The Journal of Family Practice

Although the concept of the living will was first proposed in 1969,¹ the idea caught on slowly. In fact, the first scholarly article discussing the topic didn’t appear until 16 years later.²

In contrast, an informal search of PubMed reveals that at least 38 articles on advance directives and end-of-life care have been published during the first 7 months of 2017. And a feature article in this month’s issue of JFP makes one more. Why is there such strong interest now in an issue that seldom arose when I began practice in 1978?

More complex, less personalized medicine. As medical care has become more sophisticated, there is a great deal more we can do to keep people alive as they approach the end of life, and a great many more decisions to be made.

Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Additionally, people are much less likely today to be cared for in their dying days by a family physician who knows them, their wishes, and their family well. In my early years in small-town practice, I was present when my patients were dying, and I usually knew their family members. Family meetings were easy to arrange, and we quickly came to a consensus about what to do and what not to do. If I was not available, one of my practice partners was. We cared for our patients in the office, nursing home, and hospital. Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Getting more people to participate. Consequently, it is important to understand patients’ wishes for end-of-life care and to document those wishes in writing, using things like a POLST (Physician Orders for Life-Sustaining Treatment) form. Although randomized trials support the value of advance care planning, especially in primary care settings,^3,4 two-thirds of Americans have not completed an advance directive.⁵ Rolnick suggests we “delegalize” the process to remove barriers and make it easier for people to execute such documents and integrate them into health care systems.⁶

Make it part of your office routine. A 70-year-old patient of mine with advanced COPD arrived at his office visit last month with advance directive and POLST forms in hand. We had an excellent, frank conversation, spiced with humor that he supplied, about his wishes for end-of-life care. Just like so many other tasks that we must squeeze into our busy schedules, this is one that we should hard-wire into our office systems and routines.

Although the concept of the living will was first proposed in 1969,¹ the idea caught on slowly. In fact, the first scholarly article discussing the topic didn’t appear until 16 years later.²

In contrast, an informal search of PubMed reveals that at least 38 articles on advance directives and end-of-life care have been published during the first 7 months of 2017. And a feature article in this month’s issue of JFP makes one more. Why is there such strong interest now in an issue that seldom arose when I began practice in 1978?

More complex, less personalized medicine. As medical care has become more sophisticated, there is a great deal more we can do to keep people alive as they approach the end of life, and a great many more decisions to be made.

Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Additionally, people are much less likely today to be cared for in their dying days by a family physician who knows them, their wishes, and their family well. In my early years in small-town practice, I was present when my patients were dying, and I usually knew their family members. Family meetings were easy to arrange, and we quickly came to a consensus about what to do and what not to do. If I was not available, one of my practice partners was. We cared for our patients in the office, nursing home, and hospital. Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Getting more people to participate. Consequently, it is important to understand patients’ wishes for end-of-life care and to document those wishes in writing, using things like a POLST (Physician Orders for Life-Sustaining Treatment) form. Although randomized trials support the value of advance care planning, especially in primary care settings,^3,4 two-thirds of Americans have not completed an advance directive.⁵ Rolnick suggests we “delegalize” the process to remove barriers and make it easier for people to execute such documents and integrate them into health care systems.⁶

Make it part of your office routine. A 70-year-old patient of mine with advanced COPD arrived at his office visit last month with advance directive and POLST forms in hand. We had an excellent, frank conversation, spiced with humor that he supplied, about his wishes for end-of-life care. Just like so many other tasks that we must squeeze into our busy schedules, this is one that we should hard-wire into our office systems and routines.

Although the concept of the living will was first proposed in 1969,¹ the idea caught on slowly. In fact, the first scholarly article discussing the topic didn’t appear until 16 years later.²

In contrast, an informal search of PubMed reveals that at least 38 articles on advance directives and end-of-life care have been published during the first 7 months of 2017. And a feature article in this month’s issue of JFP makes one more. Why is there such strong interest now in an issue that seldom arose when I began practice in 1978?

More complex, less personalized medicine. As medical care has become more sophisticated, there is a great deal more we can do to keep people alive as they approach the end of life, and a great many more decisions to be made.

Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Additionally, people are much less likely today to be cared for in their dying days by a family physician who knows them, their wishes, and their family well. In my early years in small-town practice, I was present when my patients were dying, and I usually knew their family members. Family meetings were easy to arrange, and we quickly came to a consensus about what to do and what not to do. If I was not available, one of my practice partners was. We cared for our patients in the office, nursing home, and hospital. Now, most dying hospitalized patients are cared for by hospitalists who may be meeting the patient for the first time.

Getting more people to participate. Consequently, it is important to understand patients’ wishes for end-of-life care and to document those wishes in writing, using things like a POLST (Physician Orders for Life-Sustaining Treatment) form. Although randomized trials support the value of advance care planning, especially in primary care settings,^3,4 two-thirds of Americans have not completed an advance directive.⁵ Rolnick suggests we “delegalize” the process to remove barriers and make it easier for people to execute such documents and integrate them into health care systems.⁶

Make it part of your office routine. A 70-year-old patient of mine with advanced COPD arrived at his office visit last month with advance directive and POLST forms in hand. We had an excellent, frank conversation, spiced with humor that he supplied, about his wishes for end-of-life care. Just like so many other tasks that we must squeeze into our busy schedules, this is one that we should hard-wire into our office systems and routines.

THE CASE

A 31-year-old woman presented to her obstetrician’s office at 16 weeks’ gestation with a 2-day history of low-grade fever and an erythematous rash measuring 1 x 4 cm on her right groin. She had a medical history of a penicillin allergy (urticaria) and her outdoor activities included gardening and picnicking.

We suspected that she was experiencing an allergic reaction and recommended an antihistamine (diphenhydramine). The patient returned 4 days later with new symptoms including headache, photophobia, neck pain, unilateral large joint pain, and periorbital cellulitis, as well as expansion of her rash. She was afebrile and an examination revealed that the 1 x 4 cm rash on her groin had grown; it was now a demarcated erythematous rash with faint central clearing measuring 5 x 8 cm. Right periorbital erythema and nuchal rigidity were also noted.

Because of her expanding rash and nuchal rigidity, we suspected Lyme meningitis and we referred her to the emergency department. Within 24 hours, the rash had spread to her abdomen, thigh, and wrist, and was consistent with erythema migrans.

Laboratory evaluation revealed an increased number of white bloods cells (13.5 million cells/mcL; normal range 4.5-11.0 million cells/mcL), and an increased number of neutrophils (10.8 million cells/mcL; normal range 1.8-8 million cells/mcL), indicating leukocytosis with a left shift. Lab tests also revealed a low hemoglobin level (10.6 g/dL; normal range 12-16 g/dL) and a mean corpuscular volume of 85.6 fL/red cell (normal range 80-100 fL/red cell), indicating microcytic anemia. A lumbar puncture was negative for disseminated Lyme disease by Gram stain, culture, and polymerase chain reaction.

THE DIAGNOSIS

A diagnosis of Lyme disease was confirmed with a positive Lyme titer serology via an enzyme-linked immunosorbent assay. The rash and other symptoms responded promptly to intravenous ceftriaxone 2 g, and the patient was discharged home on oral cefuroxime 500 mg bid for 14 days.

DISCUSSION

Lyme disease is the most common vector-borne illness in the United States, concentrated heavily in the Northeast and upper Midwest.¹ The most recent information released by the Centers for Disease Control and Prevention (CDC) lists Vermont, Maine, Pennsylvania, Rhode Island, Connecticut, New Jersey, Massachusetts, Delaware, New Hampshire, and Minnesota as the states with the highest incidence of Lyme disease.²

The number of reported cases in the United States has increased over the past 2 decades, from approximately 11,000 in 1995 to about 28,000 in 2015.³ Over the past year, we have seen several cases of Lyme disease in the obstetric population of our own practice.

Prompt treatment is crucial. Pregnant women who are acutely infected with Borrelia burgdorferi (the primary cause of Lyme disease) and do not receive treatment have experienced multiple adverse pregnancy outcomes, including preterm delivery, infants born with rash, and stillbirth.⁴ Additional concern exists for fetal cardiac anomalies, with data showing that there are twice as many cardiac defects in children born to mothers residing in endemic regions.⁵

What animal studies have taught us about Lyme disease

The potential causal relationship between Lyme disease and fetal demise was first studied in 2007. This case report involved the stillbirth of a full-term baby from an acutely infected woman who did not receive treatment. She experienced erythema multiforme 6 weeks prior to delivery.⁶

A fetal echocardiogram is reasonable in pregnant women acutely infected with Lyme disease during the first trimester, given the high potential for fetal cardiac anomalies.

The vast majority of research on Lyme disease in pregnancy comes from work with mice and dogs. These studies confirmed that acute infection with Lyme disease is associated with an increased risk of adverse fetal outcomes, specifically fetal death.⁷

Silver et al further researched the association using murine models in the 1980s. They found that fetal death occurred in 12% of acutely infected mice, compared with none of the mice that were chronically infected.⁷

In 2010, Lakos and Solymosi examined the effects of Lyme disease on pregnancy outcomes in acutely infected women. Seven out of 95 pregnant women acutely infected with B burgdorferi experienced fetal demise, further supporting the association seen in animal experiments.⁸

Treating pregnant patients

Doxycycline and tetracycline, which are routinely used to treat Lyme disease, are not appropriate in the obstetric population. The CDC recommends up to a 3-week course of antibiotics; the standard regimen is amoxicillin 500 mg by mouth tid. For women who are allergic to penicillin, as was the case with our patient, cefuroxime 500 mg by mouth bid is the treatment of choice.⁹

Our patient underwent a detailed ultrasound at 21 weeks, which revealed normal fetal anatomy and no evidence of cardiac malformations. The remainder of her pregnancy was uncomplicated, and she gave birth vaginally at 41 weeks to a baby boy weighing 3700 g.

THE TAKEAWAY

There is a need to increase awareness of Lyme disease in pregnancy on a national level. It is the responsibility of every practitioner caring for obstetric patients in endemic regions to address new-onset rash promptly. There have been cases of women who experienced erythema migrans and arthralgias after exposure to a tick bite, later delivering infants with cardiac anomalies such as atrial and ventricular septal defects.¹⁰ In obstetric patients acutely infected during the first trimester, a fetal echocardiogram is reasonable, given the demonstrated high potential for fetal cardiac anomalies.

Preventing adverse fetal outcomes requires early treatment with antibiotics. The CDC maintains that there have been no life-threatening adverse fetal effects from Lyme disease seen in women who are appropriately treated, as well as no transmission of Lyme disease in the breast milk of lactating mothers.⁹

THE CASE

A 31-year-old woman presented to her obstetrician’s office at 16 weeks’ gestation with a 2-day history of low-grade fever and an erythematous rash measuring 1 x 4 cm on her right groin. She had a medical history of a penicillin allergy (urticaria) and her outdoor activities included gardening and picnicking.

We suspected that she was experiencing an allergic reaction and recommended an antihistamine (diphenhydramine). The patient returned 4 days later with new symptoms including headache, photophobia, neck pain, unilateral large joint pain, and periorbital cellulitis, as well as expansion of her rash. She was afebrile and an examination revealed that the 1 x 4 cm rash on her groin had grown; it was now a demarcated erythematous rash with faint central clearing measuring 5 x 8 cm. Right periorbital erythema and nuchal rigidity were also noted.

Because of her expanding rash and nuchal rigidity, we suspected Lyme meningitis and we referred her to the emergency department. Within 24 hours, the rash had spread to her abdomen, thigh, and wrist, and was consistent with erythema migrans.

Laboratory evaluation revealed an increased number of white bloods cells (13.5 million cells/mcL; normal range 4.5-11.0 million cells/mcL), and an increased number of neutrophils (10.8 million cells/mcL; normal range 1.8-8 million cells/mcL), indicating leukocytosis with a left shift. Lab tests also revealed a low hemoglobin level (10.6 g/dL; normal range 12-16 g/dL) and a mean corpuscular volume of 85.6 fL/red cell (normal range 80-100 fL/red cell), indicating microcytic anemia. A lumbar puncture was negative for disseminated Lyme disease by Gram stain, culture, and polymerase chain reaction.

THE DIAGNOSIS

A diagnosis of Lyme disease was confirmed with a positive Lyme titer serology via an enzyme-linked immunosorbent assay. The rash and other symptoms responded promptly to intravenous ceftriaxone 2 g, and the patient was discharged home on oral cefuroxime 500 mg bid for 14 days.

DISCUSSION

Lyme disease is the most common vector-borne illness in the United States, concentrated heavily in the Northeast and upper Midwest.¹ The most recent information released by the Centers for Disease Control and Prevention (CDC) lists Vermont, Maine, Pennsylvania, Rhode Island, Connecticut, New Jersey, Massachusetts, Delaware, New Hampshire, and Minnesota as the states with the highest incidence of Lyme disease.²

The number of reported cases in the United States has increased over the past 2 decades, from approximately 11,000 in 1995 to about 28,000 in 2015.³ Over the past year, we have seen several cases of Lyme disease in the obstetric population of our own practice.

Prompt treatment is crucial. Pregnant women who are acutely infected with Borrelia burgdorferi (the primary cause of Lyme disease) and do not receive treatment have experienced multiple adverse pregnancy outcomes, including preterm delivery, infants born with rash, and stillbirth.⁴ Additional concern exists for fetal cardiac anomalies, with data showing that there are twice as many cardiac defects in children born to mothers residing in endemic regions.⁵

What animal studies have taught us about Lyme disease

The potential causal relationship between Lyme disease and fetal demise was first studied in 2007. This case report involved the stillbirth of a full-term baby from an acutely infected woman who did not receive treatment. She experienced erythema multiforme 6 weeks prior to delivery.⁶

A fetal echocardiogram is reasonable in pregnant women acutely infected with Lyme disease during the first trimester, given the high potential for fetal cardiac anomalies.

The vast majority of research on Lyme disease in pregnancy comes from work with mice and dogs. These studies confirmed that acute infection with Lyme disease is associated with an increased risk of adverse fetal outcomes, specifically fetal death.⁷

Silver et al further researched the association using murine models in the 1980s. They found that fetal death occurred in 12% of acutely infected mice, compared with none of the mice that were chronically infected.⁷

In 2010, Lakos and Solymosi examined the effects of Lyme disease on pregnancy outcomes in acutely infected women. Seven out of 95 pregnant women acutely infected with B burgdorferi experienced fetal demise, further supporting the association seen in animal experiments.⁸

Treating pregnant patients

Doxycycline and tetracycline, which are routinely used to treat Lyme disease, are not appropriate in the obstetric population. The CDC recommends up to a 3-week course of antibiotics; the standard regimen is amoxicillin 500 mg by mouth tid. For women who are allergic to penicillin, as was the case with our patient, cefuroxime 500 mg by mouth bid is the treatment of choice.⁹

Our patient underwent a detailed ultrasound at 21 weeks, which revealed normal fetal anatomy and no evidence of cardiac malformations. The remainder of her pregnancy was uncomplicated, and she gave birth vaginally at 41 weeks to a baby boy weighing 3700 g.

THE TAKEAWAY

There is a need to increase awareness of Lyme disease in pregnancy on a national level. It is the responsibility of every practitioner caring for obstetric patients in endemic regions to address new-onset rash promptly. There have been cases of women who experienced erythema migrans and arthralgias after exposure to a tick bite, later delivering infants with cardiac anomalies such as atrial and ventricular septal defects.¹⁰ In obstetric patients acutely infected during the first trimester, a fetal echocardiogram is reasonable, given the demonstrated high potential for fetal cardiac anomalies.

Preventing adverse fetal outcomes requires early treatment with antibiotics. The CDC maintains that there have been no life-threatening adverse fetal effects from Lyme disease seen in women who are appropriately treated, as well as no transmission of Lyme disease in the breast milk of lactating mothers.⁹

THE CASE

A 31-year-old woman presented to her obstetrician’s office at 16 weeks’ gestation with a 2-day history of low-grade fever and an erythematous rash measuring 1 x 4 cm on her right groin. She had a medical history of a penicillin allergy (urticaria) and her outdoor activities included gardening and picnicking.

We suspected that she was experiencing an allergic reaction and recommended an antihistamine (diphenhydramine). The patient returned 4 days later with new symptoms including headache, photophobia, neck pain, unilateral large joint pain, and periorbital cellulitis, as well as expansion of her rash. She was afebrile and an examination revealed that the 1 x 4 cm rash on her groin had grown; it was now a demarcated erythematous rash with faint central clearing measuring 5 x 8 cm. Right periorbital erythema and nuchal rigidity were also noted.

Because of her expanding rash and nuchal rigidity, we suspected Lyme meningitis and we referred her to the emergency department. Within 24 hours, the rash had spread to her abdomen, thigh, and wrist, and was consistent with erythema migrans.

Laboratory evaluation revealed an increased number of white bloods cells (13.5 million cells/mcL; normal range 4.5-11.0 million cells/mcL), and an increased number of neutrophils (10.8 million cells/mcL; normal range 1.8-8 million cells/mcL), indicating leukocytosis with a left shift. Lab tests also revealed a low hemoglobin level (10.6 g/dL; normal range 12-16 g/dL) and a mean corpuscular volume of 85.6 fL/red cell (normal range 80-100 fL/red cell), indicating microcytic anemia. A lumbar puncture was negative for disseminated Lyme disease by Gram stain, culture, and polymerase chain reaction.

THE DIAGNOSIS

A diagnosis of Lyme disease was confirmed with a positive Lyme titer serology via an enzyme-linked immunosorbent assay. The rash and other symptoms responded promptly to intravenous ceftriaxone 2 g, and the patient was discharged home on oral cefuroxime 500 mg bid for 14 days.

DISCUSSION

Lyme disease is the most common vector-borne illness in the United States, concentrated heavily in the Northeast and upper Midwest.¹ The most recent information released by the Centers for Disease Control and Prevention (CDC) lists Vermont, Maine, Pennsylvania, Rhode Island, Connecticut, New Jersey, Massachusetts, Delaware, New Hampshire, and Minnesota as the states with the highest incidence of Lyme disease.²

The number of reported cases in the United States has increased over the past 2 decades, from approximately 11,000 in 1995 to about 28,000 in 2015.³ Over the past year, we have seen several cases of Lyme disease in the obstetric population of our own practice.

Prompt treatment is crucial. Pregnant women who are acutely infected with Borrelia burgdorferi (the primary cause of Lyme disease) and do not receive treatment have experienced multiple adverse pregnancy outcomes, including preterm delivery, infants born with rash, and stillbirth.⁴ Additional concern exists for fetal cardiac anomalies, with data showing that there are twice as many cardiac defects in children born to mothers residing in endemic regions.⁵

What animal studies have taught us about Lyme disease

The potential causal relationship between Lyme disease and fetal demise was first studied in 2007. This case report involved the stillbirth of a full-term baby from an acutely infected woman who did not receive treatment. She experienced erythema multiforme 6 weeks prior to delivery.⁶

A fetal echocardiogram is reasonable in pregnant women acutely infected with Lyme disease during the first trimester, given the high potential for fetal cardiac anomalies.

The vast majority of research on Lyme disease in pregnancy comes from work with mice and dogs. These studies confirmed that acute infection with Lyme disease is associated with an increased risk of adverse fetal outcomes, specifically fetal death.⁷

Silver et al further researched the association using murine models in the 1980s. They found that fetal death occurred in 12% of acutely infected mice, compared with none of the mice that were chronically infected.⁷

In 2010, Lakos and Solymosi examined the effects of Lyme disease on pregnancy outcomes in acutely infected women. Seven out of 95 pregnant women acutely infected with B burgdorferi experienced fetal demise, further supporting the association seen in animal experiments.⁸

Treating pregnant patients

Doxycycline and tetracycline, which are routinely used to treat Lyme disease, are not appropriate in the obstetric population. The CDC recommends up to a 3-week course of antibiotics; the standard regimen is amoxicillin 500 mg by mouth tid. For women who are allergic to penicillin, as was the case with our patient, cefuroxime 500 mg by mouth bid is the treatment of choice.⁹

Our patient underwent a detailed ultrasound at 21 weeks, which revealed normal fetal anatomy and no evidence of cardiac malformations. The remainder of her pregnancy was uncomplicated, and she gave birth vaginally at 41 weeks to a baby boy weighing 3700 g.

THE TAKEAWAY

There is a need to increase awareness of Lyme disease in pregnancy on a national level. It is the responsibility of every practitioner caring for obstetric patients in endemic regions to address new-onset rash promptly. There have been cases of women who experienced erythema migrans and arthralgias after exposure to a tick bite, later delivering infants with cardiac anomalies such as atrial and ventricular septal defects.¹⁰ In obstetric patients acutely infected during the first trimester, a fetal echocardiogram is reasonable, given the demonstrated high potential for fetal cardiac anomalies.

Preventing adverse fetal outcomes requires early treatment with antibiotics. The CDC maintains that there have been no life-threatening adverse fetal effects from Lyme disease seen in women who are appropriately treated, as well as no transmission of Lyme disease in the breast milk of lactating mothers.⁹

A 66-year-old white woman presented to her primary care clinic with concerns about hair loss, which began 2 years ago. Recently, she had noticed some “bumps” on her cheeks, as well.

On physical examination, the physician noted hair loss in a symmetric 2-cm band-like distribution across her frontal and temporal scalp (FIGURES 1 and 2). In both areas, there was moderate perifollicular erythema, scale, and what appeared to be scarring.

The patient had lost most of her eyebrow hairs, and had prominent temporal veins (FIGURE 2) and flesh-colored papules on her cheeks. She had no significant medical history, was emotionally stable, and recently had a satisfactory health care maintenance exam. The postmenopausal patient’s last menses was 15 years earlier, and she was not taking hormone replacement.

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

Diagnosis: Frontal fibrosing alopecia

The patient was referred to our dermatology clinic, which specializes in hair loss. Based on the clinical findings, we suspected that this was a case of frontal fibrosing alopecia (FFA), a primary lymphocytic cicatricial (scarring) alopecia. A dermatopathologist confirmed the diagnosis via histologic review.

A condition on the rise. The incidence of FFA has been steadily increasing internationally since the condition was first described in 1994.¹ Among patients referred to a specialty clinic for hair loss, diagnosis of FFA has increased from 1.6% in 2000 to 17% in 2011.²

FFA is characterized by symmetric band-like hair loss with evidence of scarring in the frontal and temporal regions of the scalp. (The extent of hair loss can be assessed by retracting the patient’s hair and having the patient raise his or her eyebrows and wrinkle the forehead in a surprised look.) FFA is accompanied by eyebrow loss in 73% to 95% of patients.^2,3 Mild to severe perifollicular (and possibly more generalized) erythema and scale are usually present. In addition, erythematous or skin-colored papules may appear on the face,³ and marked exaggeration of the temporal veins is a common finding.

More than 80% of patients with frontal fibrosing alopecia are postmenopausal women.

Most patients with FFA (83%) are postmenopausal women and nearly all (98.6%) have Fitzpatrick skin type 1 or 2 (white skin that burns easily and doesn’t readily tan).⁴ Other pertinent findings include the absence of oral lesions, nail changes, or other skin diseases.

A subtype of another condition? Because they are similar histologically, some consider FFA to be a subtype of lichen planopilaris. (See “Scarring alopecia in a woman with psoriasis,” J Fam Pract. 2015;64:E1-E3.)

A punch biopsy to confirm the diagnosis of FFA should be taken from the leading edge of the hair loss and, ideally, reviewed by a dermatopathologist. Histologic examination will reveal a lichenoid lymphocytic infiltrate (predominantly around the hair follicle where the follicular stem cells reside), resulting in fibrosis and scarring.⁵

Rule out other causes of hair loss

In addition to confirming the diagnosis with histologic examination, you’ll also need to have ruled out the following conditions in the differential.

Alopecia areata may mimic the ophiasis (band-like) pattern of hair loss seen with FFA, but it is a non-scarring disorder that typically lacks any signs of inflammation.

Female pattern hair loss is characterized by a decrease in hair density and thinning. The condition is non-scarring and usually involves the frontal and vertex (crown) regions of the scalp.

Discoid lupus erythematosus is characterized by circular scarring hair loss with a central patch of inflammation, as well as depigmentation.

Central centrifugal cicatricial alopecia predominantly affects black women and is characterized by circular hair loss of the vertex, with perifollicular inflammation and scarring.

Traction alopecia can occur in the same location as FFA, but is not usually associated with perifollicular inflammation. This condition can cause scarring if traction has been longstanding and persistent. There is usually a history of certain hairstyles (such as braiding) associated with chronic tension on hair fibers.

Numerous Tx strategies exist, but they are not well studied

Because there are no published randomized clinical trials on treatment for FFA, few evidence-based treatment strategies exist.⁶ In addition, the prognosis is variable. Experts have employed numerous treatment strategies, including topical and intralesional steroids, immunosuppressive medications, antibiotics, and anti-androgen therapy, with varying results.^4,6 For most primary care physicians, it’s best to refer patients to a dermatologist to initiate treatment.

Intralesional steroids such as triamcinolone acetonide (5-10 mg/cc), as well as high-potency topical steroids, are generally helpful to stabilize the disease. There is also some evidence of benefit from oral dutasteride or finasteride at variable doses.⁶ Immunosuppressants such as hydroxychloroquine may also be used as second-line treatments, although the benefit-to-risk ratio needs to be taken into consideration.⁷

Early detection is key. In general, treatment should be initiated as soon as possible to prevent disease progression and reduce permanent scarring and hair loss. The Lichen Planopilaris Activity Index⁷ is a tool that clinicians can use to measure disease severity and track changes in disease activity through patient report of symptoms and measurements of scalp inflammation.

Our patient was started on a regimen of topical high-potency steroids (clobetasol foam, 0.05%), with targeted, intralesional injection of steroids (10 mg/cc of triamcinolone acetonide) to areas with the most inflammation. The patient was advised to use ketoconazole 2% shampoo while showering.

These interventions decreased our patient’s symptoms dramatically. Her scalp erythema and scale improved, but the hair did not regrow. One year later, her hairline was clinically stable with no evidence of disease progression. She continues to see a dermatologist annually.

CORRESPONDENCE
David V. Power, MB, MPH, Department of Family Medicine and Community Health, University of Minnesota, 516 Delaware St. SE, Minneapolis, MN 55455; [email protected].

A 66-year-old white woman presented to her primary care clinic with concerns about hair loss, which began 2 years ago. Recently, she had noticed some “bumps” on her cheeks, as well.

On physical examination, the physician noted hair loss in a symmetric 2-cm band-like distribution across her frontal and temporal scalp (FIGURES 1 and 2). In both areas, there was moderate perifollicular erythema, scale, and what appeared to be scarring.

The patient had lost most of her eyebrow hairs, and had prominent temporal veins (FIGURE 2) and flesh-colored papules on her cheeks. She had no significant medical history, was emotionally stable, and recently had a satisfactory health care maintenance exam. The postmenopausal patient’s last menses was 15 years earlier, and she was not taking hormone replacement.

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

Diagnosis: Frontal fibrosing alopecia

The patient was referred to our dermatology clinic, which specializes in hair loss. Based on the clinical findings, we suspected that this was a case of frontal fibrosing alopecia (FFA), a primary lymphocytic cicatricial (scarring) alopecia. A dermatopathologist confirmed the diagnosis via histologic review.

A condition on the rise. The incidence of FFA has been steadily increasing internationally since the condition was first described in 1994.¹ Among patients referred to a specialty clinic for hair loss, diagnosis of FFA has increased from 1.6% in 2000 to 17% in 2011.²

FFA is characterized by symmetric band-like hair loss with evidence of scarring in the frontal and temporal regions of the scalp. (The extent of hair loss can be assessed by retracting the patient’s hair and having the patient raise his or her eyebrows and wrinkle the forehead in a surprised look.) FFA is accompanied by eyebrow loss in 73% to 95% of patients.^2,3 Mild to severe perifollicular (and possibly more generalized) erythema and scale are usually present. In addition, erythematous or skin-colored papules may appear on the face,³ and marked exaggeration of the temporal veins is a common finding.

More than 80% of patients with frontal fibrosing alopecia are postmenopausal women.

Most patients with FFA (83%) are postmenopausal women and nearly all (98.6%) have Fitzpatrick skin type 1 or 2 (white skin that burns easily and doesn’t readily tan).⁴ Other pertinent findings include the absence of oral lesions, nail changes, or other skin diseases.

A subtype of another condition? Because they are similar histologically, some consider FFA to be a subtype of lichen planopilaris. (See “Scarring alopecia in a woman with psoriasis,” J Fam Pract. 2015;64:E1-E3.)

A punch biopsy to confirm the diagnosis of FFA should be taken from the leading edge of the hair loss and, ideally, reviewed by a dermatopathologist. Histologic examination will reveal a lichenoid lymphocytic infiltrate (predominantly around the hair follicle where the follicular stem cells reside), resulting in fibrosis and scarring.⁵

Rule out other causes of hair loss

In addition to confirming the diagnosis with histologic examination, you’ll also need to have ruled out the following conditions in the differential.

Alopecia areata may mimic the ophiasis (band-like) pattern of hair loss seen with FFA, but it is a non-scarring disorder that typically lacks any signs of inflammation.

Female pattern hair loss is characterized by a decrease in hair density and thinning. The condition is non-scarring and usually involves the frontal and vertex (crown) regions of the scalp.

Discoid lupus erythematosus is characterized by circular scarring hair loss with a central patch of inflammation, as well as depigmentation.

Central centrifugal cicatricial alopecia predominantly affects black women and is characterized by circular hair loss of the vertex, with perifollicular inflammation and scarring.

Traction alopecia can occur in the same location as FFA, but is not usually associated with perifollicular inflammation. This condition can cause scarring if traction has been longstanding and persistent. There is usually a history of certain hairstyles (such as braiding) associated with chronic tension on hair fibers.

Numerous Tx strategies exist, but they are not well studied

Because there are no published randomized clinical trials on treatment for FFA, few evidence-based treatment strategies exist.⁶ In addition, the prognosis is variable. Experts have employed numerous treatment strategies, including topical and intralesional steroids, immunosuppressive medications, antibiotics, and anti-androgen therapy, with varying results.^4,6 For most primary care physicians, it’s best to refer patients to a dermatologist to initiate treatment.

Intralesional steroids such as triamcinolone acetonide (5-10 mg/cc), as well as high-potency topical steroids, are generally helpful to stabilize the disease. There is also some evidence of benefit from oral dutasteride or finasteride at variable doses.⁶ Immunosuppressants such as hydroxychloroquine may also be used as second-line treatments, although the benefit-to-risk ratio needs to be taken into consideration.⁷

Early detection is key. In general, treatment should be initiated as soon as possible to prevent disease progression and reduce permanent scarring and hair loss. The Lichen Planopilaris Activity Index⁷ is a tool that clinicians can use to measure disease severity and track changes in disease activity through patient report of symptoms and measurements of scalp inflammation.

Our patient was started on a regimen of topical high-potency steroids (clobetasol foam, 0.05%), with targeted, intralesional injection of steroids (10 mg/cc of triamcinolone acetonide) to areas with the most inflammation. The patient was advised to use ketoconazole 2% shampoo while showering.

These interventions decreased our patient’s symptoms dramatically. Her scalp erythema and scale improved, but the hair did not regrow. One year later, her hairline was clinically stable with no evidence of disease progression. She continues to see a dermatologist annually.

CORRESPONDENCE
David V. Power, MB, MPH, Department of Family Medicine and Community Health, University of Minnesota, 516 Delaware St. SE, Minneapolis, MN 55455; [email protected].

A 66-year-old white woman presented to her primary care clinic with concerns about hair loss, which began 2 years ago. Recently, she had noticed some “bumps” on her cheeks, as well.

On physical examination, the physician noted hair loss in a symmetric 2-cm band-like distribution across her frontal and temporal scalp (FIGURES 1 and 2). In both areas, there was moderate perifollicular erythema, scale, and what appeared to be scarring.

The patient had lost most of her eyebrow hairs, and had prominent temporal veins (FIGURE 2) and flesh-colored papules on her cheeks. She had no significant medical history, was emotionally stable, and recently had a satisfactory health care maintenance exam. The postmenopausal patient’s last menses was 15 years earlier, and she was not taking hormone replacement.

WHAT IS YOUR DIAGNOSIS?
HOW WOULD YOU TREAT THIS PATIENT?

Diagnosis: Frontal fibrosing alopecia

The patient was referred to our dermatology clinic, which specializes in hair loss. Based on the clinical findings, we suspected that this was a case of frontal fibrosing alopecia (FFA), a primary lymphocytic cicatricial (scarring) alopecia. A dermatopathologist confirmed the diagnosis via histologic review.

A condition on the rise. The incidence of FFA has been steadily increasing internationally since the condition was first described in 1994.¹ Among patients referred to a specialty clinic for hair loss, diagnosis of FFA has increased from 1.6% in 2000 to 17% in 2011.²

FFA is characterized by symmetric band-like hair loss with evidence of scarring in the frontal and temporal regions of the scalp. (The extent of hair loss can be assessed by retracting the patient’s hair and having the patient raise his or her eyebrows and wrinkle the forehead in a surprised look.) FFA is accompanied by eyebrow loss in 73% to 95% of patients.^2,3 Mild to severe perifollicular (and possibly more generalized) erythema and scale are usually present. In addition, erythematous or skin-colored papules may appear on the face,³ and marked exaggeration of the temporal veins is a common finding.

More than 80% of patients with frontal fibrosing alopecia are postmenopausal women.

Most patients with FFA (83%) are postmenopausal women and nearly all (98.6%) have Fitzpatrick skin type 1 or 2 (white skin that burns easily and doesn’t readily tan).⁴ Other pertinent findings include the absence of oral lesions, nail changes, or other skin diseases.

A subtype of another condition? Because they are similar histologically, some consider FFA to be a subtype of lichen planopilaris. (See “Scarring alopecia in a woman with psoriasis,” J Fam Pract. 2015;64:E1-E3.)

A punch biopsy to confirm the diagnosis of FFA should be taken from the leading edge of the hair loss and, ideally, reviewed by a dermatopathologist. Histologic examination will reveal a lichenoid lymphocytic infiltrate (predominantly around the hair follicle where the follicular stem cells reside), resulting in fibrosis and scarring.⁵

Rule out other causes of hair loss

In addition to confirming the diagnosis with histologic examination, you’ll also need to have ruled out the following conditions in the differential.

Alopecia areata may mimic the ophiasis (band-like) pattern of hair loss seen with FFA, but it is a non-scarring disorder that typically lacks any signs of inflammation.

Female pattern hair loss is characterized by a decrease in hair density and thinning. The condition is non-scarring and usually involves the frontal and vertex (crown) regions of the scalp.

Discoid lupus erythematosus is characterized by circular scarring hair loss with a central patch of inflammation, as well as depigmentation.

Central centrifugal cicatricial alopecia predominantly affects black women and is characterized by circular hair loss of the vertex, with perifollicular inflammation and scarring.

Traction alopecia can occur in the same location as FFA, but is not usually associated with perifollicular inflammation. This condition can cause scarring if traction has been longstanding and persistent. There is usually a history of certain hairstyles (such as braiding) associated with chronic tension on hair fibers.

Numerous Tx strategies exist, but they are not well studied

Because there are no published randomized clinical trials on treatment for FFA, few evidence-based treatment strategies exist.⁶ In addition, the prognosis is variable. Experts have employed numerous treatment strategies, including topical and intralesional steroids, immunosuppressive medications, antibiotics, and anti-androgen therapy, with varying results.^4,6 For most primary care physicians, it’s best to refer patients to a dermatologist to initiate treatment.

Intralesional steroids such as triamcinolone acetonide (5-10 mg/cc), as well as high-potency topical steroids, are generally helpful to stabilize the disease. There is also some evidence of benefit from oral dutasteride or finasteride at variable doses.⁶ Immunosuppressants such as hydroxychloroquine may also be used as second-line treatments, although the benefit-to-risk ratio needs to be taken into consideration.⁷

Early detection is key. In general, treatment should be initiated as soon as possible to prevent disease progression and reduce permanent scarring and hair loss. The Lichen Planopilaris Activity Index⁷ is a tool that clinicians can use to measure disease severity and track changes in disease activity through patient report of symptoms and measurements of scalp inflammation.

Our patient was started on a regimen of topical high-potency steroids (clobetasol foam, 0.05%), with targeted, intralesional injection of steroids (10 mg/cc of triamcinolone acetonide) to areas with the most inflammation. The patient was advised to use ketoconazole 2% shampoo while showering.

These interventions decreased our patient’s symptoms dramatically. Her scalp erythema and scale improved, but the hair did not regrow. One year later, her hairline was clinically stable with no evidence of disease progression. She continues to see a dermatologist annually.

CORRESPONDENCE
David V. Power, MB, MPH, Department of Family Medicine and Community Health, University of Minnesota, 516 Delaware St. SE, Minneapolis, MN 55455; [email protected].

THE CASE

A 46-year-old man presented to the emergency department (ED) with sudden-onset right-sided visual loss. He had a history of asthma, but no family history of hypercoagulability, deep vein thrombosis (DVT), or stroke. The patient had an active lifestyle that included scuba diving, mountain biking, and hockey (coaching and playing). The physical examination revealed a right homonymous upper quadrantanopia. The neurologic examination was within normal limits, except for the visual deficit and unequal pupil size. A computerized tomography scan of the patient’s head did not reveal any lesions.

Based on the patient’s clinical picture, the ED physician prescribed alteplase, a tissue plasminogen activator (tPA), and admitted him to the intensive care unit for monitoring.

Subsequent magnetic resonance imaging (MRI) of the brain showed multiple small areas of acute infarct in the posterior circulation territory bilaterally, with involvement of small portions of the bilateral cerebellar hemispheres and parts of the left occipital lobe (FIGURE 1A and 1B).

An electrocardiogram showed no evidence of atrial fibrillation, and hypercoagulability studies were within normal limits. There was no evidence of May-Thurner anatomy, and an ultrasound of the lower extremities showed no DVT.

THE DIAGNOSIS

An echocardiogram with bubble study confirmed a diagnosis of patent foramen ovale (PFO) with bidirectional flow, a normal ejection fraction, and no evidence of left ventricular or left atrial thrombus. We started the patient on the anticoagulant enoxaparin 70 mg bid bridged with warfarin 5 mg/d.

This case illustrates the importance of lifestyle and occupation when evaluating patients with cryptogenic stroke associated with patent foramen ovale.

Taking the patient’s active lifestyle into consideration, he was approved for PFO closure by the PFO committee and underwent closure. Following treatment, the patient was left with a residual 2-mm blind spot in the right visual field. At a 2-year follow-up visit, he showed no new focal deficits or recurrent symptoms.

DISCUSSION

Since 1988 when Lechat et al reported increased incidence of PFO in young stroke patients,¹ many studies have supported the association between PFO and cryptogenic stroke (CS) in young adults.² Because it remained controversial as to whether PFO is a risk factor for stroke or transient ischemic attack recurrence,³ researchers investigated PFO closure as a preventive measure to decrease stroke recurrence in patients with both CS and PFO.

A 2012 meta-analysis showed possible benefits of closure compared with medical management using antiplatelet or anticoagulation therapies.⁴ However, these results were not supported by results of other studies. These include the CLOSURE I trial,⁵ which compared device closure of PFO with medical therapy, and the RESPECT⁶ and PC trials,⁷ which did not show a significant difference in the primary end point of recurrent stroke between patients who received medical therapy and those who had PFO closure.

American Heart Association/American Stroke Association’s 2011 guidelines recommend only antiplatelet therapy for patients with CS and PFO.⁸ While there is consensus that surgical closure is not better than a medical approach to patients with CS and PFO, cases should be individualized, as a patient’s clinical or social factors may dictate otherwise.

Lifestyle may warrant PFO closure

No previous studies have considered occupation or hobbies as an indication for PFO closure in patients with CS. Our patient’s active lifestyle, particularly his scuba diving and participation in contact sports, made him a poor candidate for anticoagulation. Scuba diving is associated with decompression sickness and air emboli, which can be a mechanism of cerebral ischemia, especially in patients with a right-to-left shunt, such as with PFO.⁹

We did not observe a strong temporal relationship between diving and stroke in our patient. MRI findings suggested that he had multiple minor embolic events over time, which is consistent with a prior case report.⁹ This suggested air emboli as a possible source of stroke, in which case, our patient might not benefit from antiplatelet or anticoagulation therapy.

THE TAKEAWAY

This case illustrates the importance of a thorough social history and knowledge of the patient’s hobbies, occupation, and preferences in evaluating and treating individuals with CS associated with PFO. The current literature does not provide complete answers to the cause, diagnosis, and management of CS; additional research is needed.

The work-up involved in defining the etiology of stroke includes, but is not limited to, head and brain imaging, an echocardiogram, hypercoagulability tests, and vascular imaging. The work of Sanna et al showed that approximately 12% of patients with CS have atrial fibrillation when monitored over a one-year period, suggesting atrial fibrillation as a possible cause in some cases.¹⁰

As the case described here demonstrates, further research is warranted regarding how a patient’s occupation and lifestyle factor into decision-making for patients with PFO.

THE CASE

A 46-year-old man presented to the emergency department (ED) with sudden-onset right-sided visual loss. He had a history of asthma, but no family history of hypercoagulability, deep vein thrombosis (DVT), or stroke. The patient had an active lifestyle that included scuba diving, mountain biking, and hockey (coaching and playing). The physical examination revealed a right homonymous upper quadrantanopia. The neurologic examination was within normal limits, except for the visual deficit and unequal pupil size. A computerized tomography scan of the patient’s head did not reveal any lesions.

Based on the patient’s clinical picture, the ED physician prescribed alteplase, a tissue plasminogen activator (tPA), and admitted him to the intensive care unit for monitoring.

Subsequent magnetic resonance imaging (MRI) of the brain showed multiple small areas of acute infarct in the posterior circulation territory bilaterally, with involvement of small portions of the bilateral cerebellar hemispheres and parts of the left occipital lobe (FIGURE 1A and 1B).

An electrocardiogram showed no evidence of atrial fibrillation, and hypercoagulability studies were within normal limits. There was no evidence of May-Thurner anatomy, and an ultrasound of the lower extremities showed no DVT.

THE DIAGNOSIS

An echocardiogram with bubble study confirmed a diagnosis of patent foramen ovale (PFO) with bidirectional flow, a normal ejection fraction, and no evidence of left ventricular or left atrial thrombus. We started the patient on the anticoagulant enoxaparin 70 mg bid bridged with warfarin 5 mg/d.

This case illustrates the importance of lifestyle and occupation when evaluating patients with cryptogenic stroke associated with patent foramen ovale.

Taking the patient’s active lifestyle into consideration, he was approved for PFO closure by the PFO committee and underwent closure. Following treatment, the patient was left with a residual 2-mm blind spot in the right visual field. At a 2-year follow-up visit, he showed no new focal deficits or recurrent symptoms.

DISCUSSION

Since 1988 when Lechat et al reported increased incidence of PFO in young stroke patients,¹ many studies have supported the association between PFO and cryptogenic stroke (CS) in young adults.² Because it remained controversial as to whether PFO is a risk factor for stroke or transient ischemic attack recurrence,³ researchers investigated PFO closure as a preventive measure to decrease stroke recurrence in patients with both CS and PFO.

A 2012 meta-analysis showed possible benefits of closure compared with medical management using antiplatelet or anticoagulation therapies.⁴ However, these results were not supported by results of other studies. These include the CLOSURE I trial,⁵ which compared device closure of PFO with medical therapy, and the RESPECT⁶ and PC trials,⁷ which did not show a significant difference in the primary end point of recurrent stroke between patients who received medical therapy and those who had PFO closure.

American Heart Association/American Stroke Association’s 2011 guidelines recommend only antiplatelet therapy for patients with CS and PFO.⁸ While there is consensus that surgical closure is not better than a medical approach to patients with CS and PFO, cases should be individualized, as a patient’s clinical or social factors may dictate otherwise.

Lifestyle may warrant PFO closure

No previous studies have considered occupation or hobbies as an indication for PFO closure in patients with CS. Our patient’s active lifestyle, particularly his scuba diving and participation in contact sports, made him a poor candidate for anticoagulation. Scuba diving is associated with decompression sickness and air emboli, which can be a mechanism of cerebral ischemia, especially in patients with a right-to-left shunt, such as with PFO.⁹

We did not observe a strong temporal relationship between diving and stroke in our patient. MRI findings suggested that he had multiple minor embolic events over time, which is consistent with a prior case report.⁹ This suggested air emboli as a possible source of stroke, in which case, our patient might not benefit from antiplatelet or anticoagulation therapy.

THE TAKEAWAY

This case illustrates the importance of a thorough social history and knowledge of the patient’s hobbies, occupation, and preferences in evaluating and treating individuals with CS associated with PFO. The current literature does not provide complete answers to the cause, diagnosis, and management of CS; additional research is needed.

The work-up involved in defining the etiology of stroke includes, but is not limited to, head and brain imaging, an echocardiogram, hypercoagulability tests, and vascular imaging. The work of Sanna et al showed that approximately 12% of patients with CS have atrial fibrillation when monitored over a one-year period, suggesting atrial fibrillation as a possible cause in some cases.¹⁰

As the case described here demonstrates, further research is warranted regarding how a patient’s occupation and lifestyle factor into decision-making for patients with PFO.

THE CASE

A 46-year-old man presented to the emergency department (ED) with sudden-onset right-sided visual loss. He had a history of asthma, but no family history of hypercoagulability, deep vein thrombosis (DVT), or stroke. The patient had an active lifestyle that included scuba diving, mountain biking, and hockey (coaching and playing). The physical examination revealed a right homonymous upper quadrantanopia. The neurologic examination was within normal limits, except for the visual deficit and unequal pupil size. A computerized tomography scan of the patient’s head did not reveal any lesions.

Based on the patient’s clinical picture, the ED physician prescribed alteplase, a tissue plasminogen activator (tPA), and admitted him to the intensive care unit for monitoring.

Subsequent magnetic resonance imaging (MRI) of the brain showed multiple small areas of acute infarct in the posterior circulation territory bilaterally, with involvement of small portions of the bilateral cerebellar hemispheres and parts of the left occipital lobe (FIGURE 1A and 1B).

An electrocardiogram showed no evidence of atrial fibrillation, and hypercoagulability studies were within normal limits. There was no evidence of May-Thurner anatomy, and an ultrasound of the lower extremities showed no DVT.

THE DIAGNOSIS

An echocardiogram with bubble study confirmed a diagnosis of patent foramen ovale (PFO) with bidirectional flow, a normal ejection fraction, and no evidence of left ventricular or left atrial thrombus. We started the patient on the anticoagulant enoxaparin 70 mg bid bridged with warfarin 5 mg/d.

This case illustrates the importance of lifestyle and occupation when evaluating patients with cryptogenic stroke associated with patent foramen ovale.

Taking the patient’s active lifestyle into consideration, he was approved for PFO closure by the PFO committee and underwent closure. Following treatment, the patient was left with a residual 2-mm blind spot in the right visual field. At a 2-year follow-up visit, he showed no new focal deficits or recurrent symptoms.

DISCUSSION

Since 1988 when Lechat et al reported increased incidence of PFO in young stroke patients,¹ many studies have supported the association between PFO and cryptogenic stroke (CS) in young adults.² Because it remained controversial as to whether PFO is a risk factor for stroke or transient ischemic attack recurrence,³ researchers investigated PFO closure as a preventive measure to decrease stroke recurrence in patients with both CS and PFO.

A 2012 meta-analysis showed possible benefits of closure compared with medical management using antiplatelet or anticoagulation therapies.⁴ However, these results were not supported by results of other studies. These include the CLOSURE I trial,⁵ which compared device closure of PFO with medical therapy, and the RESPECT⁶ and PC trials,⁷ which did not show a significant difference in the primary end point of recurrent stroke between patients who received medical therapy and those who had PFO closure.

American Heart Association/American Stroke Association’s 2011 guidelines recommend only antiplatelet therapy for patients with CS and PFO.⁸ While there is consensus that surgical closure is not better than a medical approach to patients with CS and PFO, cases should be individualized, as a patient’s clinical or social factors may dictate otherwise.

Lifestyle may warrant PFO closure

No previous studies have considered occupation or hobbies as an indication for PFO closure in patients with CS. Our patient’s active lifestyle, particularly his scuba diving and participation in contact sports, made him a poor candidate for anticoagulation. Scuba diving is associated with decompression sickness and air emboli, which can be a mechanism of cerebral ischemia, especially in patients with a right-to-left shunt, such as with PFO.⁹

We did not observe a strong temporal relationship between diving and stroke in our patient. MRI findings suggested that he had multiple minor embolic events over time, which is consistent with a prior case report.⁹ This suggested air emboli as a possible source of stroke, in which case, our patient might not benefit from antiplatelet or anticoagulation therapy.

THE TAKEAWAY

This case illustrates the importance of a thorough social history and knowledge of the patient’s hobbies, occupation, and preferences in evaluating and treating individuals with CS associated with PFO. The current literature does not provide complete answers to the cause, diagnosis, and management of CS; additional research is needed.

The work-up involved in defining the etiology of stroke includes, but is not limited to, head and brain imaging, an echocardiogram, hypercoagulability tests, and vascular imaging. The work of Sanna et al showed that approximately 12% of patients with CS have atrial fibrillation when monitored over a one-year period, suggesting atrial fibrillation as a possible cause in some cases.¹⁰

As the case described here demonstrates, further research is warranted regarding how a patient’s occupation and lifestyle factor into decision-making for patients with PFO.

CASE A 57-year-old man presented for evaluation of painless, intermittent passage of bright red blood per rectum for several months. His bowel habits were otherwise unchanged, averaging 2 soft bowel movements daily without straining. His medical history was significant for radiation therapy for prostate cancer 18 months earlier and a recent finding of mild microcytic anemia. A colonoscopy 7 years ago was negative for polyps, diverticula, or other lesions. He denied any family history of colon cancer or other gastrointestinal disorders. He wanted to know what he could do to stop the bleeding or if further testing would be needed.

Next steps?

Radiation therapy and its effect on the GI tract

In 1895, Dr. Wilhelm Roentgen first introduced the use of x-rays for diagnostic radiographic purposes. A year later, Dr. Emil Gruble made the first attempt to use radiation therapy (XRT) to treat cancer. In 1897, Dr. David Walsh described the first case of XRT-induced tissue injury in the British Medical Journal.¹

Since then, XRT has been used extensively to treat cancer, and its delivery techniques have improved and diversified. Like chemotherapy, XRT has its greatest effect on rapidly dividing cells, but as a result, the adverse effects of therapy are also greatest on rapidly dividing normal tissues, as well as others in the radiation field.

A large proportion of cancer patients will receive XRT, yet XRT-related costs account for less than 5% of total cancer care expenditure, suggesting cost effectiveness.^2,3 However, even with the great progress achieved in the delivery of XRT, it continues to have its share of acute and chronic complications, among the most common of which is gastrointestinal (GI) tract toxicity. These adverse effects are often first reported to, diagnosed, or treated by the primary care provider, who frequently remains pivotally involved in the patient’s longitudinal care.

Radiation therapy's adverse effects are often first reported to, diagnosed, or treated by family physicians, who frequently remain centrally involved in longitudinal care.

Approximately 50% to 75% of patients undergoing XRT will have some degree of GI symptoms of acute injury, but the majority will recover fully within a few weeks following completion of treatment.^4-6 However, in about 5% of patients,^4-6 there will be long-term consequences of varying degrees that may develop as soon as one year or as long as 10 years after XRT. These can pose substantial challenges for patients, as well as both the primary care provider and consulting specialists.

In the review that follows, we detail the potential acute and chronic complications of XRT on the GI tract and how best to manage them. But first, a word about the related terminology.

Getting a handle on XRT-related injury terminology

The preferred terms used to describe injury to normal tissue as a result of XRT include “XRT-related injury” or “pelvic radiation disease” (when the injury is confined to intrapelvic tissues); organ-specific descriptors such as “radiation enteropathy” or “XRT-induced esophageal stricture” are also used and are acceptable.^4,7,8

Terms such as “radiation enteritis” or “radiation proctitis” are considered misnomers since there is no significant histologic inflammation. Indeed, as we will discuss, acute injury is largely due to epithelial cellular injury and cell death (necrosis), while chronic injury is primarily the consequence of ongoing tissue ischemia, fibrosis, and other pathophysiologic processes.

Acute vs chronic XRT-related tissue injury

From a pathobiologic and clinical perspective, XRT-related injury can be categorized as either acute or chronic.^8-12 Acute XRT-related injury involves direct cellular necrosis of the epithelial cells and damage (eg, irreparable DNA alterations) to stem cells. This acute injury prevents appropriate cellular regeneration, which results in denuded mucosa, mucosal ulcerations, and even perforation in severe cases.¹⁰ Acute injury starts 2 to 3 weeks after initiating XRT and typically resolves within 2 to 3 months following completion of treatment.

Chronic XRT toxicity is pathophysiologically complex and multifactorial.^10-12 It includes: obliterative endarteritis of submucosal arterioles with chronic tissue ischemia, eosinophil infiltration, fibroblast proliferation and pathologic fibrosis, neovascularization with friable telangiectasia formation, and bowel serosal injury that promotes formation of dense adhesions.¹³ Its pathogenesis remains incompletely understood.

Several treatment- and patient-related variables can impact the occurrence and nature of tissue injury secondary to XRT and are summarized in the Table.^4,9-13 Newer forms of radiotherapy such as proton beam and Yttrium-90 radioembolization may also cause radiation injury,¹⁴ but to a lesser degree than conventional external beam XRT, in part because of improved dose targeting. We will not discuss these modalities in this review.

Can’t something be done to prevent injury in the first place?

There are no convincing evidence-based preventive or therapeutic treatments that address the underlying mechanisms of either the acute or chronic phases of XRT-related GI tract injury, although hyperbaric oxygen (which we’ll discuss in greater detail shortly) may be a promising option.^{8,11,12,15-17} It’s believed that hyperbaric oxygen may prove useful by facilitating angiogenesis and improving tissue oxygenation.^8,11,15-17 Unfortunately, this treatment is not widely available, and the frequency and duration required for optimal results is unclear.

Numerous pharmacologic radioprotectants have been suggested or evaluated in small studies, but none have an established role in addressing XRT-related injury. Given these voids, emphasis on symptom management and empathic, supportive care is essential.¹⁸

A look at injuries and Tx options by organs affected

The esophagus

Injurious effects on the esophagus are seen following XRT for lung, mediastinal, hypopharyngeal, or esophageal cancers.^19,20 The total XRT dose and regimen may vary, but a typical course may involve 10 gray (ie, 1000 rads) per week (2 gray per day) for 5 weeks. The maximum tolerated dose by the esophagus is approximately 6 gray, above which most patients will have long-term complications; however, some patients may experience toxicity at even lower doses.

Acute complications of esophageal XRT-related injury include mucosal ulcerations, which can present as chest pain and odynophagia. The mucosal pathology can cause dysmotility, which results in dysphagia for both liquids and solids.^19-21

If severe symptoms develop during treatment, the dose per session can be reduced and/or the sessions can be delayed. Some patients require temporary gastrostomy feeding tubes until symptoms resolve. Mucosal ulcerations can become a chronic issue as well. The mainstay of treatment is symptomatic relief with topical anesthetics and anti-acid medications.

Chronic symptoms are more varied and can be difficult to manage^14,15 and include the following:

• Strictures. Esophageal dysphagia develops in nearly two-thirds of patients postradiation and, in many cases, is due to stricture formation.²² Symptoms may range from mild dysphagia with solids to complete esophageal obstruction.²³ Barium esophagography can be helpful to delineate esophageal stricture morphology and determine treatment options.

For the majority of patients, serial endoscopic dilation with a balloon catheter or bougie (or other endoscopic techniques) achieves adequate esophageal patency to alleviate symptoms; this may need to be repeated periodically to maintain patency, as nearly one-third of patients will experience recurrent stricturing.^21,23

• Tracheo-esophageal fistulae. This complication can lead to pneumonia and generally has a poor prognosis.

Fistulae are chiefly treated endoscopically with esophageal, and occasionally, tracheobronchial stent placement. As with esophageal strictures, barium imaging can help plan the therapeutic approach. Percutaneous feeding may be required in some patients as a bridge or when fistula closure cannot be achieved.

• Secondary esophageal carcinogenesis. This dreaded complication develops in up to 2% to 3% of patients at 10 years post-XRT.¹⁹

Pharmacologic therapy for esophageal symptoms is generally unsuccessful, although acid suppression therapy may help as an adjuvant treatment to endoscopic dilation for esophageal strictures. Surgery is seldom attempted because of the fibrotic/ischemic tissues and high postoperative morbidity/mortality.

The stomach

The stomach is relatively resistant to XRT injury. Although XRT therapy can cause a transient decrease in acid output, there are rarely significant short- or long-term consequences with conventional therapeutic dosing (less than 50 gray).¹¹

The liver

Hepatic resistance to radiation is relatively high; however, liver toxicity has been reported at low doses, an effect that is seen largely following bone marrow transplantation.²⁴ Acute histologic XRT-related liver injury changes consist of severe pan-lobar congestion leading to hemorrhagic necrosis, cell atrophy, and perivascular fibrosis, as well as sclerosis of central and sublobular hepatic veins. The majority of patients will show reversal of the histologic changes within 3 months; however, approximately 25% to 40% of patients,²⁵ depending on total XRT dose to the liver and other technical factors, will experience progressive and chronic changes resulting in liver atrophy, severe perivascular injury, and fibrosis of the portal vein or bile ducts.

Besides hyperbaric oxygen, there are no evidence-based preventive or therapeutic treatments that address the underlying mechanisms of radiation-related GI tract injury.

The clinical symptoms of acute liver injury may include right upper quadrant pain, ascites, jaundice, veno-occlusive disease, or Budd-Chiari syndrome.²⁵ The major chronic complication of XRT-related liver injury is progressive fibrosis, which may advance to cirrhosis.

Small bowel

The small bowel is the most radiosensitive GI tract organ due to high cell turnover, which makes it very susceptible to XRT-related injury.^4,8,10,26-28 Under 3 gray, ≤20% of patients will develop radiation enteropathy, while at >5 gray, the incidence rises progressively with dose, and a majority of patients will be symptomatic.²⁹ The degree to which the bowel is healthy before XRT can be an important factor in developing enteropathy. Parenthetically, treatment with a full bladder may also help displace some of the loops from the field of XRT and decrease injury.

Acute XRT-related injury of the small bowel includes mucosal necrosis (ie, direct cell death) and ulcerations that may present as diarrhea, pain, malabsorption, weight loss, bleeding, and perforation.^4,8,10,26-28 Fortunately, in most patients, these are self-limited and can be managed symptomatically. Loperamide is the first-line medication for diarrhea, although Lomotil (diphenoxylate/atropine) may also be used if necessary.^4,8,10,26-28 Nutrition may be challenging in severe cases, and if dietary modifications and supplementation do not prove sufficient, home parenteral nutrition is required.

Over time, chronic small bowel pathology may develop, including strictures in 3% to 15%, fistulae in 0.6% to 4.8%, secondary neoplasia in up to 10%, dysmotility- or adhesion-related small intestinal bacterial overgrowth in up to 45%, and malabsorption with associated nutritional deficiency in up to 63%.^26-28 Other common XRT-related complications are chronic pain, which could be due to adhesions or ischemia, small intestinal bacterial overgrowth, or partial bowel obstruction, and telangiectasias that result with acute or chronic blood loss.¹³

Imaging of small bowel disease to diagnose the various manifestations of radiation enteropathy is challenging. Conventional X-rays may be difficult to interpret. Therefore, computerized tomography or magnetic resonance enterography, capsule endoscopy, or balloon-assisted enteroscopy is preferred—depending on availability, local expertise, and the suspected pre-procedure diagnosis.

Telangiectasias are not seen on cross-sectional imaging but can be seen with capsule endoscopy (which should not be ordered if stricture is suspected unless a patency capsule has been tried). Single or double balloon enteroscopy (specialized endoscopes intended for reaching the mid and distal ileum), which has been used to treat strictures or telangiectasia in healthy tissues,²⁹ can be difficult or impossible in post-XRT patients because adhesions may limit progress of the scope to the area of interest, and forceful advancement of the scope increases the risk of perforation.

Small bowel telangiectasias can cause chronic occult blood loss, which often requires iron supplementation; acute bleeding may require blood transfusion and hospitalization. Of note, choosing an iron formulation that is well tolerated is critical to avoid (additional) unpleasant GI tract adverse effects. We typically recommend elemental iron with Vitamin C to augment absorption or ferrous gluconate; some patients will require intravenous iron infusion.

Surgery may be advisable to address complications such as fistulous tracts, complex strictures, or bowel obstruction; how-ever, operating on radiated abdominal tissues and ischemic bowel is associated with high morbidity and mortality.^4,25,28,30 The surgeon may encounter dense adhesions that make an otherwise “simple” surgery problematic.

For example, it may be difficult to access the desired region and determine the borders of healthy tissue; wide excisions are, thus, often performed, which may result in small bowel failure (ie, short gut syndrome) and a mortality rate in excess of 30%.³¹ In addition, the ischemic post-XRT tissues may not heal well even if the intended surgery is completed; indeed, anastomotic leaks, failures, and infections are not uncommon. Moreover, another 30% will have other postoperative complications, 40% to 60% may require more than one laparotomy, and 50% of those who recover from the initial surgery will develop recurrence of the fistulous tract or stricture.^4,25,28,30

No drug therapy has proven effective for prevention or mechanistically-driven treatment of XRT-induced small bowel injury. Hyperbaric oxygen therapy may be the most promising medical treatment, with early response in 53% of cases and long-term response of 66% to 73% for global symptomatic relief.³² It has been used successfully for treatment of pain, diarrhea, malabsorption, and hemorrhage from mucosal ulcerations, stenosis, and fistulous tracts. When available, it should be considered as a potential therapeutic intervention.

Colon

Injury to the colon is seen in 10% to 20% of patients following XRT for prostate, bladder, cervical, or uterine cancer.³³ The maximum tolerated dose of the colon is slightly higher than for the small intestine.³⁴ The rectosigmoid area is the area most commonly implicated, but depending on the field of radiation, injury can be more extensive/proximal.

The small bowel is the most radiosensitive GI tract organ, due to high cell turnover, making it highly susceptible to radiation therapy-related injury.

Acute XRT injury of the colon produces acute mucosal necrosis, which may manifest as bowel dysmotility, diarrhea, cramps, tenesmus, or hematochezia. Sigmoidoscopy or colonoscopy will show mucosal edema, erosions, and ulcerations with a purplish/red discoloration. A barium enema will show spasm of the affected area with so-called “thumbprinting,” which indicates mucosal edema. The onset of symptoms is generally within 3 weeks of XRT initiation; symptoms are self-limited in most cases. Management is centered on symptom relief; loperamide and Lomotil are first-line agents for diarrheal symptoms.

Chronic XRT-related colopathy is the result of chronic tissue ischemia and fibrosis. This may lead to dysmotility resulting in abnormal bowel habits (ranging from constipation to diarrhea) or sigmoid stenosis/stricture resulting in an inability to evacuate the bowel. For the latter, it is important to note that fiber supplementation may not be optimal, since increasing the fecal caliber makes it more difficult to pass through the stenotic, colonic segment.

Emollients such as small doses of mineral oil will not increase the fecal caliber, but will soften fecal matter so that it can be passed with greater ease. MiraLAX may be effective, as well, but can increase the sense of urgency and contribute to incontinence in some. Lactulose can be effective, but it causes excessive gassiness/bloating that may result in abdominal pain and episodes of incontinence.

Bleeding from telangiectasias is another chronic complication of XRT-related colonic injury. Argon plasma coagulation (APC) via flexible sigmoidoscopy or colonoscopy is typically the primary therapeutic approach, reported to have a success rate of up to 90% in healthy tissues.^33,35 Even with endoscopic treatment, as mentioned earlier in the context of small bowel XRT-related telangiectasias, iron supplementation is often needed to replete stores, and choice of iron agent is important.

Furthermore, it is essential to recognize that repeat endoscopic sessions may be needed to fully treat telangiectasias, and recrudescence of bleeding months or years later should raise suspicion for recurrent telangiectasia formation (and need for repeat treatment). As with other organs, there may be a role for hyperbaric oxygen, even in difficult-to-treat cases.^36,37

Colonic fibrosis/stenosis and fistulous tract formation, as in the small bowel, are also seen in this population of patients. Endoscopic dilation can be considered, and stenting may be reasonable for short and/or distal strictures. Surgical approaches for fistulous tracts and strictures can be high-risk and associated with poor outcomes, mostly because of the underlying chronic tissue ischemia and fibrosis,^4,8,27,30,34 as discussed in the small bowel section.

Rectum

The rectum has tolerance to XRT similar to the colon,³⁸ but because of its anatomical location, rectal radiation injury is more common, and is typically seen after XRT for prostate, bladder, cervical, or uterine cancer. Acute rectal radiation injury is seen in 50% to 78% of patients,³⁶ and symptoms are similar to that of injury to the sigmoid (eg, tenesmus, loose evacuations, hematochezia), all of which are consequences of direct radiation injury to the mucosa.

Use of mesenchymal stem cells has also been described for rectal and other fistulae, but use is mostly experimental.

Chronic rectal radiation injury may present in a variety of ways. Tenesmus and incontinence are seen in 8% to 20% of patients, frequent defecation in 50%, urgency in 47%, and rectal cancer in up to 2% to 3% after 10 years.^36,37 Other complications include anorectal strictures, fissures, fistulae, and bleeding from rectal telangiectasias. While anoscopy can diagnose many of these, flexible sigmoidoscopy is needed to examine more proximal rectal sites as well as for treatment. Treatment of these chronic complications of XRT is analogous to those of the colon⁷ with the following exceptions:

• Anorectal strictures. In contrast to sigmoid strictures, these are generally more amenable to dilatation. If symptoms recur frequently, patients may be instructed on self-dilatations at home.
• Bleeding from rectal telangiectasias. In the rare cases where endoscopic APC is not feasible or successful, an alternative treatment would be radiofrequency ablation or the application of 2% to 10% formalin intra-rectally. This is reported to have up to a 93% success rate;³⁷ however, because formalin can also cause rectal pain, spasm, ulcerations, or stenosis, it is not a first-line therapy.
• Tenesmus, urgency, and incontinence. These represent a therapeutic challenge, often with no satisfactory outcomes. An array of empiric treatments may be used for symptomatic relief, including but not limited to, a trial of loperamide or fiber supplementation, which may be helpful for frequent evacuation.
• Fistulous tracts associated with rectal radiation. Endoscopic clip closure of XRT-related and other fistulous tracts is an option. This has been attempted via a variety of techniques, but results depend on the size and location of the fistulous tract, as well as other characteristics of the fistula and its surrounding tissue.^7,38,39 Use of mesenchymal stem cells has also been described for rectal and other fistulae,⁴⁰ but its indications have yet to be elucidated, and current use is mostly experimental.

CASE The patient’s recent-onset symptoms and clinical history were most suggestive of radiation proctopathy; a shared decision was made to pursue endoscopic evaluation with possible therapeutic intervention. 

Given that data were not available about the quality of the colon preparation during the exam 7 years earlier, and to rule out a more proximal colonic lesion, the patient was scheduled for colonoscopy. This revealed numerous telangiectasias and moderate friability involving the distal third of the rectum, consistent with radiation proctopathy. The telangiectasias were treated with APC. Follow-up flexible sigmoidoscopy 2 months later showed a few remaining scattered telangiectasias, which were also treated with APC.

The patient has been clinically well, without evidence of bleeding for 6 months and with resolution of anemia_.

CORRESPONDENCE
James H. Tabibian, Division of Gastroenterology, Department of Medicine, 14445 Olive View Dr., 2B-182, Sylmar, CA 91342; [email protected].

CASE A 57-year-old man presented for evaluation of painless, intermittent passage of bright red blood per rectum for several months. His bowel habits were otherwise unchanged, averaging 2 soft bowel movements daily without straining. His medical history was significant for radiation therapy for prostate cancer 18 months earlier and a recent finding of mild microcytic anemia. A colonoscopy 7 years ago was negative for polyps, diverticula, or other lesions. He denied any family history of colon cancer or other gastrointestinal disorders. He wanted to know what he could do to stop the bleeding or if further testing would be needed.

Next steps?

Radiation therapy and its effect on the GI tract

In 1895, Dr. Wilhelm Roentgen first introduced the use of x-rays for diagnostic radiographic purposes. A year later, Dr. Emil Gruble made the first attempt to use radiation therapy (XRT) to treat cancer. In 1897, Dr. David Walsh described the first case of XRT-induced tissue injury in the British Medical Journal.¹

Since then, XRT has been used extensively to treat cancer, and its delivery techniques have improved and diversified. Like chemotherapy, XRT has its greatest effect on rapidly dividing cells, but as a result, the adverse effects of therapy are also greatest on rapidly dividing normal tissues, as well as others in the radiation field.

A large proportion of cancer patients will receive XRT, yet XRT-related costs account for less than 5% of total cancer care expenditure, suggesting cost effectiveness.^2,3 However, even with the great progress achieved in the delivery of XRT, it continues to have its share of acute and chronic complications, among the most common of which is gastrointestinal (GI) tract toxicity. These adverse effects are often first reported to, diagnosed, or treated by the primary care provider, who frequently remains pivotally involved in the patient’s longitudinal care.

Radiation therapy's adverse effects are often first reported to, diagnosed, or treated by family physicians, who frequently remain centrally involved in longitudinal care.

Approximately 50% to 75% of patients undergoing XRT will have some degree of GI symptoms of acute injury, but the majority will recover fully within a few weeks following completion of treatment.^4-6 However, in about 5% of patients,^4-6 there will be long-term consequences of varying degrees that may develop as soon as one year or as long as 10 years after XRT. These can pose substantial challenges for patients, as well as both the primary care provider and consulting specialists.

In the review that follows, we detail the potential acute and chronic complications of XRT on the GI tract and how best to manage them. But first, a word about the related terminology.

Getting a handle on XRT-related injury terminology

The preferred terms used to describe injury to normal tissue as a result of XRT include “XRT-related injury” or “pelvic radiation disease” (when the injury is confined to intrapelvic tissues); organ-specific descriptors such as “radiation enteropathy” or “XRT-induced esophageal stricture” are also used and are acceptable.^4,7,8

Terms such as “radiation enteritis” or “radiation proctitis” are considered misnomers since there is no significant histologic inflammation. Indeed, as we will discuss, acute injury is largely due to epithelial cellular injury and cell death (necrosis), while chronic injury is primarily the consequence of ongoing tissue ischemia, fibrosis, and other pathophysiologic processes.

Acute vs chronic XRT-related tissue injury

From a pathobiologic and clinical perspective, XRT-related injury can be categorized as either acute or chronic.^8-12 Acute XRT-related injury involves direct cellular necrosis of the epithelial cells and damage (eg, irreparable DNA alterations) to stem cells. This acute injury prevents appropriate cellular regeneration, which results in denuded mucosa, mucosal ulcerations, and even perforation in severe cases.¹⁰ Acute injury starts 2 to 3 weeks after initiating XRT and typically resolves within 2 to 3 months following completion of treatment.

Chronic XRT toxicity is pathophysiologically complex and multifactorial.^10-12 It includes: obliterative endarteritis of submucosal arterioles with chronic tissue ischemia, eosinophil infiltration, fibroblast proliferation and pathologic fibrosis, neovascularization with friable telangiectasia formation, and bowel serosal injury that promotes formation of dense adhesions.¹³ Its pathogenesis remains incompletely understood.

Several treatment- and patient-related variables can impact the occurrence and nature of tissue injury secondary to XRT and are summarized in the Table.^4,9-13 Newer forms of radiotherapy such as proton beam and Yttrium-90 radioembolization may also cause radiation injury,¹⁴ but to a lesser degree than conventional external beam XRT, in part because of improved dose targeting. We will not discuss these modalities in this review.

Can’t something be done to prevent injury in the first place?

There are no convincing evidence-based preventive or therapeutic treatments that address the underlying mechanisms of either the acute or chronic phases of XRT-related GI tract injury, although hyperbaric oxygen (which we’ll discuss in greater detail shortly) may be a promising option.^{8,11,12,15-17} It’s believed that hyperbaric oxygen may prove useful by facilitating angiogenesis and improving tissue oxygenation.^8,11,15-17 Unfortunately, this treatment is not widely available, and the frequency and duration required for optimal results is unclear.

Numerous pharmacologic radioprotectants have been suggested or evaluated in small studies, but none have an established role in addressing XRT-related injury. Given these voids, emphasis on symptom management and empathic, supportive care is essential.¹⁸

A look at injuries and Tx options by organs affected

The esophagus

Injurious effects on the esophagus are seen following XRT for lung, mediastinal, hypopharyngeal, or esophageal cancers.^19,20 The total XRT dose and regimen may vary, but a typical course may involve 10 gray (ie, 1000 rads) per week (2 gray per day) for 5 weeks. The maximum tolerated dose by the esophagus is approximately 6 gray, above which most patients will have long-term complications; however, some patients may experience toxicity at even lower doses.

Acute complications of esophageal XRT-related injury include mucosal ulcerations, which can present as chest pain and odynophagia. The mucosal pathology can cause dysmotility, which results in dysphagia for both liquids and solids.^19-21

If severe symptoms develop during treatment, the dose per session can be reduced and/or the sessions can be delayed. Some patients require temporary gastrostomy feeding tubes until symptoms resolve. Mucosal ulcerations can become a chronic issue as well. The mainstay of treatment is symptomatic relief with topical anesthetics and anti-acid medications.

Chronic symptoms are more varied and can be difficult to manage^14,15 and include the following:

• Strictures. Esophageal dysphagia develops in nearly two-thirds of patients postradiation and, in many cases, is due to stricture formation.²² Symptoms may range from mild dysphagia with solids to complete esophageal obstruction.²³ Barium esophagography can be helpful to delineate esophageal stricture morphology and determine treatment options.

For the majority of patients, serial endoscopic dilation with a balloon catheter or bougie (or other endoscopic techniques) achieves adequate esophageal patency to alleviate symptoms; this may need to be repeated periodically to maintain patency, as nearly one-third of patients will experience recurrent stricturing.^21,23

• Tracheo-esophageal fistulae. This complication can lead to pneumonia and generally has a poor prognosis.

Fistulae are chiefly treated endoscopically with esophageal, and occasionally, tracheobronchial stent placement. As with esophageal strictures, barium imaging can help plan the therapeutic approach. Percutaneous feeding may be required in some patients as a bridge or when fistula closure cannot be achieved.

• Secondary esophageal carcinogenesis. This dreaded complication develops in up to 2% to 3% of patients at 10 years post-XRT.¹⁹

Pharmacologic therapy for esophageal symptoms is generally unsuccessful, although acid suppression therapy may help as an adjuvant treatment to endoscopic dilation for esophageal strictures. Surgery is seldom attempted because of the fibrotic/ischemic tissues and high postoperative morbidity/mortality.

The stomach

The stomach is relatively resistant to XRT injury. Although XRT therapy can cause a transient decrease in acid output, there are rarely significant short- or long-term consequences with conventional therapeutic dosing (less than 50 gray).¹¹

The liver

Hepatic resistance to radiation is relatively high; however, liver toxicity has been reported at low doses, an effect that is seen largely following bone marrow transplantation.²⁴ Acute histologic XRT-related liver injury changes consist of severe pan-lobar congestion leading to hemorrhagic necrosis, cell atrophy, and perivascular fibrosis, as well as sclerosis of central and sublobular hepatic veins. The majority of patients will show reversal of the histologic changes within 3 months; however, approximately 25% to 40% of patients,²⁵ depending on total XRT dose to the liver and other technical factors, will experience progressive and chronic changes resulting in liver atrophy, severe perivascular injury, and fibrosis of the portal vein or bile ducts.

Besides hyperbaric oxygen, there are no evidence-based preventive or therapeutic treatments that address the underlying mechanisms of radiation-related GI tract injury.

The clinical symptoms of acute liver injury may include right upper quadrant pain, ascites, jaundice, veno-occlusive disease, or Budd-Chiari syndrome.²⁵ The major chronic complication of XRT-related liver injury is progressive fibrosis, which may advance to cirrhosis.

Small bowel

The small bowel is the most radiosensitive GI tract organ due to high cell turnover, which makes it very susceptible to XRT-related injury.^4,8,10,26-28 Under 3 gray, ≤20% of patients will develop radiation enteropathy, while at >5 gray, the incidence rises progressively with dose, and a majority of patients will be symptomatic.²⁹ The degree to which the bowel is healthy before XRT can be an important factor in developing enteropathy. Parenthetically, treatment with a full bladder may also help displace some of the loops from the field of XRT and decrease injury.

Acute XRT-related injury of the small bowel includes mucosal necrosis (ie, direct cell death) and ulcerations that may present as diarrhea, pain, malabsorption, weight loss, bleeding, and perforation.^4,8,10,26-28 Fortunately, in most patients, these are self-limited and can be managed symptomatically. Loperamide is the first-line medication for diarrhea, although Lomotil (diphenoxylate/atropine) may also be used if necessary.^4,8,10,26-28 Nutrition may be challenging in severe cases, and if dietary modifications and supplementation do not prove sufficient, home parenteral nutrition is required.

Over time, chronic small bowel pathology may develop, including strictures in 3% to 15%, fistulae in 0.6% to 4.8%, secondary neoplasia in up to 10%, dysmotility- or adhesion-related small intestinal bacterial overgrowth in up to 45%, and malabsorption with associated nutritional deficiency in up to 63%.^26-28 Other common XRT-related complications are chronic pain, which could be due to adhesions or ischemia, small intestinal bacterial overgrowth, or partial bowel obstruction, and telangiectasias that result with acute or chronic blood loss.¹³

Imaging of small bowel disease to diagnose the various manifestations of radiation enteropathy is challenging. Conventional X-rays may be difficult to interpret. Therefore, computerized tomography or magnetic resonance enterography, capsule endoscopy, or balloon-assisted enteroscopy is preferred—depending on availability, local expertise, and the suspected pre-procedure diagnosis.

Telangiectasias are not seen on cross-sectional imaging but can be seen with capsule endoscopy (which should not be ordered if stricture is suspected unless a patency capsule has been tried). Single or double balloon enteroscopy (specialized endoscopes intended for reaching the mid and distal ileum), which has been used to treat strictures or telangiectasia in healthy tissues,²⁹ can be difficult or impossible in post-XRT patients because adhesions may limit progress of the scope to the area of interest, and forceful advancement of the scope increases the risk of perforation.

Small bowel telangiectasias can cause chronic occult blood loss, which often requires iron supplementation; acute bleeding may require blood transfusion and hospitalization. Of note, choosing an iron formulation that is well tolerated is critical to avoid (additional) unpleasant GI tract adverse effects. We typically recommend elemental iron with Vitamin C to augment absorption or ferrous gluconate; some patients will require intravenous iron infusion.

Surgery may be advisable to address complications such as fistulous tracts, complex strictures, or bowel obstruction; how-ever, operating on radiated abdominal tissues and ischemic bowel is associated with high morbidity and mortality.^4,25,28,30 The surgeon may encounter dense adhesions that make an otherwise “simple” surgery problematic.

For example, it may be difficult to access the desired region and determine the borders of healthy tissue; wide excisions are, thus, often performed, which may result in small bowel failure (ie, short gut syndrome) and a mortality rate in excess of 30%.³¹ In addition, the ischemic post-XRT tissues may not heal well even if the intended surgery is completed; indeed, anastomotic leaks, failures, and infections are not uncommon. Moreover, another 30% will have other postoperative complications, 40% to 60% may require more than one laparotomy, and 50% of those who recover from the initial surgery will develop recurrence of the fistulous tract or stricture.^4,25,28,30

No drug therapy has proven effective for prevention or mechanistically-driven treatment of XRT-induced small bowel injury. Hyperbaric oxygen therapy may be the most promising medical treatment, with early response in 53% of cases and long-term response of 66% to 73% for global symptomatic relief.³² It has been used successfully for treatment of pain, diarrhea, malabsorption, and hemorrhage from mucosal ulcerations, stenosis, and fistulous tracts. When available, it should be considered as a potential therapeutic intervention.

Colon

Injury to the colon is seen in 10% to 20% of patients following XRT for prostate, bladder, cervical, or uterine cancer.³³ The maximum tolerated dose of the colon is slightly higher than for the small intestine.³⁴ The rectosigmoid area is the area most commonly implicated, but depending on the field of radiation, injury can be more extensive/proximal.

The small bowel is the most radiosensitive GI tract organ, due to high cell turnover, making it highly susceptible to radiation therapy-related injury.

Acute XRT injury of the colon produces acute mucosal necrosis, which may manifest as bowel dysmotility, diarrhea, cramps, tenesmus, or hematochezia. Sigmoidoscopy or colonoscopy will show mucosal edema, erosions, and ulcerations with a purplish/red discoloration. A barium enema will show spasm of the affected area with so-called “thumbprinting,” which indicates mucosal edema. The onset of symptoms is generally within 3 weeks of XRT initiation; symptoms are self-limited in most cases. Management is centered on symptom relief; loperamide and Lomotil are first-line agents for diarrheal symptoms.

Chronic XRT-related colopathy is the result of chronic tissue ischemia and fibrosis. This may lead to dysmotility resulting in abnormal bowel habits (ranging from constipation to diarrhea) or sigmoid stenosis/stricture resulting in an inability to evacuate the bowel. For the latter, it is important to note that fiber supplementation may not be optimal, since increasing the fecal caliber makes it more difficult to pass through the stenotic, colonic segment.

Emollients such as small doses of mineral oil will not increase the fecal caliber, but will soften fecal matter so that it can be passed with greater ease. MiraLAX may be effective, as well, but can increase the sense of urgency and contribute to incontinence in some. Lactulose can be effective, but it causes excessive gassiness/bloating that may result in abdominal pain and episodes of incontinence.

Bleeding from telangiectasias is another chronic complication of XRT-related colonic injury. Argon plasma coagulation (APC) via flexible sigmoidoscopy or colonoscopy is typically the primary therapeutic approach, reported to have a success rate of up to 90% in healthy tissues.^33,35 Even with endoscopic treatment, as mentioned earlier in the context of small bowel XRT-related telangiectasias, iron supplementation is often needed to replete stores, and choice of iron agent is important.

Furthermore, it is essential to recognize that repeat endoscopic sessions may be needed to fully treat telangiectasias, and recrudescence of bleeding months or years later should raise suspicion for recurrent telangiectasia formation (and need for repeat treatment). As with other organs, there may be a role for hyperbaric oxygen, even in difficult-to-treat cases.^36,37

Colonic fibrosis/stenosis and fistulous tract formation, as in the small bowel, are also seen in this population of patients. Endoscopic dilation can be considered, and stenting may be reasonable for short and/or distal strictures. Surgical approaches for fistulous tracts and strictures can be high-risk and associated with poor outcomes, mostly because of the underlying chronic tissue ischemia and fibrosis,^4,8,27,30,34 as discussed in the small bowel section.

Rectum

The rectum has tolerance to XRT similar to the colon,³⁸ but because of its anatomical location, rectal radiation injury is more common, and is typically seen after XRT for prostate, bladder, cervical, or uterine cancer. Acute rectal radiation injury is seen in 50% to 78% of patients,³⁶ and symptoms are similar to that of injury to the sigmoid (eg, tenesmus, loose evacuations, hematochezia), all of which are consequences of direct radiation injury to the mucosa.

Use of mesenchymal stem cells has also been described for rectal and other fistulae, but use is mostly experimental.

Chronic rectal radiation injury may present in a variety of ways. Tenesmus and incontinence are seen in 8% to 20% of patients, frequent defecation in 50%, urgency in 47%, and rectal cancer in up to 2% to 3% after 10 years.^36,37 Other complications include anorectal strictures, fissures, fistulae, and bleeding from rectal telangiectasias. While anoscopy can diagnose many of these, flexible sigmoidoscopy is needed to examine more proximal rectal sites as well as for treatment. Treatment of these chronic complications of XRT is analogous to those of the colon⁷ with the following exceptions:

• Anorectal strictures. In contrast to sigmoid strictures, these are generally more amenable to dilatation. If symptoms recur frequently, patients may be instructed on self-dilatations at home.
• Bleeding from rectal telangiectasias. In the rare cases where endoscopic APC is not feasible or successful, an alternative treatment would be radiofrequency ablation or the application of 2% to 10% formalin intra-rectally. This is reported to have up to a 93% success rate;³⁷ however, because formalin can also cause rectal pain, spasm, ulcerations, or stenosis, it is not a first-line therapy.
• Tenesmus, urgency, and incontinence. These represent a therapeutic challenge, often with no satisfactory outcomes. An array of empiric treatments may be used for symptomatic relief, including but not limited to, a trial of loperamide or fiber supplementation, which may be helpful for frequent evacuation.
• Fistulous tracts associated with rectal radiation. Endoscopic clip closure of XRT-related and other fistulous tracts is an option. This has been attempted via a variety of techniques, but results depend on the size and location of the fistulous tract, as well as other characteristics of the fistula and its surrounding tissue.^7,38,39 Use of mesenchymal stem cells has also been described for rectal and other fistulae,⁴⁰ but its indications have yet to be elucidated, and current use is mostly experimental.

CASE The patient’s recent-onset symptoms and clinical history were most suggestive of radiation proctopathy; a shared decision was made to pursue endoscopic evaluation with possible therapeutic intervention. 

Given that data were not available about the quality of the colon preparation during the exam 7 years earlier, and to rule out a more proximal colonic lesion, the patient was scheduled for colonoscopy. This revealed numerous telangiectasias and moderate friability involving the distal third of the rectum, consistent with radiation proctopathy. The telangiectasias were treated with APC. Follow-up flexible sigmoidoscopy 2 months later showed a few remaining scattered telangiectasias, which were also treated with APC.

The patient has been clinically well, without evidence of bleeding for 6 months and with resolution of anemia_.

CORRESPONDENCE
James H. Tabibian, Division of Gastroenterology, Department of Medicine, 14445 Olive View Dr., 2B-182, Sylmar, CA 91342; [email protected].

CASE A 57-year-old man presented for evaluation of painless, intermittent passage of bright red blood per rectum for several months. His bowel habits were otherwise unchanged, averaging 2 soft bowel movements daily without straining. His medical history was significant for radiation therapy for prostate cancer 18 months earlier and a recent finding of mild microcytic anemia. A colonoscopy 7 years ago was negative for polyps, diverticula, or other lesions. He denied any family history of colon cancer or other gastrointestinal disorders. He wanted to know what he could do to stop the bleeding or if further testing would be needed.

Next steps?

Radiation therapy and its effect on the GI tract

In 1895, Dr. Wilhelm Roentgen first introduced the use of x-rays for diagnostic radiographic purposes. A year later, Dr. Emil Gruble made the first attempt to use radiation therapy (XRT) to treat cancer. In 1897, Dr. David Walsh described the first case of XRT-induced tissue injury in the British Medical Journal.¹

Since then, XRT has been used extensively to treat cancer, and its delivery techniques have improved and diversified. Like chemotherapy, XRT has its greatest effect on rapidly dividing cells, but as a result, the adverse effects of therapy are also greatest on rapidly dividing normal tissues, as well as others in the radiation field.

A large proportion of cancer patients will receive XRT, yet XRT-related costs account for less than 5% of total cancer care expenditure, suggesting cost effectiveness.^2,3 However, even with the great progress achieved in the delivery of XRT, it continues to have its share of acute and chronic complications, among the most common of which is gastrointestinal (GI) tract toxicity. These adverse effects are often first reported to, diagnosed, or treated by the primary care provider, who frequently remains pivotally involved in the patient’s longitudinal care.

Radiation therapy's adverse effects are often first reported to, diagnosed, or treated by family physicians, who frequently remain centrally involved in longitudinal care.

Approximately 50% to 75% of patients undergoing XRT will have some degree of GI symptoms of acute injury, but the majority will recover fully within a few weeks following completion of treatment.^4-6 However, in about 5% of patients,^4-6 there will be long-term consequences of varying degrees that may develop as soon as one year or as long as 10 years after XRT. These can pose substantial challenges for patients, as well as both the primary care provider and consulting specialists.

In the review that follows, we detail the potential acute and chronic complications of XRT on the GI tract and how best to manage them. But first, a word about the related terminology.

Getting a handle on XRT-related injury terminology

The preferred terms used to describe injury to normal tissue as a result of XRT include “XRT-related injury” or “pelvic radiation disease” (when the injury is confined to intrapelvic tissues); organ-specific descriptors such as “radiation enteropathy” or “XRT-induced esophageal stricture” are also used and are acceptable.^4,7,8

Terms such as “radiation enteritis” or “radiation proctitis” are considered misnomers since there is no significant histologic inflammation. Indeed, as we will discuss, acute injury is largely due to epithelial cellular injury and cell death (necrosis), while chronic injury is primarily the consequence of ongoing tissue ischemia, fibrosis, and other pathophysiologic processes.

Acute vs chronic XRT-related tissue injury

From a pathobiologic and clinical perspective, XRT-related injury can be categorized as either acute or chronic.^8-12 Acute XRT-related injury involves direct cellular necrosis of the epithelial cells and damage (eg, irreparable DNA alterations) to stem cells. This acute injury prevents appropriate cellular regeneration, which results in denuded mucosa, mucosal ulcerations, and even perforation in severe cases.¹⁰ Acute injury starts 2 to 3 weeks after initiating XRT and typically resolves within 2 to 3 months following completion of treatment.

Chronic XRT toxicity is pathophysiologically complex and multifactorial.^10-12 It includes: obliterative endarteritis of submucosal arterioles with chronic tissue ischemia, eosinophil infiltration, fibroblast proliferation and pathologic fibrosis, neovascularization with friable telangiectasia formation, and bowel serosal injury that promotes formation of dense adhesions.¹³ Its pathogenesis remains incompletely understood.

Several treatment- and patient-related variables can impact the occurrence and nature of tissue injury secondary to XRT and are summarized in the Table.^4,9-13 Newer forms of radiotherapy such as proton beam and Yttrium-90 radioembolization may also cause radiation injury,¹⁴ but to a lesser degree than conventional external beam XRT, in part because of improved dose targeting. We will not discuss these modalities in this review.

Can’t something be done to prevent injury in the first place?

There are no convincing evidence-based preventive or therapeutic treatments that address the underlying mechanisms of either the acute or chronic phases of XRT-related GI tract injury, although hyperbaric oxygen (which we’ll discuss in greater detail shortly) may be a promising option.^{8,11,12,15-17} It’s believed that hyperbaric oxygen may prove useful by facilitating angiogenesis and improving tissue oxygenation.^8,11,15-17 Unfortunately, this treatment is not widely available, and the frequency and duration required for optimal results is unclear.

Numerous pharmacologic radioprotectants have been suggested or evaluated in small studies, but none have an established role in addressing XRT-related injury. Given these voids, emphasis on symptom management and empathic, supportive care is essential.¹⁸

A look at injuries and Tx options by organs affected

The esophagus

Injurious effects on the esophagus are seen following XRT for lung, mediastinal, hypopharyngeal, or esophageal cancers.^19,20 The total XRT dose and regimen may vary, but a typical course may involve 10 gray (ie, 1000 rads) per week (2 gray per day) for 5 weeks. The maximum tolerated dose by the esophagus is approximately 6 gray, above which most patients will have long-term complications; however, some patients may experience toxicity at even lower doses.

Acute complications of esophageal XRT-related injury include mucosal ulcerations, which can present as chest pain and odynophagia. The mucosal pathology can cause dysmotility, which results in dysphagia for both liquids and solids.^19-21

If severe symptoms develop during treatment, the dose per session can be reduced and/or the sessions can be delayed. Some patients require temporary gastrostomy feeding tubes until symptoms resolve. Mucosal ulcerations can become a chronic issue as well. The mainstay of treatment is symptomatic relief with topical anesthetics and anti-acid medications.

Chronic symptoms are more varied and can be difficult to manage^14,15 and include the following:

• Strictures. Esophageal dysphagia develops in nearly two-thirds of patients postradiation and, in many cases, is due to stricture formation.²² Symptoms may range from mild dysphagia with solids to complete esophageal obstruction.²³ Barium esophagography can be helpful to delineate esophageal stricture morphology and determine treatment options.

For the majority of patients, serial endoscopic dilation with a balloon catheter or bougie (or other endoscopic techniques) achieves adequate esophageal patency to alleviate symptoms; this may need to be repeated periodically to maintain patency, as nearly one-third of patients will experience recurrent stricturing.^21,23

• Tracheo-esophageal fistulae. This complication can lead to pneumonia and generally has a poor prognosis.

Fistulae are chiefly treated endoscopically with esophageal, and occasionally, tracheobronchial stent placement. As with esophageal strictures, barium imaging can help plan the therapeutic approach. Percutaneous feeding may be required in some patients as a bridge or when fistula closure cannot be achieved.

• Secondary esophageal carcinogenesis. This dreaded complication develops in up to 2% to 3% of patients at 10 years post-XRT.¹⁹

Pharmacologic therapy for esophageal symptoms is generally unsuccessful, although acid suppression therapy may help as an adjuvant treatment to endoscopic dilation for esophageal strictures. Surgery is seldom attempted because of the fibrotic/ischemic tissues and high postoperative morbidity/mortality.

The stomach

The stomach is relatively resistant to XRT injury. Although XRT therapy can cause a transient decrease in acid output, there are rarely significant short- or long-term consequences with conventional therapeutic dosing (less than 50 gray).¹¹

The liver

Hepatic resistance to radiation is relatively high; however, liver toxicity has been reported at low doses, an effect that is seen largely following bone marrow transplantation.²⁴ Acute histologic XRT-related liver injury changes consist of severe pan-lobar congestion leading to hemorrhagic necrosis, cell atrophy, and perivascular fibrosis, as well as sclerosis of central and sublobular hepatic veins. The majority of patients will show reversal of the histologic changes within 3 months; however, approximately 25% to 40% of patients,²⁵ depending on total XRT dose to the liver and other technical factors, will experience progressive and chronic changes resulting in liver atrophy, severe perivascular injury, and fibrosis of the portal vein or bile ducts.

Besides hyperbaric oxygen, there are no evidence-based preventive or therapeutic treatments that address the underlying mechanisms of radiation-related GI tract injury.

The clinical symptoms of acute liver injury may include right upper quadrant pain, ascites, jaundice, veno-occlusive disease, or Budd-Chiari syndrome.²⁵ The major chronic complication of XRT-related liver injury is progressive fibrosis, which may advance to cirrhosis.

Small bowel

The small bowel is the most radiosensitive GI tract organ due to high cell turnover, which makes it very susceptible to XRT-related injury.^4,8,10,26-28 Under 3 gray, ≤20% of patients will develop radiation enteropathy, while at >5 gray, the incidence rises progressively with dose, and a majority of patients will be symptomatic.²⁹ The degree to which the bowel is healthy before XRT can be an important factor in developing enteropathy. Parenthetically, treatment with a full bladder may also help displace some of the loops from the field of XRT and decrease injury.

Acute XRT-related injury of the small bowel includes mucosal necrosis (ie, direct cell death) and ulcerations that may present as diarrhea, pain, malabsorption, weight loss, bleeding, and perforation.^4,8,10,26-28 Fortunately, in most patients, these are self-limited and can be managed symptomatically. Loperamide is the first-line medication for diarrhea, although Lomotil (diphenoxylate/atropine) may also be used if necessary.^4,8,10,26-28 Nutrition may be challenging in severe cases, and if dietary modifications and supplementation do not prove sufficient, home parenteral nutrition is required.

Over time, chronic small bowel pathology may develop, including strictures in 3% to 15%, fistulae in 0.6% to 4.8%, secondary neoplasia in up to 10%, dysmotility- or adhesion-related small intestinal bacterial overgrowth in up to 45%, and malabsorption with associated nutritional deficiency in up to 63%.^26-28 Other common XRT-related complications are chronic pain, which could be due to adhesions or ischemia, small intestinal bacterial overgrowth, or partial bowel obstruction, and telangiectasias that result with acute or chronic blood loss.¹³

Imaging of small bowel disease to diagnose the various manifestations of radiation enteropathy is challenging. Conventional X-rays may be difficult to interpret. Therefore, computerized tomography or magnetic resonance enterography, capsule endoscopy, or balloon-assisted enteroscopy is preferred—depending on availability, local expertise, and the suspected pre-procedure diagnosis.

Telangiectasias are not seen on cross-sectional imaging but can be seen with capsule endoscopy (which should not be ordered if stricture is suspected unless a patency capsule has been tried). Single or double balloon enteroscopy (specialized endoscopes intended for reaching the mid and distal ileum), which has been used to treat strictures or telangiectasia in healthy tissues,²⁹ can be difficult or impossible in post-XRT patients because adhesions may limit progress of the scope to the area of interest, and forceful advancement of the scope increases the risk of perforation.

Small bowel telangiectasias can cause chronic occult blood loss, which often requires iron supplementation; acute bleeding may require blood transfusion and hospitalization. Of note, choosing an iron formulation that is well tolerated is critical to avoid (additional) unpleasant GI tract adverse effects. We typically recommend elemental iron with Vitamin C to augment absorption or ferrous gluconate; some patients will require intravenous iron infusion.

Surgery may be advisable to address complications such as fistulous tracts, complex strictures, or bowel obstruction; how-ever, operating on radiated abdominal tissues and ischemic bowel is associated with high morbidity and mortality.^4,25,28,30 The surgeon may encounter dense adhesions that make an otherwise “simple” surgery problematic.

For example, it may be difficult to access the desired region and determine the borders of healthy tissue; wide excisions are, thus, often performed, which may result in small bowel failure (ie, short gut syndrome) and a mortality rate in excess of 30%.³¹ In addition, the ischemic post-XRT tissues may not heal well even if the intended surgery is completed; indeed, anastomotic leaks, failures, and infections are not uncommon. Moreover, another 30% will have other postoperative complications, 40% to 60% may require more than one laparotomy, and 50% of those who recover from the initial surgery will develop recurrence of the fistulous tract or stricture.^4,25,28,30

No drug therapy has proven effective for prevention or mechanistically-driven treatment of XRT-induced small bowel injury. Hyperbaric oxygen therapy may be the most promising medical treatment, with early response in 53% of cases and long-term response of 66% to 73% for global symptomatic relief.³² It has been used successfully for treatment of pain, diarrhea, malabsorption, and hemorrhage from mucosal ulcerations, stenosis, and fistulous tracts. When available, it should be considered as a potential therapeutic intervention.

Colon

Injury to the colon is seen in 10% to 20% of patients following XRT for prostate, bladder, cervical, or uterine cancer.³³ The maximum tolerated dose of the colon is slightly higher than for the small intestine.³⁴ The rectosigmoid area is the area most commonly implicated, but depending on the field of radiation, injury can be more extensive/proximal.

The small bowel is the most radiosensitive GI tract organ, due to high cell turnover, making it highly susceptible to radiation therapy-related injury.

Acute XRT injury of the colon produces acute mucosal necrosis, which may manifest as bowel dysmotility, diarrhea, cramps, tenesmus, or hematochezia. Sigmoidoscopy or colonoscopy will show mucosal edema, erosions, and ulcerations with a purplish/red discoloration. A barium enema will show spasm of the affected area with so-called “thumbprinting,” which indicates mucosal edema. The onset of symptoms is generally within 3 weeks of XRT initiation; symptoms are self-limited in most cases. Management is centered on symptom relief; loperamide and Lomotil are first-line agents for diarrheal symptoms.

Chronic XRT-related colopathy is the result of chronic tissue ischemia and fibrosis. This may lead to dysmotility resulting in abnormal bowel habits (ranging from constipation to diarrhea) or sigmoid stenosis/stricture resulting in an inability to evacuate the bowel. For the latter, it is important to note that fiber supplementation may not be optimal, since increasing the fecal caliber makes it more difficult to pass through the stenotic, colonic segment.

Emollients such as small doses of mineral oil will not increase the fecal caliber, but will soften fecal matter so that it can be passed with greater ease. MiraLAX may be effective, as well, but can increase the sense of urgency and contribute to incontinence in some. Lactulose can be effective, but it causes excessive gassiness/bloating that may result in abdominal pain and episodes of incontinence.

Bleeding from telangiectasias is another chronic complication of XRT-related colonic injury. Argon plasma coagulation (APC) via flexible sigmoidoscopy or colonoscopy is typically the primary therapeutic approach, reported to have a success rate of up to 90% in healthy tissues.^33,35 Even with endoscopic treatment, as mentioned earlier in the context of small bowel XRT-related telangiectasias, iron supplementation is often needed to replete stores, and choice of iron agent is important.

Furthermore, it is essential to recognize that repeat endoscopic sessions may be needed to fully treat telangiectasias, and recrudescence of bleeding months or years later should raise suspicion for recurrent telangiectasia formation (and need for repeat treatment). As with other organs, there may be a role for hyperbaric oxygen, even in difficult-to-treat cases.^36,37

Colonic fibrosis/stenosis and fistulous tract formation, as in the small bowel, are also seen in this population of patients. Endoscopic dilation can be considered, and stenting may be reasonable for short and/or distal strictures. Surgical approaches for fistulous tracts and strictures can be high-risk and associated with poor outcomes, mostly because of the underlying chronic tissue ischemia and fibrosis,^4,8,27,30,34 as discussed in the small bowel section.

Rectum

The rectum has tolerance to XRT similar to the colon,³⁸ but because of its anatomical location, rectal radiation injury is more common, and is typically seen after XRT for prostate, bladder, cervical, or uterine cancer. Acute rectal radiation injury is seen in 50% to 78% of patients,³⁶ and symptoms are similar to that of injury to the sigmoid (eg, tenesmus, loose evacuations, hematochezia), all of which are consequences of direct radiation injury to the mucosa.

Use of mesenchymal stem cells has also been described for rectal and other fistulae, but use is mostly experimental.

Chronic rectal radiation injury may present in a variety of ways. Tenesmus and incontinence are seen in 8% to 20% of patients, frequent defecation in 50%, urgency in 47%, and rectal cancer in up to 2% to 3% after 10 years.^36,37 Other complications include anorectal strictures, fissures, fistulae, and bleeding from rectal telangiectasias. While anoscopy can diagnose many of these, flexible sigmoidoscopy is needed to examine more proximal rectal sites as well as for treatment. Treatment of these chronic complications of XRT is analogous to those of the colon⁷ with the following exceptions:

• Anorectal strictures. In contrast to sigmoid strictures, these are generally more amenable to dilatation. If symptoms recur frequently, patients may be instructed on self-dilatations at home.
• Bleeding from rectal telangiectasias. In the rare cases where endoscopic APC is not feasible or successful, an alternative treatment would be radiofrequency ablation or the application of 2% to 10% formalin intra-rectally. This is reported to have up to a 93% success rate;³⁷ however, because formalin can also cause rectal pain, spasm, ulcerations, or stenosis, it is not a first-line therapy.
• Tenesmus, urgency, and incontinence. These represent a therapeutic challenge, often with no satisfactory outcomes. An array of empiric treatments may be used for symptomatic relief, including but not limited to, a trial of loperamide or fiber supplementation, which may be helpful for frequent evacuation.
• Fistulous tracts associated with rectal radiation. Endoscopic clip closure of XRT-related and other fistulous tracts is an option. This has been attempted via a variety of techniques, but results depend on the size and location of the fistulous tract, as well as other characteristics of the fistula and its surrounding tissue.^7,38,39 Use of mesenchymal stem cells has also been described for rectal and other fistulae,⁴⁰ but its indications have yet to be elucidated, and current use is mostly experimental.

CASE The patient’s recent-onset symptoms and clinical history were most suggestive of radiation proctopathy; a shared decision was made to pursue endoscopic evaluation with possible therapeutic intervention. 

Given that data were not available about the quality of the colon preparation during the exam 7 years earlier, and to rule out a more proximal colonic lesion, the patient was scheduled for colonoscopy. This revealed numerous telangiectasias and moderate friability involving the distal third of the rectum, consistent with radiation proctopathy. The telangiectasias were treated with APC. Follow-up flexible sigmoidoscopy 2 months later showed a few remaining scattered telangiectasias, which were also treated with APC.

The patient has been clinically well, without evidence of bleeding for 6 months and with resolution of anemia_.

CORRESPONDENCE
James H. Tabibian, Division of Gastroenterology, Department of Medicine, 14445 Olive View Dr., 2B-182, Sylmar, CA 91342; [email protected].

ILLUSTRATIVE CASE

A 66-year-old man with a history of hypertension and diabetes mellitus type 2 is hospitalized for palpitations and dizziness, and is given a diagnosis of atrial fibrillation (AF). His heart rate is successfully controlled with a beta-blocker. His CHA₂DS₂-VASc score is 3, meaning he is a candidate for anticoagulation. Which agent should you start?

Thromboembolism in patients with AF results in stroke and death and can be decreased with appropriate use of antithrombotic therapy. Evidence-based guidelines recommend patients with AF at intermediate or high risk of stroke (CHADS₂ score ≥ 2 or prior history of cardioembolic stroke or transient ischemic attack) receive antithrombotic therapy with oral anticoagulation, rather than receive no therapy or therapy with antiplatelets.^2,3

The American College of Chest Physicians also recommends the use of the direct oral anticoagulant (DOAC) dabigatran over warfarin for those patients with nonvalvular AF with an estimated glomerular filtration rate (eGFR) ≥15 mL/min/1.73 m².³

A meta-analysis of large randomized controlled trials (RCTs) of individual DOACs (dabigatran [a direct thrombin inhibitor], rivaroxaban, apixaban, and edoxaban [factor Xa inhibitors]) revealed similar or lower rates of ischemic stroke and major bleeding (except gastrointestinal bleeds; relative risk=1.25; 95% CI, 1.01 to 1.55) when compared with warfarin (at an international normalized ratio [INR] goal of 2-3).⁴ In addition, 3 separate meta-analyses that pooled results from large RCTs involving dabigatran, apixaban, and rivaroxaban also concluded that these medications result in a significant reduction in embolic stroke and reduced the risk of major bleeds and hemorrhagic stroke when compared with warfarin.^5-7

Rivaroxaban may be more effective than warfarin at preventing ischemic stroke and systemic emboli, and apixaban and dabigatran have a lower risk of bleeding.

However, we know less about the comparative effectiveness and safety of the DOACs when they are used in clinical practice, and it is not clear which, if any of these agents, are superior to others. Moreover, only about half of the patients in the United States with AF who are eligible to take DOACs are currently managed with them.⁸

STUDY SUMMARY

One DOAC is better than warfarin at one thing; 2 others are better at another

This large cohort study examined the effectiveness of 3 DOACs compared with warfarin in 61,678 patients with AF by combining data from 3 Danish national databases. The patients had newly diagnosed AF (without valvular disease or venous thromboembolism) and were prescribed standard doses of DOACs (dabigatran 150 bid [N=12,701], rivaroxaban 20 mg/d [N=7192], apixaban 5 mg bid [N=6349]) or dose-adjusted warfarin to an INR goal of 2 to 3 (N=35,436). Patients were followed for an average of 1.9 years.

Ischemic stroke, systemic emboli. In the first year of observation, there were 1702 ischemic strokes or systemic emboli. The incidence of ischemic stroke or systemic embolism was either the same or better for each of the 3 DOAC treatments than for warfarin (DOACs, 2.9-3.9 events per 100 person-years; warfarin, 3.3 events per 100 person-years; no P value provided). Ischemic stroke or systemic emboli events occurred less frequently in the rivaroxaban group compared with warfarin at one year (hazard ratio [HR]=0.83; 95% confidence interval [CI], 0.69-0.99) and after 2.5 years (HR=0.80; 95% CI, 0.69-0.94). The rates of ischemic stroke and systemic emboli for both apixaban and dabigatran were not significantly different than that for warfarin at one year and 2.5 years.

Bleeding events (defined as intracranial, major gastrointestinal, and traumatic intracranial) were lower in the apixaban group (HR=0.63; 95% CI, 0.53-0.76) and dabigatran group (HR=0.61; 95% CI, 0.51-0.74) than in the warfarin group at one year. Significant reductions remained after 2.5 years. There was no difference in bleeding events between rivaroxaban and warfarin.

Risk of death. Compared with warfarin, the risk of death after one year of treatment was lower in the apixaban (HR=0.65; 95% CI, 0.56-0.75) and dabigatran (HR=0.63; 95% CI, 0.48-0.82) groups, and there was no significant difference in the rivaroxaban group (HR=0.92; 95% CI, 0.82-1.03).

WHAT’S NEW

No agent “has it all,” but DOACs have advantages

This comparative effectiveness and safety analysis reveals that all of the DOACs are at least as effective as warfarin in preventing ischemic stroke and systemic emboli, and that rivaroxaban may be more effective, and that apixaban and dabigatran have a lower risk of bleeding than warfarin.

CAVEATS

This non-randomized cohort trial lacked INR data

This study was a non-randomized cohort trial. And, while propensity weighting helps, the researchers were unable to completely control for underlying risk factors or unknown confounders.

INR data for patients on warfarin was not provided, so it is not clear how often patients were out of therapeutic range, which could affect the stroke and bleeding results in the warfarin group. This, however, is seen with routine use of warfarin. We feel that this study reflects the challenge of maintaining patients in warfarin’s narrow therapeutic range.

CHALLENGES TO IMPLEMENTATION

It comes down to cost

Cost could be a barrier, as health insurance coverage for DOACs varies. Patients with high-deductible health insurance plans, or who find themselves in the Medicare “donut hole,” may be at a particular disadvantage.

ACKNOWLEDGEMENT

The PURLs Surveillance System was supported in part by Grant Number UL1RR024999 from the National Center For Research Resources, a Clinical Translational Science Award to the University of Chicago. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.