One hypothesis is that the condition is caused by cervical nerve-root impingement at the level of C5-C8 that leads to nerve damage²; another is that chronic exposure to sunlight causes injury to peripheral cutaneous nerves.² Brachioradial pruritus is localized to the dorsolateral forearm; it can also involve the neck, back, shoulder, upper arm, and chest, unilaterally and bilaterally. This pruritus can be intermittent and become worse upon exposure to sunlight.²

Continue to: Notalgia paresthetica

Notalgia paresthetica. This condition might also cause neuropathic pruritus as a consequence of nerve impingement. The itch of notalgia paresthesia is located on the skin, medial to the scapular border on the upper or mid-back.² It has been postulated that the itch is caused by nerve entrapment of the posterior rami of spinal nerves arising from T2-T6.⁹ However, another theory suggests that the itch is caused by damage to peripheral nerves.⁹ The itch of notalgia paresthetica can wax and wane.²

Poststroke pruritus. Brain lesions, most often caused by stroke, can cause neuropathic itch. One of the best-known syndromes related to poststroke itch is Wallenberg syndrome (ischemia from a lateral medullary infarction), which typically presents with itch, thermalgic hypoesthesia of the face, cerebellar dysfunction, nausea, and vomiting.⁷

Shingles. More than one-half of patients who develop postherpetic neuralgia as a consequence of a herpes zoster infection also develop neuropathic pruritus.⁹ It is thought that postherpetic pruritus shares a comparable pathophysiology with postherpetic neuralgia, in which neurons involved in itch stimuli become damaged.⁷

Diabetes mellitus. Pruritus from diabetes can be classified as systemic or neuropathic. Diabetes is one of the most common causes of small-fiber polyneuropathy, which can cause neuropathic pruritus.¹³

Multiple sclerosis. Central nervous system lesions that affect sensory pathways can lead to neuropathic itch in multiple sclerosis. Patients can have severe episodes of generalized pruritus. It has been hypothesized that the neuropathic itch in multiple sclerosis is induced by activation of artificial synapses in demyelinated areas.²

Continue to: Psychogenic pruritus

Chronic pruritus can be a comorbidity of psychiatric illness. A retrospective study found that pruritus occurs in 32% to 42% of psychiatric inpatients.¹⁴ Depression, anxiety, bipolar disorders, obsessive–compulsive disorders, somatoform disorders, psychosis, and substance abuse all have a strong link to psychogenic excoriation.¹⁵ Psychogenic excoriation, which can cause secondary skin lesions, occurs in psychiatric patients who excessively pick and scratch normal skin because they perceive an itch sensation or have a delusion of infestation.² Affected skin can be marked by scattered crusted lesions (FIGURE) anywhere on the body that the patient can reach—most commonly, the extremities.²

Delusion of infestation. Patients with a delusion of infestation have a strong belief that their body is infected by some kind of insect or microorganism.¹⁶ Before a diagnosis of delusion of infestation can be made, other organic causes must be excluded, including withdrawal from such substances as cocaine, amphetamines, and alcohol.¹⁶ Patients with a delusion of infestation can have, and maintain, a symptomatic response with continuing use of an atypical antipsychotic agent, including risperidone and olanzapine.¹⁷

Evaluation and diagnostic work-up

A thorough medical history, review of systems, medication review, social history, and family history are important when evaluating a patient with chronic pruritus.¹⁸ These items can be valuable in formulating a differential diagnosis, even before a physical examination.

Physical examination. The physical exam should include detailed inspection of the entire skin and hair¹⁸; such a comprehensive physical exam can determine whether the source of the itch is cutaneous.⁷ This, in turn, can help further narrow the differential diagnosis. It is crucial that the physical exam include palpation of the liver, spleen, lymph nodes, and thyroid for organomegaly,⁸ which could indicate a serious systemic condition, such as lymphoma.

The ice-pack sign—in which an ice pack applied to the pruritic area provides immediate relief—is considered pathognomonic for brachioradial pruritus.

The ice-pack sign—in which an ice pack is applied to the pruritic area, the patient experiences immediate relief of pruritus, and the itch returns soon after the ice pack is removed—is considered pathognomonic for brachioradial pruritus.¹⁹

Continue to: Chronic pruritus with abnormal findings...

Chronic pruritus with abnormal findings on the physical exam should prompt an initial work-up.¹⁸ Also consider an initial work-up for a patient with chronic pruritus whose symptom has not been relieved with conservative treatment.¹⁸

Laboratory testing. The initial laboratory work-up could include any of the following evaluations: complete blood count, measurement of thyroid-stimulating hormone, comprehensive metabolic panel (liver function, renal function, and the serum glucose level) and the erythrocyte sedimentation rate (TABLE 2).¹⁸ If warranted by the evaluation and physical exam, blood work can also include serologic studies for human immunodeficiency virus infection and ­hepatitis.¹⁷

Imaging. Chest radiography should be performed if there is suspicion of malignancy, such as lymphoma.⁷ Although brachioradial pruritus and notalgia paresthetica have been postulated to be caused by impingement of spinal nerves, obtaining spinal imaging, such as magnetic resonance imaging, as part of the initial work-up is not recommended; because spinal images might not show evidence of spinal disease, obtaining spinal imaging is not a requirement before treating brachioradial pruritus and notalgia paresthetica. Do consider spinal imaging, however, for patients in whom brachioradial pruritus or notalgia paresthetica is suspected and conservative treatment has not produced a response.

Treatment: Nondrug approaches, topicals, systemic agents

Start conservatively. Treatment of pruritus should begin with behavior modification and nonpharmacotherapeutic options (TABLE 3⁸). Educate the patient that scratching might cause secondary skin lesions; empowering them with that knowledge is sometimes enough to help break the scratching cycling—especially if the patient combines behavior modification with proper skin hydration with an emollient. To prevent secondary skin lesions through involuntary scratching, consider recommending that lesions be covered with an occlusive dressing or protective clothing.¹³

Stress has been shown to make chronic itch worse; therefore, stress-reduction activities, such as exercise, meditation, and yoga, might be helpful.20 For patients in whom pruritus has a psychological component, referral to a psychiatrist or psychologist might be therapeutic.

Continue to: When a patient complains...

When a patient complains of severe pruritus at first presentation, consider pharmacotherapy in conjunction with nonpharmacotherapeutic options. Several of the more effective topical therapies for pruritus^a are listed in TABLE 4.²⁰ Well-known systemic agents for this purpose are reviewed below and listed in TABLE 5.⁷

Systemic treatment

Antihistamines. A staple in the treatment of pruritus for many years, antihistamines are not effective for all causes; however, they are effective in treating paraneoplastic pruritus.²⁰ First-generation antihistamines, with their sedating effect, can be useful for patients who experience generalized pruritus at night.²⁰

Anticonvulsants. Gabapentin and pregabalin are analogs of the neurotransmitter gamma-aminobutyric acid.²⁰ This drug class is helpful in neuropathic pruritus specifically caused by impingements, such as brachioradial pruritus and notalgia paresthetica.²⁰ In addition, of all systemic therapies used to treat uremic pruritus, gabapentin has, in clinical trials, most consistently been found effective for uremic pruritus.⁶ (Note: Use renal dosing of gabapentin in patients with renal failure.)

Antidepressants. Selective serotonin reuptake inhibitors (SSRIs; eg, fluvoxamine, paroxetine, and sertraline) might cause itch to subside by increasing the serotonin level, which, in turn, works to decrease inflammatory substances that cause itch.⁷ SSRIs have been used to treat patients with psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.⁷

Start conservatively: Use behavior modification and nonpharmacotherapeutic options for pruritus first.

Tricyclic antidepressants (eg, amitriptyline and doxepin) lessen the itch by antagonizing histamine receptors and through anticholinergic mechanisms. Tricyclics are best used in the treatment of psychogenic and nocturnal itch.⁷

Continue to: Mirtazapine...

Mirtazapine, a tetracyclic antidepressant, works in patients with uremic pruritus, psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.¹

Substance P antagonist. Aprepitant, a neurokinin receptor I antagonist, is a newer agent that inhibits binding of the itch mediator substance P to the neurokinin receptor. The drug has been found helpful in patients with drug-induced, paraneoplastic, and brachioradial pruritus.⁷

Opioid-receptor agents. Naltrexone, as a mu opioid-receptor antagonist, has shown promise as a treatment for uremic pruritus and cholestatic pruritus. Nalfurafine, a kappa opioid-receptor agonist, is emerging as a possible therapy for uremic pruritus.⁷

Bile-acid sequestrants. A few small studies have shown that treatment with a bile-acid sequestrant, such as cholestyramine and ursodiol, induces moderate improvement in symptoms in patients with cholestatic pruritus.²¹

CORRESPONDENCE
Matasha Russell, MD, Department of Family and Community Medicine, The University of Texas Health Science Center at Houston, McGovern Medical School, 6431 Fannin Street, JJL 324, Houston, TX 77030; [email protected].

Pruritus, defined as a sensation that induces a desire to scratch¹ and classified as acute or chronic (lasting > 6 weeks),² is one of the most common complaints among primary care patients: Approximately 1% of ambulatory visits in the United States are linked to pruritus.³

Chronic pruritus impairs quality of life; its impact has been compared to that of chronic pain.⁴ Treatment should therefore be instituted promptly. Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition, as we describe here. When persistent pruritus is refractory to treatment, systemic causes should be fully explored.

In this article, we discuss the pathogenesis and management of pruritus without skin eruption in the adult nonpregnant patient. We also present practice recommendations to help you determine whether your patient’s pruritus is indicative of a serious systemic condition.

An incomplete understanding of the pathophysiology of pruritus

The pathophysiology of pruritus is not fully understood. It is generally recognized, however, that pruritus starts in the peripheral nerves located in the dermal–epidermal junction of the skin.⁵ The sensation is then transmitted along unmyelinated slow-conducting C fibers to the dorsal horn of the spinal cord.^5,6 There are 2 types of C fibers that transmit the itch impulse⁶: A histamine-dependent type and a non-­histamine-dependent type, which might explain why pruritus can be refractory to antihistamine treatment.⁶

Once the itch impulse has moved from the spinal cord, it travels along the spinothalamic tract up to the contralateral thalamus.¹ From there, the impulse ascends to the cerebral cortex.¹ In the cortex, the impulse triggers multiple areas of the brain, such as those responsible for sensation, motor function, reward, memory, and emotion.⁷

Although this condition might appear benign, chronic pruritus can be a symptom of a serious condition.

Several chemical mediators have been found to be peripheral and central inducers of pruritus: histamine, endogenous opioids, substance P, and serotonin.² There are indications that certain receptors, such as mu-opioid receptors and kappa-opioid receptors, are key contributors to itch as well.²

A diverse etiology

The International Forum for the Study of Itch (IFSI) has established 6 main categories of causes of pruritus(TABLE 1)²:

• dermatologic
• systemic
• neurologic
• psychogenic
• mixed
• other.

Continue to: In this review...

In this review, we focus on the work-up and management of 3 of those categories: systemic, neurologic, and psychogenic causes of pruritus.

Systemic causes

Research has shown that 14% to 24% of patients who seek the care of a dermatologist for chronic itch without skin lesions have a systemic illness.⁸

Renal disease. Approximately 40% of patients with end-stage renal disease who are on hemodialysis or peritoneal dialysis have uremic pruritus.2 The itch is mostly generalized but can be pronounced on the back. For most patients, the itch is worse at night, causing a major impact on quality of life.⁶

Liver disease. In hepatic disease, there is often impairment in the secretion of bile, which can lead to cholestatic pruritus.² This condition commonly affects the hands and feet first; later, it becomes generalized.² Cholestatic pruritus can be elicited by tight-fitting clothing. Relief is not achieved by scratching.⁹ This type of itch effects 70% of patients with primary biliary cirrhosis and 15% of patients with hepatitis C infection.⁹

Hematologic disorders. Pruritus is a hallmark symptom of polycythemia rubra vera. Almost 50% of patients with this disorder report pruritus that occurs after exposure to water⁹; aquagenic pruritus can precede the formal diagnosis of polycythemia rubra vera by years.² It has been speculated that platelet aggregation in this disorder leads to release of serotonin and histamine, which, in turn, causes itch.⁹

Continue to: Endocrine disorders

Endocrine disorders. Approximately 4% to 11% of patients with thyrotoxicosis have pruritus.¹ It has been suggested that vasodilation, increased skin temperature, and a decreased itch threshold from untreated Graves disease might be inciting factors.

Malignancy. In generalized chronic pruritus without a known cause, strongly consider the likelihood of underlying malignancy^8,10; for 10% of these patients, their chronic pruritus is a paraneoplastic sign. Paraneoplastic pruritus is characterized as an itch that predates clinical onset, or occurs early in the course, of a malignancy.⁹ The condition is most strongly linked to cancers of the liver, gallbladder, biliary tract, hematologic system, and skin.¹¹

Palpate the liver, spleen, lymph nodes, and thyroid for organomegaly, which could indicate a serious systemic condition as the cause of pruritus.

Chronic pruritus affects 30% of patients with Hodgkin lymphoma.⁹ General pruritus can precede this diagnosis by months, even years.¹ In Hodgkin lymphoma patients who are in remission, a return of pruritic symptoms can be a harbinger of recurrence.⁹

Neurologic causes

A recent study found that 8% to 15% of patients referred to a dermatology clinic for chronic pruritus without skin eruption had underlying neurologic pathology.¹² Although the specific mechanisms of neuropathic itch are still poorly understood, it has been theorized that the itch emanates from neuronal damage, which can come from peripheral or central nervous system lesions.⁹

Brachioradial pruritus. There are divergent theories about the etiology of brachioradial pruritus. One hypothesis is that the condition is caused by cervical nerve-root impingement at the level of C5-C8 that leads to nerve damage²; another is that chronic exposure to sunlight causes injury to peripheral cutaneous nerves.² Brachioradial pruritus is localized to the dorsolateral forearm; it can also involve the neck, back, shoulder, upper arm, and chest, unilaterally and bilaterally. This pruritus can be intermittent and become worse upon exposure to sunlight.²

Continue to: Notalgia paresthetica

Notalgia paresthetica. This condition might also cause neuropathic pruritus as a consequence of nerve impingement. The itch of notalgia paresthesia is located on the skin, medial to the scapular border on the upper or mid-back.² It has been postulated that the itch is caused by nerve entrapment of the posterior rami of spinal nerves arising from T2-T6.⁹ However, another theory suggests that the itch is caused by damage to peripheral nerves.⁹ The itch of notalgia paresthetica can wax and wane.²

Poststroke pruritus. Brain lesions, most often caused by stroke, can cause neuropathic itch. One of the best-known syndromes related to poststroke itch is Wallenberg syndrome (ischemia from a lateral medullary infarction), which typically presents with itch, thermalgic hypoesthesia of the face, cerebellar dysfunction, nausea, and vomiting.⁷

Shingles. More than one-half of patients who develop postherpetic neuralgia as a consequence of a herpes zoster infection also develop neuropathic pruritus.⁹ It is thought that postherpetic pruritus shares a comparable pathophysiology with postherpetic neuralgia, in which neurons involved in itch stimuli become damaged.⁷

Diabetes mellitus. Pruritus from diabetes can be classified as systemic or neuropathic. Diabetes is one of the most common causes of small-fiber polyneuropathy, which can cause neuropathic pruritus.¹³

Multiple sclerosis. Central nervous system lesions that affect sensory pathways can lead to neuropathic itch in multiple sclerosis. Patients can have severe episodes of generalized pruritus. It has been hypothesized that the neuropathic itch in multiple sclerosis is induced by activation of artificial synapses in demyelinated areas.²

Continue to: Psychogenic pruritus

Chronic pruritus can be a comorbidity of psychiatric illness. A retrospective study found that pruritus occurs in 32% to 42% of psychiatric inpatients.¹⁴ Depression, anxiety, bipolar disorders, obsessive–compulsive disorders, somatoform disorders, psychosis, and substance abuse all have a strong link to psychogenic excoriation.¹⁵ Psychogenic excoriation, which can cause secondary skin lesions, occurs in psychiatric patients who excessively pick and scratch normal skin because they perceive an itch sensation or have a delusion of infestation.² Affected skin can be marked by scattered crusted lesions (FIGURE) anywhere on the body that the patient can reach—most commonly, the extremities.²

Delusion of infestation. Patients with a delusion of infestation have a strong belief that their body is infected by some kind of insect or microorganism.¹⁶ Before a diagnosis of delusion of infestation can be made, other organic causes must be excluded, including withdrawal from such substances as cocaine, amphetamines, and alcohol.¹⁶ Patients with a delusion of infestation can have, and maintain, a symptomatic response with continuing use of an atypical antipsychotic agent, including risperidone and olanzapine.¹⁷

Evaluation and diagnostic work-up

A thorough medical history, review of systems, medication review, social history, and family history are important when evaluating a patient with chronic pruritus.¹⁸ These items can be valuable in formulating a differential diagnosis, even before a physical examination.

Physical examination. The physical exam should include detailed inspection of the entire skin and hair¹⁸; such a comprehensive physical exam can determine whether the source of the itch is cutaneous.⁷ This, in turn, can help further narrow the differential diagnosis. It is crucial that the physical exam include palpation of the liver, spleen, lymph nodes, and thyroid for organomegaly,⁸ which could indicate a serious systemic condition, such as lymphoma.

The ice-pack sign—in which an ice pack applied to the pruritic area provides immediate relief—is considered pathognomonic for brachioradial pruritus.

The ice-pack sign—in which an ice pack is applied to the pruritic area, the patient experiences immediate relief of pruritus, and the itch returns soon after the ice pack is removed—is considered pathognomonic for brachioradial pruritus.¹⁹

Continue to: Chronic pruritus with abnormal findings...

Chronic pruritus with abnormal findings on the physical exam should prompt an initial work-up.¹⁸ Also consider an initial work-up for a patient with chronic pruritus whose symptom has not been relieved with conservative treatment.¹⁸

Laboratory testing. The initial laboratory work-up could include any of the following evaluations: complete blood count, measurement of thyroid-stimulating hormone, comprehensive metabolic panel (liver function, renal function, and the serum glucose level) and the erythrocyte sedimentation rate (TABLE 2).¹⁸ If warranted by the evaluation and physical exam, blood work can also include serologic studies for human immunodeficiency virus infection and ­hepatitis.¹⁷

Imaging. Chest radiography should be performed if there is suspicion of malignancy, such as lymphoma.⁷ Although brachioradial pruritus and notalgia paresthetica have been postulated to be caused by impingement of spinal nerves, obtaining spinal imaging, such as magnetic resonance imaging, as part of the initial work-up is not recommended; because spinal images might not show evidence of spinal disease, obtaining spinal imaging is not a requirement before treating brachioradial pruritus and notalgia paresthetica. Do consider spinal imaging, however, for patients in whom brachioradial pruritus or notalgia paresthetica is suspected and conservative treatment has not produced a response.

Treatment: Nondrug approaches, topicals, systemic agents

Start conservatively. Treatment of pruritus should begin with behavior modification and nonpharmacotherapeutic options (TABLE 3⁸). Educate the patient that scratching might cause secondary skin lesions; empowering them with that knowledge is sometimes enough to help break the scratching cycling—especially if the patient combines behavior modification with proper skin hydration with an emollient. To prevent secondary skin lesions through involuntary scratching, consider recommending that lesions be covered with an occlusive dressing or protective clothing.¹³

Stress has been shown to make chronic itch worse; therefore, stress-reduction activities, such as exercise, meditation, and yoga, might be helpful.20 For patients in whom pruritus has a psychological component, referral to a psychiatrist or psychologist might be therapeutic.

Continue to: When a patient complains...

When a patient complains of severe pruritus at first presentation, consider pharmacotherapy in conjunction with nonpharmacotherapeutic options. Several of the more effective topical therapies for pruritus^a are listed in TABLE 4.²⁰ Well-known systemic agents for this purpose are reviewed below and listed in TABLE 5.⁷

Systemic treatment

Antihistamines. A staple in the treatment of pruritus for many years, antihistamines are not effective for all causes; however, they are effective in treating paraneoplastic pruritus.²⁰ First-generation antihistamines, with their sedating effect, can be useful for patients who experience generalized pruritus at night.²⁰

Anticonvulsants. Gabapentin and pregabalin are analogs of the neurotransmitter gamma-aminobutyric acid.²⁰ This drug class is helpful in neuropathic pruritus specifically caused by impingements, such as brachioradial pruritus and notalgia paresthetica.²⁰ In addition, of all systemic therapies used to treat uremic pruritus, gabapentin has, in clinical trials, most consistently been found effective for uremic pruritus.⁶ (Note: Use renal dosing of gabapentin in patients with renal failure.)

Antidepressants. Selective serotonin reuptake inhibitors (SSRIs; eg, fluvoxamine, paroxetine, and sertraline) might cause itch to subside by increasing the serotonin level, which, in turn, works to decrease inflammatory substances that cause itch.⁷ SSRIs have been used to treat patients with psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.⁷

Start conservatively: Use behavior modification and nonpharmacotherapeutic options for pruritus first.

Tricyclic antidepressants (eg, amitriptyline and doxepin) lessen the itch by antagonizing histamine receptors and through anticholinergic mechanisms. Tricyclics are best used in the treatment of psychogenic and nocturnal itch.⁷

Continue to: Mirtazapine...

Mirtazapine, a tetracyclic antidepressant, works in patients with uremic pruritus, psychogenic pruritus, cholestatic pruritus, and paraneoplastic pruritus.¹

Substance P antagonist. Aprepitant, a neurokinin receptor I antagonist, is a newer agent that inhibits binding of the itch mediator substance P to the neurokinin receptor. The drug has been found helpful in patients with drug-induced, paraneoplastic, and brachioradial pruritus.⁷

Opioid-receptor agents. Naltrexone, as a mu opioid-receptor antagonist, has shown promise as a treatment for uremic pruritus and cholestatic pruritus. Nalfurafine, a kappa opioid-receptor agonist, is emerging as a possible therapy for uremic pruritus.⁷

Bile-acid sequestrants. A few small studies have shown that treatment with a bile-acid sequestrant, such as cholestyramine and ursodiol, induces moderate improvement in symptoms in patients with cholestatic pruritus.²¹

CORRESPONDENCE
Matasha Russell, MD, Department of Family and Community Medicine, The University of Texas Health Science Center at Houston, McGovern Medical School, 6431 Fannin Street, JJL 324, Houston, TX 77030; [email protected].

Low back pain in not uncommon in children and adolescents.^1-3 Although the prevalence of low back pain in children < 7 years is low, it increases with age, with studies reporting lifetime prevalence at age 12 years between 16% and 18% and rates as high as 66% by 16 years of age.^4,5 Although children and adolescents usually have pain that is transient and benign without a defined cause, structural causes of low back pain should be considered in school-aged children with pain that persists for > 3 to 6 weeks. ⁴ The most common structural causes of adolescent low back pain are reviewed here.

Etiology: A mixed bag

Back pain in school-aged children is most commonly due to muscular strain, overuse, or poor posture. The pain is often transient in nature and responds to rest and postural education.^4,6 A herniated disc is an uncommon finding in younger school-aged children, but incidence increases slightly among older adolescents, particularly those who are active in collision sports and/or weight-lifting.^7,8 Pain caused by a herniated disc often radiates along the distribution of the sciatic nerve and worsens during lumbar flexion.

Spondylolysis and spondylolisthesis are important causes of back pain in children. Spondylolysis is defined as a defect or abnormality of the pars interarticularis and surrounding lamina and pedicle. Spondylolisthesis, which is less common, is defined as the translation or “slippage” of one vertebral segment in relation to the next caudal segment. These conditions commonly occur as a result of repetitive stress.

In a prospective study of adolescents < 19 years with low back pain for > 2 weeks, the prevalence of spondylolysis was 39.7%.⁹ Adolescent athletes with symptomatic low back pain are more likely to have spondylolysis than nonathletes (32% vs 2%, respectively).^2,10 Pain is often made worse by extension of the spine. Spondylolysis and spondylolisthesis can be congenital or acquired, and both can be asymptomatic. Children and teens who are athletes are at higher risk for symptomatic spondylolysis and spondylolisthesis.^10-12 This is especially true for those involved in gymnastics, dance, football, and/or volleyball, where a repetitive load is placed onto an extended spine.

Idiopathic scoliosis is an abnormal lateral curvature of the spine that usually develops during adolescence and worsens with growth. Historically, painful scoliosis was considered rare, but more recently researchers determined that children with scoliosis have a higher rate of pain compared to their peers.^13,14 School-aged children with scoliosis were found to be at 2 times the risk of low back pain compared to those without scoliosis.¹³ It is important to identify scoliosis in adolescents so that progression can be monitored.

Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side.

Screening for scoliosis in primary care is somewhat controversial. The US Preventive Services Task Force (USPSTF) finds insufficient evidence for screening asymptomatic adolescents for scoliosis.¹⁵ This recommendation is based on the fact that there is little evidence on the effect of screening on long-term outcomes. Screening may also lead to unnecessary radiation. Conversely, a position statement released by the Scoliosis Research Society, the Pediatric Orthopedic Society of North America, the American Association of Orthopedic Surgeons, and the American Academy of Pediatrics recommends scoliosis screening during routine pediatric office visits.¹⁶ Screening for girls is recommended at ages 10 and 12 years, and for boys, once between ages 13 and 14 years. The statement highlights evidence showing that focused screening by appropriate personnel has value in detecting a clinically significant curve (> 20°).

Scheuermann disease is a rare cause of back pain in children that usually develops during adolescence and results in increasing thoracic kyphosis. An autosomal dominant mutation plays a role in this disease of the growth cartilage endplate; repetitive strain on the growth cartilage is also a contributing factor.^17,18 An atypical variant manifests with kyphosis in the thoracolumbar region.¹⁷

Continue to: Other causes of low back pain

Other causes of low back pain—including inflammatory arthritis, infection (eg, discitis), and tumor—are rare in children but must always be considered, especially in the setting of persistent symptoms.^4,19-21 More on the features of these conditions is listed in TABLE 1.^{1-7,13-15,17-30}

History: Focus on onset, timing, and duration of symptoms

As with adults, obtaining a history that includes the onset, timing, and duration of symptoms is key in the evaluation of low back pain in children, as is obtaining a history of the patient’s activities; sports that repetitively load the lumbar spine in an extended position increase the risk of injury.¹⁰

Specific risk factors for low back pain in children and adolescents are poorly understood.^4,9,31 Pain can be associated with trauma, or it can have a more progressive or insidious onset. Generally, pain that is present for up to 6 weeks and is intermittent or improving has a self-limited course. Pain that persists beyond 3 to 6 weeks or is worsening is more likely to have an anatomical cause that needs further evaluation.^2,3,10,21

Identifying exacerbating and alleviating factors can provide useful information. Pain that is worse with lumbar flexion is more likely to come from muscular strain or disc pathology. Pain with extension is more likely due to a structural cause such as spondylolysis/spondylolisthesis, scoliosis, or Scheuermann disease.^{2,4,10,17,18,21} See TABLE 2 for red flag symptoms that indicate the need for imaging and further work-up.

The physical exam: Visualize, assess range of motion, and reproduce pain

The physical examination of any patient with low back pain should include direct visualization and inspection of the back, spine, and pelvis; palpation of the spine and paraspinal regions; assessment of lumbar range of motion and of the lumbar nerve roots, including tests of sensation, strength, and deep tendon reflexes; and an evaluation of the patient’s posture, which can provide clues to underlying causes of pain.

Continue to: Increased thoracic kyphosis...

Increased thoracic kyphosis that is not reversible is concerning for Scheuermann disease.^9,17,18 A significant elevation in one shoulder or side of the pelvis can be indicative of scoliosis. Increased lumbar lordosis may predispose a patient to spondylolysis.

In patients with spondylolysis, lumbar extension will usually reproduce pain, which is often unilateral. Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side. The sensitivity of the Stork test for unilateral spondylolysis is approximately 50%.³² (For more information on the Stork test, see www.physio-pedia.com/Stork_test.)

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain. Lumbar flexion with concomitant radicular pain is associated with disc pathology.⁸ Pain with a straight-leg raise is also associated with disk pathology, especially if raising the contralateral leg increases pain.⁸

Using a scoliometer. Evaluate the flexed spine for the presence of asymmetry, which can indicate scoliosis.³³ If asymmetry is present, use a scoliometer to determine the degree of asymmetry. Zero to 5° is considered clinically insignificant; monitor and reevaluate these patients at subsequent visits.^34,35 Ten degrees or more of asymmetry with a scoliometer should prompt you to order radiographs.^35,36 A smartphone-based scoliometer for iPhones was evaluated in 1 study and was shown to have reasonable reliability and validity for clinical use.³⁷

Deformity of the lower extremities. Because low back pain may be caused by biomechanical or structural deformity of the lower extremities, examine the flexibility of the hip flexors, gluteal musculature, hamstrings, and the iliotibial band.³⁸ In addition, evaluate for leg-length discrepancy and lower-extremity malalignment, such as femoral anteversion, tibial torsion, or pes planus.

Continue to: Imaging

Imaging: Know when it’s needed

Although imaging of the lumbar spine is often unnecessary in the presence of acute low back pain in children, always consider imaging in the setting of bony tenderness, pain that wakes a patient from sleep, and in the setting of other red flag symptoms (see TABLE 2). Low back pain in children that is reproducible with lumbar extension is concerning for spondylolysis or spondylolisthesis. If the pain with extension persists beyond 3 to 6 weeks, order imaging starting with radiographs.^2,39

Traditionally, 4 views of the spine—­anteroposterior (AP), lateral, and oblique (one right and one left)—were obtained, but recent evidence indicates that 2 views (AP and lateral) have similar sensitivity and specificity to 4 views with significantly reduced radiation exposure.^2,39 Because the sensitivity of plain films is relatively low, consider more advanced imaging if spondylolysis or spondylolisthesis is strongly suspected. Recent studies indicate that magnetic resonance imaging (MRI) may be as effective as computed tomography (CT) or bone scan and has the advantage of lower radiation (FIGURE 1).^2,22

Similarly, order radiographs if there is > 10° of asymmetry noted on physical exam using a scoliometer.^15,23 Calculate the Cobb angle to determine the severity of scoliosis. Refer patients with angles ≥ 20° to a pediatric orthopedist for monitoring of progression and consideration of bracing (FIGURE 2).^23,34 For patients with curvatures between 10° and 19°, repeat imaging every 6 to 12 months. Because scoliosis is a risk factor for spondylolysis, evaluate radiographs in the setting of painful scoliosis for the presence of a spondylolysis.^34,35

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain.

If excessive kyphosis is noted on exam, order radiographs to evaluate for Scheuermann disease. Classic imaging findings include Schmorl nodes, vertebral endplate changes, and anterior wedging (FIGURE 3).^17,18

In the absence of the above concerns, defer imaging of the lumbar spine until after adequate rest and rehabilitation have been attempted.

Continue to: Treatment typically involves restor physical therapy

Most cases of low back pain in children and adolescents are benign and self-limited. Many children with low back pain can be treated with relative rest from the offending activity. For children with more persistent pain, physical therapy (PT) is often indicated. Similar to that for adults, there is little evidence for specific PT programs to help children with low back pain. Rehabilitation should be individualized based on the condition being treated.

Medications. There have been no high-quality studies on the benefit of medications to treat low back pain in children. Studies have shown nonsteroidal anti-inflammatory drugs (NSAIDs) have value in adults, and they are likely safe for use in children,⁴⁰ but the risk of opiate abuse is significantly increased in adolescents who have been prescribed opiate pain medication prior to 12th grade.⁴¹

Lumbar disc herniation. Although still relatively rare, lumbar disc herniation is more common in older children and adolescents than in younger children and is treated similarly to that in adults.⁸ Range-of-motion exercise to restore lumbar motion is often first-line treatment. Research has shown that exercises that strengthen the abdominal or “core” musculature help prevent the return of low back pain.^24,25

In the case of spondylolysis or spondylolisthesis, rest from activity is generally required for a minimum of 4 to 6 weeks. Rehabilitation in the form of range of motion, especially into the lumbar extension, and spinal stabilization exercises are effective for both reducing pain and restoring range-­of-motion and strength.⁴² Have patients avoid heavy backpacks, which can reproduce pain. Children often benefit from leaving a second set of schoolbooks at home. For most patients with spondylolysis, conservative treatment with rehabilitation is equal to or better than surgical intervention in returning the patient to his/her pre-injury activity level.^26,43,44 When returning athletes to their sport, aggressive PT, defined as rest for < 10 weeks prior to initiating PT, is superior to delaying PT beyond 10 weeks of rest.²⁷

Idiopathic scoliosis. Much of the literature on the treatment of scoliosis is focused on limiting progression of the scoliotic curvature. Researchers thought that more severe curves were associated with more severe pain, but a recent systematic review showed that back pain can occur in patients with even small curvatures.²⁸ Treatment for patients with smaller degrees of curvature is similar to that for mechanical low back pain. PT may have a role in the treatment of scoliosis, but there is little evidence in the literature of its effectiveness.

Continue to: A Cochrane review showed...

Always consider imaging in the setting of bony tenderness, pain that wakes the patient from sleep, and when there is > 10° of asymmetry on physical exam using a scoliometer.

A Cochrane review showed that PT and exercise-based treatments had no effect on back pain or disability in patients with scoliosis.²⁹ And outpatient PT alone, in the absence of bracing, does not arrest progression of the scoliotic curvature.³⁵ One trial did demonstrate that an intensive inpatient treatment program of 4 to 6 weeks for patients with curvature of at least 40° reduced progression of curvature compared to an untreated control group at 1 year.³⁴ The outcomes of functional mobility and pain were not measured. Follow-up data on curve progression beyond 1 year are not available. Unfortunately, intensive inpatient treatment is not readily available or cost-effective for most patients with scoliosis.

Scheuermann disease. The mainstay of treatment for mild Scheuermann disease is advising the patient to avoid repetitive loading of the spine. Patients should avoid sports such as competitive weight-lifting, gymnastics, and football. Lower impact athletics are encouraged. Refer patients with pain to PT to address posture and core stabilization. Patients with severe kyphosis may require surgery.^17,18

Bracing: Rarely helpful for low back pain

The use of lumbar braces or corsets is rarely helpful for low back pain in children. Bracing in the setting of spondylolysis is controversial.One study indicated that bracing in combination with activity restriction and lumbar extension exercise is superior to activity restriction and lumbar flexion exercises alone.⁴³ But a meta-analysis did not demonstrate a significant difference in recovery when bracing was added.⁴⁴ Bracing may help to reduce pain initially in patients with spondylolysis who have pain at rest. Bracing is not recommended for patients with pain that abates with activity modification.

Scoliosis and Scheuermann kyphosis. Treatment of adolescent idiopathic scoliosis usually consists of observation and periodic reevaluation. Bracing is a mainstay of the nonsurgical management of scoliosis and is appropriate for curves of 20° to 40°; studies have reported successful control of curve progression in > 70% of patients.³⁶ According to 1 study, the number of cases of scoliosis needed to treat with bracing to prevent 1 surgery is 3.³⁰ Surgery is often indicated for patients with curvatures > 40°, although this is also debated.³³

Bracing is used rarely for Scheuermann kyphosis but may be helpful in more severe or painful cases.¹⁷

CORRESPONDENCE
Shawn F. Phillips, MD, MSPT, 500 University Drive H154, Hershey, PA, 17033; [email protected].

Low back pain in not uncommon in children and adolescents.^1-3 Although the prevalence of low back pain in children < 7 years is low, it increases with age, with studies reporting lifetime prevalence at age 12 years between 16% and 18% and rates as high as 66% by 16 years of age.^4,5 Although children and adolescents usually have pain that is transient and benign without a defined cause, structural causes of low back pain should be considered in school-aged children with pain that persists for > 3 to 6 weeks. ⁴ The most common structural causes of adolescent low back pain are reviewed here.

Etiology: A mixed bag

Back pain in school-aged children is most commonly due to muscular strain, overuse, or poor posture. The pain is often transient in nature and responds to rest and postural education.^4,6 A herniated disc is an uncommon finding in younger school-aged children, but incidence increases slightly among older adolescents, particularly those who are active in collision sports and/or weight-lifting.^7,8 Pain caused by a herniated disc often radiates along the distribution of the sciatic nerve and worsens during lumbar flexion.

Spondylolysis and spondylolisthesis are important causes of back pain in children. Spondylolysis is defined as a defect or abnormality of the pars interarticularis and surrounding lamina and pedicle. Spondylolisthesis, which is less common, is defined as the translation or “slippage” of one vertebral segment in relation to the next caudal segment. These conditions commonly occur as a result of repetitive stress.

In a prospective study of adolescents < 19 years with low back pain for > 2 weeks, the prevalence of spondylolysis was 39.7%.⁹ Adolescent athletes with symptomatic low back pain are more likely to have spondylolysis than nonathletes (32% vs 2%, respectively).^2,10 Pain is often made worse by extension of the spine. Spondylolysis and spondylolisthesis can be congenital or acquired, and both can be asymptomatic. Children and teens who are athletes are at higher risk for symptomatic spondylolysis and spondylolisthesis.^10-12 This is especially true for those involved in gymnastics, dance, football, and/or volleyball, where a repetitive load is placed onto an extended spine.

Idiopathic scoliosis is an abnormal lateral curvature of the spine that usually develops during adolescence and worsens with growth. Historically, painful scoliosis was considered rare, but more recently researchers determined that children with scoliosis have a higher rate of pain compared to their peers.^13,14 School-aged children with scoliosis were found to be at 2 times the risk of low back pain compared to those without scoliosis.¹³ It is important to identify scoliosis in adolescents so that progression can be monitored.

Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side.

Screening for scoliosis in primary care is somewhat controversial. The US Preventive Services Task Force (USPSTF) finds insufficient evidence for screening asymptomatic adolescents for scoliosis.¹⁵ This recommendation is based on the fact that there is little evidence on the effect of screening on long-term outcomes. Screening may also lead to unnecessary radiation. Conversely, a position statement released by the Scoliosis Research Society, the Pediatric Orthopedic Society of North America, the American Association of Orthopedic Surgeons, and the American Academy of Pediatrics recommends scoliosis screening during routine pediatric office visits.¹⁶ Screening for girls is recommended at ages 10 and 12 years, and for boys, once between ages 13 and 14 years. The statement highlights evidence showing that focused screening by appropriate personnel has value in detecting a clinically significant curve (> 20°).

Scheuermann disease is a rare cause of back pain in children that usually develops during adolescence and results in increasing thoracic kyphosis. An autosomal dominant mutation plays a role in this disease of the growth cartilage endplate; repetitive strain on the growth cartilage is also a contributing factor.^17,18 An atypical variant manifests with kyphosis in the thoracolumbar region.¹⁷

Continue to: Other causes of low back pain

Other causes of low back pain—including inflammatory arthritis, infection (eg, discitis), and tumor—are rare in children but must always be considered, especially in the setting of persistent symptoms.^4,19-21 More on the features of these conditions is listed in TABLE 1.^{1-7,13-15,17-30}

History: Focus on onset, timing, and duration of symptoms

As with adults, obtaining a history that includes the onset, timing, and duration of symptoms is key in the evaluation of low back pain in children, as is obtaining a history of the patient’s activities; sports that repetitively load the lumbar spine in an extended position increase the risk of injury.¹⁰

Specific risk factors for low back pain in children and adolescents are poorly understood.^4,9,31 Pain can be associated with trauma, or it can have a more progressive or insidious onset. Generally, pain that is present for up to 6 weeks and is intermittent or improving has a self-limited course. Pain that persists beyond 3 to 6 weeks or is worsening is more likely to have an anatomical cause that needs further evaluation.^2,3,10,21

Identifying exacerbating and alleviating factors can provide useful information. Pain that is worse with lumbar flexion is more likely to come from muscular strain or disc pathology. Pain with extension is more likely due to a structural cause such as spondylolysis/spondylolisthesis, scoliosis, or Scheuermann disease.^{2,4,10,17,18,21} See TABLE 2 for red flag symptoms that indicate the need for imaging and further work-up.

The physical exam: Visualize, assess range of motion, and reproduce pain

The physical examination of any patient with low back pain should include direct visualization and inspection of the back, spine, and pelvis; palpation of the spine and paraspinal regions; assessment of lumbar range of motion and of the lumbar nerve roots, including tests of sensation, strength, and deep tendon reflexes; and an evaluation of the patient’s posture, which can provide clues to underlying causes of pain.

Continue to: Increased thoracic kyphosis...

Increased thoracic kyphosis that is not reversible is concerning for Scheuermann disease.^9,17,18 A significant elevation in one shoulder or side of the pelvis can be indicative of scoliosis. Increased lumbar lordosis may predispose a patient to spondylolysis.

In patients with spondylolysis, lumbar extension will usually reproduce pain, which is often unilateral. Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side. The sensitivity of the Stork test for unilateral spondylolysis is approximately 50%.³² (For more information on the Stork test, see www.physio-pedia.com/Stork_test.)

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain. Lumbar flexion with concomitant radicular pain is associated with disc pathology.⁸ Pain with a straight-leg raise is also associated with disk pathology, especially if raising the contralateral leg increases pain.⁸

Using a scoliometer. Evaluate the flexed spine for the presence of asymmetry, which can indicate scoliosis.³³ If asymmetry is present, use a scoliometer to determine the degree of asymmetry. Zero to 5° is considered clinically insignificant; monitor and reevaluate these patients at subsequent visits.^34,35 Ten degrees or more of asymmetry with a scoliometer should prompt you to order radiographs.^35,36 A smartphone-based scoliometer for iPhones was evaluated in 1 study and was shown to have reasonable reliability and validity for clinical use.³⁷

Deformity of the lower extremities. Because low back pain may be caused by biomechanical or structural deformity of the lower extremities, examine the flexibility of the hip flexors, gluteal musculature, hamstrings, and the iliotibial band.³⁸ In addition, evaluate for leg-length discrepancy and lower-extremity malalignment, such as femoral anteversion, tibial torsion, or pes planus.

Continue to: Imaging

Imaging: Know when it’s needed

Although imaging of the lumbar spine is often unnecessary in the presence of acute low back pain in children, always consider imaging in the setting of bony tenderness, pain that wakes a patient from sleep, and in the setting of other red flag symptoms (see TABLE 2). Low back pain in children that is reproducible with lumbar extension is concerning for spondylolysis or spondylolisthesis. If the pain with extension persists beyond 3 to 6 weeks, order imaging starting with radiographs.^2,39

Traditionally, 4 views of the spine—­anteroposterior (AP), lateral, and oblique (one right and one left)—were obtained, but recent evidence indicates that 2 views (AP and lateral) have similar sensitivity and specificity to 4 views with significantly reduced radiation exposure.^2,39 Because the sensitivity of plain films is relatively low, consider more advanced imaging if spondylolysis or spondylolisthesis is strongly suspected. Recent studies indicate that magnetic resonance imaging (MRI) may be as effective as computed tomography (CT) or bone scan and has the advantage of lower radiation (FIGURE 1).^2,22

Similarly, order radiographs if there is > 10° of asymmetry noted on physical exam using a scoliometer.^15,23 Calculate the Cobb angle to determine the severity of scoliosis. Refer patients with angles ≥ 20° to a pediatric orthopedist for monitoring of progression and consideration of bracing (FIGURE 2).^23,34 For patients with curvatures between 10° and 19°, repeat imaging every 6 to 12 months. Because scoliosis is a risk factor for spondylolysis, evaluate radiographs in the setting of painful scoliosis for the presence of a spondylolysis.^34,35

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain.

If excessive kyphosis is noted on exam, order radiographs to evaluate for Scheuermann disease. Classic imaging findings include Schmorl nodes, vertebral endplate changes, and anterior wedging (FIGURE 3).^17,18

In the absence of the above concerns, defer imaging of the lumbar spine until after adequate rest and rehabilitation have been attempted.

Continue to: Treatment typically involves restor physical therapy

Most cases of low back pain in children and adolescents are benign and self-limited. Many children with low back pain can be treated with relative rest from the offending activity. For children with more persistent pain, physical therapy (PT) is often indicated. Similar to that for adults, there is little evidence for specific PT programs to help children with low back pain. Rehabilitation should be individualized based on the condition being treated.

Medications. There have been no high-quality studies on the benefit of medications to treat low back pain in children. Studies have shown nonsteroidal anti-inflammatory drugs (NSAIDs) have value in adults, and they are likely safe for use in children,⁴⁰ but the risk of opiate abuse is significantly increased in adolescents who have been prescribed opiate pain medication prior to 12th grade.⁴¹

Lumbar disc herniation. Although still relatively rare, lumbar disc herniation is more common in older children and adolescents than in younger children and is treated similarly to that in adults.⁸ Range-of-motion exercise to restore lumbar motion is often first-line treatment. Research has shown that exercises that strengthen the abdominal or “core” musculature help prevent the return of low back pain.^24,25

In the case of spondylolysis or spondylolisthesis, rest from activity is generally required for a minimum of 4 to 6 weeks. Rehabilitation in the form of range of motion, especially into the lumbar extension, and spinal stabilization exercises are effective for both reducing pain and restoring range-­of-motion and strength.⁴² Have patients avoid heavy backpacks, which can reproduce pain. Children often benefit from leaving a second set of schoolbooks at home. For most patients with spondylolysis, conservative treatment with rehabilitation is equal to or better than surgical intervention in returning the patient to his/her pre-injury activity level.^26,43,44 When returning athletes to their sport, aggressive PT, defined as rest for < 10 weeks prior to initiating PT, is superior to delaying PT beyond 10 weeks of rest.²⁷

Idiopathic scoliosis. Much of the literature on the treatment of scoliosis is focused on limiting progression of the scoliotic curvature. Researchers thought that more severe curves were associated with more severe pain, but a recent systematic review showed that back pain can occur in patients with even small curvatures.²⁸ Treatment for patients with smaller degrees of curvature is similar to that for mechanical low back pain. PT may have a role in the treatment of scoliosis, but there is little evidence in the literature of its effectiveness.

Continue to: A Cochrane review showed...

Always consider imaging in the setting of bony tenderness, pain that wakes the patient from sleep, and when there is > 10° of asymmetry on physical exam using a scoliometer.

A Cochrane review showed that PT and exercise-based treatments had no effect on back pain or disability in patients with scoliosis.²⁹ And outpatient PT alone, in the absence of bracing, does not arrest progression of the scoliotic curvature.³⁵ One trial did demonstrate that an intensive inpatient treatment program of 4 to 6 weeks for patients with curvature of at least 40° reduced progression of curvature compared to an untreated control group at 1 year.³⁴ The outcomes of functional mobility and pain were not measured. Follow-up data on curve progression beyond 1 year are not available. Unfortunately, intensive inpatient treatment is not readily available or cost-effective for most patients with scoliosis.

Scheuermann disease. The mainstay of treatment for mild Scheuermann disease is advising the patient to avoid repetitive loading of the spine. Patients should avoid sports such as competitive weight-lifting, gymnastics, and football. Lower impact athletics are encouraged. Refer patients with pain to PT to address posture and core stabilization. Patients with severe kyphosis may require surgery.^17,18

Bracing: Rarely helpful for low back pain

The use of lumbar braces or corsets is rarely helpful for low back pain in children. Bracing in the setting of spondylolysis is controversial.One study indicated that bracing in combination with activity restriction and lumbar extension exercise is superior to activity restriction and lumbar flexion exercises alone.⁴³ But a meta-analysis did not demonstrate a significant difference in recovery when bracing was added.⁴⁴ Bracing may help to reduce pain initially in patients with spondylolysis who have pain at rest. Bracing is not recommended for patients with pain that abates with activity modification.

Scoliosis and Scheuermann kyphosis. Treatment of adolescent idiopathic scoliosis usually consists of observation and periodic reevaluation. Bracing is a mainstay of the nonsurgical management of scoliosis and is appropriate for curves of 20° to 40°; studies have reported successful control of curve progression in > 70% of patients.³⁶ According to 1 study, the number of cases of scoliosis needed to treat with bracing to prevent 1 surgery is 3.³⁰ Surgery is often indicated for patients with curvatures > 40°, although this is also debated.³³

Bracing is used rarely for Scheuermann kyphosis but may be helpful in more severe or painful cases.¹⁷

CORRESPONDENCE
Shawn F. Phillips, MD, MSPT, 500 University Drive H154, Hershey, PA, 17033; [email protected].

Low back pain in not uncommon in children and adolescents.^1-3 Although the prevalence of low back pain in children < 7 years is low, it increases with age, with studies reporting lifetime prevalence at age 12 years between 16% and 18% and rates as high as 66% by 16 years of age.^4,5 Although children and adolescents usually have pain that is transient and benign without a defined cause, structural causes of low back pain should be considered in school-aged children with pain that persists for > 3 to 6 weeks. ⁴ The most common structural causes of adolescent low back pain are reviewed here.

Etiology: A mixed bag

Back pain in school-aged children is most commonly due to muscular strain, overuse, or poor posture. The pain is often transient in nature and responds to rest and postural education.^4,6 A herniated disc is an uncommon finding in younger school-aged children, but incidence increases slightly among older adolescents, particularly those who are active in collision sports and/or weight-lifting.^7,8 Pain caused by a herniated disc often radiates along the distribution of the sciatic nerve and worsens during lumbar flexion.

Spondylolysis and spondylolisthesis are important causes of back pain in children. Spondylolysis is defined as a defect or abnormality of the pars interarticularis and surrounding lamina and pedicle. Spondylolisthesis, which is less common, is defined as the translation or “slippage” of one vertebral segment in relation to the next caudal segment. These conditions commonly occur as a result of repetitive stress.

In a prospective study of adolescents < 19 years with low back pain for > 2 weeks, the prevalence of spondylolysis was 39.7%.⁹ Adolescent athletes with symptomatic low back pain are more likely to have spondylolysis than nonathletes (32% vs 2%, respectively).^2,10 Pain is often made worse by extension of the spine. Spondylolysis and spondylolisthesis can be congenital or acquired, and both can be asymptomatic. Children and teens who are athletes are at higher risk for symptomatic spondylolysis and spondylolisthesis.^10-12 This is especially true for those involved in gymnastics, dance, football, and/or volleyball, where a repetitive load is placed onto an extended spine.

Idiopathic scoliosis is an abnormal lateral curvature of the spine that usually develops during adolescence and worsens with growth. Historically, painful scoliosis was considered rare, but more recently researchers determined that children with scoliosis have a higher rate of pain compared to their peers.^13,14 School-aged children with scoliosis were found to be at 2 times the risk of low back pain compared to those without scoliosis.¹³ It is important to identify scoliosis in adolescents so that progression can be monitored.

Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side.

Screening for scoliosis in primary care is somewhat controversial. The US Preventive Services Task Force (USPSTF) finds insufficient evidence for screening asymptomatic adolescents for scoliosis.¹⁵ This recommendation is based on the fact that there is little evidence on the effect of screening on long-term outcomes. Screening may also lead to unnecessary radiation. Conversely, a position statement released by the Scoliosis Research Society, the Pediatric Orthopedic Society of North America, the American Association of Orthopedic Surgeons, and the American Academy of Pediatrics recommends scoliosis screening during routine pediatric office visits.¹⁶ Screening for girls is recommended at ages 10 and 12 years, and for boys, once between ages 13 and 14 years. The statement highlights evidence showing that focused screening by appropriate personnel has value in detecting a clinically significant curve (> 20°).

Scheuermann disease is a rare cause of back pain in children that usually develops during adolescence and results in increasing thoracic kyphosis. An autosomal dominant mutation plays a role in this disease of the growth cartilage endplate; repetitive strain on the growth cartilage is also a contributing factor.^17,18 An atypical variant manifests with kyphosis in the thoracolumbar region.¹⁷

Continue to: Other causes of low back pain

Other causes of low back pain—including inflammatory arthritis, infection (eg, discitis), and tumor—are rare in children but must always be considered, especially in the setting of persistent symptoms.^4,19-21 More on the features of these conditions is listed in TABLE 1.^{1-7,13-15,17-30}

History: Focus on onset, timing, and duration of symptoms

As with adults, obtaining a history that includes the onset, timing, and duration of symptoms is key in the evaluation of low back pain in children, as is obtaining a history of the patient’s activities; sports that repetitively load the lumbar spine in an extended position increase the risk of injury.¹⁰

Specific risk factors for low back pain in children and adolescents are poorly understood.^4,9,31 Pain can be associated with trauma, or it can have a more progressive or insidious onset. Generally, pain that is present for up to 6 weeks and is intermittent or improving has a self-limited course. Pain that persists beyond 3 to 6 weeks or is worsening is more likely to have an anatomical cause that needs further evaluation.^2,3,10,21

Identifying exacerbating and alleviating factors can provide useful information. Pain that is worse with lumbar flexion is more likely to come from muscular strain or disc pathology. Pain with extension is more likely due to a structural cause such as spondylolysis/spondylolisthesis, scoliosis, or Scheuermann disease.^{2,4,10,17,18,21} See TABLE 2 for red flag symptoms that indicate the need for imaging and further work-up.

The physical exam: Visualize, assess range of motion, and reproduce pain

The physical examination of any patient with low back pain should include direct visualization and inspection of the back, spine, and pelvis; palpation of the spine and paraspinal regions; assessment of lumbar range of motion and of the lumbar nerve roots, including tests of sensation, strength, and deep tendon reflexes; and an evaluation of the patient’s posture, which can provide clues to underlying causes of pain.

Continue to: Increased thoracic kyphosis...

Increased thoracic kyphosis that is not reversible is concerning for Scheuermann disease.^9,17,18 A significant elevation in one shoulder or side of the pelvis can be indicative of scoliosis. Increased lumbar lordosis may predispose a patient to spondylolysis.

In patients with spondylolysis, lumbar extension will usually reproduce pain, which is often unilateral. Hyperextension in a single-leg stance, commonly known as the Stork test, is positive for unilateral spondylolysis when it reproduces pain on the ipsilateral side. The sensitivity of the Stork test for unilateral spondylolysis is approximately 50%.³² (For more information on the Stork test, see www.physio-pedia.com/Stork_test.)

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain. Lumbar flexion with concomitant radicular pain is associated with disc pathology.⁸ Pain with a straight-leg raise is also associated with disk pathology, especially if raising the contralateral leg increases pain.⁸

Using a scoliometer. Evaluate the flexed spine for the presence of asymmetry, which can indicate scoliosis.³³ If asymmetry is present, use a scoliometer to determine the degree of asymmetry. Zero to 5° is considered clinically insignificant; monitor and reevaluate these patients at subsequent visits.^34,35 Ten degrees or more of asymmetry with a scoliometer should prompt you to order radiographs.^35,36 A smartphone-based scoliometer for iPhones was evaluated in 1 study and was shown to have reasonable reliability and validity for clinical use.³⁷

Deformity of the lower extremities. Because low back pain may be caused by biomechanical or structural deformity of the lower extremities, examine the flexibility of the hip flexors, gluteal musculature, hamstrings, and the iliotibial band.³⁸ In addition, evaluate for leg-length discrepancy and lower-extremity malalignment, such as femoral anteversion, tibial torsion, or pes planus.

Continue to: Imaging

Imaging: Know when it’s needed

Although imaging of the lumbar spine is often unnecessary in the presence of acute low back pain in children, always consider imaging in the setting of bony tenderness, pain that wakes a patient from sleep, and in the setting of other red flag symptoms (see TABLE 2). Low back pain in children that is reproducible with lumbar extension is concerning for spondylolysis or spondylolisthesis. If the pain with extension persists beyond 3 to 6 weeks, order imaging starting with radiographs.^2,39

Traditionally, 4 views of the spine—­anteroposterior (AP), lateral, and oblique (one right and one left)—were obtained, but recent evidence indicates that 2 views (AP and lateral) have similar sensitivity and specificity to 4 views with significantly reduced radiation exposure.^2,39 Because the sensitivity of plain films is relatively low, consider more advanced imaging if spondylolysis or spondylolisthesis is strongly suspected. Recent studies indicate that magnetic resonance imaging (MRI) may be as effective as computed tomography (CT) or bone scan and has the advantage of lower radiation (FIGURE 1).^2,22

Similarly, order radiographs if there is > 10° of asymmetry noted on physical exam using a scoliometer.^15,23 Calculate the Cobb angle to determine the severity of scoliosis. Refer patients with angles ≥ 20° to a pediatric orthopedist for monitoring of progression and consideration of bracing (FIGURE 2).^23,34 For patients with curvatures between 10° and 19°, repeat imaging every 6 to 12 months. Because scoliosis is a risk factor for spondylolysis, evaluate radiographs in the setting of painful scoliosis for the presence of a spondylolysis.^34,35

Pain reproduced with lumbar flexion is less concerning for bony pathology and is most often related to soft-tissue strain.

If excessive kyphosis is noted on exam, order radiographs to evaluate for Scheuermann disease. Classic imaging findings include Schmorl nodes, vertebral endplate changes, and anterior wedging (FIGURE 3).^17,18

In the absence of the above concerns, defer imaging of the lumbar spine until after adequate rest and rehabilitation have been attempted.

Continue to: Treatment typically involves restor physical therapy

Most cases of low back pain in children and adolescents are benign and self-limited. Many children with low back pain can be treated with relative rest from the offending activity. For children with more persistent pain, physical therapy (PT) is often indicated. Similar to that for adults, there is little evidence for specific PT programs to help children with low back pain. Rehabilitation should be individualized based on the condition being treated.

Medications. There have been no high-quality studies on the benefit of medications to treat low back pain in children. Studies have shown nonsteroidal anti-inflammatory drugs (NSAIDs) have value in adults, and they are likely safe for use in children,⁴⁰ but the risk of opiate abuse is significantly increased in adolescents who have been prescribed opiate pain medication prior to 12th grade.⁴¹

Lumbar disc herniation. Although still relatively rare, lumbar disc herniation is more common in older children and adolescents than in younger children and is treated similarly to that in adults.⁸ Range-of-motion exercise to restore lumbar motion is often first-line treatment. Research has shown that exercises that strengthen the abdominal or “core” musculature help prevent the return of low back pain.^24,25

In the case of spondylolysis or spondylolisthesis, rest from activity is generally required for a minimum of 4 to 6 weeks. Rehabilitation in the form of range of motion, especially into the lumbar extension, and spinal stabilization exercises are effective for both reducing pain and restoring range-­of-motion and strength.⁴² Have patients avoid heavy backpacks, which can reproduce pain. Children often benefit from leaving a second set of schoolbooks at home. For most patients with spondylolysis, conservative treatment with rehabilitation is equal to or better than surgical intervention in returning the patient to his/her pre-injury activity level.^26,43,44 When returning athletes to their sport, aggressive PT, defined as rest for < 10 weeks prior to initiating PT, is superior to delaying PT beyond 10 weeks of rest.²⁷

Idiopathic scoliosis. Much of the literature on the treatment of scoliosis is focused on limiting progression of the scoliotic curvature. Researchers thought that more severe curves were associated with more severe pain, but a recent systematic review showed that back pain can occur in patients with even small curvatures.²⁸ Treatment for patients with smaller degrees of curvature is similar to that for mechanical low back pain. PT may have a role in the treatment of scoliosis, but there is little evidence in the literature of its effectiveness.

Continue to: A Cochrane review showed...

Always consider imaging in the setting of bony tenderness, pain that wakes the patient from sleep, and when there is > 10° of asymmetry on physical exam using a scoliometer.

A Cochrane review showed that PT and exercise-based treatments had no effect on back pain or disability in patients with scoliosis.²⁹ And outpatient PT alone, in the absence of bracing, does not arrest progression of the scoliotic curvature.³⁵ One trial did demonstrate that an intensive inpatient treatment program of 4 to 6 weeks for patients with curvature of at least 40° reduced progression of curvature compared to an untreated control group at 1 year.³⁴ The outcomes of functional mobility and pain were not measured. Follow-up data on curve progression beyond 1 year are not available. Unfortunately, intensive inpatient treatment is not readily available or cost-effective for most patients with scoliosis.

Scheuermann disease. The mainstay of treatment for mild Scheuermann disease is advising the patient to avoid repetitive loading of the spine. Patients should avoid sports such as competitive weight-lifting, gymnastics, and football. Lower impact athletics are encouraged. Refer patients with pain to PT to address posture and core stabilization. Patients with severe kyphosis may require surgery.^17,18

Bracing: Rarely helpful for low back pain

The use of lumbar braces or corsets is rarely helpful for low back pain in children. Bracing in the setting of spondylolysis is controversial.One study indicated that bracing in combination with activity restriction and lumbar extension exercise is superior to activity restriction and lumbar flexion exercises alone.⁴³ But a meta-analysis did not demonstrate a significant difference in recovery when bracing was added.⁴⁴ Bracing may help to reduce pain initially in patients with spondylolysis who have pain at rest. Bracing is not recommended for patients with pain that abates with activity modification.

Scoliosis and Scheuermann kyphosis. Treatment of adolescent idiopathic scoliosis usually consists of observation and periodic reevaluation. Bracing is a mainstay of the nonsurgical management of scoliosis and is appropriate for curves of 20° to 40°; studies have reported successful control of curve progression in > 70% of patients.³⁶ According to 1 study, the number of cases of scoliosis needed to treat with bracing to prevent 1 surgery is 3.³⁰ Surgery is often indicated for patients with curvatures > 40°, although this is also debated.³³

Bracing is used rarely for Scheuermann kyphosis but may be helpful in more severe or painful cases.¹⁷

CORRESPONDENCE
Shawn F. Phillips, MD, MSPT, 500 University Drive H154, Hershey, PA, 17033; [email protected].

Neurofibromatosis type 1 (NF1) is an autosomal dominant inherited disorder that is estimated to occur in 1:2500 births and to have a prevalence of 1:2000 to 1:4000.^1,2 It was first described in 1882 by Friedrich Daniel Von Recklinghausen, who identified patients and their relatives with signs of neuroectodermal abnormalities (café-au-lait macules [CALMs], axillary and inguinal freckling, and neurofibromas).

NF1 may begin insidiously in childhood and evolves as the patient ages. It is associated with intracranial, intraspinal, and intraorbital neoplasms, although other organs and tissues can also be involved.

The family physician might be the first one to recognize the signs of this condition during a well-child exam and is in a unique position to coordinate a multidisciplinary approach to care.

A mutated allele and early manifestations on the skin

NF1 has been attributed to genetic mosaicism and is classified as segmental, generalized, or (less frequently) gonadal. The disorder results from germline mutations in the NF1 tumor-suppressor gene on chromosome 17, known to codify the cytoplasmic protein called neurofibromin.³ The penetrance of NF1 is complete, which means that 100% of patients with the mutated allele will develop the disease.

Patients typically have symptoms by the third decade of life, although many will show signs of the disease in early childhood. CALMs are the earliest expression of NF1. They manifest in the first 2 years of life and are found in almost all affected patients. The lesions are well defined and measure 10 to 40 mm. They are typically light brown, although they may darken with sun exposure.

Histologically, the lesions will show macromelanosomes and high concentrations of melanin but do not represent an increased risk for malignancy.⁴ Not all isolated CALMs are a sign of NF1. While children younger than 29 months with 6 or more CALMs have a high risk for NF1 (80.4%; 95% confidence interval [CI], 74.6% to 86.2%), those who are older than 29 months with at least 1 atypical CALM or fewer than 6 CALMs have just a 0.9% (95% CI, 0% to 2.6%) risk for constitutional NF1.⁵

Freckles are also observed in 90% of patients with NF1; these tend to develop after the third year of life. The breast and trunk are the most commonly affected areas in adults. The pathophysiology is unknown, but this freckling is believed to be related to skin friction, high humidity, and ambient temperature.⁶

Continue to: Neurofibromas are benign...

Some patients show disfiguration when hundreds of neurofibromas are present.

Neurofibromas are benign subcutaneous palpable lesions that grow within peripheral nerve tissue, including spinal, subcutaneous, plexiform, or dermal encapsulated nerves. Originating in Schwann cells, they are composed of fibroblasts, mast cells, macrophages, endothelial cells, and other perineural cells. Some patients show disfiguration when hundreds of these masses are present (FIGURE). These tumors increase in number as the patient ages or during pregnancy, which is thought to be secondary to hormonal changes.⁷ They are sometimes painful and can be pruritic. Their appearance can also cause patient distress.

The diagnosis is a clinical one

Suspicion for NF1 should be high in patients presenting with the dermatologic findings described, although CALMs and freckling are not exclusive to NF1. Diagnostic criteria for NF1, which distinguish it from other conditions, were first outlined in a National Institutes of Health Consensus Development Conference Statement in 1987.⁸ The list of criteria has subsequently been expanded.

While the presence of at least 2 criteria is required for diagnosis,² NF1 should be suspected in individuals who have any of the following findings^8,9:

• the presence of at least 6 CALMs that are > 5 mm in prepubertal children and > 15 mm in adults
• 2 or more neurofibromas of any type, or at least one plexiform neurofibroma
• axillary or groin freckling
• optic pathway glioma
• 2 or more Lisch nodules (iris hamartomas seen on slit-lamp examination)
• bony dysplasia (sphenoid wing dysplasia, bowing of long bone ± pseudarthrosis)
• first-degree relative with NF1.

What you’ll see as the disease progresses

NF1 can affect a variety of systems, and potential complications of the disease are numerous and varied (see TABLE⁹). Here is some of what you may see as the patient’s disease progresses to various organ systems:

Learning disabilities and other cognitive and behavioral problems, such as attention-deficit/hyperactivity disorder, may affect up to 70% of children with NF1. Additionally, children with NF1 have visual/spatial problems, impaired visual motor integration, and language deficits.¹⁰ The etiology of cognitive impairment in NF1 is unknown.¹¹  

Continue to: Hypertension

Hypertension is common and may contribute to premature death in patients with NF1. Up to 27% of patients will have significant cardiovascular anomalies, including pulmonary valve stenosis, hypertrophic cardiomyopathy in patients with complete deletions of the NF1 gene, intracardiac neurofibromas, renal artery stenosis, coarctation of the aorta, and cerebral infarctions.¹² Renal artery stenosis occurs in approximately 2% of the NF1 population, and the diagnosis should be considered in hypertensive children, young adults, pregnant women, older individuals with refractory hypertension, and those with an abdominal bruit.¹³

Psychological issues. The disfigurement caused by neurofibromas and the uncertainty of an unpredictable disease course can cause psychological manifestations for patients with NF1. Anxiety and depression are common. Not surprisingly, patients with more severe disease report more adverse psychological effects.

Orthopedic deformities. Spinal deformities are the most common skeletal manifestation of NF1, with an incidence estimated from 10% to 25% in various studies. Bone mineral density, as measured by age- and gender-adjusted Z-scores, is significantly lower in NF1 patients than in the general population.¹⁴ Children may develop bowing of the long bones, particularly the tibia, and pseudarthrosis, a false joint in a long bone. Children with NF1 need yearly assessment of the spine. Patients with clinical evidence of scoliosis should be referred to Orthopedics for further evaluation.

Eye issues. A majority of adult patients develop neurofibroma-like nodules in the iris known as Lisch nodules. The nodules are not thought to cause any ophthalmologic complications. Patients may also develop palpebral neurofibroma, which may become large and sporadically show malignant transformation. Optic nerve glioma may cause strabismus and proptosis, and a large number of patients will also develop glaucoma and globe enlargement.¹⁵

Gastrointestinal lesions and cancer. Neurofibromas can grow in the stomach, liver, mesentery, retroperitoneum, and bowel. Adenocarcinoma developed in 23% of patients.¹⁶ Gastrointestinal tract bleeding, pseudo-obstruction, and protein-losing enteropathy also may occur.¹⁷

Continue to: Central nervous system manifestations

Central nervous system manifestations. Neurological manifestations have been observed in 55% of patients with NF1.¹⁸ These include headache, hydrocephalus, epilepsy, lacunar stroke, white matter disease, intraspinal neurofibroma, facial palsy, radiculopathy, and polyneuropathy. Tumors include optic pathway tumors, meningioma, and cerebral glioma. Glioma is the predominant tumor type in NF1 and occurs in all parts of the nervous system, with a predilection for the optic pathways, brainstem, and cerebellum.¹⁸

Malignant peripheral nerve sheath tumors. There is an 8% to 13% lifetime risk for malignant peripheral nerve sheath tumors (MPNST), predominantly in individuals between the ages of 20 and 35.^19,20 Any change in neurofibroma from soft to hard, or a rapid increase in the size, is suspicious for MPNST. Other symptoms include persistent pain lasting for longer than a month, pain that disturbs sleep, and new neurological deficits. These cancers can be hard to detect, leading to poor prognosis secondary to metastasis.^19,20 The greatest risk factors for MPNST are pain associated with a mass and the presence of cutaneous and subcutaneous neurofibromas.²¹

Treatment is symptom based, but there is a new option

Treatment is individualized to the patient’s symptoms. Neurofibromas that are disfiguring, disruptive, or malignant may be surgically removed.

In April 2020, the US Food and Drug Administration (FDA) approved selumetinib (Koselugo) for the treatment of pediatric patients (ages ≥ 2 years) with NF1 who have symptomatic, inoperable plexiform neurofibromas (PNs).²² In a clinical trial, patients received selumetinib 25 mg/m² orally twice a day until they demonstrated disease progression or experienced “unacceptable” adverse events.^22,23 The overall response rate was 66%, defined as “the percentage of patients with a complete response and those who experienced more than a 20% reduction in PN volume on MRI that was confirmed on a subsequent MRI within 3 to 6 months.”²²

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms.

Of note, all patients had a partial, not complete, response. Common adverse effects included vomiting, rash, abdominal pain, diarrhea, and nausea.²³ Selumetinib may also cause more serious adverse effects, including cardiomyopathy and ocular toxicity. Prior to treatment initiation and at regular intervals during treatment, patients should undergo cardiac and ophthalmic evaluation.^22,23 Selumetinib was granted priority review and orphan drug status by the FDA.²²

Continue to: You play a key role in ongoing monitoring

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms; optimization of quality of life; and identification and treatment of comorbidities. Family physicians are well positioned to monitor patients with NF1 for age-specific disease manifestations and potential complications.⁹ All patients require:

• an annual physical examination by a physician who is familiar with the individual and with the disease
• annual ophthalmologic examination in early childhood; less frequent examination in older children and adults
• regular blood pressure monitoring
• other studies (eg, MRI) only as indicated on the basis of clinically apparent signs or symptoms
• monitoring by an appropriate specialist if there are abnormalities of the central nervous, skeletal, or cardiovascular systems
• referral to a neurologist for any unexplained neurological signs and symptoms. Referral should be urgent if there are acute symptoms of progressive sensory disturbance, motor deficit and incoordination, or sphincter disturbances since these might indicate an intracranial lesion or spinal cord compression. Headaches on waking, morning vomiting, and altered consciousness are suggestive of raised intracranial pressure.

Children with NF1 benefit from coordinated care between the FP and a pediatrician or other specialist familiar with the disease. In addition to providing usual well care, perform regular assessment of development and school performance. Pay careful attention to the cardiovascular system (particularly blood pressure) and evaluate for scoliosis.

Young adults should be continually monitored for all complications, especially hypertension. This population requires continued education about NF1 and its possible complications and may benefit from counseling about disease inheritance. Screen for anxiety and depression; offer psychological support.

Provide adult patients with education about complications, especially malignant peripheral nerve sheath tumors and spinal cord compression.

Adults require monitoring based on patient preference and disease severity. For this population, blood pressure should be measured annually, or more frequently if the patient’s values indicate borderline hypertension. Provide education about complications, especially MPNSTs and spinal cord compression. Patients who have abnormalities of the central nervous, skeletal, or cardiovascular systems should be monitored by an appropriate specialist. If desired, the patient may be referred to a geneticist, especially if he or she expresses concern about inheritance. Cutaneous neurofibromas can be removed if they cause discomfort, although removal occasionally results in neurological deficit.

CORRESPONDENCE
T. Grant Phillips, MD, Associate Director, UPMC Altoona Family Physicians Residency, 501 Howard Avenue, Altoona, PA 16601-4899; [email protected]

Neurofibromatosis type 1 (NF1) is an autosomal dominant inherited disorder that is estimated to occur in 1:2500 births and to have a prevalence of 1:2000 to 1:4000.^1,2 It was first described in 1882 by Friedrich Daniel Von Recklinghausen, who identified patients and their relatives with signs of neuroectodermal abnormalities (café-au-lait macules [CALMs], axillary and inguinal freckling, and neurofibromas).

NF1 may begin insidiously in childhood and evolves as the patient ages. It is associated with intracranial, intraspinal, and intraorbital neoplasms, although other organs and tissues can also be involved.

The family physician might be the first one to recognize the signs of this condition during a well-child exam and is in a unique position to coordinate a multidisciplinary approach to care.

A mutated allele and early manifestations on the skin

NF1 has been attributed to genetic mosaicism and is classified as segmental, generalized, or (less frequently) gonadal. The disorder results from germline mutations in the NF1 tumor-suppressor gene on chromosome 17, known to codify the cytoplasmic protein called neurofibromin.³ The penetrance of NF1 is complete, which means that 100% of patients with the mutated allele will develop the disease.

Patients typically have symptoms by the third decade of life, although many will show signs of the disease in early childhood. CALMs are the earliest expression of NF1. They manifest in the first 2 years of life and are found in almost all affected patients. The lesions are well defined and measure 10 to 40 mm. They are typically light brown, although they may darken with sun exposure.

Histologically, the lesions will show macromelanosomes and high concentrations of melanin but do not represent an increased risk for malignancy.⁴ Not all isolated CALMs are a sign of NF1. While children younger than 29 months with 6 or more CALMs have a high risk for NF1 (80.4%; 95% confidence interval [CI], 74.6% to 86.2%), those who are older than 29 months with at least 1 atypical CALM or fewer than 6 CALMs have just a 0.9% (95% CI, 0% to 2.6%) risk for constitutional NF1.⁵

Freckles are also observed in 90% of patients with NF1; these tend to develop after the third year of life. The breast and trunk are the most commonly affected areas in adults. The pathophysiology is unknown, but this freckling is believed to be related to skin friction, high humidity, and ambient temperature.⁶

Continue to: Neurofibromas are benign...

Some patients show disfiguration when hundreds of neurofibromas are present.

Neurofibromas are benign subcutaneous palpable lesions that grow within peripheral nerve tissue, including spinal, subcutaneous, plexiform, or dermal encapsulated nerves. Originating in Schwann cells, they are composed of fibroblasts, mast cells, macrophages, endothelial cells, and other perineural cells. Some patients show disfiguration when hundreds of these masses are present (FIGURE). These tumors increase in number as the patient ages or during pregnancy, which is thought to be secondary to hormonal changes.⁷ They are sometimes painful and can be pruritic. Their appearance can also cause patient distress.

The diagnosis is a clinical one

Suspicion for NF1 should be high in patients presenting with the dermatologic findings described, although CALMs and freckling are not exclusive to NF1. Diagnostic criteria for NF1, which distinguish it from other conditions, were first outlined in a National Institutes of Health Consensus Development Conference Statement in 1987.⁸ The list of criteria has subsequently been expanded.

While the presence of at least 2 criteria is required for diagnosis,² NF1 should be suspected in individuals who have any of the following findings^8,9:

• the presence of at least 6 CALMs that are > 5 mm in prepubertal children and > 15 mm in adults
• 2 or more neurofibromas of any type, or at least one plexiform neurofibroma
• axillary or groin freckling
• optic pathway glioma
• 2 or more Lisch nodules (iris hamartomas seen on slit-lamp examination)
• bony dysplasia (sphenoid wing dysplasia, bowing of long bone ± pseudarthrosis)
• first-degree relative with NF1.

What you’ll see as the disease progresses

NF1 can affect a variety of systems, and potential complications of the disease are numerous and varied (see TABLE⁹). Here is some of what you may see as the patient’s disease progresses to various organ systems:

Learning disabilities and other cognitive and behavioral problems, such as attention-deficit/hyperactivity disorder, may affect up to 70% of children with NF1. Additionally, children with NF1 have visual/spatial problems, impaired visual motor integration, and language deficits.¹⁰ The etiology of cognitive impairment in NF1 is unknown.¹¹  

Continue to: Hypertension

Hypertension is common and may contribute to premature death in patients with NF1. Up to 27% of patients will have significant cardiovascular anomalies, including pulmonary valve stenosis, hypertrophic cardiomyopathy in patients with complete deletions of the NF1 gene, intracardiac neurofibromas, renal artery stenosis, coarctation of the aorta, and cerebral infarctions.¹² Renal artery stenosis occurs in approximately 2% of the NF1 population, and the diagnosis should be considered in hypertensive children, young adults, pregnant women, older individuals with refractory hypertension, and those with an abdominal bruit.¹³

Psychological issues. The disfigurement caused by neurofibromas and the uncertainty of an unpredictable disease course can cause psychological manifestations for patients with NF1. Anxiety and depression are common. Not surprisingly, patients with more severe disease report more adverse psychological effects.

Orthopedic deformities. Spinal deformities are the most common skeletal manifestation of NF1, with an incidence estimated from 10% to 25% in various studies. Bone mineral density, as measured by age- and gender-adjusted Z-scores, is significantly lower in NF1 patients than in the general population.¹⁴ Children may develop bowing of the long bones, particularly the tibia, and pseudarthrosis, a false joint in a long bone. Children with NF1 need yearly assessment of the spine. Patients with clinical evidence of scoliosis should be referred to Orthopedics for further evaluation.

Eye issues. A majority of adult patients develop neurofibroma-like nodules in the iris known as Lisch nodules. The nodules are not thought to cause any ophthalmologic complications. Patients may also develop palpebral neurofibroma, which may become large and sporadically show malignant transformation. Optic nerve glioma may cause strabismus and proptosis, and a large number of patients will also develop glaucoma and globe enlargement.¹⁵

Gastrointestinal lesions and cancer. Neurofibromas can grow in the stomach, liver, mesentery, retroperitoneum, and bowel. Adenocarcinoma developed in 23% of patients.¹⁶ Gastrointestinal tract bleeding, pseudo-obstruction, and protein-losing enteropathy also may occur.¹⁷

Continue to: Central nervous system manifestations

Central nervous system manifestations. Neurological manifestations have been observed in 55% of patients with NF1.¹⁸ These include headache, hydrocephalus, epilepsy, lacunar stroke, white matter disease, intraspinal neurofibroma, facial palsy, radiculopathy, and polyneuropathy. Tumors include optic pathway tumors, meningioma, and cerebral glioma. Glioma is the predominant tumor type in NF1 and occurs in all parts of the nervous system, with a predilection for the optic pathways, brainstem, and cerebellum.¹⁸

Malignant peripheral nerve sheath tumors. There is an 8% to 13% lifetime risk for malignant peripheral nerve sheath tumors (MPNST), predominantly in individuals between the ages of 20 and 35.^19,20 Any change in neurofibroma from soft to hard, or a rapid increase in the size, is suspicious for MPNST. Other symptoms include persistent pain lasting for longer than a month, pain that disturbs sleep, and new neurological deficits. These cancers can be hard to detect, leading to poor prognosis secondary to metastasis.^19,20 The greatest risk factors for MPNST are pain associated with a mass and the presence of cutaneous and subcutaneous neurofibromas.²¹

Treatment is symptom based, but there is a new option

Treatment is individualized to the patient’s symptoms. Neurofibromas that are disfiguring, disruptive, or malignant may be surgically removed.

In April 2020, the US Food and Drug Administration (FDA) approved selumetinib (Koselugo) for the treatment of pediatric patients (ages ≥ 2 years) with NF1 who have symptomatic, inoperable plexiform neurofibromas (PNs).²² In a clinical trial, patients received selumetinib 25 mg/m² orally twice a day until they demonstrated disease progression or experienced “unacceptable” adverse events.^22,23 The overall response rate was 66%, defined as “the percentage of patients with a complete response and those who experienced more than a 20% reduction in PN volume on MRI that was confirmed on a subsequent MRI within 3 to 6 months.”²²

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms.

Of note, all patients had a partial, not complete, response. Common adverse effects included vomiting, rash, abdominal pain, diarrhea, and nausea.²³ Selumetinib may also cause more serious adverse effects, including cardiomyopathy and ocular toxicity. Prior to treatment initiation and at regular intervals during treatment, patients should undergo cardiac and ophthalmic evaluation.^22,23 Selumetinib was granted priority review and orphan drug status by the FDA.²²

Continue to: You play a key role in ongoing monitoring

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms; optimization of quality of life; and identification and treatment of comorbidities. Family physicians are well positioned to monitor patients with NF1 for age-specific disease manifestations and potential complications.⁹ All patients require:

• an annual physical examination by a physician who is familiar with the individual and with the disease
• annual ophthalmologic examination in early childhood; less frequent examination in older children and adults
• regular blood pressure monitoring
• other studies (eg, MRI) only as indicated on the basis of clinically apparent signs or symptoms
• monitoring by an appropriate specialist if there are abnormalities of the central nervous, skeletal, or cardiovascular systems
• referral to a neurologist for any unexplained neurological signs and symptoms. Referral should be urgent if there are acute symptoms of progressive sensory disturbance, motor deficit and incoordination, or sphincter disturbances since these might indicate an intracranial lesion or spinal cord compression. Headaches on waking, morning vomiting, and altered consciousness are suggestive of raised intracranial pressure.

Children with NF1 benefit from coordinated care between the FP and a pediatrician or other specialist familiar with the disease. In addition to providing usual well care, perform regular assessment of development and school performance. Pay careful attention to the cardiovascular system (particularly blood pressure) and evaluate for scoliosis.

Young adults should be continually monitored for all complications, especially hypertension. This population requires continued education about NF1 and its possible complications and may benefit from counseling about disease inheritance. Screen for anxiety and depression; offer psychological support.

Provide adult patients with education about complications, especially malignant peripheral nerve sheath tumors and spinal cord compression.

Adults require monitoring based on patient preference and disease severity. For this population, blood pressure should be measured annually, or more frequently if the patient’s values indicate borderline hypertension. Provide education about complications, especially MPNSTs and spinal cord compression. Patients who have abnormalities of the central nervous, skeletal, or cardiovascular systems should be monitored by an appropriate specialist. If desired, the patient may be referred to a geneticist, especially if he or she expresses concern about inheritance. Cutaneous neurofibromas can be removed if they cause discomfort, although removal occasionally results in neurological deficit.

CORRESPONDENCE
T. Grant Phillips, MD, Associate Director, UPMC Altoona Family Physicians Residency, 501 Howard Avenue, Altoona, PA 16601-4899; [email protected]

Neurofibromatosis type 1 (NF1) is an autosomal dominant inherited disorder that is estimated to occur in 1:2500 births and to have a prevalence of 1:2000 to 1:4000.^1,2 It was first described in 1882 by Friedrich Daniel Von Recklinghausen, who identified patients and their relatives with signs of neuroectodermal abnormalities (café-au-lait macules [CALMs], axillary and inguinal freckling, and neurofibromas).

NF1 may begin insidiously in childhood and evolves as the patient ages. It is associated with intracranial, intraspinal, and intraorbital neoplasms, although other organs and tissues can also be involved.

The family physician might be the first one to recognize the signs of this condition during a well-child exam and is in a unique position to coordinate a multidisciplinary approach to care.

A mutated allele and early manifestations on the skin

NF1 has been attributed to genetic mosaicism and is classified as segmental, generalized, or (less frequently) gonadal. The disorder results from germline mutations in the NF1 tumor-suppressor gene on chromosome 17, known to codify the cytoplasmic protein called neurofibromin.³ The penetrance of NF1 is complete, which means that 100% of patients with the mutated allele will develop the disease.

Patients typically have symptoms by the third decade of life, although many will show signs of the disease in early childhood. CALMs are the earliest expression of NF1. They manifest in the first 2 years of life and are found in almost all affected patients. The lesions are well defined and measure 10 to 40 mm. They are typically light brown, although they may darken with sun exposure.

Histologically, the lesions will show macromelanosomes and high concentrations of melanin but do not represent an increased risk for malignancy.⁴ Not all isolated CALMs are a sign of NF1. While children younger than 29 months with 6 or more CALMs have a high risk for NF1 (80.4%; 95% confidence interval [CI], 74.6% to 86.2%), those who are older than 29 months with at least 1 atypical CALM or fewer than 6 CALMs have just a 0.9% (95% CI, 0% to 2.6%) risk for constitutional NF1.⁵

Freckles are also observed in 90% of patients with NF1; these tend to develop after the third year of life. The breast and trunk are the most commonly affected areas in adults. The pathophysiology is unknown, but this freckling is believed to be related to skin friction, high humidity, and ambient temperature.⁶

Continue to: Neurofibromas are benign...

Some patients show disfiguration when hundreds of neurofibromas are present.

Neurofibromas are benign subcutaneous palpable lesions that grow within peripheral nerve tissue, including spinal, subcutaneous, plexiform, or dermal encapsulated nerves. Originating in Schwann cells, they are composed of fibroblasts, mast cells, macrophages, endothelial cells, and other perineural cells. Some patients show disfiguration when hundreds of these masses are present (FIGURE). These tumors increase in number as the patient ages or during pregnancy, which is thought to be secondary to hormonal changes.⁷ They are sometimes painful and can be pruritic. Their appearance can also cause patient distress.

The diagnosis is a clinical one

Suspicion for NF1 should be high in patients presenting with the dermatologic findings described, although CALMs and freckling are not exclusive to NF1. Diagnostic criteria for NF1, which distinguish it from other conditions, were first outlined in a National Institutes of Health Consensus Development Conference Statement in 1987.⁸ The list of criteria has subsequently been expanded.

While the presence of at least 2 criteria is required for diagnosis,² NF1 should be suspected in individuals who have any of the following findings^8,9:

• the presence of at least 6 CALMs that are > 5 mm in prepubertal children and > 15 mm in adults
• 2 or more neurofibromas of any type, or at least one plexiform neurofibroma
• axillary or groin freckling
• optic pathway glioma
• 2 or more Lisch nodules (iris hamartomas seen on slit-lamp examination)
• bony dysplasia (sphenoid wing dysplasia, bowing of long bone ± pseudarthrosis)
• first-degree relative with NF1.

What you’ll see as the disease progresses

NF1 can affect a variety of systems, and potential complications of the disease are numerous and varied (see TABLE⁹). Here is some of what you may see as the patient’s disease progresses to various organ systems:

Learning disabilities and other cognitive and behavioral problems, such as attention-deficit/hyperactivity disorder, may affect up to 70% of children with NF1. Additionally, children with NF1 have visual/spatial problems, impaired visual motor integration, and language deficits.¹⁰ The etiology of cognitive impairment in NF1 is unknown.¹¹  

Continue to: Hypertension

Hypertension is common and may contribute to premature death in patients with NF1. Up to 27% of patients will have significant cardiovascular anomalies, including pulmonary valve stenosis, hypertrophic cardiomyopathy in patients with complete deletions of the NF1 gene, intracardiac neurofibromas, renal artery stenosis, coarctation of the aorta, and cerebral infarctions.¹² Renal artery stenosis occurs in approximately 2% of the NF1 population, and the diagnosis should be considered in hypertensive children, young adults, pregnant women, older individuals with refractory hypertension, and those with an abdominal bruit.¹³

Psychological issues. The disfigurement caused by neurofibromas and the uncertainty of an unpredictable disease course can cause psychological manifestations for patients with NF1. Anxiety and depression are common. Not surprisingly, patients with more severe disease report more adverse psychological effects.

Orthopedic deformities. Spinal deformities are the most common skeletal manifestation of NF1, with an incidence estimated from 10% to 25% in various studies. Bone mineral density, as measured by age- and gender-adjusted Z-scores, is significantly lower in NF1 patients than in the general population.¹⁴ Children may develop bowing of the long bones, particularly the tibia, and pseudarthrosis, a false joint in a long bone. Children with NF1 need yearly assessment of the spine. Patients with clinical evidence of scoliosis should be referred to Orthopedics for further evaluation.

Eye issues. A majority of adult patients develop neurofibroma-like nodules in the iris known as Lisch nodules. The nodules are not thought to cause any ophthalmologic complications. Patients may also develop palpebral neurofibroma, which may become large and sporadically show malignant transformation. Optic nerve glioma may cause strabismus and proptosis, and a large number of patients will also develop glaucoma and globe enlargement.¹⁵

Gastrointestinal lesions and cancer. Neurofibromas can grow in the stomach, liver, mesentery, retroperitoneum, and bowel. Adenocarcinoma developed in 23% of patients.¹⁶ Gastrointestinal tract bleeding, pseudo-obstruction, and protein-losing enteropathy also may occur.¹⁷

Continue to: Central nervous system manifestations

Central nervous system manifestations. Neurological manifestations have been observed in 55% of patients with NF1.¹⁸ These include headache, hydrocephalus, epilepsy, lacunar stroke, white matter disease, intraspinal neurofibroma, facial palsy, radiculopathy, and polyneuropathy. Tumors include optic pathway tumors, meningioma, and cerebral glioma. Glioma is the predominant tumor type in NF1 and occurs in all parts of the nervous system, with a predilection for the optic pathways, brainstem, and cerebellum.¹⁸

Malignant peripheral nerve sheath tumors. There is an 8% to 13% lifetime risk for malignant peripheral nerve sheath tumors (MPNST), predominantly in individuals between the ages of 20 and 35.^19,20 Any change in neurofibroma from soft to hard, or a rapid increase in the size, is suspicious for MPNST. Other symptoms include persistent pain lasting for longer than a month, pain that disturbs sleep, and new neurological deficits. These cancers can be hard to detect, leading to poor prognosis secondary to metastasis.^19,20 The greatest risk factors for MPNST are pain associated with a mass and the presence of cutaneous and subcutaneous neurofibromas.²¹

Treatment is symptom based, but there is a new option

Treatment is individualized to the patient’s symptoms. Neurofibromas that are disfiguring, disruptive, or malignant may be surgically removed.

In April 2020, the US Food and Drug Administration (FDA) approved selumetinib (Koselugo) for the treatment of pediatric patients (ages ≥ 2 years) with NF1 who have symptomatic, inoperable plexiform neurofibromas (PNs).²² In a clinical trial, patients received selumetinib 25 mg/m² orally twice a day until they demonstrated disease progression or experienced “unacceptable” adverse events.^22,23 The overall response rate was 66%, defined as “the percentage of patients with a complete response and those who experienced more than a 20% reduction in PN volume on MRI that was confirmed on a subsequent MRI within 3 to 6 months.”²²

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms.

Of note, all patients had a partial, not complete, response. Common adverse effects included vomiting, rash, abdominal pain, diarrhea, and nausea.²³ Selumetinib may also cause more serious adverse effects, including cardiomyopathy and ocular toxicity. Prior to treatment initiation and at regular intervals during treatment, patients should undergo cardiac and ophthalmic evaluation.^22,23 Selumetinib was granted priority review and orphan drug status by the FDA.²²

Continue to: You play a key role in ongoing monitoring

In light of the condition’s heterogeneity, the goals of care include early recognition and treatment of complications, especially neoplasms; optimization of quality of life; and identification and treatment of comorbidities. Family physicians are well positioned to monitor patients with NF1 for age-specific disease manifestations and potential complications.⁹ All patients require:

• an annual physical examination by a physician who is familiar with the individual and with the disease
• annual ophthalmologic examination in early childhood; less frequent examination in older children and adults
• regular blood pressure monitoring
• other studies (eg, MRI) only as indicated on the basis of clinically apparent signs or symptoms
• monitoring by an appropriate specialist if there are abnormalities of the central nervous, skeletal, or cardiovascular systems
• referral to a neurologist for any unexplained neurological signs and symptoms. Referral should be urgent if there are acute symptoms of progressive sensory disturbance, motor deficit and incoordination, or sphincter disturbances since these might indicate an intracranial lesion or spinal cord compression. Headaches on waking, morning vomiting, and altered consciousness are suggestive of raised intracranial pressure.

Children with NF1 benefit from coordinated care between the FP and a pediatrician or other specialist familiar with the disease. In addition to providing usual well care, perform regular assessment of development and school performance. Pay careful attention to the cardiovascular system (particularly blood pressure) and evaluate for scoliosis.

Young adults should be continually monitored for all complications, especially hypertension. This population requires continued education about NF1 and its possible complications and may benefit from counseling about disease inheritance. Screen for anxiety and depression; offer psychological support.

Provide adult patients with education about complications, especially malignant peripheral nerve sheath tumors and spinal cord compression.

Adults require monitoring based on patient preference and disease severity. For this population, blood pressure should be measured annually, or more frequently if the patient’s values indicate borderline hypertension. Provide education about complications, especially MPNSTs and spinal cord compression. Patients who have abnormalities of the central nervous, skeletal, or cardiovascular systems should be monitored by an appropriate specialist. If desired, the patient may be referred to a geneticist, especially if he or she expresses concern about inheritance. Cutaneous neurofibromas can be removed if they cause discomfort, although removal occasionally results in neurological deficit.

CORRESPONDENCE
T. Grant Phillips, MD, Associate Director, UPMC Altoona Family Physicians Residency, 501 Howard Avenue, Altoona, PA 16601-4899; [email protected]

Venous thromboembolism (VTE) is a common and dangerous disease, affecting 0.1%-0.2% of the population annually—a rate that might be underreported.¹ VTE is a collective term for venous blood clots, including (1) deep vein thrombosis (DVT) of peripheral veins and (2) pulmonary embolism, which occurs after a clot travels through the heart and becomes lodged in the pulmonary vasculature. Two-thirds of VTE cases present clinically as DVT²; most mortality from VTE disease is caused by the 20% of cases of pulmonary embolism that present as sudden death.¹

VTE is comparable to myocardial infarction (MI) in incidence and severity. In 2008, 208 of every 100,000 people had an MI, with a 30-day mortality of 16/100,000³; VTE disease has an annual incidence of 161 of every 100,000 people and a 28-day mortality of 18/100,000.⁴ Although the incidence and severity of MI are steadily decreasing, the rate of VTE appears constant.^3,5 The high mortality of VTE suggests that primary prevention, which we discuss in this article, is valuable (see “Key points: Primary prevention of venous thromboembolism”).

Risk factors

Virchow’s triad of venous stasis, vascular injury, and hypercoagulability describes predisposing factors for VTE.⁶ Although venous valves promote blood flow, they produce isolated low-flow areas adjacent to valves that become concentrated and locally hypoxic, increasing the risk of clotting.⁷ The great majority of DVTs (≥ 96%) occur in the lower extremity,⁸ starting in the calf; there, 75% of cases resolve spontaneously before they extend into the deep veins of the proximal leg.⁷ One-half of DVTs that do move into the proximal leg eventually embolize.⁷

Major risk factors for VTE comprise inherited conditions, medical history, medical therapeutics, and behaviors (TABLE 1).^9-11 Unlike the preventive management of coronary artery disease (CAD), there is no simple, generalized prevention algorithm to address VTE risk factors.

Risk factors for VTE and CAD overlap. Risk factors for atherosclerosis—­obesity, diabetes, smoking, hypertension, ­hyperlipidemia—also increase the risk of VTE (TABLE 1).^9-11 The association between risk factors for VTE and atherosclerosis is demonstrated by a doubling of the risk of MI and stroke in the year following VTE.¹¹ Lifestyle changes are expected to reduce the risk of VTE, as they do for acute CAD, but studies are lacking to confirm this connection. There is no prospective evidence showing that weight loss or control of diabetes or hypertension reduces the risk of VTE.¹² Smoking cessation does appear to reduce risk: Former smokers have the same VTE risk as never-smokers.¹³

Thrombophilia testing: Not generally useful

Inherited and acquired thrombophilic conditions define a group of disorders in which the risk of VTE is increased. Although thrombophilia testing was once considered for primary and secondary prevention of VTE, such testing is rarely used now because proof of benefit is lacking: A large case–control study showed that thrombophilia testing did not predict recurrence after a first VTE.¹⁴ Guidelines of the American College of Chest Physicians (ACCP) do not address thrombophilia, and the American Society of Hematology recommends against thrombophilia testing after a provoked VTE.^15,16

Primary prophylaxis of patients with a family history of VTE and inherited thrombophilia is controversial. Patients with both a family history of VTE and demonstrated thrombophilia do have double the average incidence of VTE, but this increased risk does not offset the significant bleeding risk associated with anticoagulation.¹⁷ Recommendations for thrombophilia testing are limited to certain situations in pregnancy, discussed in a bit.^16,18,19

Continue to: Primary prevention of VTE in the clinic

There is no single, overarching preventive strategy for VTE in an ambulatory patient (although statins, discussed in a moment, offer some benefit, broadly). There are, however, distinct behavioral characteristics and medical circumstances for which opportunities exist to reduce VTE risk—for example, when a person engages in long-distance travel, receives hormonal therapy, is pregnant, or has cancer. In each scenario, recognizing and mitigating risk are important.

Statins offer a (slight) benefit

There is evidence that statins reduce the risk of VTE—slightly^20-23:

• A large randomized, controlled trial showed that rosuvastatin, 20 mg/d, reduced the rate of VTE, compared to placebo; however, the 2-year number needed to treat (NNT) was 349.²⁰ The VTE benefit is minimal, however, compared to primary prevention of cardiovascular disease with statins (5-year NNT = 56).²¹ The sole significant adverse event associated with statins was new-onset type 2 diabetes (5-year number needed to harm = 235).²¹
• A subsequent meta-analysis confirmed a small reduction in VTE risk with statins.²² In its 2012 guidelines, ACCP declined to issue a recommendation on the use of statins for VTE prevention.²³ When considering statins for primary cardiovascular disease prevention, take the additional VTE prevention into account.

Simple strategies can help prevent travel-related VTE

Travel is a common inciting factor for VTE. A systematic review showed that VTE risk triples after travel of ≥ 4 hours, increasing by 20% with each additional 2 hours.²⁴ Most VTE occurs in travelers who have other VTE risk factors.²⁵ Based on case–control studies,²³ guidelines recommend these preventive measures:

• frequent calf exercises
• sitting in an aisle seat during air travel
• keeping hydrated.

A Cochrane review showed that graded compression stockings reduce asymptomatic DVT in travelers by a factor of 10, in high- and low-risk patients.²⁶

VTE risk varies with type of hormonal contraception

Most contraceptives increase VTE risk (TABLE 2^27,28). Risk with combined oral contraceptives varies with the amount of estrogen and progesterone. To reduce VTE risk with oral contraceptives, patients can use an agent that contains a lower dose of estrogen or one in which levonorgestrel replaces other progesterones.²⁷

Continue to: Studies suggest that the levonorgestrel-releasing...

Studies suggest that the levonorgestrel-releasing intrauterine device and progestin-only pills are not associated with an increase in VTE risk.²⁷ Although the quality of evidence varies, most nonoral hormonal contraceptives have been determined to carry a risk of VTE that is similar to that of combined oral contraceptives.²⁸

In hormone replacement, avoid pills to lower risk

Hormone replacement therapy (HRT) for postmenopausal women increases VTE risk when administered in oral form, with combined estrogen and progestin HRT doubling the risk and estrogen-only formulations having a lower risk.²⁹ VTE risk is highest in the first 6 months of HRT, declining to that of a non-HRT user within 5 years.²⁹ Neither transdermal HRT nor estrogen creams increase the risk of VTE, according to a systematic review.³⁰ The estradiol-containing vaginal ring also does not confer increased risk.²⁹

Pregnancy, thrombophilia, and VTE prevention

VTE affects as many as 0.2% of pregnancies but causes 9% of pregnancy-related deaths.¹⁸ The severity of VTE in pregnancy led the American College of Obstetricians and Gynecologists (ACOG) to recommend primary VTE prophylaxis in patients with certain thrombophilias.¹⁸ Thrombophilia testing is recommended in patients with proven high-risk thrombophilia in a first-degree relative.¹⁸ ACOG recognizes 5 thrombophilias considered to carry a high risk of VTE in pregnancy¹⁸:

• homozygous Factor V Leiden
• homozygous prothrombin G20210A mutation
• antithrombin deficiency
• heterozygous Factor V Leiden and prothrombin G20210A mutation
• antiphospholipid antibody syndrome.

ACOG recommends limiting thrombophilia testing to (1) any specific thrombophilia carried by a relative and (2) possibly, the antiphospholipid antibodies anticardiolipin and lupus anticoagulant.^18,19 Antiphospholipid testing is recommended when there is a history of stillbirth, 3 early pregnancy losses, or delivery earlier than 34 weeks secondary to preeclampsia.¹⁹

Primary VTE prophylaxis is recommended for pregnant patients with a high-risk thrombophilia; low-molecular-weight heparin (LMWH) is safe and its effects are predictable.¹⁸ Because postpartum risk of VTE is higher than antepartum risk, postpartum prophylaxis is also recommended with lower-risk thrombophilias¹⁸; a vitamin K antagonist or LMWH can be used.¹⁸ ACCP and ACOG recommendations for VTE prophylaxis in pregnancy differ slightly (TABLE 3^16,18,19).

Continue to: Cancer increases risks of VTE and bleeding

Cancer increases VTE risk > 6-fold³¹; metastases, chemotherapy, and radiotherapy further increase risk. Cancer also greatly increases the risk of bleeding: Cancer patients with VTE have an annual major bleeding rate ≥ 20%.³² Guidelines do not recommend primary VTE prophylaxis for cancer, although American Society of Clinical Oncology guidelines discuss consideration of prophylaxis for select, high-risk patients,^33,34 including those with multiple myeloma, metastatic gastrointestinal cancer, or metastatic brain cancer.^31,34 Recent evidence (discussed in a moment) supports the use of apixaban for primary VTE prevention during chemotherapy for high-risk cancer.

The Khorana Risk Score (TABLE 4^35,36) for VTE was developed and validated for use in patients with solid cancer³⁵: A score of 2 conveys nearly a 10% risk of VTE over 6 months.³⁶ A recent study of 550 cancer patients with a Khorana score of ≥ 2—the first evidence of risk-guided primary VTE prevention in cancer—showed that primary prophylaxis with 2.5 mg of apixaban, bid, reduced the risk of VTE (NNT = 17); however, the number needed to harm (for major bleeding) was 59.³⁷ Mortality was not changed with apixaban treatment.³⁷

Primary VTE prevention in med-surg hospitalizations

The risk of VTE increases significantly during hospitalization, although not enough to justify universal prophylaxis. Recommended prevention strategies for different classes of hospitalized patients are summarized below.

In medically hospitalized patients, risk is stratified with a risk-assessment model. Medically hospitalized patients have, on average, a VTE risk of 1.2%²³; 12 risk-assessment models designed to stratify risk were recently compared.³⁸ Two models, the Caprini Score (TABLE 5)³⁹ and the IMPROVE VTE Risk Calculator,⁴⁰ were best able to identify low-risk patients (negative predictive value, > 99%).³⁸ American Society of Hematology guidelines recommend IMPROVE VTE or the Padua Prediction Score for risk stratification.⁴¹ While the Caprini score only designates 11% of eventual VTE cases as low risk, both the IMPROVE VTE and Padua scores miss more than 35% of eventual VTE.³⁸

There is no prospective evidence that weight loss or control of diabetes or hypertension reduces the risk of VTE; smoking cessation does appear to reduce risk.

Because LMWH prophylaxis has been shown to reduce VTE by 40% without increasing the risk of major bleeding, using Caprini should prevent 2 VTEs for every 1000 patients, without an increase in major bleeding and with 13 additional minor bleeding events.⁴²

Continue to: Critically ill patients

Critically ill patients are assumed to be at high risk of VTE and do not require stratification.²³ For high-risk patients, prophylaxis with LMWH, low-dose unfractionated heparin (LDUH), or fondaparinux is recommended for the duration of admission.²³ For patients at high risk of both VTE and bleeding, mechanical prophylaxis with intermittent pneumatic compression (IPC) is recommended instead of LMWH, LDUH, or fondaparinux.²³

Surgery, like trauma (see next page), increases the risk of VTE and has been well studied. Prophylaxis after orthopedic surgery differs from that of other types of surgery.

In orthopedic surgery, risk depends on the procedure. For major orthopedic surgery, including total hip or knee arthroplasty and hip fracture surgery, VTE prophylaxis is recommended for 35 days postsurgically.⁴³ LMWH is the preferred agent, although many other means have been shown to be beneficial.⁴⁴ A recent systematic review demonstrated that aspirin is not inferior to other medications after hip or knee arthroplasty.⁴⁵ No mechanical or pharmacotherapeutic prophylaxis is generally recommended after nonmajor orthopedic surgery.⁴³

Taking a statin can reduce the risk of VTE— slightly.

Nonorthopedic surgery is stratified by risk factors, using Caprini⁴⁴ (TABLE 5³⁹). For medium-risk patients (Caprini score, 3-4) LDUH, LMWH, or IPC is recommended; for high-risk patients (Caprini score, ≥ 5) preventive treatment should combine pharmacotherapeutic and mechanical prophylaxis.⁴⁶ A recent meta-analysis, comprising 14,776 patients, showed that surgical patients with a Caprini score ≥ 7 had a reduced incidence of VTE when given chemoprophylaxis, whereas patients whose score is < 7 do not benefit from chemoprophylaxis.⁴³ When bleeding risk is high, IPC is recommended as sole therapy.⁴³ Prophylaxis is not recommended when risk (determined by the Caprini score) is low.⁴⁶

Post-hospitalization. Risk of VTE can persist for as long as 90 days after hospitalization; this finding has led to evaluation of the benefit of prolonged chemoprophylaxis.²³ Extended-duration LMWH prophylaxis decreases the incidence of VTE, but at the cost of increased risk of major bleeding.⁴⁷ Based on this evidence, guidelines recommend against prolonged-duration anticoagulation.²³ A 2016 trial showed that 35 days of the direct-acting anticoagulant betrixaban reduced the risk of symptomatic VTE events, compared to 10 days of LMWH (NNT = 167), without increased risk of bleeding.⁴⁸ This is a limited benefit, however, that is unlikely to change guideline recommendations.

Continue to: Trauma

Trauma: VTE risk increases with severity

Trauma increases the risk of VTE considerably. A national study showed that 1.5% of admitted trauma patients experienced VTE during hospitalization and that 1.2% were readmitted for VTE within 1 year.⁴⁹ As many as 32% of trauma patients admitted to the intensive care unit experience VTE despite appropriate prophylaxis.⁵⁰ A Cochrane Review⁵¹ found that:

• prophylaxis significantly reduces DVT risk
• pharmacotherapeutic prophylaxis is more effective than mechanical prophylaxis
• LMWH is more effective than LDUH.

Guidelines recommend that major trauma patients receive prophylaxis with LMWH, LDUH, or IPC.⁴⁶

CORRESPONDENCE
Michael J. Arnold, MD, CDR, MC, USN; Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Jacksonville, FL 32214; [email protected].