When following CT guidelines isn’t enough

AN 86-YEAR-OLD MAN ON WARFARIN FAINTED AND FELL while baby-sitting his great-grandchildren. He had transient neurologic symptoms after collapsing but appeared normal by the time paramedics arrived. He was taken by private vehicle to the hospital, where an emergency department (ED) physician examined him. After tests for a myocardial infarction revealed normal enzymes, electrocardiogram, and chest radiograph, the patient was discharged home.

He returned to the hospital the following day and underwent a computed tomography (CT) scan, which showed a large cerebral hemorrhage. He died soon afterward.

PLAINTIFF’S CLAIM The patient should have had a CT scan during the first ED visit. A scan at that visit would have found the hemorrhage in time to save the patient’s life.

THE DEFENSE No discussion with family members about a blow to the head or head trauma occurred, and a CT scan wasn’t requested. The patient didn’t meet criteria for a head scan. Even if a scan had been done at the initial visit, it might not have revealed the bleed. Moreover, the patient’s age decreased the likelihood that earlier detection would have changed the outcome.

VERDICT Confidential Utah settlements. The hospital settled for a nominal sum early in the litigation process; the physician settled for a confidential amount immediately before trial.

COMMENT Even when clear guidelines for imaging exist, taking care to weigh extenuating circumstances—in this case, that the patient was on warfarin—is critical.

Failure to document treatment refusal proves costly

A 15-YEAR-OLD BOY lost consciousness at home on Halloween and needed cardiopulmonary resuscitation. When paramedics arrived on the scene, they found the boy conscious and breathing, so they left. The boy, who had a history of drug abuse, died 8 hours later of anoxic encephalopathy caused by cocaine and opiate intoxication.

PLAINTIFF’S CLAIM The paramedics were negligent in failing to evaluate the boy’s condition properly and transport him to a hospital.

THE DEFENSE The paramedics left without assessing the boy because he and his father said they didn’t want or need medical help. (The paramedics neglected to obtain signed refusal of treatment forms.)

VERDICT $5.1 million Illinois verdict.

COMMENT Here is a $5 million verdict that hinges on the completion of forms for refusal of treatment, a remarkable reminder of the importance of documentation.

Enlarging uterus goes uninvestigated

AT AN ANNUAL GYNECOLOGIC EXAMINATION, a woman’s physician noted that her uterus had enlarged since her last visit and described it as “top size” in the chart. At the patient’s next annual exam 21 months later, the uterus had grown to 14 weeks’ gestational size.

Ten months after that, when the woman returned to her physician complaining of abdominal discomfort, her uterus was larger than at the previous examination. The physician advised her to consider a hysterectomy.

About 2 months later, the patient went to the doctor again because of increasing pelvic pressure. Her uterus was 18 to 20 weeks’ gestational size. The physician ordered an ultrasound, which showed a large mass on each ovary and no fibroids or masses within the uterus. Magnetic resonance imaging confirmed the ultrasound findings.

The doctor referred the woman to an oncological gynecologist. She subsequently underwent an abdominal hysterectomy with bilateral salpingo-oophorectomy and bilateral periaortic lymph node dissection. The pathology report described ovarian cancer with an ominous prognosis.

PLAINTIFF’S CLAIM The plaintiff alleged that the physician was negligent for failing to order testing when he first noticed the abnormal size of the uterus and at the patient’s subsequent visits. Failure to do so at the first exam and subsequent visits was negligent and allowed the cancer to advance instead of allowing for surgery and cure at an early stage.

THE DEFENSE No information about the defense is available.

VERDICT $650,000 Maryland settlement.

COMMENT It’s never a good policy to ignore a changing physical exam without good documentation, including a clear discussion of medical decision making.

Third ED visit isn’t the charm

A 39-YEAR-OLD QUADRIPLEGIC MAN went to the emergency department (ED) complaining of abdominal pain. His history included involvement in a shooting when he was 16, drug abuse, homelessness, and frequent visits to the ED, where the staff knew him to be combative and ignore medical advice. The ED physician who saw the man ordered a radiograph and other testing, then released him without a conclusive diagnosis.

A month later, the man came back to the ED by ambulance, complaining of severe abdominal pain that he’d had for 4 days. Another ED physician saw him but didn’t make a diagnosis. After 4 hours, the hospital discharged the patient by ambulance to stay with family. When the family refused to accept him, the ambulance brought him back to the hospital.

With the involvement of social services, the patient was wheeled across the street to a motel. After about 5 hours, during which the motel staff said the patient was screaming in pain, the staff called an ambulance, which brought the man back to the ED covered with bloody vomit.

The same ED physician who had seen him earlier examined him, along with another ED physician. A fecal impaction was removed manually and a soap suds enema administered. The patient seemed to improve and, after about 7 hours, was released and rolled outside with a taxi voucher.

He said the hospital staff told him he was abusing the hospital’s services and the police would be called if he returned. He was taken to the house of a family member, where he was found dead 4 hours later from a ruptured duodenal ulcer.

PLAINTIFF’S CLAIM The physician who saw the patient at the first ED visit should have diagnosed peptic ulcer disease; the doctors who saw the man at the second and third visits should have diagnosed the ruptured ulcer. The hospital violated the federal Emergency Medical Treatment and Labor Act (EMTALA) by failing to stabilize the patient before discharging him.

THE DEFENSE The patient was stable and improving each time he was discharged. The hospital denied threatening to arrest the patient if he returned to the ED after the third visit.

VERDICT $1.4 million Kentucky verdict. The first trial ended in a mistrial. All defendants except the hospital settled for undisclosed amounts before a second trial, at which the hospital was found to be 15% at fault and a $1.5 million award for punitive damages was assessed against the hospital for violating EMTALA.

The hospital appealed and the matter was returned for trial after a ruling that affirmed everything except the punitive damages. At the third trial, a jury awarded $1.4 million in punitive damages.

COMMENT Most of us have a visceral reaction when faced with a drug abusing, noncompliant patient who frequently shows up at the ED. We must remember that such patients do get sick and that in this case, despite repeated visits to the ED, a tragedy occurred.

When following CT guidelines isn’t enough

AN 86-YEAR-OLD MAN ON WARFARIN FAINTED AND FELL while baby-sitting his great-grandchildren. He had transient neurologic symptoms after collapsing but appeared normal by the time paramedics arrived. He was taken by private vehicle to the hospital, where an emergency department (ED) physician examined him. After tests for a myocardial infarction revealed normal enzymes, electrocardiogram, and chest radiograph, the patient was discharged home.

He returned to the hospital the following day and underwent a computed tomography (CT) scan, which showed a large cerebral hemorrhage. He died soon afterward.

PLAINTIFF’S CLAIM The patient should have had a CT scan during the first ED visit. A scan at that visit would have found the hemorrhage in time to save the patient’s life.

THE DEFENSE No discussion with family members about a blow to the head or head trauma occurred, and a CT scan wasn’t requested. The patient didn’t meet criteria for a head scan. Even if a scan had been done at the initial visit, it might not have revealed the bleed. Moreover, the patient’s age decreased the likelihood that earlier detection would have changed the outcome.

VERDICT Confidential Utah settlements. The hospital settled for a nominal sum early in the litigation process; the physician settled for a confidential amount immediately before trial.

COMMENT Even when clear guidelines for imaging exist, taking care to weigh extenuating circumstances—in this case, that the patient was on warfarin—is critical.

Failure to document treatment refusal proves costly

A 15-YEAR-OLD BOY lost consciousness at home on Halloween and needed cardiopulmonary resuscitation. When paramedics arrived on the scene, they found the boy conscious and breathing, so they left. The boy, who had a history of drug abuse, died 8 hours later of anoxic encephalopathy caused by cocaine and opiate intoxication.

PLAINTIFF’S CLAIM The paramedics were negligent in failing to evaluate the boy’s condition properly and transport him to a hospital.

THE DEFENSE The paramedics left without assessing the boy because he and his father said they didn’t want or need medical help. (The paramedics neglected to obtain signed refusal of treatment forms.)

VERDICT $5.1 million Illinois verdict.

COMMENT Here is a $5 million verdict that hinges on the completion of forms for refusal of treatment, a remarkable reminder of the importance of documentation.

Enlarging uterus goes uninvestigated

AT AN ANNUAL GYNECOLOGIC EXAMINATION, a woman’s physician noted that her uterus had enlarged since her last visit and described it as “top size” in the chart. At the patient’s next annual exam 21 months later, the uterus had grown to 14 weeks’ gestational size.

Ten months after that, when the woman returned to her physician complaining of abdominal discomfort, her uterus was larger than at the previous examination. The physician advised her to consider a hysterectomy.

About 2 months later, the patient went to the doctor again because of increasing pelvic pressure. Her uterus was 18 to 20 weeks’ gestational size. The physician ordered an ultrasound, which showed a large mass on each ovary and no fibroids or masses within the uterus. Magnetic resonance imaging confirmed the ultrasound findings.

The doctor referred the woman to an oncological gynecologist. She subsequently underwent an abdominal hysterectomy with bilateral salpingo-oophorectomy and bilateral periaortic lymph node dissection. The pathology report described ovarian cancer with an ominous prognosis.

PLAINTIFF’S CLAIM The plaintiff alleged that the physician was negligent for failing to order testing when he first noticed the abnormal size of the uterus and at the patient’s subsequent visits. Failure to do so at the first exam and subsequent visits was negligent and allowed the cancer to advance instead of allowing for surgery and cure at an early stage.

THE DEFENSE No information about the defense is available.

VERDICT $650,000 Maryland settlement.

COMMENT It’s never a good policy to ignore a changing physical exam without good documentation, including a clear discussion of medical decision making.

Third ED visit isn’t the charm

A 39-YEAR-OLD QUADRIPLEGIC MAN went to the emergency department (ED) complaining of abdominal pain. His history included involvement in a shooting when he was 16, drug abuse, homelessness, and frequent visits to the ED, where the staff knew him to be combative and ignore medical advice. The ED physician who saw the man ordered a radiograph and other testing, then released him without a conclusive diagnosis.

A month later, the man came back to the ED by ambulance, complaining of severe abdominal pain that he’d had for 4 days. Another ED physician saw him but didn’t make a diagnosis. After 4 hours, the hospital discharged the patient by ambulance to stay with family. When the family refused to accept him, the ambulance brought him back to the hospital.

With the involvement of social services, the patient was wheeled across the street to a motel. After about 5 hours, during which the motel staff said the patient was screaming in pain, the staff called an ambulance, which brought the man back to the ED covered with bloody vomit.

The same ED physician who had seen him earlier examined him, along with another ED physician. A fecal impaction was removed manually and a soap suds enema administered. The patient seemed to improve and, after about 7 hours, was released and rolled outside with a taxi voucher.

He said the hospital staff told him he was abusing the hospital’s services and the police would be called if he returned. He was taken to the house of a family member, where he was found dead 4 hours later from a ruptured duodenal ulcer.

PLAINTIFF’S CLAIM The physician who saw the patient at the first ED visit should have diagnosed peptic ulcer disease; the doctors who saw the man at the second and third visits should have diagnosed the ruptured ulcer. The hospital violated the federal Emergency Medical Treatment and Labor Act (EMTALA) by failing to stabilize the patient before discharging him.

THE DEFENSE The patient was stable and improving each time he was discharged. The hospital denied threatening to arrest the patient if he returned to the ED after the third visit.

VERDICT $1.4 million Kentucky verdict. The first trial ended in a mistrial. All defendants except the hospital settled for undisclosed amounts before a second trial, at which the hospital was found to be 15% at fault and a $1.5 million award for punitive damages was assessed against the hospital for violating EMTALA.

The hospital appealed and the matter was returned for trial after a ruling that affirmed everything except the punitive damages. At the third trial, a jury awarded $1.4 million in punitive damages.

COMMENT Most of us have a visceral reaction when faced with a drug abusing, noncompliant patient who frequently shows up at the ED. We must remember that such patients do get sick and that in this case, despite repeated visits to the ED, a tragedy occurred.

When following CT guidelines isn’t enough

AN 86-YEAR-OLD MAN ON WARFARIN FAINTED AND FELL while baby-sitting his great-grandchildren. He had transient neurologic symptoms after collapsing but appeared normal by the time paramedics arrived. He was taken by private vehicle to the hospital, where an emergency department (ED) physician examined him. After tests for a myocardial infarction revealed normal enzymes, electrocardiogram, and chest radiograph, the patient was discharged home.

He returned to the hospital the following day and underwent a computed tomography (CT) scan, which showed a large cerebral hemorrhage. He died soon afterward.

PLAINTIFF’S CLAIM The patient should have had a CT scan during the first ED visit. A scan at that visit would have found the hemorrhage in time to save the patient’s life.

THE DEFENSE No discussion with family members about a blow to the head or head trauma occurred, and a CT scan wasn’t requested. The patient didn’t meet criteria for a head scan. Even if a scan had been done at the initial visit, it might not have revealed the bleed. Moreover, the patient’s age decreased the likelihood that earlier detection would have changed the outcome.

VERDICT Confidential Utah settlements. The hospital settled for a nominal sum early in the litigation process; the physician settled for a confidential amount immediately before trial.

COMMENT Even when clear guidelines for imaging exist, taking care to weigh extenuating circumstances—in this case, that the patient was on warfarin—is critical.

Failure to document treatment refusal proves costly

A 15-YEAR-OLD BOY lost consciousness at home on Halloween and needed cardiopulmonary resuscitation. When paramedics arrived on the scene, they found the boy conscious and breathing, so they left. The boy, who had a history of drug abuse, died 8 hours later of anoxic encephalopathy caused by cocaine and opiate intoxication.

PLAINTIFF’S CLAIM The paramedics were negligent in failing to evaluate the boy’s condition properly and transport him to a hospital.

THE DEFENSE The paramedics left without assessing the boy because he and his father said they didn’t want or need medical help. (The paramedics neglected to obtain signed refusal of treatment forms.)

VERDICT $5.1 million Illinois verdict.

COMMENT Here is a $5 million verdict that hinges on the completion of forms for refusal of treatment, a remarkable reminder of the importance of documentation.

Enlarging uterus goes uninvestigated

AT AN ANNUAL GYNECOLOGIC EXAMINATION, a woman’s physician noted that her uterus had enlarged since her last visit and described it as “top size” in the chart. At the patient’s next annual exam 21 months later, the uterus had grown to 14 weeks’ gestational size.

Ten months after that, when the woman returned to her physician complaining of abdominal discomfort, her uterus was larger than at the previous examination. The physician advised her to consider a hysterectomy.

About 2 months later, the patient went to the doctor again because of increasing pelvic pressure. Her uterus was 18 to 20 weeks’ gestational size. The physician ordered an ultrasound, which showed a large mass on each ovary and no fibroids or masses within the uterus. Magnetic resonance imaging confirmed the ultrasound findings.

The doctor referred the woman to an oncological gynecologist. She subsequently underwent an abdominal hysterectomy with bilateral salpingo-oophorectomy and bilateral periaortic lymph node dissection. The pathology report described ovarian cancer with an ominous prognosis.

PLAINTIFF’S CLAIM The plaintiff alleged that the physician was negligent for failing to order testing when he first noticed the abnormal size of the uterus and at the patient’s subsequent visits. Failure to do so at the first exam and subsequent visits was negligent and allowed the cancer to advance instead of allowing for surgery and cure at an early stage.

THE DEFENSE No information about the defense is available.

VERDICT $650,000 Maryland settlement.

COMMENT It’s never a good policy to ignore a changing physical exam without good documentation, including a clear discussion of medical decision making.

Third ED visit isn’t the charm

A 39-YEAR-OLD QUADRIPLEGIC MAN went to the emergency department (ED) complaining of abdominal pain. His history included involvement in a shooting when he was 16, drug abuse, homelessness, and frequent visits to the ED, where the staff knew him to be combative and ignore medical advice. The ED physician who saw the man ordered a radiograph and other testing, then released him without a conclusive diagnosis.

A month later, the man came back to the ED by ambulance, complaining of severe abdominal pain that he’d had for 4 days. Another ED physician saw him but didn’t make a diagnosis. After 4 hours, the hospital discharged the patient by ambulance to stay with family. When the family refused to accept him, the ambulance brought him back to the hospital.

With the involvement of social services, the patient was wheeled across the street to a motel. After about 5 hours, during which the motel staff said the patient was screaming in pain, the staff called an ambulance, which brought the man back to the ED covered with bloody vomit.

The same ED physician who had seen him earlier examined him, along with another ED physician. A fecal impaction was removed manually and a soap suds enema administered. The patient seemed to improve and, after about 7 hours, was released and rolled outside with a taxi voucher.

He said the hospital staff told him he was abusing the hospital’s services and the police would be called if he returned. He was taken to the house of a family member, where he was found dead 4 hours later from a ruptured duodenal ulcer.

PLAINTIFF’S CLAIM The physician who saw the patient at the first ED visit should have diagnosed peptic ulcer disease; the doctors who saw the man at the second and third visits should have diagnosed the ruptured ulcer. The hospital violated the federal Emergency Medical Treatment and Labor Act (EMTALA) by failing to stabilize the patient before discharging him.

THE DEFENSE The patient was stable and improving each time he was discharged. The hospital denied threatening to arrest the patient if he returned to the ED after the third visit.

VERDICT $1.4 million Kentucky verdict. The first trial ended in a mistrial. All defendants except the hospital settled for undisclosed amounts before a second trial, at which the hospital was found to be 15% at fault and a $1.5 million award for punitive damages was assessed against the hospital for violating EMTALA.

The hospital appealed and the matter was returned for trial after a ruling that affirmed everything except the punitive damages. At the third trial, a jury awarded $1.4 million in punitive damages.

COMMENT Most of us have a visceral reaction when faced with a drug abusing, noncompliant patient who frequently shows up at the ED. We must remember that such patients do get sick and that in this case, despite repeated visits to the ED, a tragedy occurred.

ILLUSTRATIVE CASE

A 13-year-old boy comes in to your office for follow-up of anterior knee pain from OSD that has not responded to 2 months of physical therapy. he is still unable to play on his recreational soccer team. What treatment can you offer to help him return to the sport he enjoys?

OSD is characterized by inflammation of the growth plate just below the knee, the result of repetitive strain on the secondary ossification center of the tibial tuberosity.² Closure of the tibial growth plate is the definitive remedy for OSD, but the pain that some adolescents experience until that happens can be long-lasting and considerable. Nine years after diagnosis of OSD, one study found, up to 60% of patients who had received conservative treatment reported pain on kneeling and 18% had sports limitations.³

Inability to play may affect self-esteem
Adolescents whose recreational activities are limited due to OSD may experience a number of negative effects, including alienation from friends, altered peer group dynamics, and a decline in self-esteem. Surgery, which involves excision of the pain-producing ossicle with or without tuberculoplasty, relieves the pain and allows patients to return to their chosen sport in 90% to 95% of cases that have not responded to conservative treatment.^4,5 For a self-limiting (although prolonged) condition like OSD, most physicians and patients would prefer to avoid surgery and opt for a more conservative approach.

Dextrose injections have been shown to be safe and effective when used for the treatment of tendon and ligamentous disorders such as Achilles tendonitis and lateral epicondylitis, although the mechanism of action is not clear.^6,7 The study detailed in this PURL is the first prospective RCT of dextrose injections for the treatment of OSD.

STUDY SUMMARY: injections get adolescents back in the game

Topol et al¹ sought to compare the efficacy of injections of dextrose and lidocaine with lidocaine-only injections or supervised usual care in treating OSD in young athletes. Sixty-six Argentinian boys and girls ages 9 to 17 years, all of whom had anterior knee pain and participated in kicking or jumping sports on organized teams, were considered for the study. The absence of either patellofemoral crepitus or proximal patellar tendon tenderness was a prerequisite for participation, as was reproduction of the anterior knee pain and localization of pain precisely to the tibial tuberosity during a single leg squat to confirm the OSD diagnosis.

After diagnosis, the patients completed ≥2 months of formal and gently progressive hamstring stretching, quads strengthening, and gradual reintroduction into their respective sports. Those who experienced pain during team play that persisted for ≥3 months—54 patients, all but 3 of whom were male, with a total of 65 knees requiring treatment—were included in the study. Participants were randomized to the usual care group or to one of the injection groups, which was blinded to patients, guardians, and physicians.

The injection groups received a solution of lidocaine 1%, alone or with 12.5% dextrose, at the start of the study and again at 1 and 2 months. Adequate injection was determined by complete pain relief during a single leg squat, which was also used to determine both proximal and distal points of tenderness. Both injection groups received 0.5-mL injections with a 27-gauge needle, repeated at approximately 1-cm intervals for a total of 3 to 4 midline injections. After 5 minutes, the leg squat was repeated to detect any remaining pain, and painful areas were injected until the patient could do a pain-free leg squat.

Because pain reduction may precede full healing, those in both the lidocaine-only and the dextrose-lidocaine groups received injections on all 3 occasions even if they were pain free. They were instructed to avoid running for a week after the initial treatment and then to run as tolerated. Subsequent treatments required a 3-day rest from running. Participants were able to return to their sport after the second injection and rest period.

Patients in all 3 groups received handouts explaining hamstring stretches and quadriceps strengthening exercises. The usual care group received individual instruction from a physical therapist. They were also given a video and returned at least once, both to ensure that they were performing the exercises correctly and to encourage compliance.

The primary outcome involved the Nirschl Pain Phase Scale (NPPS), a 7-point measure of sports-related symptoms and level of participation. Scores of 4 to 7 represent sports limitation resulting from pain. Scores <4 (which may involve soreness or pain but participation in the sport is unlimited) and 0 (asymptomatic participation) were the threshold goals for the study.

The groups were similar at baseline, and follow-up was 100%. At 3 months, NPPS scores improved more in dextrose-treated knees than in either the lidocaine-treated knees (3.9 vs 2.4; P=.004) or those who received usual care (3.9 vs 1.2; P=.001), and lidocaine alone was significantly better than usual care (2.4 vs 1.2; P=.024). More than 90% of participants in both injection groups achieved unlimited sports participation by 3 months. However, knees treated with dextrose were significantly more likely than lidocaine-treated knees to allow asymptomatic participation (NPPS=0), with 14 of 21 knees and 5 of 22 knees, respectively, being pain-free. After one year, more dextrose-treated knees than lidocaine-treated knees were asymptomatic with sports participation (32 of 38 vs 6 of 13; P=.024).

There were no reported adverse effects during this study and fewer than 10% of subjects required acetaminophen for postinjection pain control.

WHAT’S NEW: OSD can be safely and effectively treated

This study found dextrose injections to be safe, well tolerated, and effective in treating patients with intractable OSD symptoms. The results suggest that the duration of both the sports limitation and sports-related symptoms may be reduced with dextrose injections in adolescent athletes with recalcitrant OSD.

CAVEATS: Lack of validated measure, controls

NPPS is not an ideal measure of OSD symptoms because it has not been validated. The failure to use a validated measure of tendinopathy symptoms (eg, the Victorian Institute of Sport Assessment-Patella⁸) is a significant limitation of this trial. The athletes included in this study had already failed to respond to the usual treatment, which suggests that injections should be reserved for those who have tried hamstring stretching and quad-strengthening exercises.

CHALLENGES TO IMPLEMENTATION: Patient and provider comfort may be an issue

Although the injections in this study were well tolerated, there is a risk of infection, bleeding, and pain with any injection or invasive procedure. In addition, adolescents often have difficulty tolerating injections, especially repeated needlesticks like those called for in the proposed treatment. The nonviscous nature of dextrose allows 27- to 30-gauge needles to be used, which may make the injections easier for teens to tolerate. Some physicians may be hesitant to start these young patients on a new injectable therapy.

Click here to view PURL METHODOLOGY

ILLUSTRATIVE CASE

A 13-year-old boy comes in to your office for follow-up of anterior knee pain from OSD that has not responded to 2 months of physical therapy. he is still unable to play on his recreational soccer team. What treatment can you offer to help him return to the sport he enjoys?

OSD is characterized by inflammation of the growth plate just below the knee, the result of repetitive strain on the secondary ossification center of the tibial tuberosity.² Closure of the tibial growth plate is the definitive remedy for OSD, but the pain that some adolescents experience until that happens can be long-lasting and considerable. Nine years after diagnosis of OSD, one study found, up to 60% of patients who had received conservative treatment reported pain on kneeling and 18% had sports limitations.³

Inability to play may affect self-esteem
Adolescents whose recreational activities are limited due to OSD may experience a number of negative effects, including alienation from friends, altered peer group dynamics, and a decline in self-esteem. Surgery, which involves excision of the pain-producing ossicle with or without tuberculoplasty, relieves the pain and allows patients to return to their chosen sport in 90% to 95% of cases that have not responded to conservative treatment.^4,5 For a self-limiting (although prolonged) condition like OSD, most physicians and patients would prefer to avoid surgery and opt for a more conservative approach.

Dextrose injections have been shown to be safe and effective when used for the treatment of tendon and ligamentous disorders such as Achilles tendonitis and lateral epicondylitis, although the mechanism of action is not clear.^6,7 The study detailed in this PURL is the first prospective RCT of dextrose injections for the treatment of OSD.

STUDY SUMMARY: injections get adolescents back in the game

Topol et al¹ sought to compare the efficacy of injections of dextrose and lidocaine with lidocaine-only injections or supervised usual care in treating OSD in young athletes. Sixty-six Argentinian boys and girls ages 9 to 17 years, all of whom had anterior knee pain and participated in kicking or jumping sports on organized teams, were considered for the study. The absence of either patellofemoral crepitus or proximal patellar tendon tenderness was a prerequisite for participation, as was reproduction of the anterior knee pain and localization of pain precisely to the tibial tuberosity during a single leg squat to confirm the OSD diagnosis.

After diagnosis, the patients completed ≥2 months of formal and gently progressive hamstring stretching, quads strengthening, and gradual reintroduction into their respective sports. Those who experienced pain during team play that persisted for ≥3 months—54 patients, all but 3 of whom were male, with a total of 65 knees requiring treatment—were included in the study. Participants were randomized to the usual care group or to one of the injection groups, which was blinded to patients, guardians, and physicians.

The injection groups received a solution of lidocaine 1%, alone or with 12.5% dextrose, at the start of the study and again at 1 and 2 months. Adequate injection was determined by complete pain relief during a single leg squat, which was also used to determine both proximal and distal points of tenderness. Both injection groups received 0.5-mL injections with a 27-gauge needle, repeated at approximately 1-cm intervals for a total of 3 to 4 midline injections. After 5 minutes, the leg squat was repeated to detect any remaining pain, and painful areas were injected until the patient could do a pain-free leg squat.

Because pain reduction may precede full healing, those in both the lidocaine-only and the dextrose-lidocaine groups received injections on all 3 occasions even if they were pain free. They were instructed to avoid running for a week after the initial treatment and then to run as tolerated. Subsequent treatments required a 3-day rest from running. Participants were able to return to their sport after the second injection and rest period.

Patients in all 3 groups received handouts explaining hamstring stretches and quadriceps strengthening exercises. The usual care group received individual instruction from a physical therapist. They were also given a video and returned at least once, both to ensure that they were performing the exercises correctly and to encourage compliance.

The primary outcome involved the Nirschl Pain Phase Scale (NPPS), a 7-point measure of sports-related symptoms and level of participation. Scores of 4 to 7 represent sports limitation resulting from pain. Scores <4 (which may involve soreness or pain but participation in the sport is unlimited) and 0 (asymptomatic participation) were the threshold goals for the study.

The groups were similar at baseline, and follow-up was 100%. At 3 months, NPPS scores improved more in dextrose-treated knees than in either the lidocaine-treated knees (3.9 vs 2.4; P=.004) or those who received usual care (3.9 vs 1.2; P=.001), and lidocaine alone was significantly better than usual care (2.4 vs 1.2; P=.024). More than 90% of participants in both injection groups achieved unlimited sports participation by 3 months. However, knees treated with dextrose were significantly more likely than lidocaine-treated knees to allow asymptomatic participation (NPPS=0), with 14 of 21 knees and 5 of 22 knees, respectively, being pain-free. After one year, more dextrose-treated knees than lidocaine-treated knees were asymptomatic with sports participation (32 of 38 vs 6 of 13; P=.024).

There were no reported adverse effects during this study and fewer than 10% of subjects required acetaminophen for postinjection pain control.

WHAT’S NEW: OSD can be safely and effectively treated

This study found dextrose injections to be safe, well tolerated, and effective in treating patients with intractable OSD symptoms. The results suggest that the duration of both the sports limitation and sports-related symptoms may be reduced with dextrose injections in adolescent athletes with recalcitrant OSD.

CAVEATS: Lack of validated measure, controls

NPPS is not an ideal measure of OSD symptoms because it has not been validated. The failure to use a validated measure of tendinopathy symptoms (eg, the Victorian Institute of Sport Assessment-Patella⁸) is a significant limitation of this trial. The athletes included in this study had already failed to respond to the usual treatment, which suggests that injections should be reserved for those who have tried hamstring stretching and quad-strengthening exercises.

CHALLENGES TO IMPLEMENTATION: Patient and provider comfort may be an issue

Although the injections in this study were well tolerated, there is a risk of infection, bleeding, and pain with any injection or invasive procedure. In addition, adolescents often have difficulty tolerating injections, especially repeated needlesticks like those called for in the proposed treatment. The nonviscous nature of dextrose allows 27- to 30-gauge needles to be used, which may make the injections easier for teens to tolerate. Some physicians may be hesitant to start these young patients on a new injectable therapy.

Click here to view PURL METHODOLOGY

ILLUSTRATIVE CASE

A 13-year-old boy comes in to your office for follow-up of anterior knee pain from OSD that has not responded to 2 months of physical therapy. he is still unable to play on his recreational soccer team. What treatment can you offer to help him return to the sport he enjoys?

OSD is characterized by inflammation of the growth plate just below the knee, the result of repetitive strain on the secondary ossification center of the tibial tuberosity.² Closure of the tibial growth plate is the definitive remedy for OSD, but the pain that some adolescents experience until that happens can be long-lasting and considerable. Nine years after diagnosis of OSD, one study found, up to 60% of patients who had received conservative treatment reported pain on kneeling and 18% had sports limitations.³

Inability to play may affect self-esteem
Adolescents whose recreational activities are limited due to OSD may experience a number of negative effects, including alienation from friends, altered peer group dynamics, and a decline in self-esteem. Surgery, which involves excision of the pain-producing ossicle with or without tuberculoplasty, relieves the pain and allows patients to return to their chosen sport in 90% to 95% of cases that have not responded to conservative treatment.^4,5 For a self-limiting (although prolonged) condition like OSD, most physicians and patients would prefer to avoid surgery and opt for a more conservative approach.

Dextrose injections have been shown to be safe and effective when used for the treatment of tendon and ligamentous disorders such as Achilles tendonitis and lateral epicondylitis, although the mechanism of action is not clear.^6,7 The study detailed in this PURL is the first prospective RCT of dextrose injections for the treatment of OSD.

STUDY SUMMARY: injections get adolescents back in the game

Topol et al¹ sought to compare the efficacy of injections of dextrose and lidocaine with lidocaine-only injections or supervised usual care in treating OSD in young athletes. Sixty-six Argentinian boys and girls ages 9 to 17 years, all of whom had anterior knee pain and participated in kicking or jumping sports on organized teams, were considered for the study. The absence of either patellofemoral crepitus or proximal patellar tendon tenderness was a prerequisite for participation, as was reproduction of the anterior knee pain and localization of pain precisely to the tibial tuberosity during a single leg squat to confirm the OSD diagnosis.

After diagnosis, the patients completed ≥2 months of formal and gently progressive hamstring stretching, quads strengthening, and gradual reintroduction into their respective sports. Those who experienced pain during team play that persisted for ≥3 months—54 patients, all but 3 of whom were male, with a total of 65 knees requiring treatment—were included in the study. Participants were randomized to the usual care group or to one of the injection groups, which was blinded to patients, guardians, and physicians.

The injection groups received a solution of lidocaine 1%, alone or with 12.5% dextrose, at the start of the study and again at 1 and 2 months. Adequate injection was determined by complete pain relief during a single leg squat, which was also used to determine both proximal and distal points of tenderness. Both injection groups received 0.5-mL injections with a 27-gauge needle, repeated at approximately 1-cm intervals for a total of 3 to 4 midline injections. After 5 minutes, the leg squat was repeated to detect any remaining pain, and painful areas were injected until the patient could do a pain-free leg squat.

Because pain reduction may precede full healing, those in both the lidocaine-only and the dextrose-lidocaine groups received injections on all 3 occasions even if they were pain free. They were instructed to avoid running for a week after the initial treatment and then to run as tolerated. Subsequent treatments required a 3-day rest from running. Participants were able to return to their sport after the second injection and rest period.

Patients in all 3 groups received handouts explaining hamstring stretches and quadriceps strengthening exercises. The usual care group received individual instruction from a physical therapist. They were also given a video and returned at least once, both to ensure that they were performing the exercises correctly and to encourage compliance.

The primary outcome involved the Nirschl Pain Phase Scale (NPPS), a 7-point measure of sports-related symptoms and level of participation. Scores of 4 to 7 represent sports limitation resulting from pain. Scores <4 (which may involve soreness or pain but participation in the sport is unlimited) and 0 (asymptomatic participation) were the threshold goals for the study.

The groups were similar at baseline, and follow-up was 100%. At 3 months, NPPS scores improved more in dextrose-treated knees than in either the lidocaine-treated knees (3.9 vs 2.4; P=.004) or those who received usual care (3.9 vs 1.2; P=.001), and lidocaine alone was significantly better than usual care (2.4 vs 1.2; P=.024). More than 90% of participants in both injection groups achieved unlimited sports participation by 3 months. However, knees treated with dextrose were significantly more likely than lidocaine-treated knees to allow asymptomatic participation (NPPS=0), with 14 of 21 knees and 5 of 22 knees, respectively, being pain-free. After one year, more dextrose-treated knees than lidocaine-treated knees were asymptomatic with sports participation (32 of 38 vs 6 of 13; P=.024).

There were no reported adverse effects during this study and fewer than 10% of subjects required acetaminophen for postinjection pain control.

WHAT’S NEW: OSD can be safely and effectively treated

This study found dextrose injections to be safe, well tolerated, and effective in treating patients with intractable OSD symptoms. The results suggest that the duration of both the sports limitation and sports-related symptoms may be reduced with dextrose injections in adolescent athletes with recalcitrant OSD.

CAVEATS: Lack of validated measure, controls

NPPS is not an ideal measure of OSD symptoms because it has not been validated. The failure to use a validated measure of tendinopathy symptoms (eg, the Victorian Institute of Sport Assessment-Patella⁸) is a significant limitation of this trial. The athletes included in this study had already failed to respond to the usual treatment, which suggests that injections should be reserved for those who have tried hamstring stretching and quad-strengthening exercises.

CHALLENGES TO IMPLEMENTATION: Patient and provider comfort may be an issue

Although the injections in this study were well tolerated, there is a risk of infection, bleeding, and pain with any injection or invasive procedure. In addition, adolescents often have difficulty tolerating injections, especially repeated needlesticks like those called for in the proposed treatment. The nonviscous nature of dextrose allows 27- to 30-gauge needles to be used, which may make the injections easier for teens to tolerate. Some physicians may be hesitant to start these young patients on a new injectable therapy.

Click here to view PURL METHODOLOGY

Evidence summary

An RCT found that MVM supplementation for one academic year didn’t improve academic achievement more than placebo in 640 children, 8 to 12 years of age, from low-income urban families.¹ Scores on the Terra Nova academic achievement test of reading, math, language, science, and social sciences didn’t differ between students taking MVM supplements or placebo.

Another RCT that compared MVM supplementation with placebo among 245 children between 6 and 12 years of age found no clinically significant improvements in IQ scores overall. However, within a small subset, more children who took MVM showed a clinical increase in IQ than children who took placebo.²

Investigators randomized children to daily MVM supplementation (50% of the US recommended daily allowance) or placebo for 3 months, then measured their Wechsler IQ scores. Overall, the MVM group scored 2.5 points higher (95% confidence interval [CI], 1.85-3.15) than the placebo group (a 15-point change is clinically significant).

More children taking MVM supplements (44) than placebo (25) showed increases in nonverbal IQ scores of 15 or more points (35% compared with 21%; P<.01). The authors speculate that this result may be attributable to the fact that one in 7 schoolchildren was undernourished. A major weakness of the study was its 16% attrition rate.

Fortified milk reduced disease in young children in India

A community-based, double-blind RCT found that milk fortified with vitamins A, C, and E plus minerals reduced common illnesses over the course of a year more than unfortified milk among 633 children 1 to 3 years of age living in a peri-urban area of India.³

Children who drank fortified milk had fewer days of fever (9.1 compared with 9.7 days for placebo; P=.005), a lower incidence of diarrhea (odds ratio [OR]=0.82; 95% CI, 0.73-0.93), and a decreased rate of lower respiratory illness (OR=0.74; 95% CI, 0.57-0.97). Children 2 years and younger showed the greatest effect.

Asthma and food allergies: The data are mixed

An inception cohort study found an association between early MV use and a higher risk of asthma and food allergies.⁴ Investigators evaluated more than 8000 American women and their newborns over the first 3 years of life. The study population included more families with low socioeconomic status (50%), blacks (51%), and infants born before 37 weeks’ gestation (23%) than the general US population.

Exclusively formula-fed infants who took MV in the first 6 months were more likely to develop asthma (OR=1.27; 95% CI, 1.04-1.56) and food allergies (OR=1.6; 95% CI, 1.2-2.2) than formula-fed infants who didn’t take MV.

However, a birth cohort study of 2470 Swedish children that analyzed health data from parental questionnaires and compared serum immunoglobulin E (IgE) concentrations at 8 years of age found no association between MV use within the past 12 months and clinical allergic disease or specific IgE concentrations.⁵ Children who took MV at age 4 years or earlier had lower rates of IgE sensitization to food allergens at 8 years (OR=0.61; 95% CI, 0.39-0.97).

Recommendations

According to the American Academy of Pediatrics Committee on Nutrition,⁶ healthy children who are growing normally and consume a varied diet don’t need routine supplementation with vitamins and minerals. The Committee states that if parents wish to give their children supplements, a standard pediatric multivitamin generally poses no risk.

Evidence summary

An RCT found that MVM supplementation for one academic year didn’t improve academic achievement more than placebo in 640 children, 8 to 12 years of age, from low-income urban families.¹ Scores on the Terra Nova academic achievement test of reading, math, language, science, and social sciences didn’t differ between students taking MVM supplements or placebo.

Another RCT that compared MVM supplementation with placebo among 245 children between 6 and 12 years of age found no clinically significant improvements in IQ scores overall. However, within a small subset, more children who took MVM showed a clinical increase in IQ than children who took placebo.²

Investigators randomized children to daily MVM supplementation (50% of the US recommended daily allowance) or placebo for 3 months, then measured their Wechsler IQ scores. Overall, the MVM group scored 2.5 points higher (95% confidence interval [CI], 1.85-3.15) than the placebo group (a 15-point change is clinically significant).

More children taking MVM supplements (44) than placebo (25) showed increases in nonverbal IQ scores of 15 or more points (35% compared with 21%; P<.01). The authors speculate that this result may be attributable to the fact that one in 7 schoolchildren was undernourished. A major weakness of the study was its 16% attrition rate.

Fortified milk reduced disease in young children in India

A community-based, double-blind RCT found that milk fortified with vitamins A, C, and E plus minerals reduced common illnesses over the course of a year more than unfortified milk among 633 children 1 to 3 years of age living in a peri-urban area of India.³

Children who drank fortified milk had fewer days of fever (9.1 compared with 9.7 days for placebo; P=.005), a lower incidence of diarrhea (odds ratio [OR]=0.82; 95% CI, 0.73-0.93), and a decreased rate of lower respiratory illness (OR=0.74; 95% CI, 0.57-0.97). Children 2 years and younger showed the greatest effect.

Asthma and food allergies: The data are mixed

An inception cohort study found an association between early MV use and a higher risk of asthma and food allergies.⁴ Investigators evaluated more than 8000 American women and their newborns over the first 3 years of life. The study population included more families with low socioeconomic status (50%), blacks (51%), and infants born before 37 weeks’ gestation (23%) than the general US population.

Exclusively formula-fed infants who took MV in the first 6 months were more likely to develop asthma (OR=1.27; 95% CI, 1.04-1.56) and food allergies (OR=1.6; 95% CI, 1.2-2.2) than formula-fed infants who didn’t take MV.

However, a birth cohort study of 2470 Swedish children that analyzed health data from parental questionnaires and compared serum immunoglobulin E (IgE) concentrations at 8 years of age found no association between MV use within the past 12 months and clinical allergic disease or specific IgE concentrations.⁵ Children who took MV at age 4 years or earlier had lower rates of IgE sensitization to food allergens at 8 years (OR=0.61; 95% CI, 0.39-0.97).

Recommendations

According to the American Academy of Pediatrics Committee on Nutrition,⁶ healthy children who are growing normally and consume a varied diet don’t need routine supplementation with vitamins and minerals. The Committee states that if parents wish to give their children supplements, a standard pediatric multivitamin generally poses no risk.

Evidence summary

An RCT found that MVM supplementation for one academic year didn’t improve academic achievement more than placebo in 640 children, 8 to 12 years of age, from low-income urban families.¹ Scores on the Terra Nova academic achievement test of reading, math, language, science, and social sciences didn’t differ between students taking MVM supplements or placebo.

Another RCT that compared MVM supplementation with placebo among 245 children between 6 and 12 years of age found no clinically significant improvements in IQ scores overall. However, within a small subset, more children who took MVM showed a clinical increase in IQ than children who took placebo.²

Investigators randomized children to daily MVM supplementation (50% of the US recommended daily allowance) or placebo for 3 months, then measured their Wechsler IQ scores. Overall, the MVM group scored 2.5 points higher (95% confidence interval [CI], 1.85-3.15) than the placebo group (a 15-point change is clinically significant).

More children taking MVM supplements (44) than placebo (25) showed increases in nonverbal IQ scores of 15 or more points (35% compared with 21%; P<.01). The authors speculate that this result may be attributable to the fact that one in 7 schoolchildren was undernourished. A major weakness of the study was its 16% attrition rate.

Fortified milk reduced disease in young children in India

A community-based, double-blind RCT found that milk fortified with vitamins A, C, and E plus minerals reduced common illnesses over the course of a year more than unfortified milk among 633 children 1 to 3 years of age living in a peri-urban area of India.³

Children who drank fortified milk had fewer days of fever (9.1 compared with 9.7 days for placebo; P=.005), a lower incidence of diarrhea (odds ratio [OR]=0.82; 95% CI, 0.73-0.93), and a decreased rate of lower respiratory illness (OR=0.74; 95% CI, 0.57-0.97). Children 2 years and younger showed the greatest effect.

Asthma and food allergies: The data are mixed

An inception cohort study found an association between early MV use and a higher risk of asthma and food allergies.⁴ Investigators evaluated more than 8000 American women and their newborns over the first 3 years of life. The study population included more families with low socioeconomic status (50%), blacks (51%), and infants born before 37 weeks’ gestation (23%) than the general US population.

Exclusively formula-fed infants who took MV in the first 6 months were more likely to develop asthma (OR=1.27; 95% CI, 1.04-1.56) and food allergies (OR=1.6; 95% CI, 1.2-2.2) than formula-fed infants who didn’t take MV.

However, a birth cohort study of 2470 Swedish children that analyzed health data from parental questionnaires and compared serum immunoglobulin E (IgE) concentrations at 8 years of age found no association between MV use within the past 12 months and clinical allergic disease or specific IgE concentrations.⁵ Children who took MV at age 4 years or earlier had lower rates of IgE sensitization to food allergens at 8 years (OR=0.61; 95% CI, 0.39-0.97).

Recommendations

According to the American Academy of Pediatrics Committee on Nutrition,⁶ healthy children who are growing normally and consume a varied diet don’t need routine supplementation with vitamins and minerals. The Committee states that if parents wish to give their children supplements, a standard pediatric multivitamin generally poses no risk.

Evidence summary

A 2008 Cochrane review examined 7 RCTs comparing ICS with placebo in 1989 patients 30 to 52 years of age with mild asthma or COPD. The reviewers found no evidence of increased bone turnover, decreased bone mineral density, or increased vertebral fracture in the ICS group compared with the placebo group at 2 to 3 years’ follow-up (odds ratio [OR] for fracture=1.87; 95% confidence interval [CI], 0.5-7.0).

Steroid doses ranged from 200 to 4000 mcg beclomethasone equivalent ICS per day.¹ A 100-mcg beclomethasone equivalent ICS dose is 50 mcg fluticasone, 80 mcg budesonide, or 200 mcg triamcinolone.²

A 2008 meta-analysis of 11 RCTs that examined a number of adverse effects of ICS in adult patients with COPD found 3 studies (8131 patients) that reported no significant increase in fracture risk at 36 months in the ICS group compared with the placebo group (OR=1.09; 95% CI, 0.89-1.33). Steroid doses ranged from 1000 to 2000 mcg beclomethasone equivalent ICS per day.³

Some studies suggest an association between dose and risk
A 2008 meta-analysis that included patients with COPD or asthma, average age 43 to 81 years, showed no difference in fracture risk overall at 1 to 4 years’ follow-up (OR=1.02; 95% CI, 0.96-1.08). This analysis examined 4 RCTs, 6 case-control studies, and 3 cohort studies.⁴

A subgroup analysis of patients taking higher-dose ICS (>1500 mcg beclomethasone equivalent ICS per day) that pooled data from case-control and cohort studies suggested an increased risk of fracture (OR=1.30; 95% CI, 1.07-1.58).⁴

Investigators identified a possible dose-dependent relationship in another meta-analysis of 5 case-control studies (43,783 cases, 259,936 controls).⁵ The meta-analysis included 4 of the studies examined in the previously discussed meta-analysis.⁴

The investigators found a relative risk of 1.12 (95% CI, 1.0-1.26) for nonvertebral fracture for each 1000-mcg increase in beclomethasone equivalent ICS dose per day.⁵ Longer follow-up time wasn’t associated with greater fracture risk.

But the relationship isn’t clear
Although some non-RCT studies discussed here show that higher doses of steroids may lead to increased fracture risk, the strength of this association isn’t clear. The authors of the Cochrane review and the meta-analyses point out that a significant number of confounding factors can put asthma and COPD patients at increased risk for fracture. They include age, smoking status, inactivity, and severity of underlying lung disease. The fact that different authors controlled differently for these factors introduced heterogeneity into the meta-analyses described here.^1,3-5

Recommendations

Guidelines for the Diagnosis and Management of Asthma from the National Heart, Lung, and Blood Institute state that “most benefit is achieved with relatively low doses of ICS, whereas the risk of adverse effects increases with dose. … ICS use may be associated with a dose-dependent reduction in bone mineral content, although low or medium doses appear to have no major adverse effect. Elderly patients may be more at risk due to preexisting osteoporosis, changes in estrogen levels that affect calcium utilization, and a sedentary lifestyle.”⁶

Evidence summary

A 2008 Cochrane review examined 7 RCTs comparing ICS with placebo in 1989 patients 30 to 52 years of age with mild asthma or COPD. The reviewers found no evidence of increased bone turnover, decreased bone mineral density, or increased vertebral fracture in the ICS group compared with the placebo group at 2 to 3 years’ follow-up (odds ratio [OR] for fracture=1.87; 95% confidence interval [CI], 0.5-7.0).

Steroid doses ranged from 200 to 4000 mcg beclomethasone equivalent ICS per day.¹ A 100-mcg beclomethasone equivalent ICS dose is 50 mcg fluticasone, 80 mcg budesonide, or 200 mcg triamcinolone.²

A 2008 meta-analysis of 11 RCTs that examined a number of adverse effects of ICS in adult patients with COPD found 3 studies (8131 patients) that reported no significant increase in fracture risk at 36 months in the ICS group compared with the placebo group (OR=1.09; 95% CI, 0.89-1.33). Steroid doses ranged from 1000 to 2000 mcg beclomethasone equivalent ICS per day.³

Some studies suggest an association between dose and risk
A 2008 meta-analysis that included patients with COPD or asthma, average age 43 to 81 years, showed no difference in fracture risk overall at 1 to 4 years’ follow-up (OR=1.02; 95% CI, 0.96-1.08). This analysis examined 4 RCTs, 6 case-control studies, and 3 cohort studies.⁴

A subgroup analysis of patients taking higher-dose ICS (>1500 mcg beclomethasone equivalent ICS per day) that pooled data from case-control and cohort studies suggested an increased risk of fracture (OR=1.30; 95% CI, 1.07-1.58).⁴

Investigators identified a possible dose-dependent relationship in another meta-analysis of 5 case-control studies (43,783 cases, 259,936 controls).⁵ The meta-analysis included 4 of the studies examined in the previously discussed meta-analysis.⁴

The investigators found a relative risk of 1.12 (95% CI, 1.0-1.26) for nonvertebral fracture for each 1000-mcg increase in beclomethasone equivalent ICS dose per day.⁵ Longer follow-up time wasn’t associated with greater fracture risk.

But the relationship isn’t clear
Although some non-RCT studies discussed here show that higher doses of steroids may lead to increased fracture risk, the strength of this association isn’t clear. The authors of the Cochrane review and the meta-analyses point out that a significant number of confounding factors can put asthma and COPD patients at increased risk for fracture. They include age, smoking status, inactivity, and severity of underlying lung disease. The fact that different authors controlled differently for these factors introduced heterogeneity into the meta-analyses described here.^1,3-5

Recommendations

Guidelines for the Diagnosis and Management of Asthma from the National Heart, Lung, and Blood Institute state that “most benefit is achieved with relatively low doses of ICS, whereas the risk of adverse effects increases with dose. … ICS use may be associated with a dose-dependent reduction in bone mineral content, although low or medium doses appear to have no major adverse effect. Elderly patients may be more at risk due to preexisting osteoporosis, changes in estrogen levels that affect calcium utilization, and a sedentary lifestyle.”⁶

Evidence summary

A 2008 Cochrane review examined 7 RCTs comparing ICS with placebo in 1989 patients 30 to 52 years of age with mild asthma or COPD. The reviewers found no evidence of increased bone turnover, decreased bone mineral density, or increased vertebral fracture in the ICS group compared with the placebo group at 2 to 3 years’ follow-up (odds ratio [OR] for fracture=1.87; 95% confidence interval [CI], 0.5-7.0).

Steroid doses ranged from 200 to 4000 mcg beclomethasone equivalent ICS per day.¹ A 100-mcg beclomethasone equivalent ICS dose is 50 mcg fluticasone, 80 mcg budesonide, or 200 mcg triamcinolone.²

A 2008 meta-analysis of 11 RCTs that examined a number of adverse effects of ICS in adult patients with COPD found 3 studies (8131 patients) that reported no significant increase in fracture risk at 36 months in the ICS group compared with the placebo group (OR=1.09; 95% CI, 0.89-1.33). Steroid doses ranged from 1000 to 2000 mcg beclomethasone equivalent ICS per day.³

Some studies suggest an association between dose and risk
A 2008 meta-analysis that included patients with COPD or asthma, average age 43 to 81 years, showed no difference in fracture risk overall at 1 to 4 years’ follow-up (OR=1.02; 95% CI, 0.96-1.08). This analysis examined 4 RCTs, 6 case-control studies, and 3 cohort studies.⁴

A subgroup analysis of patients taking higher-dose ICS (>1500 mcg beclomethasone equivalent ICS per day) that pooled data from case-control and cohort studies suggested an increased risk of fracture (OR=1.30; 95% CI, 1.07-1.58).⁴

Investigators identified a possible dose-dependent relationship in another meta-analysis of 5 case-control studies (43,783 cases, 259,936 controls).⁵ The meta-analysis included 4 of the studies examined in the previously discussed meta-analysis.⁴

The investigators found a relative risk of 1.12 (95% CI, 1.0-1.26) for nonvertebral fracture for each 1000-mcg increase in beclomethasone equivalent ICS dose per day.⁵ Longer follow-up time wasn’t associated with greater fracture risk.

But the relationship isn’t clear
Although some non-RCT studies discussed here show that higher doses of steroids may lead to increased fracture risk, the strength of this association isn’t clear. The authors of the Cochrane review and the meta-analyses point out that a significant number of confounding factors can put asthma and COPD patients at increased risk for fracture. They include age, smoking status, inactivity, and severity of underlying lung disease. The fact that different authors controlled differently for these factors introduced heterogeneity into the meta-analyses described here.^1,3-5

Recommendations

Guidelines for the Diagnosis and Management of Asthma from the National Heart, Lung, and Blood Institute state that “most benefit is achieved with relatively low doses of ICS, whereas the risk of adverse effects increases with dose. … ICS use may be associated with a dose-dependent reduction in bone mineral content, although low or medium doses appear to have no major adverse effect. Elderly patients may be more at risk due to preexisting osteoporosis, changes in estrogen levels that affect calcium utilization, and a sedentary lifestyle.”⁶

A surge in elderly patients with diabetes has placed a large burden on extended-care facilities. According to the Centers for Medicare and Medicaid Services, the prevalence of diabetes among nursing home residents is 33.3%.¹ Between 1995 and 2004, the estimated number of long-term care residents with diabetes mellitus (DM) grew by 7.1%, from approximately 242,000 to 329,000.² The increase adds to the challenge extended-care facilities face in attempting to provide high-quality care to patients with diabetes. No well-accepted management guidelines exist for nursing home residents with DM.³

Frail older adults with DM are more likely to suffer from cardiovascular conditions than younger patients, and are at greater risk for hypoglycemic coma and serious hyperglycemia.^4,5 A high frequency of hypoglycemia, especially nocturnal hypoglycemia, has been reported among nursing home residents with diabetes.⁶ Intensive insulin therapy is associated with hypoglycemia and increased mortality.⁷ However, hyperglycemia also must be considered because it significantly impairs quality of life. Uncontrolled hyperglycemia causes osmotic diuresis, leading to polyuria, nocturia, aggravated incontinence, and disrupted sleep, as well as contributing to dehydration.⁴ All of these problems have serious implications for quality of life and overall health.

Although studies have identified poor glycemic control and hypertension as the major problems facing nursing home patients with DM,^2,6 little research has examined how therapies targeting these problems help the elderly. Solid evidence supports the effectiveness of controlling hyperglycemia, lipid levels, and blood pressure (BP), along with aspirin therapy, in preventing microvascular disease, but does not reflect research involving older patients.⁸

A study of nursing homes found that health care teams did not respond to half of all significantly abnormal laboratory test results.⁹ Physicians who are aware of the problems associated with DM in elderly patients may hesitate to treat them because of the lack of guidelines for this patient population or concerns about adverse effects. Because of the deficiency of clinical trial data in elderly patients and the heterogenicity of the population, the American Diabetes Association (ADA) suggests that “less stringent treatment goals” may be appropriate.¹⁰

A central conflict in diabetes care for nursing home residents revolves around the need for guidelines that are both generalizable and easily individualized. Some studies support the need for individualized care, particularly with regard to A1c goals, because residents vary greatly in both disease burden and frailty.^8,11 Yet individualized treatment could increase the complexity of care for nurses who must manage many patients, potentially having a negative effect on patient care.

Implementation of a treatment protocol for residents with DM is associated with a decrease in the number of hospital days for acute and chronic complications,¹² but one study found that only 15% of nursing homes had such a protocol.¹³ Ultimately, long-term care facilities may benefit from an approach that strikes a balance between individualized care and generalized goals and does not closely mimic either acute hospital care or outpatient management of diabetes.³

In the absence of specific recommendations for extended-care residents with diabetes, our study evaluated the status of care in this population on the basis of pharmacotherapy and standards of care recommended by the ADA for ambulatory adults with DM.

Methods

Data collection
We reviewed the charts of 245 patients in 14 long-term care facilities in Ohio and West Virginia. All participating facilities signed a letter of agreement to take part in the study. The study was approved by the Ohio University Institutional Review Board.

At each facility, the director of nursing supplied a list of residents with DM. To be eligible for chart review, residents had to have a physician-documented diagnosis of type 1 or type 2 DM and have lived at the facility for at least 3 of the previous 12 months. Residents in both skilled nursing care and assisted living facilities were able to participate; short-term rehabilitation residents were not.

We performed a comprehensive review of each chart, examining the medical diagnoses, medication lists, laboratory reports, and physician and consultation notes for a one-year period. Data collection focused on diabetes-related intermediate outcomes and processes of care. Intermediate outcomes included A1c tests, lipid panels, and BP readings. Processes of care included aspirin therapy, use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers, use of statins, eye exams, foot exams, and microalbumin tests. The data collected omitted information identifying the patient, physician, or facility.

We compared the collected data with the 2011 ADA standards of care: blood glucose (fasting 80-120 mg/dL; postprandial 100-140 mg/dL), A1c (<7%), BP (<130/80 mm Hg), and lipid levels (low-density lipoprotein [LDL] <100 mg/dL; high-density lipoprotein [HDL] >40 mg/dL in men and >50 mg/dL in women; triglycerides <150 mg/dL).

Data analysis
We entered the data into an Excel database by type and key format and analyzed results using SPSS software, version 14.0 (SPSS, Chicago, IL). We used percentages and means±standard deviation to describe the data.

Results

TABLE 1 lists characteristics of the patients in the study: 24.5% were male and 75.5% were female; 9 (3.7%) were diagnosed with type 1 DM; 236 (96.3%) had type 2 DM. The mean age was 81±9 years, with a range of 44 to 103 years. Approximately 96% were Caucasian. The residents’ medical care was managed by family physicians (66.1%), internists (25.7%), geriatricians (6.9%), endocrinologists (0.8%), and other physicians (0.4%). The findings that follow are all based on a one-year period unless otherwise specified.

TABLE 1
Study population profile

Diabetes management
Most of the residents (211 [86.1%]) underwent glucose monitoring. The proportion of residents who received specific diabetes interventions is detailed in TABLE 2.

Hypoglycemia. Fifty-two residents (24.6% of those receiving glucose monitoring and 21.2% of the total) experienced a hypoglycemic event; 103 (representing 48.8% of the monitored patients and 42% of the total) had hyperglycemic events. On average, each resident experienced 1±2 mild hypoglycemic episodes per month, with a maximum of 13 mild episodes for one resident. Severe hypoglycemia (< 50 mg/dL) occurred less often, on average 0.24±1 time per resident. One resident had 15 severe hypoglycemic events in a month. The mean low hypoglycemic episode was at a glucose level of 52±16 mg/dL.

Hyperglycemia occurred more often than hypoglycemia (8±14 times per month), with a mean high glucose level of 352±89 mg/dL. This study used a generous range for normal glucose readings (70-249 mg/dL), and 89% of blood glucose readings were within that range. Thirty-seven percent of residents had an A1c <7.0%.

Medication. Thirty-two (13.1%) patients received no oral medication or insulin, and were managed with lifestyle interventions alone. Sixty-four patients (26.1%) used only oral medications, 64 (26.1%) received only insulin, and 85 (34.7%) were treated with both. Of the patients receiving insulin, 108 (72%) were on a sliding scale regimen. Seventy-seven (51.7%) of the patients on insulin experienced hypoglycemia, vs 30 (20%) of those taking oral medication. Twenty-seven (31.8%) patients in the combined therapy group had hypoglycemic events.

TABLE 2
Interventions received by the study population

Preventive care
Foot and eye care. Dilated eye examinations were provided for 133 residents (54.3%). Most (76.3%) received foot examinations, and 69.4% were seen by a podiatrist.

Blood pressure. Of the 240 residents (98%) whose BP was monitored, 107 (43.7%) had readings lower than 125/85 mm Hg, a goal set by a team of diabetologists, endocrinologists, and geriatricians at Ohio University. One hundred residents (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker; 122 (49.8%) were receiving aspirin therapy. In the total population, 110 patients (44.9%) were prescribed a statin.

Lipid monitoring. Of the 190 residents (77.6%) whose lipids were monitored, only 89 (46.8%) met the LDL goal suggested by the ADA. Fifty-six (29.5%) had triglycerides <150 mg/dL.

The HDL goal recommended by the ADA is >40 mg/dL for men and >50 mg/dL for women. Three of the 24 men and 16 of the 91 women whose lipids were monitored met the HDL goal.

Discussion

Although several components of diabetes management in our study population failed to meet the ADA standards of care for ambulatory adults, some elements of care were well managed. Monthly foot exams were performed on 76.3% of patients; 69.4% were seen by a podiatrist. While the number of residents receiving foot exams had decreased by 10.7% since a previous study by our research group, the number of podiatric consults increased by 11.4%.¹⁴

Dilated eye exams were given to 54.3% of residents. More patients should be given the opportunity to have an annual eye exam. Diabetes is the leading cause of new cases of blindness among adults 20 to 74 years of age,¹⁵ and impaired vision affects patient activity levels, susceptibility to falls, and quality of life.

In addition to a good record of preventive exams, physicians were proficient in monitoring residents with diabetes with regular testing regimens. Eighty-six percent of patients underwent regular blood glucose monitoring; 84% had had their A1c tested in the past year, and 36.7% achieved the A1c goal of <7%. The average A1c reading was 6.7±1%.

While these A1c values would seem to reflect well-managed diabetes, blood glucose readings tell a different story. A comparison of A1c values and hyperglycemic events revealed a disparity between the estimated average glucose reading and the actual readings. Of the patients who underwent scheduled fingersticks, 24.6% experienced a hypoglycemic event and 48.8% had hyperglycemic events. On average, each patient had 8 hyperglycemic episodes per month. The average highest glucose reading was >350 mg/dL.

A1c is only part of the story. While A1c can be a marker of sustained hyperglycemia, it does not reflect the stability of glycemic control.¹⁶ A study by Löfgren and colleagues confirmed that elderly diabetic patients in nursing homes who have low A1c levels often suffer from hypoglycemia.⁶ Patients receiving insulin therapy are more likely to experience hypoglycemia.⁷

The mismatch between A1c and glucose readings reveals an important point about the management of diabetes in long-term care patients: A1c values do not tell the entire story about a patient’s blood glucose; thus, a physician cannot look only at A1c to assess a patient’s diabetes management. A previous study demonstrated that when physicians base treatment plans solely on A1c without consulting glucose logs or being familiar with newer treatments, adherence to evidence-based algorithms is unlikely.¹⁷

While A1c does provide information about average blood glucose levels, it does not offer perspective on hypoglycemia or glucose variability. It is vital that physicians screen the glucose log for evidence of hypo- and hyperglycemia before adjusting the patient’s treatment plan. Physicians must also keep in mind that A1c may be falsely low in elderly patients who have concomitant anemia, which lowers the value.

Controlling BP and lipids helps prevent complications. In addition to diabetes management, our study evaluated regulation of the complications of diabetes, particularly cardiovascular complications. Evidence suggests that people with DM derive the greatest mortality benefit from a treatment plan centered first on hypertension, then lipids, and finally, glycemic control.¹⁸ A renewed focus on the BP and lipid aspects of diabetes care is needed.

Our data demonstrate that, of the 240 patients who met the ADA goal of <130/80 mm Hg, only 100 (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker. Lowering BP to <130/80 mm Hg may provide further benefit in preventing diabetes-related macrovascular complications.⁸

Lipid levels are a critical gauge of cardiovascular risk. Previous studies of patients with type 2 DM have shown that treating hyperlipidemia can produce a mortality benefit within 2 to 4 years, whereas aggressive glucose management takes approximately 8 years.¹⁸

A lipid panel was performed for 77.6% of the patients in our study—an improvement over a previous study by our team in which only 33% of patients received lipid checks.¹⁴ In the current study, a mere 2.9% of patients met the ADA’s combined lipid goals (LDL <100 mg/dL; HDL >40 mg/dL in men and >50 mg/dL in women; and triglycerides<150 mg/dL). Considering the LDL goal alone, 46.8% of the 190 patients whose lipids were monitored achieved it.

One hundred ten (44.9%) of the 245 patients in our study were prescribed a statin. Various studies support the use of lipid-lowering medications to increase HDL in elderly patients with DM.⁸ Yet data suggest that the rate of statin use among older adults is suboptimal.¹⁹ Our study highlights the limited prescription of statins for elderly nursing home patients who need them.

The ADA lipid goals are reasonable for this patient population, especially considering the potential mortality benefit. Rather than adjust lipid and BP goals, standards of care should emphasize the importance of meeting these objectives and suggest means to achieve them, including greater use of statins.

One set of standards does not work for all patients. Our study demonstrates that the ADA standards of care for ambulatory adults with diabetes are not acceptable for long-term care residents with DM. Although stringent A1c goals are appropriate for ambulatory adults, the risk of hypoglycemic episodes among the older and frailer nursing home population is too great to adhere to such a strict approach.

We recommend new guidelines be established. Guidelines developed specifically for residents in extended care are necessary to ensure the proper care of these patients, particularly in the face of a steady increase in their number.

Study limitations. Future inquiries into this subject should take into account the weaknesses of this study. First, it was conducted solely in Ohio and West Virginia. A chart review covering more territory could explore regional differences in diabetes care provided by long-term facilities and provide more evidence of the need for a population-specific standards of care.

The study also failed to account for comorbid conditions, including dementia, and code status, and followed residents for only one year. More extensive reviews could examine the effects of therapy in this patient population and the relationship between mortality and treatment plan, spurring movement toward more uniform and effective care of patients with diabetes in the long-term care setting.

CORRESPONDENCE Jay Shubrook Jr, DO, Department of Family Medicine, Ohio University Heritage College of Osteopathic Medicine, Grosvenor Hall, Athens, OH 45701; [email protected]

A surge in elderly patients with diabetes has placed a large burden on extended-care facilities. According to the Centers for Medicare and Medicaid Services, the prevalence of diabetes among nursing home residents is 33.3%.¹ Between 1995 and 2004, the estimated number of long-term care residents with diabetes mellitus (DM) grew by 7.1%, from approximately 242,000 to 329,000.² The increase adds to the challenge extended-care facilities face in attempting to provide high-quality care to patients with diabetes. No well-accepted management guidelines exist for nursing home residents with DM.³

Frail older adults with DM are more likely to suffer from cardiovascular conditions than younger patients, and are at greater risk for hypoglycemic coma and serious hyperglycemia.^4,5 A high frequency of hypoglycemia, especially nocturnal hypoglycemia, has been reported among nursing home residents with diabetes.⁶ Intensive insulin therapy is associated with hypoglycemia and increased mortality.⁷ However, hyperglycemia also must be considered because it significantly impairs quality of life. Uncontrolled hyperglycemia causes osmotic diuresis, leading to polyuria, nocturia, aggravated incontinence, and disrupted sleep, as well as contributing to dehydration.⁴ All of these problems have serious implications for quality of life and overall health.

Although studies have identified poor glycemic control and hypertension as the major problems facing nursing home patients with DM,^2,6 little research has examined how therapies targeting these problems help the elderly. Solid evidence supports the effectiveness of controlling hyperglycemia, lipid levels, and blood pressure (BP), along with aspirin therapy, in preventing microvascular disease, but does not reflect research involving older patients.⁸

A study of nursing homes found that health care teams did not respond to half of all significantly abnormal laboratory test results.⁹ Physicians who are aware of the problems associated with DM in elderly patients may hesitate to treat them because of the lack of guidelines for this patient population or concerns about adverse effects. Because of the deficiency of clinical trial data in elderly patients and the heterogenicity of the population, the American Diabetes Association (ADA) suggests that “less stringent treatment goals” may be appropriate.¹⁰

A central conflict in diabetes care for nursing home residents revolves around the need for guidelines that are both generalizable and easily individualized. Some studies support the need for individualized care, particularly with regard to A1c goals, because residents vary greatly in both disease burden and frailty.^8,11 Yet individualized treatment could increase the complexity of care for nurses who must manage many patients, potentially having a negative effect on patient care.

Implementation of a treatment protocol for residents with DM is associated with a decrease in the number of hospital days for acute and chronic complications,¹² but one study found that only 15% of nursing homes had such a protocol.¹³ Ultimately, long-term care facilities may benefit from an approach that strikes a balance between individualized care and generalized goals and does not closely mimic either acute hospital care or outpatient management of diabetes.³

In the absence of specific recommendations for extended-care residents with diabetes, our study evaluated the status of care in this population on the basis of pharmacotherapy and standards of care recommended by the ADA for ambulatory adults with DM.

Methods

Data collection
We reviewed the charts of 245 patients in 14 long-term care facilities in Ohio and West Virginia. All participating facilities signed a letter of agreement to take part in the study. The study was approved by the Ohio University Institutional Review Board.

At each facility, the director of nursing supplied a list of residents with DM. To be eligible for chart review, residents had to have a physician-documented diagnosis of type 1 or type 2 DM and have lived at the facility for at least 3 of the previous 12 months. Residents in both skilled nursing care and assisted living facilities were able to participate; short-term rehabilitation residents were not.

We performed a comprehensive review of each chart, examining the medical diagnoses, medication lists, laboratory reports, and physician and consultation notes for a one-year period. Data collection focused on diabetes-related intermediate outcomes and processes of care. Intermediate outcomes included A1c tests, lipid panels, and BP readings. Processes of care included aspirin therapy, use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers, use of statins, eye exams, foot exams, and microalbumin tests. The data collected omitted information identifying the patient, physician, or facility.

We compared the collected data with the 2011 ADA standards of care: blood glucose (fasting 80-120 mg/dL; postprandial 100-140 mg/dL), A1c (<7%), BP (<130/80 mm Hg), and lipid levels (low-density lipoprotein [LDL] <100 mg/dL; high-density lipoprotein [HDL] >40 mg/dL in men and >50 mg/dL in women; triglycerides <150 mg/dL).

Data analysis
We entered the data into an Excel database by type and key format and analyzed results using SPSS software, version 14.0 (SPSS, Chicago, IL). We used percentages and means±standard deviation to describe the data.

Results

TABLE 1 lists characteristics of the patients in the study: 24.5% were male and 75.5% were female; 9 (3.7%) were diagnosed with type 1 DM; 236 (96.3%) had type 2 DM. The mean age was 81±9 years, with a range of 44 to 103 years. Approximately 96% were Caucasian. The residents’ medical care was managed by family physicians (66.1%), internists (25.7%), geriatricians (6.9%), endocrinologists (0.8%), and other physicians (0.4%). The findings that follow are all based on a one-year period unless otherwise specified.

TABLE 1
Study population profile

Diabetes management
Most of the residents (211 [86.1%]) underwent glucose monitoring. The proportion of residents who received specific diabetes interventions is detailed in TABLE 2.

Hypoglycemia. Fifty-two residents (24.6% of those receiving glucose monitoring and 21.2% of the total) experienced a hypoglycemic event; 103 (representing 48.8% of the monitored patients and 42% of the total) had hyperglycemic events. On average, each resident experienced 1±2 mild hypoglycemic episodes per month, with a maximum of 13 mild episodes for one resident. Severe hypoglycemia (< 50 mg/dL) occurred less often, on average 0.24±1 time per resident. One resident had 15 severe hypoglycemic events in a month. The mean low hypoglycemic episode was at a glucose level of 52±16 mg/dL.

Hyperglycemia occurred more often than hypoglycemia (8±14 times per month), with a mean high glucose level of 352±89 mg/dL. This study used a generous range for normal glucose readings (70-249 mg/dL), and 89% of blood glucose readings were within that range. Thirty-seven percent of residents had an A1c <7.0%.

Medication. Thirty-two (13.1%) patients received no oral medication or insulin, and were managed with lifestyle interventions alone. Sixty-four patients (26.1%) used only oral medications, 64 (26.1%) received only insulin, and 85 (34.7%) were treated with both. Of the patients receiving insulin, 108 (72%) were on a sliding scale regimen. Seventy-seven (51.7%) of the patients on insulin experienced hypoglycemia, vs 30 (20%) of those taking oral medication. Twenty-seven (31.8%) patients in the combined therapy group had hypoglycemic events.

TABLE 2
Interventions received by the study population

Preventive care
Foot and eye care. Dilated eye examinations were provided for 133 residents (54.3%). Most (76.3%) received foot examinations, and 69.4% were seen by a podiatrist.

Blood pressure. Of the 240 residents (98%) whose BP was monitored, 107 (43.7%) had readings lower than 125/85 mm Hg, a goal set by a team of diabetologists, endocrinologists, and geriatricians at Ohio University. One hundred residents (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker; 122 (49.8%) were receiving aspirin therapy. In the total population, 110 patients (44.9%) were prescribed a statin.

Lipid monitoring. Of the 190 residents (77.6%) whose lipids were monitored, only 89 (46.8%) met the LDL goal suggested by the ADA. Fifty-six (29.5%) had triglycerides <150 mg/dL.

The HDL goal recommended by the ADA is >40 mg/dL for men and >50 mg/dL for women. Three of the 24 men and 16 of the 91 women whose lipids were monitored met the HDL goal.

Discussion

Although several components of diabetes management in our study population failed to meet the ADA standards of care for ambulatory adults, some elements of care were well managed. Monthly foot exams were performed on 76.3% of patients; 69.4% were seen by a podiatrist. While the number of residents receiving foot exams had decreased by 10.7% since a previous study by our research group, the number of podiatric consults increased by 11.4%.¹⁴

Dilated eye exams were given to 54.3% of residents. More patients should be given the opportunity to have an annual eye exam. Diabetes is the leading cause of new cases of blindness among adults 20 to 74 years of age,¹⁵ and impaired vision affects patient activity levels, susceptibility to falls, and quality of life.

In addition to a good record of preventive exams, physicians were proficient in monitoring residents with diabetes with regular testing regimens. Eighty-six percent of patients underwent regular blood glucose monitoring; 84% had had their A1c tested in the past year, and 36.7% achieved the A1c goal of <7%. The average A1c reading was 6.7±1%.

While these A1c values would seem to reflect well-managed diabetes, blood glucose readings tell a different story. A comparison of A1c values and hyperglycemic events revealed a disparity between the estimated average glucose reading and the actual readings. Of the patients who underwent scheduled fingersticks, 24.6% experienced a hypoglycemic event and 48.8% had hyperglycemic events. On average, each patient had 8 hyperglycemic episodes per month. The average highest glucose reading was >350 mg/dL.

A1c is only part of the story. While A1c can be a marker of sustained hyperglycemia, it does not reflect the stability of glycemic control.¹⁶ A study by Löfgren and colleagues confirmed that elderly diabetic patients in nursing homes who have low A1c levels often suffer from hypoglycemia.⁶ Patients receiving insulin therapy are more likely to experience hypoglycemia.⁷

The mismatch between A1c and glucose readings reveals an important point about the management of diabetes in long-term care patients: A1c values do not tell the entire story about a patient’s blood glucose; thus, a physician cannot look only at A1c to assess a patient’s diabetes management. A previous study demonstrated that when physicians base treatment plans solely on A1c without consulting glucose logs or being familiar with newer treatments, adherence to evidence-based algorithms is unlikely.¹⁷

While A1c does provide information about average blood glucose levels, it does not offer perspective on hypoglycemia or glucose variability. It is vital that physicians screen the glucose log for evidence of hypo- and hyperglycemia before adjusting the patient’s treatment plan. Physicians must also keep in mind that A1c may be falsely low in elderly patients who have concomitant anemia, which lowers the value.

Controlling BP and lipids helps prevent complications. In addition to diabetes management, our study evaluated regulation of the complications of diabetes, particularly cardiovascular complications. Evidence suggests that people with DM derive the greatest mortality benefit from a treatment plan centered first on hypertension, then lipids, and finally, glycemic control.¹⁸ A renewed focus on the BP and lipid aspects of diabetes care is needed.

Our data demonstrate that, of the 240 patients who met the ADA goal of <130/80 mm Hg, only 100 (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker. Lowering BP to <130/80 mm Hg may provide further benefit in preventing diabetes-related macrovascular complications.⁸

Lipid levels are a critical gauge of cardiovascular risk. Previous studies of patients with type 2 DM have shown that treating hyperlipidemia can produce a mortality benefit within 2 to 4 years, whereas aggressive glucose management takes approximately 8 years.¹⁸

A lipid panel was performed for 77.6% of the patients in our study—an improvement over a previous study by our team in which only 33% of patients received lipid checks.¹⁴ In the current study, a mere 2.9% of patients met the ADA’s combined lipid goals (LDL <100 mg/dL; HDL >40 mg/dL in men and >50 mg/dL in women; and triglycerides<150 mg/dL). Considering the LDL goal alone, 46.8% of the 190 patients whose lipids were monitored achieved it.

One hundred ten (44.9%) of the 245 patients in our study were prescribed a statin. Various studies support the use of lipid-lowering medications to increase HDL in elderly patients with DM.⁸ Yet data suggest that the rate of statin use among older adults is suboptimal.¹⁹ Our study highlights the limited prescription of statins for elderly nursing home patients who need them.

The ADA lipid goals are reasonable for this patient population, especially considering the potential mortality benefit. Rather than adjust lipid and BP goals, standards of care should emphasize the importance of meeting these objectives and suggest means to achieve them, including greater use of statins.

One set of standards does not work for all patients. Our study demonstrates that the ADA standards of care for ambulatory adults with diabetes are not acceptable for long-term care residents with DM. Although stringent A1c goals are appropriate for ambulatory adults, the risk of hypoglycemic episodes among the older and frailer nursing home population is too great to adhere to such a strict approach.

We recommend new guidelines be established. Guidelines developed specifically for residents in extended care are necessary to ensure the proper care of these patients, particularly in the face of a steady increase in their number.

Study limitations. Future inquiries into this subject should take into account the weaknesses of this study. First, it was conducted solely in Ohio and West Virginia. A chart review covering more territory could explore regional differences in diabetes care provided by long-term facilities and provide more evidence of the need for a population-specific standards of care.

The study also failed to account for comorbid conditions, including dementia, and code status, and followed residents for only one year. More extensive reviews could examine the effects of therapy in this patient population and the relationship between mortality and treatment plan, spurring movement toward more uniform and effective care of patients with diabetes in the long-term care setting.

CORRESPONDENCE Jay Shubrook Jr, DO, Department of Family Medicine, Ohio University Heritage College of Osteopathic Medicine, Grosvenor Hall, Athens, OH 45701; [email protected]

A surge in elderly patients with diabetes has placed a large burden on extended-care facilities. According to the Centers for Medicare and Medicaid Services, the prevalence of diabetes among nursing home residents is 33.3%.¹ Between 1995 and 2004, the estimated number of long-term care residents with diabetes mellitus (DM) grew by 7.1%, from approximately 242,000 to 329,000.² The increase adds to the challenge extended-care facilities face in attempting to provide high-quality care to patients with diabetes. No well-accepted management guidelines exist for nursing home residents with DM.³

Frail older adults with DM are more likely to suffer from cardiovascular conditions than younger patients, and are at greater risk for hypoglycemic coma and serious hyperglycemia.^4,5 A high frequency of hypoglycemia, especially nocturnal hypoglycemia, has been reported among nursing home residents with diabetes.⁶ Intensive insulin therapy is associated with hypoglycemia and increased mortality.⁷ However, hyperglycemia also must be considered because it significantly impairs quality of life. Uncontrolled hyperglycemia causes osmotic diuresis, leading to polyuria, nocturia, aggravated incontinence, and disrupted sleep, as well as contributing to dehydration.⁴ All of these problems have serious implications for quality of life and overall health.

Although studies have identified poor glycemic control and hypertension as the major problems facing nursing home patients with DM,^2,6 little research has examined how therapies targeting these problems help the elderly. Solid evidence supports the effectiveness of controlling hyperglycemia, lipid levels, and blood pressure (BP), along with aspirin therapy, in preventing microvascular disease, but does not reflect research involving older patients.⁸

A study of nursing homes found that health care teams did not respond to half of all significantly abnormal laboratory test results.⁹ Physicians who are aware of the problems associated with DM in elderly patients may hesitate to treat them because of the lack of guidelines for this patient population or concerns about adverse effects. Because of the deficiency of clinical trial data in elderly patients and the heterogenicity of the population, the American Diabetes Association (ADA) suggests that “less stringent treatment goals” may be appropriate.¹⁰

A central conflict in diabetes care for nursing home residents revolves around the need for guidelines that are both generalizable and easily individualized. Some studies support the need for individualized care, particularly with regard to A1c goals, because residents vary greatly in both disease burden and frailty.^8,11 Yet individualized treatment could increase the complexity of care for nurses who must manage many patients, potentially having a negative effect on patient care.

Implementation of a treatment protocol for residents with DM is associated with a decrease in the number of hospital days for acute and chronic complications,¹² but one study found that only 15% of nursing homes had such a protocol.¹³ Ultimately, long-term care facilities may benefit from an approach that strikes a balance between individualized care and generalized goals and does not closely mimic either acute hospital care or outpatient management of diabetes.³

In the absence of specific recommendations for extended-care residents with diabetes, our study evaluated the status of care in this population on the basis of pharmacotherapy and standards of care recommended by the ADA for ambulatory adults with DM.

Methods

Data collection
We reviewed the charts of 245 patients in 14 long-term care facilities in Ohio and West Virginia. All participating facilities signed a letter of agreement to take part in the study. The study was approved by the Ohio University Institutional Review Board.

At each facility, the director of nursing supplied a list of residents with DM. To be eligible for chart review, residents had to have a physician-documented diagnosis of type 1 or type 2 DM and have lived at the facility for at least 3 of the previous 12 months. Residents in both skilled nursing care and assisted living facilities were able to participate; short-term rehabilitation residents were not.

We performed a comprehensive review of each chart, examining the medical diagnoses, medication lists, laboratory reports, and physician and consultation notes for a one-year period. Data collection focused on diabetes-related intermediate outcomes and processes of care. Intermediate outcomes included A1c tests, lipid panels, and BP readings. Processes of care included aspirin therapy, use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers, use of statins, eye exams, foot exams, and microalbumin tests. The data collected omitted information identifying the patient, physician, or facility.

We compared the collected data with the 2011 ADA standards of care: blood glucose (fasting 80-120 mg/dL; postprandial 100-140 mg/dL), A1c (<7%), BP (<130/80 mm Hg), and lipid levels (low-density lipoprotein [LDL] <100 mg/dL; high-density lipoprotein [HDL] >40 mg/dL in men and >50 mg/dL in women; triglycerides <150 mg/dL).

Data analysis
We entered the data into an Excel database by type and key format and analyzed results using SPSS software, version 14.0 (SPSS, Chicago, IL). We used percentages and means±standard deviation to describe the data.

Results

TABLE 1 lists characteristics of the patients in the study: 24.5% were male and 75.5% were female; 9 (3.7%) were diagnosed with type 1 DM; 236 (96.3%) had type 2 DM. The mean age was 81±9 years, with a range of 44 to 103 years. Approximately 96% were Caucasian. The residents’ medical care was managed by family physicians (66.1%), internists (25.7%), geriatricians (6.9%), endocrinologists (0.8%), and other physicians (0.4%). The findings that follow are all based on a one-year period unless otherwise specified.

TABLE 1
Study population profile

Diabetes management
Most of the residents (211 [86.1%]) underwent glucose monitoring. The proportion of residents who received specific diabetes interventions is detailed in TABLE 2.

Hypoglycemia. Fifty-two residents (24.6% of those receiving glucose monitoring and 21.2% of the total) experienced a hypoglycemic event; 103 (representing 48.8% of the monitored patients and 42% of the total) had hyperglycemic events. On average, each resident experienced 1±2 mild hypoglycemic episodes per month, with a maximum of 13 mild episodes for one resident. Severe hypoglycemia (< 50 mg/dL) occurred less often, on average 0.24±1 time per resident. One resident had 15 severe hypoglycemic events in a month. The mean low hypoglycemic episode was at a glucose level of 52±16 mg/dL.

Hyperglycemia occurred more often than hypoglycemia (8±14 times per month), with a mean high glucose level of 352±89 mg/dL. This study used a generous range for normal glucose readings (70-249 mg/dL), and 89% of blood glucose readings were within that range. Thirty-seven percent of residents had an A1c <7.0%.

Medication. Thirty-two (13.1%) patients received no oral medication or insulin, and were managed with lifestyle interventions alone. Sixty-four patients (26.1%) used only oral medications, 64 (26.1%) received only insulin, and 85 (34.7%) were treated with both. Of the patients receiving insulin, 108 (72%) were on a sliding scale regimen. Seventy-seven (51.7%) of the patients on insulin experienced hypoglycemia, vs 30 (20%) of those taking oral medication. Twenty-seven (31.8%) patients in the combined therapy group had hypoglycemic events.

TABLE 2
Interventions received by the study population

Preventive care
Foot and eye care. Dilated eye examinations were provided for 133 residents (54.3%). Most (76.3%) received foot examinations, and 69.4% were seen by a podiatrist.

Blood pressure. Of the 240 residents (98%) whose BP was monitored, 107 (43.7%) had readings lower than 125/85 mm Hg, a goal set by a team of diabetologists, endocrinologists, and geriatricians at Ohio University. One hundred residents (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker; 122 (49.8%) were receiving aspirin therapy. In the total population, 110 patients (44.9%) were prescribed a statin.

Lipid monitoring. Of the 190 residents (77.6%) whose lipids were monitored, only 89 (46.8%) met the LDL goal suggested by the ADA. Fifty-six (29.5%) had triglycerides <150 mg/dL.

The HDL goal recommended by the ADA is >40 mg/dL for men and >50 mg/dL for women. Three of the 24 men and 16 of the 91 women whose lipids were monitored met the HDL goal.

Discussion

Although several components of diabetes management in our study population failed to meet the ADA standards of care for ambulatory adults, some elements of care were well managed. Monthly foot exams were performed on 76.3% of patients; 69.4% were seen by a podiatrist. While the number of residents receiving foot exams had decreased by 10.7% since a previous study by our research group, the number of podiatric consults increased by 11.4%.¹⁴

Dilated eye exams were given to 54.3% of residents. More patients should be given the opportunity to have an annual eye exam. Diabetes is the leading cause of new cases of blindness among adults 20 to 74 years of age,¹⁵ and impaired vision affects patient activity levels, susceptibility to falls, and quality of life.

In addition to a good record of preventive exams, physicians were proficient in monitoring residents with diabetes with regular testing regimens. Eighty-six percent of patients underwent regular blood glucose monitoring; 84% had had their A1c tested in the past year, and 36.7% achieved the A1c goal of <7%. The average A1c reading was 6.7±1%.

While these A1c values would seem to reflect well-managed diabetes, blood glucose readings tell a different story. A comparison of A1c values and hyperglycemic events revealed a disparity between the estimated average glucose reading and the actual readings. Of the patients who underwent scheduled fingersticks, 24.6% experienced a hypoglycemic event and 48.8% had hyperglycemic events. On average, each patient had 8 hyperglycemic episodes per month. The average highest glucose reading was >350 mg/dL.

A1c is only part of the story. While A1c can be a marker of sustained hyperglycemia, it does not reflect the stability of glycemic control.¹⁶ A study by Löfgren and colleagues confirmed that elderly diabetic patients in nursing homes who have low A1c levels often suffer from hypoglycemia.⁶ Patients receiving insulin therapy are more likely to experience hypoglycemia.⁷

The mismatch between A1c and glucose readings reveals an important point about the management of diabetes in long-term care patients: A1c values do not tell the entire story about a patient’s blood glucose; thus, a physician cannot look only at A1c to assess a patient’s diabetes management. A previous study demonstrated that when physicians base treatment plans solely on A1c without consulting glucose logs or being familiar with newer treatments, adherence to evidence-based algorithms is unlikely.¹⁷

While A1c does provide information about average blood glucose levels, it does not offer perspective on hypoglycemia or glucose variability. It is vital that physicians screen the glucose log for evidence of hypo- and hyperglycemia before adjusting the patient’s treatment plan. Physicians must also keep in mind that A1c may be falsely low in elderly patients who have concomitant anemia, which lowers the value.

Controlling BP and lipids helps prevent complications. In addition to diabetes management, our study evaluated regulation of the complications of diabetes, particularly cardiovascular complications. Evidence suggests that people with DM derive the greatest mortality benefit from a treatment plan centered first on hypertension, then lipids, and finally, glycemic control.¹⁸ A renewed focus on the BP and lipid aspects of diabetes care is needed.

Our data demonstrate that, of the 240 patients who met the ADA goal of <130/80 mm Hg, only 100 (40.8%) were taking an angiotensin-converting enzyme inhibitor or an angiotensin II receptor blocker. Lowering BP to <130/80 mm Hg may provide further benefit in preventing diabetes-related macrovascular complications.⁸

Lipid levels are a critical gauge of cardiovascular risk. Previous studies of patients with type 2 DM have shown that treating hyperlipidemia can produce a mortality benefit within 2 to 4 years, whereas aggressive glucose management takes approximately 8 years.¹⁸

A lipid panel was performed for 77.6% of the patients in our study—an improvement over a previous study by our team in which only 33% of patients received lipid checks.¹⁴ In the current study, a mere 2.9% of patients met the ADA’s combined lipid goals (LDL <100 mg/dL; HDL >40 mg/dL in men and >50 mg/dL in women; and triglycerides<150 mg/dL). Considering the LDL goal alone, 46.8% of the 190 patients whose lipids were monitored achieved it.

One hundred ten (44.9%) of the 245 patients in our study were prescribed a statin. Various studies support the use of lipid-lowering medications to increase HDL in elderly patients with DM.⁸ Yet data suggest that the rate of statin use among older adults is suboptimal.¹⁹ Our study highlights the limited prescription of statins for elderly nursing home patients who need them.

The ADA lipid goals are reasonable for this patient population, especially considering the potential mortality benefit. Rather than adjust lipid and BP goals, standards of care should emphasize the importance of meeting these objectives and suggest means to achieve them, including greater use of statins.

One set of standards does not work for all patients. Our study demonstrates that the ADA standards of care for ambulatory adults with diabetes are not acceptable for long-term care residents with DM. Although stringent A1c goals are appropriate for ambulatory adults, the risk of hypoglycemic episodes among the older and frailer nursing home population is too great to adhere to such a strict approach.

We recommend new guidelines be established. Guidelines developed specifically for residents in extended care are necessary to ensure the proper care of these patients, particularly in the face of a steady increase in their number.

Study limitations. Future inquiries into this subject should take into account the weaknesses of this study. First, it was conducted solely in Ohio and West Virginia. A chart review covering more territory could explore regional differences in diabetes care provided by long-term facilities and provide more evidence of the need for a population-specific standards of care.

The study also failed to account for comorbid conditions, including dementia, and code status, and followed residents for only one year. More extensive reviews could examine the effects of therapy in this patient population and the relationship between mortality and treatment plan, spurring movement toward more uniform and effective care of patients with diabetes in the long-term care setting.

CORRESPONDENCE Jay Shubrook Jr, DO, Department of Family Medicine, Ohio University Heritage College of Osteopathic Medicine, Grosvenor Hall, Athens, OH 45701; [email protected]

What to include in the workup

Whether you’re doing an initial evaluation of a patient with high blood pressure (BP) or examining a patient with resistant hypertension, the history should focus on the duration of hypertension, previous BP levels, and comorbid conditions. It is also important to take a targeted family history, inquiring about hypertension as well as genetic disorders that increase the likelihood of secondary hypertension.

Inherited diseases associated with secondary hypertension include polycystic kidney disease, multiple endocrine neoplasia type 2 (MEN2), and von Hippel-Lindau syndrome.^12,13 All are inherited in an autosomal dominant pattern. Patients with von Hippel-Lindau syndrome may present with multiple tumors, which can develop in the eyes, brain, adrenal glands, pancreas, liver, spinal cord, kidneys, or other parts of the body. Pheochromocytoma is a manifestation of both MEN2 and von Hippel-Lindau syndrome, and some specialists recommend that everyone with a family history of either condition undergo screening for pheochromocytoma.¹⁴

Table
Secondary hypertension: What you’ll see, what to test for^8-11

BP measurement is key
The physical examination should start with a calculation of body mass index, as well as a careful measurement of BP. The patient should be seated quietly in a chair for ≥5 minutes, with both feet on the floor and the arm being tested supported at heart level.

Unfortunately, reliability on the office BP measurement can be a confounding factor in the diagnosis of hypertension. “White coat hypertension”—in which BP is persistently elevated in the office and persistently normal in nonclinical settings—should be considered in patients who have high BP but no other signs or symptoms, and ambulatory monitoring used to rule out hypertension.^15,16

Physicians also need to consider the opposite effect: Masked hypertension, characterized by normal office readings and elevated ambulatory readings, is more serious, of course, with patients at higher risk for end organ damage from unrecognized hypertension.^17,18 Asking patients who self-monitor what type of BP readings they’re getting can be helpful in identifying masked hypertension. Ambulatory monitoring may be used to identify this condition, as well.

Other components in the physical workup include a fundoscopic exam; assessment of the thorax for murmurs and the abdomen for enlarged kidneys, masses, and abnormal aortic pulsation; auscultation for abdominal and carotid bruits; palpation of the thyroid gland; and palpation of the lower extremities for edema and pulses.

Include these tests in the workup
Routine tests for a patient with hypertension include:

• electrocardiogram
• blood glucose and hematocrit
• serum potassium, creatinine, and fasting lipid profiles
• urinalysis with measurement of microalbumin.

Microalbuminuria, a sensitive marker of early renal disease, is defined as a urinary albumin excretion between 30 and 300 mg/d.¹⁹ The albumin-creatinine ratio (30-300 mcg/mg), measured in spot urine specimens, is a more convenient way to detect it.²⁰

Suspicious findings prompt further testing. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) recommends specific testing—much of it detailed below—if any aspect of the initial evaluation raises suspicion of a secondary cause or the patient has hypertension that’s of sudden onset or hard to control.²¹ (According to the National Heart, Lung, and Blood Institute, JNC 8 is due for release later this year.)

Kidney disease may be a consequence or a cause

The overall prevalence of hypertension in patients with renal disease is 60%,²² but varies according to the type of nephropathy. Eighty-seven percent of patients with diabetic nephropathy also have hypertension, and hypertension and diabetes are the 2 most common causes of end-stage renal disease.^23,24

A combination of 2 or more drugs is usually needed to achieve the target BP of <130/80 mm Hg in patients with diabetes.²¹ ACE inhibitors and angiotensin receptor blockers have been found to slow the progression of diabetic nephropathy.^25-27

Is renal artery stenosis to blame?
Renal artery stenosis is the most common form of secondary hypertension that’s reversible, affecting about 0.2% to 3.1% of hypertensive patients.^5,6,28 The condition leads to renal ischemia, thereby stimulating the renin-angiotensin-aldosterone axis and causing secondary hyperaldosteronism.

In younger patients, especially women between 15 and 50 years of age, fibromuscular disease is the most common cause of renovascular hypertension.^29,30 In older patients, atherosclerosis (which accounts for 90% of renovascular hypertension) is more likely.^29,30

The choice of initial imaging tests includes duplex renal ultrasonography, magnetic resonance angiography (MRA), and spiral computed tomographic angiography. Contrast angiography is the gold standard, but it carries a risk of contrast-induced nephropathy. Duplex ultrasonography and MRA do not use iodinated contrast media, and are safe for patients with chronic kidney disease.⁸

Treatment. Percutaneous transluminal renal artery angioplasty is a treatment option for patients with renal artery stenosis. Angioplasty achieves higher cure rates for patients with fibromuscular dysplasia than for those with atherosclerotic renal artery stenosis.³¹ Most patients referred for renal artery revascularization have atherosclerosis. Because they’re generally older individuals with comorbidities, the benefits of stent revascularization for this group is controversial. Such patients require antihypertensive therapy with drugs that block the renin-angiotensin system.³²

Endocrine disorders must be ruled out

Primary hyperaldosteronism is thought to be present in 0.3% to 1.4% of patients with hypertension.^5,6 The prevalence varies widely from one source to another, however, and may be as high as 5% to 20% among patients with resistant hypertension.^33,34

Hyperaldosteronism is related to either an aldosterone-secreting adrenal adenoma (in about 40% of cases) or bilateral adrenal hyperplasia (in the remaining 60%), and leads to increased sodium reabsorption and, typically, to a loss of potassium.³⁵

Renin-secreting tumor, which usually arises from the juxtaglomerular cells of the kidney, is a rare cause of hyperaldosteronism. Extrarenal renin-secreting tumors have also been reported.³⁶

What should raise your suspicion. Suspect hyperaldosteronism in patients who have both hypertension and hypokalemia, but keep in mind that not all patients with hyperaldosteronism have low serum potassium.³⁷ Further laboratory evaluation should include a simultaneous measurement of plasma aldosterone (PA) and plasma renin activity (PRA). Patients with hyperaldosteronism will have elevated PA and suppressed PRA.

Testing considerations. It is important to ensure that the PA/PRA test is performed in the morning, with the patient in an upright position.³⁶ He or she should be on a high sodium diet in preparation for the test, consuming 2 to 3 grams of sodium per meal for ≥2 days.³⁷

In patients with a positive PA/PRA ratio (≥20), a 24-hour urinary aldosterone excretion test should be performed. A finding >12 to 14 mcg, along with a PRA <1.0 ng/mL per hour, confirms the diagnosis of primary hyperaldosteronism.^18,37 Computed tomography or magnetic resonance imaging of the adrenal glands will distinguish between aldosterone-producing adenoma and bilateral adrenal hyperplasia.

Treatment. Laparoscopic adrenalectomy is the accepted surgical treatment of primary hyperaldosteronism.³⁷ Patients with bilateral disease due to idiopathic hyperaldosteronism are not candidates for surgery and should be treated medically, with potassium-sparing diuretics such as spironolactone.

Cushing’s syndrome is marked by rapid weight gain
High BP may be a manifestation of Cushing’s syndrome, which affects 0.1% to 0.5% of patients with hypertension.^5-7 Other signs and symptoms of Cushing’s syndrome include fatigue, weakness, hirsutism, amenorrhea, moon facies, dorsal hump, purple striae, truncal obesity, and hypokalemia. Rapid weight gain is the most common manifestation, and typically the one for which patients seek medical attention.³⁸

The most widely used screening test for Cushing’s syndrome is a 24-hour urine collection measuring urinary-free cortisol.⁹ Normal urinary cortisol excretion is 20 to 100 mcg/dL in 24 hours; most patients with Cushing’s syndrome produce >250 mcg/dL in that time frame.⁹

Once hypercortisolism is established, determination of the cause is the next step. A serum adrenocorticotropic hormone (ACTH) level is needed to localize it. Normal (9-52 pg/mL) or elevated ACTH indicates a pituitary or ectopic source, while low levels of ACTH are an indication of an adrenal source.^9,39

Treatment. Surgical resection of the tumor is often curative. For pituitary tumors (Cushing’s disease), transsphenoidal resection is the standard of care.³⁹ For adrenal adenomas, unilateral adrenalectomy is the best option.³⁹

Pheochromocytomas: Most are adrenal, sporadic, and benign
Pheochromocytomas—neuroendocrine, catecholamine-secreting tumors that develop from the adrenal medulla—are another cause of secondary hypertension. Catecholamines include norepinephrine and epinephrine and, rarely, dopamine secreted either intermittently or continuously. The prevalence of pheochromocytoma is 0.1% to 0.3% among patients with hypertension.^5,6,28 A “rule of 10” (90:10 ratio) is often applied to pheochromocytomas because of the following:

• 90% of pheochromocytomas are located in the adrenal glands; the remaining 10% are extra-adrenal and can occur anywhere along the sympathetic chain⁴⁰
• 90% are sporadic; 10% are familial⁴¹
• 90% are benign; 10% are malignant⁴⁰
• 90% are found in adults; 10% affect children.⁴²

Signs and symptoms of pheochromocytomas include palpitations, headache, dyspnea, diaphoresis, and flushing, as well as paroxysmal hypertension.⁴⁰ Measurement of 24-hour urinary catecholamines and their metabolites has been the screening test of choice,⁴³ but recent evidence suggests that measurement of plasma metanephrine and normetanephrine is a far more sensitive screen.¹⁰

Treatment. Surgical resection is the treatment of choice. Alpha blockade is started 7 to 10 days preoperatively;^44,45 a beta-blocker is added only after an adequate alpha blockade has been established, as unopposed alpha stimulation could precipitate a hypertensive crisis. Laparoscopic adrenalectomy is routinely performed for small (<5 cm) pheochromocytomas.^46,47

Don’t forget these (relatively) common secondary causes

Obstructive sleep apnea (OSA) is one of the most common conditions associated with resistant hypertension.⁴⁸ Signs and symptoms include snoring, daytime somnolence, and obesity (body mass index ≥30 kg/m²).

OSA involves upper airway collapse during inspiration, causing intermittent hypoxemia with resultant sympathetic nervous system activation.¹¹ The underlying mechanism of OSA-induced hypertension is strongly related to sympathetic activation.⁴⁹ Overnight polysomnography is required for diagnosis.¹¹

Continuous positive airway pressure is the treatment of choice for patients unable to lose weight.¹¹

Pregnancy-induced hypertension is the most common medical problem encountered in pregnancy. It occurs in up to 15% of pregnancies, either during the pregnancy itself or postpartum. Postpartum hypertension may be related to preexisting chronic hypertension or to the persistence of gestational hypertension or preeclampsia, which usually occurs after 20 weeks’ gestation and is characterized by the presence of hypertension and proteinuria.⁵⁰ Methyldopa and labetalol are commonly used treatments for hypertension during pregnancy.⁵¹

Drug-induced hypertension. Several drugs can cause or exacerbate hypertension, rendering it resistant to therapy. A careful review of the patient’s medication regimen is essential. Generally, drug-induced hypertension falls into 2 broad categories based on mechanism of action: volume overload and sympathetic activity.^52,53

Corticosteroids can elevate BP in a dose-dependent manner, as a result of volume overload. Glycyrrhizic acid, the main ingredient in licorice, produces a state of excess mineralocorticoid, with a similar effect. Estrogen-containing oral contraceptives can cause an increased synthesis of angiotensin in the liver, resulting in greater aldosterone secretion and volume overload.

Drugs that stimulate sympathetic activity include cocaine, ephedrine, amphetamine, phenylephrine, phenylpropanolamine, caffeine, and alcohol. Nonsteroidal anti-inflammatory drugs may interfere with the action of ACE inhibitors and cause renal vasoconstriction, leading to sodium and water retention and hypertension.⁵⁴

Discontinuation of the medication in question is preferable. In many cases, an agent that does not affect BP can be found to replace it.

If the patient is a child
Hypertension is uncommon in young people. However, coarctation of the aorta, a congenital narrowing associated with secondary hypertension, is typically diagnosed in childhood. In rare cases, the condition remains undetected well into adulthood.⁵⁵ Clinical signs include weak femoral pulses, visible pulsations in the neck, a systolic murmur of the anterior and posterior thorax, and elevated BP in the upper extremities with low BP in the lower extremities.

Thus, once hypertension is confirmed in a young patient, BP should be measured in both arms and legs.⁵⁶ Delayed or absent femoral pulses and a difference in systolic BP of ≥20 mm Hg between the arms and legs provide evidence of aortic coarctation.⁵⁷ In adults, stenting is the initial treatment for this condition because of the morbidity associated with surgery.⁵⁷ Stenting is an option for children with this condition, as well.⁵⁸

CORRESPONDENCE Bernard M. Karnath, MD, University of Texas Medical Branch at Galveston, 301 University Boulevard, Galveston, TX 77555; [email protected]

What to include in the workup

Whether you’re doing an initial evaluation of a patient with high blood pressure (BP) or examining a patient with resistant hypertension, the history should focus on the duration of hypertension, previous BP levels, and comorbid conditions. It is also important to take a targeted family history, inquiring about hypertension as well as genetic disorders that increase the likelihood of secondary hypertension.

Inherited diseases associated with secondary hypertension include polycystic kidney disease, multiple endocrine neoplasia type 2 (MEN2), and von Hippel-Lindau syndrome.^12,13 All are inherited in an autosomal dominant pattern. Patients with von Hippel-Lindau syndrome may present with multiple tumors, which can develop in the eyes, brain, adrenal glands, pancreas, liver, spinal cord, kidneys, or other parts of the body. Pheochromocytoma is a manifestation of both MEN2 and von Hippel-Lindau syndrome, and some specialists recommend that everyone with a family history of either condition undergo screening for pheochromocytoma.¹⁴

Table
Secondary hypertension: What you’ll see, what to test for^8-11

BP measurement is key
The physical examination should start with a calculation of body mass index, as well as a careful measurement of BP. The patient should be seated quietly in a chair for ≥5 minutes, with both feet on the floor and the arm being tested supported at heart level.

Unfortunately, reliability on the office BP measurement can be a confounding factor in the diagnosis of hypertension. “White coat hypertension”—in which BP is persistently elevated in the office and persistently normal in nonclinical settings—should be considered in patients who have high BP but no other signs or symptoms, and ambulatory monitoring used to rule out hypertension.^15,16

Physicians also need to consider the opposite effect: Masked hypertension, characterized by normal office readings and elevated ambulatory readings, is more serious, of course, with patients at higher risk for end organ damage from unrecognized hypertension.^17,18 Asking patients who self-monitor what type of BP readings they’re getting can be helpful in identifying masked hypertension. Ambulatory monitoring may be used to identify this condition, as well.

Other components in the physical workup include a fundoscopic exam; assessment of the thorax for murmurs and the abdomen for enlarged kidneys, masses, and abnormal aortic pulsation; auscultation for abdominal and carotid bruits; palpation of the thyroid gland; and palpation of the lower extremities for edema and pulses.

Include these tests in the workup
Routine tests for a patient with hypertension include:

• electrocardiogram
• blood glucose and hematocrit
• serum potassium, creatinine, and fasting lipid profiles
• urinalysis with measurement of microalbumin.

Microalbuminuria, a sensitive marker of early renal disease, is defined as a urinary albumin excretion between 30 and 300 mg/d.¹⁹ The albumin-creatinine ratio (30-300 mcg/mg), measured in spot urine specimens, is a more convenient way to detect it.²⁰

Suspicious findings prompt further testing. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) recommends specific testing—much of it detailed below—if any aspect of the initial evaluation raises suspicion of a secondary cause or the patient has hypertension that’s of sudden onset or hard to control.²¹ (According to the National Heart, Lung, and Blood Institute, JNC 8 is due for release later this year.)

Kidney disease may be a consequence or a cause

The overall prevalence of hypertension in patients with renal disease is 60%,²² but varies according to the type of nephropathy. Eighty-seven percent of patients with diabetic nephropathy also have hypertension, and hypertension and diabetes are the 2 most common causes of end-stage renal disease.^23,24

A combination of 2 or more drugs is usually needed to achieve the target BP of <130/80 mm Hg in patients with diabetes.²¹ ACE inhibitors and angiotensin receptor blockers have been found to slow the progression of diabetic nephropathy.^25-27

Is renal artery stenosis to blame?
Renal artery stenosis is the most common form of secondary hypertension that’s reversible, affecting about 0.2% to 3.1% of hypertensive patients.^5,6,28 The condition leads to renal ischemia, thereby stimulating the renin-angiotensin-aldosterone axis and causing secondary hyperaldosteronism.

In younger patients, especially women between 15 and 50 years of age, fibromuscular disease is the most common cause of renovascular hypertension.^29,30 In older patients, atherosclerosis (which accounts for 90% of renovascular hypertension) is more likely.^29,30

The choice of initial imaging tests includes duplex renal ultrasonography, magnetic resonance angiography (MRA), and spiral computed tomographic angiography. Contrast angiography is the gold standard, but it carries a risk of contrast-induced nephropathy. Duplex ultrasonography and MRA do not use iodinated contrast media, and are safe for patients with chronic kidney disease.⁸

Treatment. Percutaneous transluminal renal artery angioplasty is a treatment option for patients with renal artery stenosis. Angioplasty achieves higher cure rates for patients with fibromuscular dysplasia than for those with atherosclerotic renal artery stenosis.³¹ Most patients referred for renal artery revascularization have atherosclerosis. Because they’re generally older individuals with comorbidities, the benefits of stent revascularization for this group is controversial. Such patients require antihypertensive therapy with drugs that block the renin-angiotensin system.³²

Endocrine disorders must be ruled out

Primary hyperaldosteronism is thought to be present in 0.3% to 1.4% of patients with hypertension.^5,6 The prevalence varies widely from one source to another, however, and may be as high as 5% to 20% among patients with resistant hypertension.^33,34

Hyperaldosteronism is related to either an aldosterone-secreting adrenal adenoma (in about 40% of cases) or bilateral adrenal hyperplasia (in the remaining 60%), and leads to increased sodium reabsorption and, typically, to a loss of potassium.³⁵

Renin-secreting tumor, which usually arises from the juxtaglomerular cells of the kidney, is a rare cause of hyperaldosteronism. Extrarenal renin-secreting tumors have also been reported.³⁶

What should raise your suspicion. Suspect hyperaldosteronism in patients who have both hypertension and hypokalemia, but keep in mind that not all patients with hyperaldosteronism have low serum potassium.³⁷ Further laboratory evaluation should include a simultaneous measurement of plasma aldosterone (PA) and plasma renin activity (PRA). Patients with hyperaldosteronism will have elevated PA and suppressed PRA.

Testing considerations. It is important to ensure that the PA/PRA test is performed in the morning, with the patient in an upright position.³⁶ He or she should be on a high sodium diet in preparation for the test, consuming 2 to 3 grams of sodium per meal for ≥2 days.³⁷

In patients with a positive PA/PRA ratio (≥20), a 24-hour urinary aldosterone excretion test should be performed. A finding >12 to 14 mcg, along with a PRA <1.0 ng/mL per hour, confirms the diagnosis of primary hyperaldosteronism.^18,37 Computed tomography or magnetic resonance imaging of the adrenal glands will distinguish between aldosterone-producing adenoma and bilateral adrenal hyperplasia.

Treatment. Laparoscopic adrenalectomy is the accepted surgical treatment of primary hyperaldosteronism.³⁷ Patients with bilateral disease due to idiopathic hyperaldosteronism are not candidates for surgery and should be treated medically, with potassium-sparing diuretics such as spironolactone.

Cushing’s syndrome is marked by rapid weight gain
High BP may be a manifestation of Cushing’s syndrome, which affects 0.1% to 0.5% of patients with hypertension.^5-7 Other signs and symptoms of Cushing’s syndrome include fatigue, weakness, hirsutism, amenorrhea, moon facies, dorsal hump, purple striae, truncal obesity, and hypokalemia. Rapid weight gain is the most common manifestation, and typically the one for which patients seek medical attention.³⁸

The most widely used screening test for Cushing’s syndrome is a 24-hour urine collection measuring urinary-free cortisol.⁹ Normal urinary cortisol excretion is 20 to 100 mcg/dL in 24 hours; most patients with Cushing’s syndrome produce >250 mcg/dL in that time frame.⁹

Once hypercortisolism is established, determination of the cause is the next step. A serum adrenocorticotropic hormone (ACTH) level is needed to localize it. Normal (9-52 pg/mL) or elevated ACTH indicates a pituitary or ectopic source, while low levels of ACTH are an indication of an adrenal source.^9,39

Treatment. Surgical resection of the tumor is often curative. For pituitary tumors (Cushing’s disease), transsphenoidal resection is the standard of care.³⁹ For adrenal adenomas, unilateral adrenalectomy is the best option.³⁹

Pheochromocytomas: Most are adrenal, sporadic, and benign
Pheochromocytomas—neuroendocrine, catecholamine-secreting tumors that develop from the adrenal medulla—are another cause of secondary hypertension. Catecholamines include norepinephrine and epinephrine and, rarely, dopamine secreted either intermittently or continuously. The prevalence of pheochromocytoma is 0.1% to 0.3% among patients with hypertension.^5,6,28 A “rule of 10” (90:10 ratio) is often applied to pheochromocytomas because of the following:

• 90% of pheochromocytomas are located in the adrenal glands; the remaining 10% are extra-adrenal and can occur anywhere along the sympathetic chain⁴⁰
• 90% are sporadic; 10% are familial⁴¹
• 90% are benign; 10% are malignant⁴⁰
• 90% are found in adults; 10% affect children.⁴²

Signs and symptoms of pheochromocytomas include palpitations, headache, dyspnea, diaphoresis, and flushing, as well as paroxysmal hypertension.⁴⁰ Measurement of 24-hour urinary catecholamines and their metabolites has been the screening test of choice,⁴³ but recent evidence suggests that measurement of plasma metanephrine and normetanephrine is a far more sensitive screen.¹⁰

Treatment. Surgical resection is the treatment of choice. Alpha blockade is started 7 to 10 days preoperatively;^44,45 a beta-blocker is added only after an adequate alpha blockade has been established, as unopposed alpha stimulation could precipitate a hypertensive crisis. Laparoscopic adrenalectomy is routinely performed for small (<5 cm) pheochromocytomas.^46,47

Don’t forget these (relatively) common secondary causes

Obstructive sleep apnea (OSA) is one of the most common conditions associated with resistant hypertension.⁴⁸ Signs and symptoms include snoring, daytime somnolence, and obesity (body mass index ≥30 kg/m²).

OSA involves upper airway collapse during inspiration, causing intermittent hypoxemia with resultant sympathetic nervous system activation.¹¹ The underlying mechanism of OSA-induced hypertension is strongly related to sympathetic activation.⁴⁹ Overnight polysomnography is required for diagnosis.¹¹

Continuous positive airway pressure is the treatment of choice for patients unable to lose weight.¹¹

Pregnancy-induced hypertension is the most common medical problem encountered in pregnancy. It occurs in up to 15% of pregnancies, either during the pregnancy itself or postpartum. Postpartum hypertension may be related to preexisting chronic hypertension or to the persistence of gestational hypertension or preeclampsia, which usually occurs after 20 weeks’ gestation and is characterized by the presence of hypertension and proteinuria.⁵⁰ Methyldopa and labetalol are commonly used treatments for hypertension during pregnancy.⁵¹

Drug-induced hypertension. Several drugs can cause or exacerbate hypertension, rendering it resistant to therapy. A careful review of the patient’s medication regimen is essential. Generally, drug-induced hypertension falls into 2 broad categories based on mechanism of action: volume overload and sympathetic activity.^52,53

Corticosteroids can elevate BP in a dose-dependent manner, as a result of volume overload. Glycyrrhizic acid, the main ingredient in licorice, produces a state of excess mineralocorticoid, with a similar effect. Estrogen-containing oral contraceptives can cause an increased synthesis of angiotensin in the liver, resulting in greater aldosterone secretion and volume overload.

Drugs that stimulate sympathetic activity include cocaine, ephedrine, amphetamine, phenylephrine, phenylpropanolamine, caffeine, and alcohol. Nonsteroidal anti-inflammatory drugs may interfere with the action of ACE inhibitors and cause renal vasoconstriction, leading to sodium and water retention and hypertension.⁵⁴

Discontinuation of the medication in question is preferable. In many cases, an agent that does not affect BP can be found to replace it.

If the patient is a child
Hypertension is uncommon in young people. However, coarctation of the aorta, a congenital narrowing associated with secondary hypertension, is typically diagnosed in childhood. In rare cases, the condition remains undetected well into adulthood.⁵⁵ Clinical signs include weak femoral pulses, visible pulsations in the neck, a systolic murmur of the anterior and posterior thorax, and elevated BP in the upper extremities with low BP in the lower extremities.

Thus, once hypertension is confirmed in a young patient, BP should be measured in both arms and legs.⁵⁶ Delayed or absent femoral pulses and a difference in systolic BP of ≥20 mm Hg between the arms and legs provide evidence of aortic coarctation.⁵⁷ In adults, stenting is the initial treatment for this condition because of the morbidity associated with surgery.⁵⁷ Stenting is an option for children with this condition, as well.⁵⁸

CORRESPONDENCE Bernard M. Karnath, MD, University of Texas Medical Branch at Galveston, 301 University Boulevard, Galveston, TX 77555; [email protected]

What to include in the workup

Whether you’re doing an initial evaluation of a patient with high blood pressure (BP) or examining a patient with resistant hypertension, the history should focus on the duration of hypertension, previous BP levels, and comorbid conditions. It is also important to take a targeted family history, inquiring about hypertension as well as genetic disorders that increase the likelihood of secondary hypertension.

Inherited diseases associated with secondary hypertension include polycystic kidney disease, multiple endocrine neoplasia type 2 (MEN2), and von Hippel-Lindau syndrome.^12,13 All are inherited in an autosomal dominant pattern. Patients with von Hippel-Lindau syndrome may present with multiple tumors, which can develop in the eyes, brain, adrenal glands, pancreas, liver, spinal cord, kidneys, or other parts of the body. Pheochromocytoma is a manifestation of both MEN2 and von Hippel-Lindau syndrome, and some specialists recommend that everyone with a family history of either condition undergo screening for pheochromocytoma.¹⁴

Table
Secondary hypertension: What you’ll see, what to test for^8-11

BP measurement is key
The physical examination should start with a calculation of body mass index, as well as a careful measurement of BP. The patient should be seated quietly in a chair for ≥5 minutes, with both feet on the floor and the arm being tested supported at heart level.

Unfortunately, reliability on the office BP measurement can be a confounding factor in the diagnosis of hypertension. “White coat hypertension”—in which BP is persistently elevated in the office and persistently normal in nonclinical settings—should be considered in patients who have high BP but no other signs or symptoms, and ambulatory monitoring used to rule out hypertension.^15,16

Physicians also need to consider the opposite effect: Masked hypertension, characterized by normal office readings and elevated ambulatory readings, is more serious, of course, with patients at higher risk for end organ damage from unrecognized hypertension.^17,18 Asking patients who self-monitor what type of BP readings they’re getting can be helpful in identifying masked hypertension. Ambulatory monitoring may be used to identify this condition, as well.

Other components in the physical workup include a fundoscopic exam; assessment of the thorax for murmurs and the abdomen for enlarged kidneys, masses, and abnormal aortic pulsation; auscultation for abdominal and carotid bruits; palpation of the thyroid gland; and palpation of the lower extremities for edema and pulses.

Include these tests in the workup
Routine tests for a patient with hypertension include:

• electrocardiogram
• blood glucose and hematocrit
• serum potassium, creatinine, and fasting lipid profiles
• urinalysis with measurement of microalbumin.

Microalbuminuria, a sensitive marker of early renal disease, is defined as a urinary albumin excretion between 30 and 300 mg/d.¹⁹ The albumin-creatinine ratio (30-300 mcg/mg), measured in spot urine specimens, is a more convenient way to detect it.²⁰

Suspicious findings prompt further testing. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) recommends specific testing—much of it detailed below—if any aspect of the initial evaluation raises suspicion of a secondary cause or the patient has hypertension that’s of sudden onset or hard to control.²¹ (According to the National Heart, Lung, and Blood Institute, JNC 8 is due for release later this year.)

Kidney disease may be a consequence or a cause

The overall prevalence of hypertension in patients with renal disease is 60%,²² but varies according to the type of nephropathy. Eighty-seven percent of patients with diabetic nephropathy also have hypertension, and hypertension and diabetes are the 2 most common causes of end-stage renal disease.^23,24

A combination of 2 or more drugs is usually needed to achieve the target BP of <130/80 mm Hg in patients with diabetes.²¹ ACE inhibitors and angiotensin receptor blockers have been found to slow the progression of diabetic nephropathy.^25-27

Is renal artery stenosis to blame?
Renal artery stenosis is the most common form of secondary hypertension that’s reversible, affecting about 0.2% to 3.1% of hypertensive patients.^5,6,28 The condition leads to renal ischemia, thereby stimulating the renin-angiotensin-aldosterone axis and causing secondary hyperaldosteronism.

In younger patients, especially women between 15 and 50 years of age, fibromuscular disease is the most common cause of renovascular hypertension.^29,30 In older patients, atherosclerosis (which accounts for 90% of renovascular hypertension) is more likely.^29,30

The choice of initial imaging tests includes duplex renal ultrasonography, magnetic resonance angiography (MRA), and spiral computed tomographic angiography. Contrast angiography is the gold standard, but it carries a risk of contrast-induced nephropathy. Duplex ultrasonography and MRA do not use iodinated contrast media, and are safe for patients with chronic kidney disease.⁸

Treatment. Percutaneous transluminal renal artery angioplasty is a treatment option for patients with renal artery stenosis. Angioplasty achieves higher cure rates for patients with fibromuscular dysplasia than for those with atherosclerotic renal artery stenosis.³¹ Most patients referred for renal artery revascularization have atherosclerosis. Because they’re generally older individuals with comorbidities, the benefits of stent revascularization for this group is controversial. Such patients require antihypertensive therapy with drugs that block the renin-angiotensin system.³²

Endocrine disorders must be ruled out

Primary hyperaldosteronism is thought to be present in 0.3% to 1.4% of patients with hypertension.^5,6 The prevalence varies widely from one source to another, however, and may be as high as 5% to 20% among patients with resistant hypertension.^33,34

Hyperaldosteronism is related to either an aldosterone-secreting adrenal adenoma (in about 40% of cases) or bilateral adrenal hyperplasia (in the remaining 60%), and leads to increased sodium reabsorption and, typically, to a loss of potassium.³⁵

Renin-secreting tumor, which usually arises from the juxtaglomerular cells of the kidney, is a rare cause of hyperaldosteronism. Extrarenal renin-secreting tumors have also been reported.³⁶

What should raise your suspicion. Suspect hyperaldosteronism in patients who have both hypertension and hypokalemia, but keep in mind that not all patients with hyperaldosteronism have low serum potassium.³⁷ Further laboratory evaluation should include a simultaneous measurement of plasma aldosterone (PA) and plasma renin activity (PRA). Patients with hyperaldosteronism will have elevated PA and suppressed PRA.

Testing considerations. It is important to ensure that the PA/PRA test is performed in the morning, with the patient in an upright position.³⁶ He or she should be on a high sodium diet in preparation for the test, consuming 2 to 3 grams of sodium per meal for ≥2 days.³⁷

In patients with a positive PA/PRA ratio (≥20), a 24-hour urinary aldosterone excretion test should be performed. A finding >12 to 14 mcg, along with a PRA <1.0 ng/mL per hour, confirms the diagnosis of primary hyperaldosteronism.^18,37 Computed tomography or magnetic resonance imaging of the adrenal glands will distinguish between aldosterone-producing adenoma and bilateral adrenal hyperplasia.

Treatment. Laparoscopic adrenalectomy is the accepted surgical treatment of primary hyperaldosteronism.³⁷ Patients with bilateral disease due to idiopathic hyperaldosteronism are not candidates for surgery and should be treated medically, with potassium-sparing diuretics such as spironolactone.

Cushing’s syndrome is marked by rapid weight gain
High BP may be a manifestation of Cushing’s syndrome, which affects 0.1% to 0.5% of patients with hypertension.^5-7 Other signs and symptoms of Cushing’s syndrome include fatigue, weakness, hirsutism, amenorrhea, moon facies, dorsal hump, purple striae, truncal obesity, and hypokalemia. Rapid weight gain is the most common manifestation, and typically the one for which patients seek medical attention.³⁸

The most widely used screening test for Cushing’s syndrome is a 24-hour urine collection measuring urinary-free cortisol.⁹ Normal urinary cortisol excretion is 20 to 100 mcg/dL in 24 hours; most patients with Cushing’s syndrome produce >250 mcg/dL in that time frame.⁹

Once hypercortisolism is established, determination of the cause is the next step. A serum adrenocorticotropic hormone (ACTH) level is needed to localize it. Normal (9-52 pg/mL) or elevated ACTH indicates a pituitary or ectopic source, while low levels of ACTH are an indication of an adrenal source.^9,39

Treatment. Surgical resection of the tumor is often curative. For pituitary tumors (Cushing’s disease), transsphenoidal resection is the standard of care.³⁹ For adrenal adenomas, unilateral adrenalectomy is the best option.³⁹

Pheochromocytomas: Most are adrenal, sporadic, and benign
Pheochromocytomas—neuroendocrine, catecholamine-secreting tumors that develop from the adrenal medulla—are another cause of secondary hypertension. Catecholamines include norepinephrine and epinephrine and, rarely, dopamine secreted either intermittently or continuously. The prevalence of pheochromocytoma is 0.1% to 0.3% among patients with hypertension.^5,6,28 A “rule of 10” (90:10 ratio) is often applied to pheochromocytomas because of the following:

• 90% of pheochromocytomas are located in the adrenal glands; the remaining 10% are extra-adrenal and can occur anywhere along the sympathetic chain⁴⁰
• 90% are sporadic; 10% are familial⁴¹
• 90% are benign; 10% are malignant⁴⁰
• 90% are found in adults; 10% affect children.⁴²

Signs and symptoms of pheochromocytomas include palpitations, headache, dyspnea, diaphoresis, and flushing, as well as paroxysmal hypertension.⁴⁰ Measurement of 24-hour urinary catecholamines and their metabolites has been the screening test of choice,⁴³ but recent evidence suggests that measurement of plasma metanephrine and normetanephrine is a far more sensitive screen.¹⁰

Treatment. Surgical resection is the treatment of choice. Alpha blockade is started 7 to 10 days preoperatively;^44,45 a beta-blocker is added only after an adequate alpha blockade has been established, as unopposed alpha stimulation could precipitate a hypertensive crisis. Laparoscopic adrenalectomy is routinely performed for small (<5 cm) pheochromocytomas.^46,47

Don’t forget these (relatively) common secondary causes

Obstructive sleep apnea (OSA) is one of the most common conditions associated with resistant hypertension.⁴⁸ Signs and symptoms include snoring, daytime somnolence, and obesity (body mass index ≥30 kg/m²).

OSA involves upper airway collapse during inspiration, causing intermittent hypoxemia with resultant sympathetic nervous system activation.¹¹ The underlying mechanism of OSA-induced hypertension is strongly related to sympathetic activation.⁴⁹ Overnight polysomnography is required for diagnosis.¹¹

Continuous positive airway pressure is the treatment of choice for patients unable to lose weight.¹¹

Pregnancy-induced hypertension is the most common medical problem encountered in pregnancy. It occurs in up to 15% of pregnancies, either during the pregnancy itself or postpartum. Postpartum hypertension may be related to preexisting chronic hypertension or to the persistence of gestational hypertension or preeclampsia, which usually occurs after 20 weeks’ gestation and is characterized by the presence of hypertension and proteinuria.⁵⁰ Methyldopa and labetalol are commonly used treatments for hypertension during pregnancy.⁵¹

Drug-induced hypertension. Several drugs can cause or exacerbate hypertension, rendering it resistant to therapy. A careful review of the patient’s medication regimen is essential. Generally, drug-induced hypertension falls into 2 broad categories based on mechanism of action: volume overload and sympathetic activity.^52,53

Corticosteroids can elevate BP in a dose-dependent manner, as a result of volume overload. Glycyrrhizic acid, the main ingredient in licorice, produces a state of excess mineralocorticoid, with a similar effect. Estrogen-containing oral contraceptives can cause an increased synthesis of angiotensin in the liver, resulting in greater aldosterone secretion and volume overload.

Drugs that stimulate sympathetic activity include cocaine, ephedrine, amphetamine, phenylephrine, phenylpropanolamine, caffeine, and alcohol. Nonsteroidal anti-inflammatory drugs may interfere with the action of ACE inhibitors and cause renal vasoconstriction, leading to sodium and water retention and hypertension.⁵⁴

Discontinuation of the medication in question is preferable. In many cases, an agent that does not affect BP can be found to replace it.

If the patient is a child
Hypertension is uncommon in young people. However, coarctation of the aorta, a congenital narrowing associated with secondary hypertension, is typically diagnosed in childhood. In rare cases, the condition remains undetected well into adulthood.⁵⁵ Clinical signs include weak femoral pulses, visible pulsations in the neck, a systolic murmur of the anterior and posterior thorax, and elevated BP in the upper extremities with low BP in the lower extremities.

Thus, once hypertension is confirmed in a young patient, BP should be measured in both arms and legs.⁵⁶ Delayed or absent femoral pulses and a difference in systolic BP of ≥20 mm Hg between the arms and legs provide evidence of aortic coarctation.⁵⁷ In adults, stenting is the initial treatment for this condition because of the morbidity associated with surgery.⁵⁷ Stenting is an option for children with this condition, as well.⁵⁸

CORRESPONDENCE Bernard M. Karnath, MD, University of Texas Medical Branch at Galveston, 301 University Boulevard, Galveston, TX 77555; [email protected]

With all of the discussion about a truly comprehensive and inclusive psychiatry, it was sad to see the emptiness and one-sidedness of Dr. Nasrallah’s June editorial (“Innovative approaches to treatment-resistant depression,” From the Editor, Current Psychiatry, June 2012, p. 4-5; http://bit.ly/1GM92oV). Depression certainly is not a unified diagnosis such as measles or appendicitis. In the face of so-called treatment-resistance, the first step is to review the psychological and biologic formulation of the patient and the reasons for his or her depression. Dr. Nasrallah does not mention the need for a review of all aspects of the patient’s life. The approaches he suggests are dreary, dull, ineffective, and unchanging. It proves that patients are best cared for by psychiatrists who practice biopsychosocial psychiatry and not merely to pay lip service to it.

Arnold Robbins, MD
Private Practice
Cambridge, MA
Clinical Professor of Psychiatry
Boston University School of Medicine
Boston, MA

Dr. Nasrallah responds

I thank Dr. Robbins for his letter. As someone trained by the father of biopsychosocial psychiatry—George Engel, MD, at the University of Rochester—I agree that it is the optimal practice of psychiatry and I practice that approach with all my patients. However, I was describing innovative interventions and paradigm shifts for truly end-of-the-road refractory depression, where all psychosocial and pharmacotherapy treatments have failed and the patient is desperate, disabled, and at high risk for suicide. None of the available interventions work with such individuals and that’s why I regard the innovative breakthroughs I described in my editorial as a promise of hope, thanks to dedicated psychiatric neuroscientists. I hope psychotherapy researchers can achieve breakthroughs for these patients as well.

Henry A. Nasrallah, MD
Editor-in-Chief

Psychotherapy for GAD

I appreciated Dr. Barry’s in-depth review of current diagnostic criteria and therapy for generalized anxiety disorder (GAD) (“Generalized anxiety disorder: Helping patients overcome worry,” Current Psychiatry, May 2012, p. 40-44; http://bit.ly/1oUymyJ). However, I want to point out an error under the “Evidence-based treatments” section labeled “Psychotherapy.” Dr. Barry states that cognitive-behavioral therapy (CBT) is the preferred form of psychotherapy for GAD. In my 40 years of practice, I have found a combination of medication—preferably selective serotonin reuptake inhibitors—and psychodynamic psychotherapy is the most effective treatment for GAD and provides more enduring relief.

Jonathan Shedler, PhD, of the University of Colorado Denver School of Medicine reported on the efficacy of psychodynamic psychotherapy vs behavioral therapy.¹ He compiled the results of meta-analyses of psychotherapy efficacy by 18 investigators covering 792 studies. The findings show a superior result for psychodynamic psychotherapy over behavioral therapy and the effects are more lasting.

Edward L. Parsons, MD
Private Practice
Westfield, NJ

The author responds

I appreciate Dr. Parsons’ comments and his valuable contribution to the dialogue on GAD. Dr. Parsons reemphasizes the importance of psychotherapy in this chronic condition. As demonstrated by the meta-analyses reviewed by Dr. Shedler, there is sufficient evidence in the medical literature to support either psychodynamic psychotherapy or CBT. The specific therapeutic recommendation should consider unique patient variables, such as therapist availability and expertise, the presence of co-occurring conditions or dynamics that would better align with a specific modality, and patient preference and psychological mindedness, to name a few. Regardless, psychotherapy is indicated in the treatment of GAD, and both CBT and psychodynamic psychotherapy are well-supported interventions.

Matthew J. Barry, DO
Lead Psychiatrist
Rochester Veterans Affairs Outpatient Clinic
Canandaigua Veterans Affairs Medical Center
Canandaigua, NY

Caution with prazosin

We welcome the article discussing the use of prazosin and antipsychotics for posttraumatic stress disorder (PTSD)-related nightmares (Graham RL, Leckband SG, Endow-Eyer RA. “PTSD nightmares: Prazosin and atypical antipsychotics,” Current Psychiatry, June 2012, p. 59-62; http://bit.ly/LVAlSo). The favorable outcomes associated with prazosin use combined with its low cost and general tolerability give it considerable potential. Prazosin may be particularly valuable given the unfavorable cardiometabolic risks associated with antipsychotic use, especially because evidence suggests individuals with PTSD have higher rates of cardiovascular disease.¹

We believe the occurrence of adverse cardiovascular effects when starting prazosin requires further attention. As an α1-adrenergic receptor antagonist, it has been linked to orthostatic hypotension and syncope.^2,3 Its cardiovascular effects may be further complicated by concomitant use of other antihypertensive medications. Therefore, we suggest a low initiation dose and gradual titration of prazosin. In individuals who initially were normotensive but then experienced hypotension following prazosin administration, we successfully used short-term sodium chloride tablets, 4 g/d. We discontinued sodium chloride after titration was completed and no postural hypotension was evident.

To minimize polypharmacy, individuals on multiple agents for hypertension may benefit from substituting prazosin for 1 of their regular anti- hypertensives. Despite the mounting evidence supporting prazosin use, it is not indicated for PTSD.

This material is the result of work supported with resources and the use of facilities of the Mental Health and Research and Development Service Lines, Atlanta Veterans Affairs Medical Center, Decatur, GA.

Arshya Vahabzadeh, MD
PGY-2, Resident Psychiatrist
Emory University School of Medicine

Erica Duncan, MD
Attending Psychiatrist
and Associate Professor
Mental Health Service Line
Atlanta Veterans Affairs Medical Center
Decatur, GA
Department of Psychiatry and Behavioral
Sciences
Emory University School of Medicine
Atlanta, GA

The authors respond

We agree with the comments by Drs. Vahabzadeh and Duncan regarding the cardiovascular adverse effects of prazosin. It is important to assess the hemodynamic status of the patient before initiating prazosin therapy, and usually, initiation is attempted only if a patient is normotensive or hypertensive because of potential orthostatic hypotension and syncope, which can occur in up to 4% of patients.¹ As noted by Drs. Vahabzadeh and Duncan, prazosin often is viewed as a dual treatment for both nightmares and blood pressure in individuals who are hypertensive prior to initiation. Prazosin therapy usually is initiated at 1 mg at bedtime and titrated by 1 to 2 mg every 3 to 5 days.² The average dose was approximately 3 mg in studies evaluating prazosin for treating PTSD-associated nightmares (dose range: 1 to 10 mg).² Until the patient is stabilized on a prazosin dose, blood pressure should be monitored daily for inpatients. Outpatients should be educated regarding the signs and symptoms of hypotension, especially dizziness and light-headedness upon standing, along with monitoring blood pressure at his or her next clinic appointment.

Prazosin does not carry an FDA indication for PTSD. Although this is important to consider, the level of evidence in terms of treatment of nightmares also is key. Aurora et al found prazosin was the only medication with a level A rating for treating PTSD-associated nightmares, indicating it as a recommended therapy option.² Because we do not have any medications indicated for PTSD-associated nightmares, it is crucial to practice evidence-based medicine and base therapy choices on available literature supporting the most effective and safe options.

Safety is an issue in many clinicians’ minds, especially when treating geriatric patients with PTSD because of the risk of hypotensive effects with prazosin leading to negative outcomes, such as falls. In Peskind et al’s open-label study of 9 older patients (mean age: 76) with intractable PTSD-associated nightmares treated with prazosin (mean dose: 2.3 mg; increased by 1 mg per week to a maximum dose of 4 mg), 8 patients experienced >50% reduction in nightmares after 8 weeks of treatment, and 1 patient experienced transient orthostasis when starting prazosin that resolved spontaneously with only mild decreases in blood pressure noted otherwise (<20 mm Hg decrease in systolic blood pressure upon standing).³ Although this study was small, it provides evidence that prazosin can be an effective and safe treatment option in geriatric patients and is devoid of the highly sedating side effects of some other treatment options.

Rebecca L. Graham, PharmD
Second-Year Psychiatric Pharmacy Resident
Veterans Affairs San Diego Healthcare System
(VASDHS)

Susan G. Leckband, RPh, BCPP
Clinical Psychiatric Pharmacist Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego

Rene A. Endow-Eyer, PharmD, BCPP
Psychiatric Clinical Pharmacy Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego
San Diego, CA

With all of the discussion about a truly comprehensive and inclusive psychiatry, it was sad to see the emptiness and one-sidedness of Dr. Nasrallah’s June editorial (“Innovative approaches to treatment-resistant depression,” From the Editor, Current Psychiatry, June 2012, p. 4-5; http://bit.ly/1GM92oV). Depression certainly is not a unified diagnosis such as measles or appendicitis. In the face of so-called treatment-resistance, the first step is to review the psychological and biologic formulation of the patient and the reasons for his or her depression. Dr. Nasrallah does not mention the need for a review of all aspects of the patient’s life. The approaches he suggests are dreary, dull, ineffective, and unchanging. It proves that patients are best cared for by psychiatrists who practice biopsychosocial psychiatry and not merely to pay lip service to it.

Arnold Robbins, MD
Private Practice
Cambridge, MA
Clinical Professor of Psychiatry
Boston University School of Medicine
Boston, MA

Dr. Nasrallah responds

I thank Dr. Robbins for his letter. As someone trained by the father of biopsychosocial psychiatry—George Engel, MD, at the University of Rochester—I agree that it is the optimal practice of psychiatry and I practice that approach with all my patients. However, I was describing innovative interventions and paradigm shifts for truly end-of-the-road refractory depression, where all psychosocial and pharmacotherapy treatments have failed and the patient is desperate, disabled, and at high risk for suicide. None of the available interventions work with such individuals and that’s why I regard the innovative breakthroughs I described in my editorial as a promise of hope, thanks to dedicated psychiatric neuroscientists. I hope psychotherapy researchers can achieve breakthroughs for these patients as well.

Henry A. Nasrallah, MD
Editor-in-Chief

Psychotherapy for GAD

I appreciated Dr. Barry’s in-depth review of current diagnostic criteria and therapy for generalized anxiety disorder (GAD) (“Generalized anxiety disorder: Helping patients overcome worry,” Current Psychiatry, May 2012, p. 40-44; http://bit.ly/1oUymyJ). However, I want to point out an error under the “Evidence-based treatments” section labeled “Psychotherapy.” Dr. Barry states that cognitive-behavioral therapy (CBT) is the preferred form of psychotherapy for GAD. In my 40 years of practice, I have found a combination of medication—preferably selective serotonin reuptake inhibitors—and psychodynamic psychotherapy is the most effective treatment for GAD and provides more enduring relief.

Jonathan Shedler, PhD, of the University of Colorado Denver School of Medicine reported on the efficacy of psychodynamic psychotherapy vs behavioral therapy.¹ He compiled the results of meta-analyses of psychotherapy efficacy by 18 investigators covering 792 studies. The findings show a superior result for psychodynamic psychotherapy over behavioral therapy and the effects are more lasting.

Edward L. Parsons, MD
Private Practice
Westfield, NJ

The author responds

I appreciate Dr. Parsons’ comments and his valuable contribution to the dialogue on GAD. Dr. Parsons reemphasizes the importance of psychotherapy in this chronic condition. As demonstrated by the meta-analyses reviewed by Dr. Shedler, there is sufficient evidence in the medical literature to support either psychodynamic psychotherapy or CBT. The specific therapeutic recommendation should consider unique patient variables, such as therapist availability and expertise, the presence of co-occurring conditions or dynamics that would better align with a specific modality, and patient preference and psychological mindedness, to name a few. Regardless, psychotherapy is indicated in the treatment of GAD, and both CBT and psychodynamic psychotherapy are well-supported interventions.

Matthew J. Barry, DO
Lead Psychiatrist
Rochester Veterans Affairs Outpatient Clinic
Canandaigua Veterans Affairs Medical Center
Canandaigua, NY

Caution with prazosin

We welcome the article discussing the use of prazosin and antipsychotics for posttraumatic stress disorder (PTSD)-related nightmares (Graham RL, Leckband SG, Endow-Eyer RA. “PTSD nightmares: Prazosin and atypical antipsychotics,” Current Psychiatry, June 2012, p. 59-62; http://bit.ly/LVAlSo). The favorable outcomes associated with prazosin use combined with its low cost and general tolerability give it considerable potential. Prazosin may be particularly valuable given the unfavorable cardiometabolic risks associated with antipsychotic use, especially because evidence suggests individuals with PTSD have higher rates of cardiovascular disease.¹

We believe the occurrence of adverse cardiovascular effects when starting prazosin requires further attention. As an α1-adrenergic receptor antagonist, it has been linked to orthostatic hypotension and syncope.^2,3 Its cardiovascular effects may be further complicated by concomitant use of other antihypertensive medications. Therefore, we suggest a low initiation dose and gradual titration of prazosin. In individuals who initially were normotensive but then experienced hypotension following prazosin administration, we successfully used short-term sodium chloride tablets, 4 g/d. We discontinued sodium chloride after titration was completed and no postural hypotension was evident.

To minimize polypharmacy, individuals on multiple agents for hypertension may benefit from substituting prazosin for 1 of their regular anti- hypertensives. Despite the mounting evidence supporting prazosin use, it is not indicated for PTSD.

This material is the result of work supported with resources and the use of facilities of the Mental Health and Research and Development Service Lines, Atlanta Veterans Affairs Medical Center, Decatur, GA.

Arshya Vahabzadeh, MD
PGY-2, Resident Psychiatrist
Emory University School of Medicine

Erica Duncan, MD
Attending Psychiatrist
and Associate Professor
Mental Health Service Line
Atlanta Veterans Affairs Medical Center
Decatur, GA
Department of Psychiatry and Behavioral
Sciences
Emory University School of Medicine
Atlanta, GA

The authors respond

We agree with the comments by Drs. Vahabzadeh and Duncan regarding the cardiovascular adverse effects of prazosin. It is important to assess the hemodynamic status of the patient before initiating prazosin therapy, and usually, initiation is attempted only if a patient is normotensive or hypertensive because of potential orthostatic hypotension and syncope, which can occur in up to 4% of patients.¹ As noted by Drs. Vahabzadeh and Duncan, prazosin often is viewed as a dual treatment for both nightmares and blood pressure in individuals who are hypertensive prior to initiation. Prazosin therapy usually is initiated at 1 mg at bedtime and titrated by 1 to 2 mg every 3 to 5 days.² The average dose was approximately 3 mg in studies evaluating prazosin for treating PTSD-associated nightmares (dose range: 1 to 10 mg).² Until the patient is stabilized on a prazosin dose, blood pressure should be monitored daily for inpatients. Outpatients should be educated regarding the signs and symptoms of hypotension, especially dizziness and light-headedness upon standing, along with monitoring blood pressure at his or her next clinic appointment.

Prazosin does not carry an FDA indication for PTSD. Although this is important to consider, the level of evidence in terms of treatment of nightmares also is key. Aurora et al found prazosin was the only medication with a level A rating for treating PTSD-associated nightmares, indicating it as a recommended therapy option.² Because we do not have any medications indicated for PTSD-associated nightmares, it is crucial to practice evidence-based medicine and base therapy choices on available literature supporting the most effective and safe options.

Safety is an issue in many clinicians’ minds, especially when treating geriatric patients with PTSD because of the risk of hypotensive effects with prazosin leading to negative outcomes, such as falls. In Peskind et al’s open-label study of 9 older patients (mean age: 76) with intractable PTSD-associated nightmares treated with prazosin (mean dose: 2.3 mg; increased by 1 mg per week to a maximum dose of 4 mg), 8 patients experienced >50% reduction in nightmares after 8 weeks of treatment, and 1 patient experienced transient orthostasis when starting prazosin that resolved spontaneously with only mild decreases in blood pressure noted otherwise (<20 mm Hg decrease in systolic blood pressure upon standing).³ Although this study was small, it provides evidence that prazosin can be an effective and safe treatment option in geriatric patients and is devoid of the highly sedating side effects of some other treatment options.

Rebecca L. Graham, PharmD
Second-Year Psychiatric Pharmacy Resident
Veterans Affairs San Diego Healthcare System
(VASDHS)

Susan G. Leckband, RPh, BCPP
Clinical Psychiatric Pharmacist Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego

Rene A. Endow-Eyer, PharmD, BCPP
Psychiatric Clinical Pharmacy Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego
San Diego, CA

With all of the discussion about a truly comprehensive and inclusive psychiatry, it was sad to see the emptiness and one-sidedness of Dr. Nasrallah’s June editorial (“Innovative approaches to treatment-resistant depression,” From the Editor, Current Psychiatry, June 2012, p. 4-5; http://bit.ly/1GM92oV). Depression certainly is not a unified diagnosis such as measles or appendicitis. In the face of so-called treatment-resistance, the first step is to review the psychological and biologic formulation of the patient and the reasons for his or her depression. Dr. Nasrallah does not mention the need for a review of all aspects of the patient’s life. The approaches he suggests are dreary, dull, ineffective, and unchanging. It proves that patients are best cared for by psychiatrists who practice biopsychosocial psychiatry and not merely to pay lip service to it.

Arnold Robbins, MD
Private Practice
Cambridge, MA
Clinical Professor of Psychiatry
Boston University School of Medicine
Boston, MA

Dr. Nasrallah responds

I thank Dr. Robbins for his letter. As someone trained by the father of biopsychosocial psychiatry—George Engel, MD, at the University of Rochester—I agree that it is the optimal practice of psychiatry and I practice that approach with all my patients. However, I was describing innovative interventions and paradigm shifts for truly end-of-the-road refractory depression, where all psychosocial and pharmacotherapy treatments have failed and the patient is desperate, disabled, and at high risk for suicide. None of the available interventions work with such individuals and that’s why I regard the innovative breakthroughs I described in my editorial as a promise of hope, thanks to dedicated psychiatric neuroscientists. I hope psychotherapy researchers can achieve breakthroughs for these patients as well.

Henry A. Nasrallah, MD
Editor-in-Chief

Psychotherapy for GAD

I appreciated Dr. Barry’s in-depth review of current diagnostic criteria and therapy for generalized anxiety disorder (GAD) (“Generalized anxiety disorder: Helping patients overcome worry,” Current Psychiatry, May 2012, p. 40-44; http://bit.ly/1oUymyJ). However, I want to point out an error under the “Evidence-based treatments” section labeled “Psychotherapy.” Dr. Barry states that cognitive-behavioral therapy (CBT) is the preferred form of psychotherapy for GAD. In my 40 years of practice, I have found a combination of medication—preferably selective serotonin reuptake inhibitors—and psychodynamic psychotherapy is the most effective treatment for GAD and provides more enduring relief.

Jonathan Shedler, PhD, of the University of Colorado Denver School of Medicine reported on the efficacy of psychodynamic psychotherapy vs behavioral therapy.¹ He compiled the results of meta-analyses of psychotherapy efficacy by 18 investigators covering 792 studies. The findings show a superior result for psychodynamic psychotherapy over behavioral therapy and the effects are more lasting.

Edward L. Parsons, MD
Private Practice
Westfield, NJ

The author responds

I appreciate Dr. Parsons’ comments and his valuable contribution to the dialogue on GAD. Dr. Parsons reemphasizes the importance of psychotherapy in this chronic condition. As demonstrated by the meta-analyses reviewed by Dr. Shedler, there is sufficient evidence in the medical literature to support either psychodynamic psychotherapy or CBT. The specific therapeutic recommendation should consider unique patient variables, such as therapist availability and expertise, the presence of co-occurring conditions or dynamics that would better align with a specific modality, and patient preference and psychological mindedness, to name a few. Regardless, psychotherapy is indicated in the treatment of GAD, and both CBT and psychodynamic psychotherapy are well-supported interventions.

Matthew J. Barry, DO
Lead Psychiatrist
Rochester Veterans Affairs Outpatient Clinic
Canandaigua Veterans Affairs Medical Center
Canandaigua, NY

Caution with prazosin

We welcome the article discussing the use of prazosin and antipsychotics for posttraumatic stress disorder (PTSD)-related nightmares (Graham RL, Leckband SG, Endow-Eyer RA. “PTSD nightmares: Prazosin and atypical antipsychotics,” Current Psychiatry, June 2012, p. 59-62; http://bit.ly/LVAlSo). The favorable outcomes associated with prazosin use combined with its low cost and general tolerability give it considerable potential. Prazosin may be particularly valuable given the unfavorable cardiometabolic risks associated with antipsychotic use, especially because evidence suggests individuals with PTSD have higher rates of cardiovascular disease.¹

We believe the occurrence of adverse cardiovascular effects when starting prazosin requires further attention. As an α1-adrenergic receptor antagonist, it has been linked to orthostatic hypotension and syncope.^2,3 Its cardiovascular effects may be further complicated by concomitant use of other antihypertensive medications. Therefore, we suggest a low initiation dose and gradual titration of prazosin. In individuals who initially were normotensive but then experienced hypotension following prazosin administration, we successfully used short-term sodium chloride tablets, 4 g/d. We discontinued sodium chloride after titration was completed and no postural hypotension was evident.

To minimize polypharmacy, individuals on multiple agents for hypertension may benefit from substituting prazosin for 1 of their regular anti- hypertensives. Despite the mounting evidence supporting prazosin use, it is not indicated for PTSD.

This material is the result of work supported with resources and the use of facilities of the Mental Health and Research and Development Service Lines, Atlanta Veterans Affairs Medical Center, Decatur, GA.

Arshya Vahabzadeh, MD
PGY-2, Resident Psychiatrist
Emory University School of Medicine

Erica Duncan, MD
Attending Psychiatrist
and Associate Professor
Mental Health Service Line
Atlanta Veterans Affairs Medical Center
Decatur, GA
Department of Psychiatry and Behavioral
Sciences
Emory University School of Medicine
Atlanta, GA

The authors respond

We agree with the comments by Drs. Vahabzadeh and Duncan regarding the cardiovascular adverse effects of prazosin. It is important to assess the hemodynamic status of the patient before initiating prazosin therapy, and usually, initiation is attempted only if a patient is normotensive or hypertensive because of potential orthostatic hypotension and syncope, which can occur in up to 4% of patients.¹ As noted by Drs. Vahabzadeh and Duncan, prazosin often is viewed as a dual treatment for both nightmares and blood pressure in individuals who are hypertensive prior to initiation. Prazosin therapy usually is initiated at 1 mg at bedtime and titrated by 1 to 2 mg every 3 to 5 days.² The average dose was approximately 3 mg in studies evaluating prazosin for treating PTSD-associated nightmares (dose range: 1 to 10 mg).² Until the patient is stabilized on a prazosin dose, blood pressure should be monitored daily for inpatients. Outpatients should be educated regarding the signs and symptoms of hypotension, especially dizziness and light-headedness upon standing, along with monitoring blood pressure at his or her next clinic appointment.

Prazosin does not carry an FDA indication for PTSD. Although this is important to consider, the level of evidence in terms of treatment of nightmares also is key. Aurora et al found prazosin was the only medication with a level A rating for treating PTSD-associated nightmares, indicating it as a recommended therapy option.² Because we do not have any medications indicated for PTSD-associated nightmares, it is crucial to practice evidence-based medicine and base therapy choices on available literature supporting the most effective and safe options.

Safety is an issue in many clinicians’ minds, especially when treating geriatric patients with PTSD because of the risk of hypotensive effects with prazosin leading to negative outcomes, such as falls. In Peskind et al’s open-label study of 9 older patients (mean age: 76) with intractable PTSD-associated nightmares treated with prazosin (mean dose: 2.3 mg; increased by 1 mg per week to a maximum dose of 4 mg), 8 patients experienced >50% reduction in nightmares after 8 weeks of treatment, and 1 patient experienced transient orthostasis when starting prazosin that resolved spontaneously with only mild decreases in blood pressure noted otherwise (<20 mm Hg decrease in systolic blood pressure upon standing).³ Although this study was small, it provides evidence that prazosin can be an effective and safe treatment option in geriatric patients and is devoid of the highly sedating side effects of some other treatment options.

Rebecca L. Graham, PharmD
Second-Year Psychiatric Pharmacy Resident
Veterans Affairs San Diego Healthcare System
(VASDHS)

Susan G. Leckband, RPh, BCPP
Clinical Psychiatric Pharmacist Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego

Rene A. Endow-Eyer, PharmD, BCPP
Psychiatric Clinical Pharmacy Specialist
VASDHS
Assistant Clinical Professor
Skaggs School of Pharmacy and
Pharmaceutical Sciences
Department of Psychiatry
University of California, San Diego
San Diego, CA

Nearly 2.5 million persons die each year in the United States.¹ For the bereaved, these deaths may be among the most painful and disruptive events they will experience. In this article, we evaluate the growing body of research on complicated grief (CG)—which also has been called prolonged grief, chronic grief, traumatic grief, and pathological grief—with an emphasis on how to identify CG and distinguish it from other adaptive and maladaptive reactions to the loss of a loved one. In addition, we review empirical evidence on treating CG, including psychotherapy, pharmacotherapy, and combined treatment approaches.

The bereavement-specific syndrome we refer to as CG currently is being reviewed for possible inclusion in DSM-5 as an official diagnosis. At press time, proposals for DSM-5 included a bereavement-related adjustment disorder within the new Trauma- and Stressor-Related Disorders category, as well as a provisional diagnosis of CG entitled Persistent Complex Bereavement-Related Disorder, which, upon acceptance, would be listed in Section III.²

What is ‘normal’ grief?

Grief is highly variable across individuals and time and may range from an absence of distress to severe and persistent pain and anguish. There’s no simple definition of “normal grief.” However, as clinicians, it’s necessary to understand the range of usual reactions. We recommend 2 considerations when evaluating grief reactions.

First, be aware that grief encompasses a range of cognitions, emotions, and behaviors. It may range from a relative lack of painful thoughts and emotions to intense and disruptive sadness, loneliness, anger, guilt, intrusive thoughts, difficulty concentrating, preoccupation with loss, social withdrawal, and a sense of being overwhelmed by the loss and its consequences. In the months after a loss, bereaved individuals may look for the deceased in a crowd, speak to them, or even experience auditory or visual hallucinations of the deceased. Nonetheless, positive feelings such as relief, peace, and happiness also are common following a loss.³ Moreover, laughter and smiling when discussing a lost loved one predicts reductions in grief symptoms over time.⁴ Overall, grief research suggests that, far from proceeding along standard and uniform stages,⁵ grief is complex and comprises a broad spectrum of thoughts, feelings, and behaviors that vary within and among individuals.

Second, note that in the absence of complications, grief progresses. For those who experience elevated levels of distress, the pain and disruption of loss initially may feel overwhelming but will subside in intensity over time for most individuals.⁵ This is not to say that an individual will never again feel sadness or longing for the deceased; elements of grief are likely to remain. Although the trajectory of grief symptoms varies among individuals and may progress in fits and starts, over time grief becomes more intermittent, less interfering, and is balanced with a sense of interest and purpose in life.

What is CG?

As research on grief experiences has grown, there’s increasing recognition that a minority of bereaved individuals experience more extreme grief symptoms that cause substantial, persistent distress and impairment despite the passage of many months or years. Shear et al⁶ proposed a set of CG diagnostic criteria (Table) in which a cluster of symptoms of intense and persistent separation distress are defined as core symptoms. Similar to other psychiatric diagnoses, the symptoms must be associated with significant distress or impairment.

Table

Proposed diagnostic criteria for complicated grief

Assessing CG symptoms

Among those with persistent elevated distress, a CG diagnosis must be considered in the context of the individual’s social and cultural environment, time since the loss, and duration of symptoms. The hallmark symptom of CG is separation distress with a focus of cognitive, behavioral, and emotional symptoms on the loss and its consequences. CG is associated with substantial distress, functional impairment, and an increased risk for suicide. See the Box for a case study.

Many individuals with CG remain undiagnosed and untreated for years despite high levels of distress and impairment and high risk for negative consequences such as suicide.⁷ Accordingly, there’s a need for greater CG screening. Clinically useful tools for assessing CG include a brief, 5-item dimensional screening assessment⁶ and the patient-rated Inventory of Complicated Grief.⁸

Distinguishing complicated and uncomplicated grief. Exhibiting CG symptoms in the first several months after a loss does not mean an individual has or will develop CG. Most bereaved adults report painful thoughts and emotions in the weeks and months following the loss, including distressed yearning, waves of intense grief, persistent and intrusive thoughts, images related to the death, somatic distress, and a feeling of being disconnected from others. For most individuals, the intensity of this response diminishes within 6 to 18 months after the loved one’s death.⁵ Although the optimal length of time to wait before establishing a diagnosis remains debatable, the earliest CG should be diagnosed is 6 months after a loss.

It’s common for grief to occasionally rise in intensity for days or weeks. This surge may occur many months or years after the loss, even in people who exhibited relatively little distress or impairment. In particular, anniversaries, holidays, or periods of stress may trigger increased grief intensity. However, these surges typically subside naturally within a short time. Accordingly, CG should be diagnosed only when symptoms persist for >1 month.

CG vs other post-loss disorders. CG, major depressive disorder (MDD), and posttraumatic stress disorder (PTSD) often are comorbid in bereaved adults. Simon et al⁹ found 72% of CG patients in a treatment- seeking sample reported a lifetime history of MDD and 53% reported a lifetime history of PTSD. However, CG can be distinguished from these disorders. In the same study, 25% of CG patients had no other axis I diagnosis.⁹ After accounting for comorbid disorders, researchers associated CG severity with work and social impairment. These findings provide clear evidence for the incremental validity of CG—ie, a CG diagnosis gives clinicians additional information that predicts impairment above and beyond other disorders. However, future research needs to further examine CG and its overlap and differentiation from MDD and PTSD.

Distinguishing CG and MDD. Intense yearning or preoccupation with the deceased is a common symptom of CG but not MDD. In addition, CG symptoms possess intentionality. For example, emotional distress such as sadness and anger are prominent features of both CG and MDD. However, in CG, these symptoms are specific to the loss or circumstances of the loss, whereas in MDD they generally are more nebulous and generalized. Similarly, CG entails proximity seeking related to the deceased, and avoidance of reminders of the deceased, whereas MDD includes a more general social withdrawal and anhedonia.

Distinguishing CG and PTSD. CG and loss-related PTSD are distinguished by the predominant emotions and focus of concern associated with each disorder. The predominant emotion in PTSD is fear, whereas in CG it is sadness and longing. In PTSD, intrusive thoughts and memories associated with the trauma generally are associated with the event itself and produce an ongoing sense of threat.¹⁰ Avoidance in PTSD is intended to reduce this threat feeling. By contrast, in CG, intrusive memories focus on the deceased or the circumstances of the death, and avoidance is aimed at preventing painful reminders of the loss or its permanence. Importantly, both syndromes may be present.

Box

Treating CG

When is treatment indicated? For years, bereavement theorists emphasized the need to work through emotions and memories related to the deceased with particular focus on negative material. However, evidence suggests that universal application of treatment to all bereaved individuals is unhelpful. In a recent meta-analysis, Neimeyer et al¹¹ found that the outcomes of grief therapy applied indiscriminately to all bereaved adults or all members of high-risk populations—such as parents whose child experienced a violent death—were no better than would be expected by the passage of time. In contrast, grief therapy applied only to those who develop elevated and persistent distress (eg, CG) led to greater and more enduring improvement in post-loss distress than was observed in control conditions.

These results suggest that most grieving individuals who do not meet criteria for CG (or other psychiatric disorders) will not require intervention. Those who do seek treatment for grief-related distress in the acute grief period should be assessed for bereavement-related depression, anxiety, and suicidality, and treated or referred to professional or community-based resources for support or counseling as clinically indicated.

Evidence for psychotherapy. For those who meet CG criteria, psychotherapy targeting the specific symptoms of CG is helpful. The evidence is strongest for CG treatment (CGT), a 16-session, manualized psychotherapy developed by M. Katherine Shear, MD.¹² CGT is based on an attachment model and cognitive-behavioral therapy (CBT) principles, and is informed by the dual-process theory proposed by Stroebe et al.¹³ According to this theory, natural healing following loss comprises 2 processes:

• a loss-oriented process in which the patient comes to terms with the loss, and
• a restoration-oriented process in which the patient reinvigorates a sense of purpose and meaning in life without the deceased.

CGT focuses on both processes. To address the former, it includes clinician-guided exercises in which the patient revisits the time of the death and planned activities in which the patient reengages with people, places, or thoughts that remind him or her of the deceased. CGT aims to allow the patient to gain an increased tolerance of the distressing thoughts and emotions associated with the loss so that these thoughts can be processed and the finality of the death and its circumstances can be accepted.

The restoration process is addressed by having patients generate and discuss personal goals and aspirations for the near and distant future, as well as scheduling pleasurable and rewarding events. This is accomplished by having patients imagine what they would want for themselves if their grief was less intense and planning concrete steps to take toward these goals. The restoration-oriented process is addressed concurrent with the loss-oriented process to encourage the oscillation between processes thought to be characteristic of a natural healing process following the loss of a loved one.

Other psychotherapy approaches (eg, support groups) may have a role for some individuals, and future research may suggest alternative approaches to CGT. To date, CGT is the most targeted evidence-based psychotherapy with randomized controlled data supporting its use in CG.

Pharmacotherapy for CG. Early research suggested that antidepressants—in particular tricyclics—may effectively reduce depressive symptoms in bereavement-related depression; their effect on CG symptoms, however, may not be as strong.¹⁴ Research on pharmacologic treatment that targets CG symptoms is developing. Because of the overlap between CG, PTSD, and MDD, researchers have hypothesized that antidepressants may be effective. Two open-label studies reported that the selective serotonin reuptake inhibitor (SSRI) escitalopram may be effective for CG.^15,16 Although a post-hoc comparison of paroxetine and nortriptyline¹⁷ showed significant reduction in CG and depressive symptoms with both agents, effects could not be separated from concomitant psychotherapy. Furthermore, an examination of naturalistic data on combining antidepressants with CGT suggested that antidepressants may improve outcomes for individuals receiving CGT.¹⁸ A multicenter, randomized controlled trial funded by the National Institute of Mental Health is examining the potential efficacy of citalopram, an SSRI, alone or in combination with CGT.¹⁹

The efficacy of benzodiazepines, which commonly are prescribed for bereaved individuals, has not been assessed in CG. However, recent research suggests they may not be useful for medically managing recent grief²⁰ and that their use in the aftermath of a loss may lead to long-term dependence in geriatric patients.²¹

Related Resources

• Center for Anxiety and Traumatic Stress Disorders. Massachusetts General Hospital. www.bostongrief.com.
• Zisook S, Shear K. Grief and bereavement: what psychiatrists need to know. World Psychiatry. 2009;8(2):67-74.
• Bonanno G. The other side of sadness: what the new science of bereavement tells us about loss. New York, NY: Basic Books; 2009.

Drug Brand Names

• Citalopram • Celexa
• Nortriptyline • Aventyl, Pamelor
• Escitalopram • Lexapro
• Paroxetine • Paxil

Disclosures

Dr. Simon receives grant or research support from the American Cancer Society, the American Foundation for Suicide Prevention, the Department of Defense, Forest Laboratories, and the National Institute of Mental Health.

Nearly 2.5 million persons die each year in the United States.¹ For the bereaved, these deaths may be among the most painful and disruptive events they will experience. In this article, we evaluate the growing body of research on complicated grief (CG)—which also has been called prolonged grief, chronic grief, traumatic grief, and pathological grief—with an emphasis on how to identify CG and distinguish it from other adaptive and maladaptive reactions to the loss of a loved one. In addition, we review empirical evidence on treating CG, including psychotherapy, pharmacotherapy, and combined treatment approaches.

The bereavement-specific syndrome we refer to as CG currently is being reviewed for possible inclusion in DSM-5 as an official diagnosis. At press time, proposals for DSM-5 included a bereavement-related adjustment disorder within the new Trauma- and Stressor-Related Disorders category, as well as a provisional diagnosis of CG entitled Persistent Complex Bereavement-Related Disorder, which, upon acceptance, would be listed in Section III.²

What is ‘normal’ grief?

Grief is highly variable across individuals and time and may range from an absence of distress to severe and persistent pain and anguish. There’s no simple definition of “normal grief.” However, as clinicians, it’s necessary to understand the range of usual reactions. We recommend 2 considerations when evaluating grief reactions.

First, be aware that grief encompasses a range of cognitions, emotions, and behaviors. It may range from a relative lack of painful thoughts and emotions to intense and disruptive sadness, loneliness, anger, guilt, intrusive thoughts, difficulty concentrating, preoccupation with loss, social withdrawal, and a sense of being overwhelmed by the loss and its consequences. In the months after a loss, bereaved individuals may look for the deceased in a crowd, speak to them, or even experience auditory or visual hallucinations of the deceased. Nonetheless, positive feelings such as relief, peace, and happiness also are common following a loss.³ Moreover, laughter and smiling when discussing a lost loved one predicts reductions in grief symptoms over time.⁴ Overall, grief research suggests that, far from proceeding along standard and uniform stages,⁵ grief is complex and comprises a broad spectrum of thoughts, feelings, and behaviors that vary within and among individuals.

Second, note that in the absence of complications, grief progresses. For those who experience elevated levels of distress, the pain and disruption of loss initially may feel overwhelming but will subside in intensity over time for most individuals.⁵ This is not to say that an individual will never again feel sadness or longing for the deceased; elements of grief are likely to remain. Although the trajectory of grief symptoms varies among individuals and may progress in fits and starts, over time grief becomes more intermittent, less interfering, and is balanced with a sense of interest and purpose in life.

What is CG?

As research on grief experiences has grown, there’s increasing recognition that a minority of bereaved individuals experience more extreme grief symptoms that cause substantial, persistent distress and impairment despite the passage of many months or years. Shear et al⁶ proposed a set of CG diagnostic criteria (Table) in which a cluster of symptoms of intense and persistent separation distress are defined as core symptoms. Similar to other psychiatric diagnoses, the symptoms must be associated with significant distress or impairment.

Table

Proposed diagnostic criteria for complicated grief

Assessing CG symptoms

Among those with persistent elevated distress, a CG diagnosis must be considered in the context of the individual’s social and cultural environment, time since the loss, and duration of symptoms. The hallmark symptom of CG is separation distress with a focus of cognitive, behavioral, and emotional symptoms on the loss and its consequences. CG is associated with substantial distress, functional impairment, and an increased risk for suicide. See the Box for a case study.

Many individuals with CG remain undiagnosed and untreated for years despite high levels of distress and impairment and high risk for negative consequences such as suicide.⁷ Accordingly, there’s a need for greater CG screening. Clinically useful tools for assessing CG include a brief, 5-item dimensional screening assessment⁶ and the patient-rated Inventory of Complicated Grief.⁸

Distinguishing complicated and uncomplicated grief. Exhibiting CG symptoms in the first several months after a loss does not mean an individual has or will develop CG. Most bereaved adults report painful thoughts and emotions in the weeks and months following the loss, including distressed yearning, waves of intense grief, persistent and intrusive thoughts, images related to the death, somatic distress, and a feeling of being disconnected from others. For most individuals, the intensity of this response diminishes within 6 to 18 months after the loved one’s death.⁵ Although the optimal length of time to wait before establishing a diagnosis remains debatable, the earliest CG should be diagnosed is 6 months after a loss.

It’s common for grief to occasionally rise in intensity for days or weeks. This surge may occur many months or years after the loss, even in people who exhibited relatively little distress or impairment. In particular, anniversaries, holidays, or periods of stress may trigger increased grief intensity. However, these surges typically subside naturally within a short time. Accordingly, CG should be diagnosed only when symptoms persist for >1 month.

CG vs other post-loss disorders. CG, major depressive disorder (MDD), and posttraumatic stress disorder (PTSD) often are comorbid in bereaved adults. Simon et al⁹ found 72% of CG patients in a treatment- seeking sample reported a lifetime history of MDD and 53% reported a lifetime history of PTSD. However, CG can be distinguished from these disorders. In the same study, 25% of CG patients had no other axis I diagnosis.⁹ After accounting for comorbid disorders, researchers associated CG severity with work and social impairment. These findings provide clear evidence for the incremental validity of CG—ie, a CG diagnosis gives clinicians additional information that predicts impairment above and beyond other disorders. However, future research needs to further examine CG and its overlap and differentiation from MDD and PTSD.

Distinguishing CG and MDD. Intense yearning or preoccupation with the deceased is a common symptom of CG but not MDD. In addition, CG symptoms possess intentionality. For example, emotional distress such as sadness and anger are prominent features of both CG and MDD. However, in CG, these symptoms are specific to the loss or circumstances of the loss, whereas in MDD they generally are more nebulous and generalized. Similarly, CG entails proximity seeking related to the deceased, and avoidance of reminders of the deceased, whereas MDD includes a more general social withdrawal and anhedonia.

Distinguishing CG and PTSD. CG and loss-related PTSD are distinguished by the predominant emotions and focus of concern associated with each disorder. The predominant emotion in PTSD is fear, whereas in CG it is sadness and longing. In PTSD, intrusive thoughts and memories associated with the trauma generally are associated with the event itself and produce an ongoing sense of threat.¹⁰ Avoidance in PTSD is intended to reduce this threat feeling. By contrast, in CG, intrusive memories focus on the deceased or the circumstances of the death, and avoidance is aimed at preventing painful reminders of the loss or its permanence. Importantly, both syndromes may be present.

Box

Treating CG

When is treatment indicated? For years, bereavement theorists emphasized the need to work through emotions and memories related to the deceased with particular focus on negative material. However, evidence suggests that universal application of treatment to all bereaved individuals is unhelpful. In a recent meta-analysis, Neimeyer et al¹¹ found that the outcomes of grief therapy applied indiscriminately to all bereaved adults or all members of high-risk populations—such as parents whose child experienced a violent death—were no better than would be expected by the passage of time. In contrast, grief therapy applied only to those who develop elevated and persistent distress (eg, CG) led to greater and more enduring improvement in post-loss distress than was observed in control conditions.

These results suggest that most grieving individuals who do not meet criteria for CG (or other psychiatric disorders) will not require intervention. Those who do seek treatment for grief-related distress in the acute grief period should be assessed for bereavement-related depression, anxiety, and suicidality, and treated or referred to professional or community-based resources for support or counseling as clinically indicated.

Evidence for psychotherapy. For those who meet CG criteria, psychotherapy targeting the specific symptoms of CG is helpful. The evidence is strongest for CG treatment (CGT), a 16-session, manualized psychotherapy developed by M. Katherine Shear, MD.¹² CGT is based on an attachment model and cognitive-behavioral therapy (CBT) principles, and is informed by the dual-process theory proposed by Stroebe et al.¹³ According to this theory, natural healing following loss comprises 2 processes:

• a loss-oriented process in which the patient comes to terms with the loss, and
• a restoration-oriented process in which the patient reinvigorates a sense of purpose and meaning in life without the deceased.

CGT focuses on both processes. To address the former, it includes clinician-guided exercises in which the patient revisits the time of the death and planned activities in which the patient reengages with people, places, or thoughts that remind him or her of the deceased. CGT aims to allow the patient to gain an increased tolerance of the distressing thoughts and emotions associated with the loss so that these thoughts can be processed and the finality of the death and its circumstances can be accepted.

The restoration process is addressed by having patients generate and discuss personal goals and aspirations for the near and distant future, as well as scheduling pleasurable and rewarding events. This is accomplished by having patients imagine what they would want for themselves if their grief was less intense and planning concrete steps to take toward these goals. The restoration-oriented process is addressed concurrent with the loss-oriented process to encourage the oscillation between processes thought to be characteristic of a natural healing process following the loss of a loved one.

Other psychotherapy approaches (eg, support groups) may have a role for some individuals, and future research may suggest alternative approaches to CGT. To date, CGT is the most targeted evidence-based psychotherapy with randomized controlled data supporting its use in CG.

Pharmacotherapy for CG. Early research suggested that antidepressants—in particular tricyclics—may effectively reduce depressive symptoms in bereavement-related depression; their effect on CG symptoms, however, may not be as strong.¹⁴ Research on pharmacologic treatment that targets CG symptoms is developing. Because of the overlap between CG, PTSD, and MDD, researchers have hypothesized that antidepressants may be effective. Two open-label studies reported that the selective serotonin reuptake inhibitor (SSRI) escitalopram may be effective for CG.^15,16 Although a post-hoc comparison of paroxetine and nortriptyline¹⁷ showed significant reduction in CG and depressive symptoms with both agents, effects could not be separated from concomitant psychotherapy. Furthermore, an examination of naturalistic data on combining antidepressants with CGT suggested that antidepressants may improve outcomes for individuals receiving CGT.¹⁸ A multicenter, randomized controlled trial funded by the National Institute of Mental Health is examining the potential efficacy of citalopram, an SSRI, alone or in combination with CGT.¹⁹

The efficacy of benzodiazepines, which commonly are prescribed for bereaved individuals, has not been assessed in CG. However, recent research suggests they may not be useful for medically managing recent grief²⁰ and that their use in the aftermath of a loss may lead to long-term dependence in geriatric patients.²¹

Related Resources

• Center for Anxiety and Traumatic Stress Disorders. Massachusetts General Hospital. www.bostongrief.com.
• Zisook S, Shear K. Grief and bereavement: what psychiatrists need to know. World Psychiatry. 2009;8(2):67-74.
• Bonanno G. The other side of sadness: what the new science of bereavement tells us about loss. New York, NY: Basic Books; 2009.

Drug Brand Names

• Citalopram • Celexa
• Nortriptyline • Aventyl, Pamelor
• Escitalopram • Lexapro
• Paroxetine • Paxil

Disclosures

Dr. Simon receives grant or research support from the American Cancer Society, the American Foundation for Suicide Prevention, the Department of Defense, Forest Laboratories, and the National Institute of Mental Health.

Nearly 2.5 million persons die each year in the United States.¹ For the bereaved, these deaths may be among the most painful and disruptive events they will experience. In this article, we evaluate the growing body of research on complicated grief (CG)—which also has been called prolonged grief, chronic grief, traumatic grief, and pathological grief—with an emphasis on how to identify CG and distinguish it from other adaptive and maladaptive reactions to the loss of a loved one. In addition, we review empirical evidence on treating CG, including psychotherapy, pharmacotherapy, and combined treatment approaches.

The bereavement-specific syndrome we refer to as CG currently is being reviewed for possible inclusion in DSM-5 as an official diagnosis. At press time, proposals for DSM-5 included a bereavement-related adjustment disorder within the new Trauma- and Stressor-Related Disorders category, as well as a provisional diagnosis of CG entitled Persistent Complex Bereavement-Related Disorder, which, upon acceptance, would be listed in Section III.²

What is ‘normal’ grief?

Grief is highly variable across individuals and time and may range from an absence of distress to severe and persistent pain and anguish. There’s no simple definition of “normal grief.” However, as clinicians, it’s necessary to understand the range of usual reactions. We recommend 2 considerations when evaluating grief reactions.

First, be aware that grief encompasses a range of cognitions, emotions, and behaviors. It may range from a relative lack of painful thoughts and emotions to intense and disruptive sadness, loneliness, anger, guilt, intrusive thoughts, difficulty concentrating, preoccupation with loss, social withdrawal, and a sense of being overwhelmed by the loss and its consequences. In the months after a loss, bereaved individuals may look for the deceased in a crowd, speak to them, or even experience auditory or visual hallucinations of the deceased. Nonetheless, positive feelings such as relief, peace, and happiness also are common following a loss.³ Moreover, laughter and smiling when discussing a lost loved one predicts reductions in grief symptoms over time.⁴ Overall, grief research suggests that, far from proceeding along standard and uniform stages,⁵ grief is complex and comprises a broad spectrum of thoughts, feelings, and behaviors that vary within and among individuals.

Second, note that in the absence of complications, grief progresses. For those who experience elevated levels of distress, the pain and disruption of loss initially may feel overwhelming but will subside in intensity over time for most individuals.⁵ This is not to say that an individual will never again feel sadness or longing for the deceased; elements of grief are likely to remain. Although the trajectory of grief symptoms varies among individuals and may progress in fits and starts, over time grief becomes more intermittent, less interfering, and is balanced with a sense of interest and purpose in life.

What is CG?

As research on grief experiences has grown, there’s increasing recognition that a minority of bereaved individuals experience more extreme grief symptoms that cause substantial, persistent distress and impairment despite the passage of many months or years. Shear et al⁶ proposed a set of CG diagnostic criteria (Table) in which a cluster of symptoms of intense and persistent separation distress are defined as core symptoms. Similar to other psychiatric diagnoses, the symptoms must be associated with significant distress or impairment.

Table

Proposed diagnostic criteria for complicated grief

Assessing CG symptoms

Among those with persistent elevated distress, a CG diagnosis must be considered in the context of the individual’s social and cultural environment, time since the loss, and duration of symptoms. The hallmark symptom of CG is separation distress with a focus of cognitive, behavioral, and emotional symptoms on the loss and its consequences. CG is associated with substantial distress, functional impairment, and an increased risk for suicide. See the Box for a case study.

Many individuals with CG remain undiagnosed and untreated for years despite high levels of distress and impairment and high risk for negative consequences such as suicide.⁷ Accordingly, there’s a need for greater CG screening. Clinically useful tools for assessing CG include a brief, 5-item dimensional screening assessment⁶ and the patient-rated Inventory of Complicated Grief.⁸

Distinguishing complicated and uncomplicated grief. Exhibiting CG symptoms in the first several months after a loss does not mean an individual has or will develop CG. Most bereaved adults report painful thoughts and emotions in the weeks and months following the loss, including distressed yearning, waves of intense grief, persistent and intrusive thoughts, images related to the death, somatic distress, and a feeling of being disconnected from others. For most individuals, the intensity of this response diminishes within 6 to 18 months after the loved one’s death.⁵ Although the optimal length of time to wait before establishing a diagnosis remains debatable, the earliest CG should be diagnosed is 6 months after a loss.

It’s common for grief to occasionally rise in intensity for days or weeks. This surge may occur many months or years after the loss, even in people who exhibited relatively little distress or impairment. In particular, anniversaries, holidays, or periods of stress may trigger increased grief intensity. However, these surges typically subside naturally within a short time. Accordingly, CG should be diagnosed only when symptoms persist for >1 month.

CG vs other post-loss disorders. CG, major depressive disorder (MDD), and posttraumatic stress disorder (PTSD) often are comorbid in bereaved adults. Simon et al⁹ found 72% of CG patients in a treatment- seeking sample reported a lifetime history of MDD and 53% reported a lifetime history of PTSD. However, CG can be distinguished from these disorders. In the same study, 25% of CG patients had no other axis I diagnosis.⁹ After accounting for comorbid disorders, researchers associated CG severity with work and social impairment. These findings provide clear evidence for the incremental validity of CG—ie, a CG diagnosis gives clinicians additional information that predicts impairment above and beyond other disorders. However, future research needs to further examine CG and its overlap and differentiation from MDD and PTSD.

Distinguishing CG and MDD. Intense yearning or preoccupation with the deceased is a common symptom of CG but not MDD. In addition, CG symptoms possess intentionality. For example, emotional distress such as sadness and anger are prominent features of both CG and MDD. However, in CG, these symptoms are specific to the loss or circumstances of the loss, whereas in MDD they generally are more nebulous and generalized. Similarly, CG entails proximity seeking related to the deceased, and avoidance of reminders of the deceased, whereas MDD includes a more general social withdrawal and anhedonia.

Distinguishing CG and PTSD. CG and loss-related PTSD are distinguished by the predominant emotions and focus of concern associated with each disorder. The predominant emotion in PTSD is fear, whereas in CG it is sadness and longing. In PTSD, intrusive thoughts and memories associated with the trauma generally are associated with the event itself and produce an ongoing sense of threat.¹⁰ Avoidance in PTSD is intended to reduce this threat feeling. By contrast, in CG, intrusive memories focus on the deceased or the circumstances of the death, and avoidance is aimed at preventing painful reminders of the loss or its permanence. Importantly, both syndromes may be present.

Box

Treating CG

When is treatment indicated? For years, bereavement theorists emphasized the need to work through emotions and memories related to the deceased with particular focus on negative material. However, evidence suggests that universal application of treatment to all bereaved individuals is unhelpful. In a recent meta-analysis, Neimeyer et al¹¹ found that the outcomes of grief therapy applied indiscriminately to all bereaved adults or all members of high-risk populations—such as parents whose child experienced a violent death—were no better than would be expected by the passage of time. In contrast, grief therapy applied only to those who develop elevated and persistent distress (eg, CG) led to greater and more enduring improvement in post-loss distress than was observed in control conditions.

These results suggest that most grieving individuals who do not meet criteria for CG (or other psychiatric disorders) will not require intervention. Those who do seek treatment for grief-related distress in the acute grief period should be assessed for bereavement-related depression, anxiety, and suicidality, and treated or referred to professional or community-based resources for support or counseling as clinically indicated.

Evidence for psychotherapy. For those who meet CG criteria, psychotherapy targeting the specific symptoms of CG is helpful. The evidence is strongest for CG treatment (CGT), a 16-session, manualized psychotherapy developed by M. Katherine Shear, MD.¹² CGT is based on an attachment model and cognitive-behavioral therapy (CBT) principles, and is informed by the dual-process theory proposed by Stroebe et al.¹³ According to this theory, natural healing following loss comprises 2 processes:

• a loss-oriented process in which the patient comes to terms with the loss, and
• a restoration-oriented process in which the patient reinvigorates a sense of purpose and meaning in life without the deceased.

CGT focuses on both processes. To address the former, it includes clinician-guided exercises in which the patient revisits the time of the death and planned activities in which the patient reengages with people, places, or thoughts that remind him or her of the deceased. CGT aims to allow the patient to gain an increased tolerance of the distressing thoughts and emotions associated with the loss so that these thoughts can be processed and the finality of the death and its circumstances can be accepted.

The restoration process is addressed by having patients generate and discuss personal goals and aspirations for the near and distant future, as well as scheduling pleasurable and rewarding events. This is accomplished by having patients imagine what they would want for themselves if their grief was less intense and planning concrete steps to take toward these goals. The restoration-oriented process is addressed concurrent with the loss-oriented process to encourage the oscillation between processes thought to be characteristic of a natural healing process following the loss of a loved one.

Other psychotherapy approaches (eg, support groups) may have a role for some individuals, and future research may suggest alternative approaches to CGT. To date, CGT is the most targeted evidence-based psychotherapy with randomized controlled data supporting its use in CG.

Pharmacotherapy for CG. Early research suggested that antidepressants—in particular tricyclics—may effectively reduce depressive symptoms in bereavement-related depression; their effect on CG symptoms, however, may not be as strong.¹⁴ Research on pharmacologic treatment that targets CG symptoms is developing. Because of the overlap between CG, PTSD, and MDD, researchers have hypothesized that antidepressants may be effective. Two open-label studies reported that the selective serotonin reuptake inhibitor (SSRI) escitalopram may be effective for CG.^15,16 Although a post-hoc comparison of paroxetine and nortriptyline¹⁷ showed significant reduction in CG and depressive symptoms with both agents, effects could not be separated from concomitant psychotherapy. Furthermore, an examination of naturalistic data on combining antidepressants with CGT suggested that antidepressants may improve outcomes for individuals receiving CGT.¹⁸ A multicenter, randomized controlled trial funded by the National Institute of Mental Health is examining the potential efficacy of citalopram, an SSRI, alone or in combination with CGT.¹⁹

The efficacy of benzodiazepines, which commonly are prescribed for bereaved individuals, has not been assessed in CG. However, recent research suggests they may not be useful for medically managing recent grief²⁰ and that their use in the aftermath of a loss may lead to long-term dependence in geriatric patients.²¹

Related Resources

• Center for Anxiety and Traumatic Stress Disorders. Massachusetts General Hospital. www.bostongrief.com.
• Zisook S, Shear K. Grief and bereavement: what psychiatrists need to know. World Psychiatry. 2009;8(2):67-74.
• Bonanno G. The other side of sadness: what the new science of bereavement tells us about loss. New York, NY: Basic Books; 2009.

Drug Brand Names

• Citalopram • Celexa
• Nortriptyline • Aventyl, Pamelor
• Escitalopram • Lexapro
• Paroxetine • Paxil

Disclosures

Dr. Simon receives grant or research support from the American Cancer Society, the American Foundation for Suicide Prevention, the Department of Defense, Forest Laboratories, and the National Institute of Mental Health.

There is considerable overlap between symptoms of adult attention-deficit/hyperactivity disorder (ADHD) and mild cognitive impairment (MCI), including problems with sustained attention or concentration, anterograde memory, and executive functioning. Differentiating these clinical syndromes based on symptomatic presentation alone can be difficult, but considering the following factors can help you make a more informed diagnosis:

Neurodevelopmental disorder history. DSM-IV-TR stipulates onset for some ADHD symptoms by age 7, although a DSM-5 Work Group is considering symptom onset as late as age 12.¹ Initial onset or a dramatic worsening of longstanding ADHD symptoms in middle-age or older adults is atypical for this neurodevelopmental disorder.

Detailed self-diagnosed symptoms. Patients with ADHD usually can give a satisfactory history of their symptoms. Patients with MCI often are less able to provide a useful history because they have prominent difficulties with anterograde memory, which may be associated with emerging anosognosia.

Educational learning difficulties. Patients with ADHD frequently have comorbid learning difficulties and substance abuse disorders, which are uncommon in MCI.

Rating scales. When in doubt, use rating scales to assess for ADHD.² Ask your patient to complete the rating scale based on how he or she remembers behaving in elementary through middle school, most of their adult life after age 20, and since symptom onset. Obtain collateral ratings from a reliable informant based on his or her knowledge of the patient’s long-term behavioral functioning.

Worsening symptoms. The typical ADHD patient will have a “positive” screen for symptoms, but will report fewer and less severe symptoms from childhood or adolescence through young adulthood and into middle and older age. Suspect MCI when your patient or an informant reports a clear worsening of symptoms in recent months or years despite a lack of evidence of a significant intervening psychiatric disorder.

Psychopharmacotherapy. Patients with MCI usually do not benefit from medications for ADHD. Patients with ADHD often report improvement in at least some of their symptoms with psychopharmacologic treatment.

When your patient’s history, rating scale assessment, and medication trials do not allow you to make a confident differential diagnosis, consider referring him or her for psychological or neuropsychological testing.

There can be overlap in psychometric test findings of middle-age and older adults with a history of ADHD and those who may have MCI. Still, MCI patients’ cognitive difficulties usually are more concerning and dramatic, including problems with spontaneous recall as well as “recognition memory.”

When findings from psychometric testing are equivocal because of possible co-occurrence, retesting in 12 to 18 months usually will help you make a reliable differential diagnosis. Specifically, progression of cognitive dysfunction—including evidence of worsening anterograde memory—is common in MCI but not in ADHD.

Current symptoms of major depressive disorder may further “muddy the waters.” However, parameters such as response to adequate medication trials, progression of cognitive dysfunction, and worsening of test-based cognitive or neuropsychological deficits over time can be useful in reaching a satisfactory differential diagnosis.

Disclosure

Dr. Pollak reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

There is considerable overlap between symptoms of adult attention-deficit/hyperactivity disorder (ADHD) and mild cognitive impairment (MCI), including problems with sustained attention or concentration, anterograde memory, and executive functioning. Differentiating these clinical syndromes based on symptomatic presentation alone can be difficult, but considering the following factors can help you make a more informed diagnosis:

Neurodevelopmental disorder history. DSM-IV-TR stipulates onset for some ADHD symptoms by age 7, although a DSM-5 Work Group is considering symptom onset as late as age 12.¹ Initial onset or a dramatic worsening of longstanding ADHD symptoms in middle-age or older adults is atypical for this neurodevelopmental disorder.

Detailed self-diagnosed symptoms. Patients with ADHD usually can give a satisfactory history of their symptoms. Patients with MCI often are less able to provide a useful history because they have prominent difficulties with anterograde memory, which may be associated with emerging anosognosia.

Educational learning difficulties. Patients with ADHD frequently have comorbid learning difficulties and substance abuse disorders, which are uncommon in MCI.

Rating scales. When in doubt, use rating scales to assess for ADHD.² Ask your patient to complete the rating scale based on how he or she remembers behaving in elementary through middle school, most of their adult life after age 20, and since symptom onset. Obtain collateral ratings from a reliable informant based on his or her knowledge of the patient’s long-term behavioral functioning.

Worsening symptoms. The typical ADHD patient will have a “positive” screen for symptoms, but will report fewer and less severe symptoms from childhood or adolescence through young adulthood and into middle and older age. Suspect MCI when your patient or an informant reports a clear worsening of symptoms in recent months or years despite a lack of evidence of a significant intervening psychiatric disorder.

Psychopharmacotherapy. Patients with MCI usually do not benefit from medications for ADHD. Patients with ADHD often report improvement in at least some of their symptoms with psychopharmacologic treatment.

When your patient’s history, rating scale assessment, and medication trials do not allow you to make a confident differential diagnosis, consider referring him or her for psychological or neuropsychological testing.

There can be overlap in psychometric test findings of middle-age and older adults with a history of ADHD and those who may have MCI. Still, MCI patients’ cognitive difficulties usually are more concerning and dramatic, including problems with spontaneous recall as well as “recognition memory.”

When findings from psychometric testing are equivocal because of possible co-occurrence, retesting in 12 to 18 months usually will help you make a reliable differential diagnosis. Specifically, progression of cognitive dysfunction—including evidence of worsening anterograde memory—is common in MCI but not in ADHD.

Current symptoms of major depressive disorder may further “muddy the waters.” However, parameters such as response to adequate medication trials, progression of cognitive dysfunction, and worsening of test-based cognitive or neuropsychological deficits over time can be useful in reaching a satisfactory differential diagnosis.

Disclosure

Dr. Pollak reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

There is considerable overlap between symptoms of adult attention-deficit/hyperactivity disorder (ADHD) and mild cognitive impairment (MCI), including problems with sustained attention or concentration, anterograde memory, and executive functioning. Differentiating these clinical syndromes based on symptomatic presentation alone can be difficult, but considering the following factors can help you make a more informed diagnosis:

Neurodevelopmental disorder history. DSM-IV-TR stipulates onset for some ADHD symptoms by age 7, although a DSM-5 Work Group is considering symptom onset as late as age 12.¹ Initial onset or a dramatic worsening of longstanding ADHD symptoms in middle-age or older adults is atypical for this neurodevelopmental disorder.

Detailed self-diagnosed symptoms. Patients with ADHD usually can give a satisfactory history of their symptoms. Patients with MCI often are less able to provide a useful history because they have prominent difficulties with anterograde memory, which may be associated with emerging anosognosia.

Educational learning difficulties. Patients with ADHD frequently have comorbid learning difficulties and substance abuse disorders, which are uncommon in MCI.

Rating scales. When in doubt, use rating scales to assess for ADHD.² Ask your patient to complete the rating scale based on how he or she remembers behaving in elementary through middle school, most of their adult life after age 20, and since symptom onset. Obtain collateral ratings from a reliable informant based on his or her knowledge of the patient’s long-term behavioral functioning.

Worsening symptoms. The typical ADHD patient will have a “positive” screen for symptoms, but will report fewer and less severe symptoms from childhood or adolescence through young adulthood and into middle and older age. Suspect MCI when your patient or an informant reports a clear worsening of symptoms in recent months or years despite a lack of evidence of a significant intervening psychiatric disorder.

Psychopharmacotherapy. Patients with MCI usually do not benefit from medications for ADHD. Patients with ADHD often report improvement in at least some of their symptoms with psychopharmacologic treatment.

When your patient’s history, rating scale assessment, and medication trials do not allow you to make a confident differential diagnosis, consider referring him or her for psychological or neuropsychological testing.

There can be overlap in psychometric test findings of middle-age and older adults with a history of ADHD and those who may have MCI. Still, MCI patients’ cognitive difficulties usually are more concerning and dramatic, including problems with spontaneous recall as well as “recognition memory.”

When findings from psychometric testing are equivocal because of possible co-occurrence, retesting in 12 to 18 months usually will help you make a reliable differential diagnosis. Specifically, progression of cognitive dysfunction—including evidence of worsening anterograde memory—is common in MCI but not in ADHD.

Current symptoms of major depressive disorder may further “muddy the waters.” However, parameters such as response to adequate medication trials, progression of cognitive dysfunction, and worsening of test-based cognitive or neuropsychological deficits over time can be useful in reaching a satisfactory differential diagnosis.

Disclosure

Dr. Pollak reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Discuss this article at www.facebook.com/CurrentPsychiatry

You’ve been treating Mr. H, a 54-year-old factory worker and tobacco user, for depression that developed after a work-related back injury and subsequent disability. His depression has had a fair response to an antidepressant. He also has been maintained on chronic opioids (morphine and oxycodone/acetaminophen) for 18 months by his primary care physician (PCP). At the end of your appointment, he asks you for a refill of the opioids because he “ran out” early because of increased night pain and resultant insomnia and “stress.” He clarifies he has asked for early refills before from his PCP, but lately he has been denied. Because you “seem to listen to me more,” he asks for your help. How should you manage Mr. H?

Opioids are among the most commonly misused prescription drugs in the United States.¹ In 2008, poisoning was the leading cause of death from injury in the United States; roughly 90% of poisonings resulted from drug exposure, and >40% of these drug poisonings were from prescription opioids.² The Centers for Disease Control and Prevention estimates that the number of emergency department (ED) visits for nonmedical use of opioids increased 111% between 2004 and 2008, from 144,600 to 305,900 visits.³ The highest number of visits were for use of oxycodone, hydrocodone, and methadone.³

Increased prescribing of opioids and overdose deaths attributable to prescribed opioids have raised concern among physicians about how to effectively treat pain as well as prevent, recognize, and manage aberrant medication-taking behaviors (AMTBs). Psychiatrists are well-positioned to screen and manage their own patients for prescription opioid use disorder (POUD) or collaborate with opioid prescribers to accomplish the same.

Clarifying terminology

Terminology used to describe POUD and related conditions often is poorly defined or loosely applied. Because emotions often enter discussions between patients and physicians about problems related to opioid therapy, nonstigmatizing and more objective terminology is needed, and clinicians are working toward standardizing this. Relevant terms are defined in Table 1.⁴

The DSM-5 Substance Use Disorders Work Group has proposed using the term opioid use disorder (OUD) to replace the term opioid dependence.⁵ The hope is that removing the word “dependence” from the diagnostic term will reduce confusion between “dependence” due to expected physical dependence (tolerance, withdrawal) on medically prescribed opioids vs true addiction (currently defined as “opioid dependence” in DSM-IV-TR). This Work Group also has proposed combining opioid abuse and opioid dependence criteria into a single diagnosis of OUD, and adding “craving” to the criteria. For the complete proposed criteria, see www.dsm5.org/ProposedRevisions/Pages/proposedrevision.aspx?rid=460.These changes are still under review. In this article, we use the term POUD.

Table 1

Terminology related to prescription opioid use disorder

POUD and chronic pain

The incidence of POUD during opioid therapy for pain is unknown.⁶ Some researchers have suggested it may be as low as 0.2%,⁷ while others estimate that rates of POUD in patients with chronic pain may be similar to those in the general population: 3% to 16%.⁸ When applying the proposed DSM-5 criteria to patients receiving long-term opioid therapy for noncancer pain, the lifetime prevalence of POUD may be as high as 35%.⁹

Prescribers may be contributing to POUD. Roughly 76% of opioids used for nonmedical purposes were prescribed to someone else, 20% were prescribed to the user, and 4% came from other sources.¹ Strategies to reduce POUD risk may be underused. In a retrospective cohort study of 1,612 patient electronic medical records from 8 primary care clinics that managed patients with long-term opioids for chronic noncancer pain (average prescribing duration of 2 years duration, ≥3 monthly prescriptions in 6 months), researchers evaluated how often prescribers used 3 risk reduction practices:

• urine drug tests
• regular office visits (≥1 every 6 months and within 30 days of changing opioid treatment)
• restricted early refills (≤1 opioid refill more than a week early).¹⁰

Risk factors for opioid misuse included age 1 early refill. Researchers found that even for high-risk patients, these strategies were used infrequently. Less than one-quarter of patients with ≥3 risk factors ever had a drug test, and those at increased risk were more likely to receive >1 early refill but no more likely to have more frequent visits. Issues such as patient entitlement, lack of physician education, and time constraints may explain why these strategies are not used more often.¹¹

No one procedure or set of variables is sufficient to identify chronic pain patients who may be at risk for POUD. However, a history of drug or alcohol use disorders may be a significant risk factor.^12,13

Few tools have been developed to help identify those at risk of AMTBs or POUD, and all have limitations.^4,14 Recommended self-report measures include the Current Opioid Misuse Measure and the Opioid Risk Tool.¹⁵ A review of studies in which these kinds of tools were developed revealed limited evidence for their use; most studies had methodological shortcomings, did not use standardized AMTB criteria, and provided little assessment of whether these tools changed clinician behaviors or improved patient outcomes.¹⁶

Evaluating AMTBs

Although diagnosing POUD in pain patients receiving chronic opioids can be challenging, assessing for AMTBs typically is helpful. Once AMTBs are identified, they can be examined to determine what drives their expression (Table 1⁴ and Table 2¹⁷). However, often it is easier to identify AMTBs than to interpret their origins; as much as 30% to 50% of patients who complain of chronic pain may have primary substance dependence to sedatives, opioids, or both.¹¹

Table 2

Aberrant medication-taking behaviors and POUD risk

Although AMTBs are common among chronic nonmalignant pain patients,^18,19 how often AMTBs reflect underlying POUD is uncertain.⁷ It is critical to interpret AMTBs with a balance of caution and care: “react therapeutically, not punitively.”²⁰ Categorizing a patient’s AMTB as more or less likely to support a POUD diagnosis can be helpful, but is not conclusive (Table 2).¹⁷ Clinical correlation often is required. No single AMTB alone is indicative of POUD. When evaluating AMTBs, the treating provider should use a nonjudgmental stance, and consider obtaining collateral data from people who can provide differing perspectives of the patient’s behaviors, such as other clinicians, significant others, family, etc. (a release of information from the patient may be required). Another source of collateral data is prescription monitoring databases. These databases typically are state-based and provide electronic access to prescription information, allowing you to search for patterns—ie, use of multiple prescribers or pharmacies, undisclosed prescriptions, etc. Interest in establishing a single, federal database has been increasing, but striking a balance between carefully monitoring for AMTBs and protecting privacy remains unresolved.

DSM-IV-TR diagnostic criteria for opioid dependence²¹ can be challenging to interpret in patients who are prescribed opioids for pain (Table 3

Assessment is primarily clinical and requires an awareness of appropriate terminology, an index of clinical suspicion, and expertise teasing apart pain, addiction, and pseudoaddiction. In our experience, it is helpful to ask a chronic pain patient whom you suspect might have POUD, “Have you ever used your prescribed opioids for reasons other than improving function or reducing pain, such as for getting a ‘high,’ managing stress, escaping from problems, etc.?” An affirmative response suggests an underlying problem with use of prescribed opioids, indicating a need for more careful questioning to determine if AMTBs or POUD coexist with chronic pain.

Drug testing can help determine if a patient is taking opioids that are not prescribed—as well as illicit drugs or alcohol—and confirm the presence of those that are prescribed. Toxicology screening should include opioids typically screened for (eg, morphine, codeine, heroin) and those for which additional tests may be required (eg, semi-synthetics such as oxycodone and hydrocodone and synthetics such as fentanyl).

Table 3

Identifying addiction in pain patients: Limitations of DSM-IV-TR

Table 4

Possible indicators of addiction in pain patients

Helping POUD patients

Goals of treatment include establishing a therapeutic alliance, educating patients about POUD, reducing relapse risk, and optimizing overall health (including pain and physical function). The ASAM Patient Placement Criteria²² provide guidance regarding level-of-care decisions. Treatment ideally includes a combination of education about POUD and its relationship to chronic pain, pharmacotherapy, psychotherapy—such as motivational enhancement therapy, 12-step facilitation therapy, cognitive-behavioral therapy, and relapse prevention—and referral to self-help groups such as Narcotics Anonymous or Pills Anonymous. Importantly, if pain is genuine, it requires treatment.

Pharmacotherapy. Methadone is recommended as the standard of care for OUD by the National Institutes of Health. Methadone is a full opioid agonist that decreases illicit opioid use, mortality, and related problems and requires highly structured treatment approaches under federal and state regulation. POUD patients may have higher rates of methadone maintenance treatment retention than heroin-dependent patients.²³ Published trials of buprenorphine for OUD have shown good treatment retention and reduction in illicit drug use and adverse events.²⁴ Buprenorphine also decreases mortality among OUD patients.

The first large-scale, randomized clinical trial of buprenorphine specifically for POUD included 653 treatment-seeking outpatients.²⁵ This study was designed to approximate clinical practice and included buprenorphine/naloxone, recommended abstinence, and self-help; one-half of participants received intensive addiction counseling. POUD patients were most likely to reduce prescription opioid misuse during buprenorphine/naloxone treatment. If tapered off buprenorphine/naloxone, even after 12 weeks of treatment, the likelihood of an unsuccessful outcome was high. Moreover, opioid dependence counseling did not seem to afford any difference in outcomes. However, despite clinical effectiveness, over the last decade only 19% of patients admitted primarily for OUD treatment (other than heroin) were planned to be offered buprenorphine or methadone.²⁶

A Cochrane review of oral naltrexone for OUD found that the drug was no better than placebo but concluded that available evidence does not allow an adequate evaluation.²⁷ Opioid antagonists may be of value to patients who do not want to take agonists or partial agonists. Extended-release naltrexone also is available to treat OUD.

See the Box below that details steps the FDA and others have taken to prevent POUD and Table 5 for precautions to incorporate when prescribing opioids long-term.

Box

Table 5

Universal precautions with chronic opioid management

CASE CONTINUED: A closer evaluation

After expressing your appreciation for Mr. H’s kind words and empathy for his chronic pain, you redirect him to his PCP. You ask him to sign a release of information so you and his other clinicians can coordinate his care. When discussing Mr. H with his PCP, you learn the patient has made limited requests for early refills and dose escalation primarily in relation to inadequate pain control and function, has genuine pain pathology, and is greatly distressed over his inability to work. No other AMTBs are present, and a check of the state prescribing database reveals that Mr. H did receive a small quantity of opioids from an ED on 1 occasion.

You and Mr. H’s PCP agree this is “pseudo-addiction” but want to watch Mr. H more closely and look for ways to coordinate his care. The PCP agrees to implement a prescribing agreement, start drug testing (including for the prescribed opioids), and reassess maximizing Mr. H’s function and pain management while you address his combined pain, depression, insomnia, and tobacco use.

Related Resources

• Ries RK, Fiellin D, Miller SC, et al, eds. Principles of addiction medicine. 4th ed. Hagerstown, MD: Lippincott Williams & Wilkins; 2009.
• Department of Veterans Affairs. Department of Defense. VA/DoD clinical practice guideline for management of opioid therapy for chronic pain. Appendix C: sample opioid pain care agreement. http://www.healthquality.va.gov/COT_312_Full-er.pdf. Published May 2010. Accessed June 21, 2012.
• Weaver M, Schnoll S. Addiction issues in prescribing opioids for chronic non-malignant pain. J Addiction Med. 2007;1(1):2-10.
• Weaver M, Heit HA, Savage S, et al. Clinical case discussion: chronic pain management. J Addiction Med. 2007;1(1):11-14.

Drug Brand Names

• Buprenorphine • Subutex
• Buprenorphine/naloxone • Suboxone
• Codeine • Tylenol with codeine, others
• Fentanyl • Duragesic, Actiq
• Hydrocodone • Lortab, Vicodin, others
• Methadone • Dolophine, Methadose
• Morphine • Roxanol
• Naltrexone extended-release • Vivitrol
• Oxycodone • OxyContin, Roxicodone

Disclosures

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Frankowski’s time toward this project was provided by the American Board of Addiction Medicine-accredited Cincinnati VA Addiction Medicine Research Fellowship, affiliated with the CeTREAD, Department of Psychiatry and Behavioral Neuroscience, University of Cincinnati, Cincinnati, OH.

The statements in this publication do not necessarily reflect the views or opinions of the Department of Veterans Affairs.

Acknowledgement

The authors thank Catherine Constance and Sandra Mason at the Cincinnati VA Medical Center for their administrative assistance.

Discuss this article at www.facebook.com/CurrentPsychiatry

You’ve been treating Mr. H, a 54-year-old factory worker and tobacco user, for depression that developed after a work-related back injury and subsequent disability. His depression has had a fair response to an antidepressant. He also has been maintained on chronic opioids (morphine and oxycodone/acetaminophen) for 18 months by his primary care physician (PCP). At the end of your appointment, he asks you for a refill of the opioids because he “ran out” early because of increased night pain and resultant insomnia and “stress.” He clarifies he has asked for early refills before from his PCP, but lately he has been denied. Because you “seem to listen to me more,” he asks for your help. How should you manage Mr. H?

Opioids are among the most commonly misused prescription drugs in the United States.¹ In 2008, poisoning was the leading cause of death from injury in the United States; roughly 90% of poisonings resulted from drug exposure, and >40% of these drug poisonings were from prescription opioids.² The Centers for Disease Control and Prevention estimates that the number of emergency department (ED) visits for nonmedical use of opioids increased 111% between 2004 and 2008, from 144,600 to 305,900 visits.³ The highest number of visits were for use of oxycodone, hydrocodone, and methadone.³

Increased prescribing of opioids and overdose deaths attributable to prescribed opioids have raised concern among physicians about how to effectively treat pain as well as prevent, recognize, and manage aberrant medication-taking behaviors (AMTBs). Psychiatrists are well-positioned to screen and manage their own patients for prescription opioid use disorder (POUD) or collaborate with opioid prescribers to accomplish the same.

Clarifying terminology

Terminology used to describe POUD and related conditions often is poorly defined or loosely applied. Because emotions often enter discussions between patients and physicians about problems related to opioid therapy, nonstigmatizing and more objective terminology is needed, and clinicians are working toward standardizing this. Relevant terms are defined in Table 1.⁴

The DSM-5 Substance Use Disorders Work Group has proposed using the term opioid use disorder (OUD) to replace the term opioid dependence.⁵ The hope is that removing the word “dependence” from the diagnostic term will reduce confusion between “dependence” due to expected physical dependence (tolerance, withdrawal) on medically prescribed opioids vs true addiction (currently defined as “opioid dependence” in DSM-IV-TR). This Work Group also has proposed combining opioid abuse and opioid dependence criteria into a single diagnosis of OUD, and adding “craving” to the criteria. For the complete proposed criteria, see www.dsm5.org/ProposedRevisions/Pages/proposedrevision.aspx?rid=460.These changes are still under review. In this article, we use the term POUD.

Table 1

Terminology related to prescription opioid use disorder

POUD and chronic pain

The incidence of POUD during opioid therapy for pain is unknown.⁶ Some researchers have suggested it may be as low as 0.2%,⁷ while others estimate that rates of POUD in patients with chronic pain may be similar to those in the general population: 3% to 16%.⁸ When applying the proposed DSM-5 criteria to patients receiving long-term opioid therapy for noncancer pain, the lifetime prevalence of POUD may be as high as 35%.⁹

Prescribers may be contributing to POUD. Roughly 76% of opioids used for nonmedical purposes were prescribed to someone else, 20% were prescribed to the user, and 4% came from other sources.¹ Strategies to reduce POUD risk may be underused. In a retrospective cohort study of 1,612 patient electronic medical records from 8 primary care clinics that managed patients with long-term opioids for chronic noncancer pain (average prescribing duration of 2 years duration, ≥3 monthly prescriptions in 6 months), researchers evaluated how often prescribers used 3 risk reduction practices:

• urine drug tests
• regular office visits (≥1 every 6 months and within 30 days of changing opioid treatment)
• restricted early refills (≤1 opioid refill more than a week early).¹⁰

Risk factors for opioid misuse included age 1 early refill. Researchers found that even for high-risk patients, these strategies were used infrequently. Less than one-quarter of patients with ≥3 risk factors ever had a drug test, and those at increased risk were more likely to receive >1 early refill but no more likely to have more frequent visits. Issues such as patient entitlement, lack of physician education, and time constraints may explain why these strategies are not used more often.¹¹

No one procedure or set of variables is sufficient to identify chronic pain patients who may be at risk for POUD. However, a history of drug or alcohol use disorders may be a significant risk factor.^12,13

Few tools have been developed to help identify those at risk of AMTBs or POUD, and all have limitations.^4,14 Recommended self-report measures include the Current Opioid Misuse Measure and the Opioid Risk Tool.¹⁵ A review of studies in which these kinds of tools were developed revealed limited evidence for their use; most studies had methodological shortcomings, did not use standardized AMTB criteria, and provided little assessment of whether these tools changed clinician behaviors or improved patient outcomes.¹⁶

Evaluating AMTBs

Although diagnosing POUD in pain patients receiving chronic opioids can be challenging, assessing for AMTBs typically is helpful. Once AMTBs are identified, they can be examined to determine what drives their expression (Table 1⁴ and Table 2¹⁷). However, often it is easier to identify AMTBs than to interpret their origins; as much as 30% to 50% of patients who complain of chronic pain may have primary substance dependence to sedatives, opioids, or both.¹¹

Table 2

Aberrant medication-taking behaviors and POUD risk