Case Reports

ABSTRACT

Stress fractures are often missed, especially in unusual clinical settings. We report on 2 patients who presented to our orthopedic surgery clinic with incidental findings of asymptomatic proximal fibular tension side stress fractures in severe longstanding varus osteoarthritic knees. Initial plain films demonstrated an expansile deformity of the proximal fibular shaft, and differential diagnosis included a healed or healing fracture versus possible neoplasm. Magnetic resonance imaging with and without gadolinium was utilized to rule out the latter prior to planned total knee arthroplasty.

Continue to: The proximal fibula...

 

The proximal fibula is a rare site for stress fractures, with most of these fractures occurring in military recruits.¹ To the authors’ knowledge, there has been only 1 documented case of a proximal fibular stress fracture in patients with severe osteoarthritis (OA) and fixed varus deformity, which mimicked L5 radiculopathy.² We are not aware of any reports of asymptomatic tension-side fibular stress fractures in varus knees. In our 2 cases, the patients were indicated for total knee arthroplasty (TKA) for varus degenerative joint disease after failing nonoperative treatment; however, further work-up was justified to rule out neoplasm after plain films revealed expansile deformities of the proximal fibular shaft. Each patient subsequently underwent magnetic resonance imaging (MRI) with and without gadolinium contrast, which demonstrated a healed and healing proximal fibular stress fracture. Magnetic resonance imaging is rarely indicated in the evaluation of degenerative joint disease, and stress fractures about a varus knee generally occur on the compression side of the tibia and are symptomatic.^3-7 The patients provided informed written consent for print and electronic publication of this case report.

CASE REPORT

The first patient was a 77-year-old male who presented with longstanding knee pain, left greater than right, exacerbated by weight-bearing activities. The patient had no improvement with physical therapy or anti-inflammatory medication. He denied any history of trauma, weakness, paresthesias, or a recent increase in activity. The patient also denied any fevers, chills, night sweats, or other constitutional symptoms. On physical examination, the patient had an antalgic gait and limited range of motion bilaterally. Examination of his right lower extremity demonstrated a fixed 5° varus deformity. No distinct point tenderness was noted.

Radiographs of the right knee demonstrated varus deformity and tricompartmental degenerative changes with severe medial joint space narrowing. An expansile deformity of the proximal right fibular shaft was also noted (Figure 1), which was not present on the films 2 years earlier (Figure 2). The absence of this deformity on previous imaging raised the suspicion of a tumor. An MRI with and without gadolinium, which was obtained to rule out a neoplastic process, showed an old, healed proximal fibular shaft fracture with chronic periosteal reaction (Figure 3). There was no marrow edema to suggest acute injury and no neoplastic lesion. He was reassured regarding the benign findings and was scheduled for a left TKA, as his pain was more severe on the left knee. The patient’s stress fracture healed without complications, and he underwent a successful left TKA. He returned approximately 6 months after his procedure with worsening right knee pain and underwent a successful TKA on the right knee as well.

The second patient was a 67-year-old male with longstanding bilateral knee pain, right greater than left, with no antecedent trauma. He denied a history of increased activity, or weakness or paresthesias. He denied any fevers, chills, night sweats, or other constitutional symptoms. One year prior to presentation at our clinic, he had received corticosteroid injections and hyaluronic acid, without relief. The patient also had a history with another surgeon of arthroscopy 1 year earlier and subchondroplasty 3 years before presentation to our clinic. On physical examination, the patient’s right knee displayed a fixed 7° varus deformity with decreased range of motion, effusion, and diffuse crepitus. Further examination revealed tenderness to palpation of the proximal fibula.

Radiographs of the right knee showed degenerative joint disease with varus deformity and medial compartment joint space narrowing. They also demonstrated an expansile deformity of mixed lucency and sclerosis involving the proximal right fibular shaft (Figure 4). Although these findings appeared to be consistent with a stress fracture, their appearance was also suspicious for a neoplasm. To rule out malignancy, an MRI with and without gadolinium was obtained that revealed a healing stress fracture of the proximal fibula (Figure 5). The patient was reassured, and plans were made to proceed with a TKA. The patient’s stress fracture healed without complications, and he underwent successful right TKA. Radiographs from the patient’s 8-week follow-up showed a healed fibular stress fracture (Figure 6).

Continue to: DISCUSSION

 

 

DISCUSSION

To our knowledge, this is the first report of incidental tension-side stress fractures in varus osteoarthritic knees. Stress fractures have been classified into 2 groups, fatigue fractures and insufficiency fractures. Fatigue fractures occur when abnormal stress is applied to normal bones, and insufficiency fractures result when normal stress is applied to abnormal bones.⁸ Stress fractures can also be classified into risk categories based on which bone is involved and the loading of the bone.⁹ Sites loaded in tension have increased risk of nonunion, progression to complete fracture, and reoccurrence compared with sites loaded in compression.⁹ Stress fractures of the fibula occur rarely, and when present, they are more commonly observed in the distal fibula in athletes and military recruits.1 Stress fractures occur rarely in patients with primary OA, and when present in this setting, obesity and malalignment are the contributing factors.³ Neither patient was obese in our case (body mass index of 27 and 28, respectively), but significant varus deformity was present in both patients. Stress fractures occurring near the knee in the setting of a varus deformity generally occur on the compression side of the tibia and are symptomatic.^3-7

Regarding malalignment, Cheung and colleagues¹⁰ reported about a case of an elderly female with OA of the knee with valgus deformity that initially developed a proximal fibular stress fracture followed by a proximal tibial stress fracture. However, both of our patients had varus deformities. Mullaji and Shetty³ documented stress fractures in 34 patients with OA, a majority with varus deformities, but did not report any isolated proximal fibular stress fractures. Manish and colleagues² reported the only documented case of an isolated proximal fibular stress fracture in a patient with osteoarthritic varus deformity. The patient presented initially with pain and paresthesias of the lower thigh and leg consistent with an L5 radiculopathy. They believed that the varus deformity and the repetitive contraction of the lateral knee muscles put increased shear forces on the fibula leading to the stress fracture. Our patients did not present with any radicular symptoms, a history of acute worsening pain, or an increased activity concerning for a stress fracture. Instead, our patients presented with progressively worsening knee pain typical of severe OA and incidental findings on imaging of tension-side fibular stress fractures. An MRI with and without gadolinium confirmed the diagnosis of a healed fracture in our first patient and a healing fracture in our second patient.

CONCLUSION

Although exceedingly rare in osteoarthritic varus knees, we presented 2 cases of MRI-confirmed proximal fibular stress fractures in this report. As demonstrated, patients may present with symptoms of OA or radicular symptoms as described by Manish and colleagues.² Presentation may also include an expansile lesion on imaging, prompting a differential diagnosis that includes a neoplasm. If present in the setting of an osteoarthritic varus knee, stress fractures of the proximal fibula should heal with conservative treatment and not affect the plan or outcome of TKA.

ABSTRACT

Stress fractures are often missed, especially in unusual clinical settings. We report on 2 patients who presented to our orthopedic surgery clinic with incidental findings of asymptomatic proximal fibular tension side stress fractures in severe longstanding varus osteoarthritic knees. Initial plain films demonstrated an expansile deformity of the proximal fibular shaft, and differential diagnosis included a healed or healing fracture versus possible neoplasm. Magnetic resonance imaging with and without gadolinium was utilized to rule out the latter prior to planned total knee arthroplasty.

Continue to: The proximal fibula...

 

The proximal fibula is a rare site for stress fractures, with most of these fractures occurring in military recruits.¹ To the authors’ knowledge, there has been only 1 documented case of a proximal fibular stress fracture in patients with severe osteoarthritis (OA) and fixed varus deformity, which mimicked L5 radiculopathy.² We are not aware of any reports of asymptomatic tension-side fibular stress fractures in varus knees. In our 2 cases, the patients were indicated for total knee arthroplasty (TKA) for varus degenerative joint disease after failing nonoperative treatment; however, further work-up was justified to rule out neoplasm after plain films revealed expansile deformities of the proximal fibular shaft. Each patient subsequently underwent magnetic resonance imaging (MRI) with and without gadolinium contrast, which demonstrated a healed and healing proximal fibular stress fracture. Magnetic resonance imaging is rarely indicated in the evaluation of degenerative joint disease, and stress fractures about a varus knee generally occur on the compression side of the tibia and are symptomatic.^3-7 The patients provided informed written consent for print and electronic publication of this case report.

CASE REPORT

The first patient was a 77-year-old male who presented with longstanding knee pain, left greater than right, exacerbated by weight-bearing activities. The patient had no improvement with physical therapy or anti-inflammatory medication. He denied any history of trauma, weakness, paresthesias, or a recent increase in activity. The patient also denied any fevers, chills, night sweats, or other constitutional symptoms. On physical examination, the patient had an antalgic gait and limited range of motion bilaterally. Examination of his right lower extremity demonstrated a fixed 5° varus deformity. No distinct point tenderness was noted.

Radiographs of the right knee demonstrated varus deformity and tricompartmental degenerative changes with severe medial joint space narrowing. An expansile deformity of the proximal right fibular shaft was also noted (Figure 1), which was not present on the films 2 years earlier (Figure 2). The absence of this deformity on previous imaging raised the suspicion of a tumor. An MRI with and without gadolinium, which was obtained to rule out a neoplastic process, showed an old, healed proximal fibular shaft fracture with chronic periosteal reaction (Figure 3). There was no marrow edema to suggest acute injury and no neoplastic lesion. He was reassured regarding the benign findings and was scheduled for a left TKA, as his pain was more severe on the left knee. The patient’s stress fracture healed without complications, and he underwent a successful left TKA. He returned approximately 6 months after his procedure with worsening right knee pain and underwent a successful TKA on the right knee as well.

The second patient was a 67-year-old male with longstanding bilateral knee pain, right greater than left, with no antecedent trauma. He denied a history of increased activity, or weakness or paresthesias. He denied any fevers, chills, night sweats, or other constitutional symptoms. One year prior to presentation at our clinic, he had received corticosteroid injections and hyaluronic acid, without relief. The patient also had a history with another surgeon of arthroscopy 1 year earlier and subchondroplasty 3 years before presentation to our clinic. On physical examination, the patient’s right knee displayed a fixed 7° varus deformity with decreased range of motion, effusion, and diffuse crepitus. Further examination revealed tenderness to palpation of the proximal fibula.

Radiographs of the right knee showed degenerative joint disease with varus deformity and medial compartment joint space narrowing. They also demonstrated an expansile deformity of mixed lucency and sclerosis involving the proximal right fibular shaft (Figure 4). Although these findings appeared to be consistent with a stress fracture, their appearance was also suspicious for a neoplasm. To rule out malignancy, an MRI with and without gadolinium was obtained that revealed a healing stress fracture of the proximal fibula (Figure 5). The patient was reassured, and plans were made to proceed with a TKA. The patient’s stress fracture healed without complications, and he underwent successful right TKA. Radiographs from the patient’s 8-week follow-up showed a healed fibular stress fracture (Figure 6).

Continue to: DISCUSSION

 

 

DISCUSSION

To our knowledge, this is the first report of incidental tension-side stress fractures in varus osteoarthritic knees. Stress fractures have been classified into 2 groups, fatigue fractures and insufficiency fractures. Fatigue fractures occur when abnormal stress is applied to normal bones, and insufficiency fractures result when normal stress is applied to abnormal bones.⁸ Stress fractures can also be classified into risk categories based on which bone is involved and the loading of the bone.⁹ Sites loaded in tension have increased risk of nonunion, progression to complete fracture, and reoccurrence compared with sites loaded in compression.⁹ Stress fractures of the fibula occur rarely, and when present, they are more commonly observed in the distal fibula in athletes and military recruits.1 Stress fractures occur rarely in patients with primary OA, and when present in this setting, obesity and malalignment are the contributing factors.³ Neither patient was obese in our case (body mass index of 27 and 28, respectively), but significant varus deformity was present in both patients. Stress fractures occurring near the knee in the setting of a varus deformity generally occur on the compression side of the tibia and are symptomatic.^3-7

Regarding malalignment, Cheung and colleagues¹⁰ reported about a case of an elderly female with OA of the knee with valgus deformity that initially developed a proximal fibular stress fracture followed by a proximal tibial stress fracture. However, both of our patients had varus deformities. Mullaji and Shetty³ documented stress fractures in 34 patients with OA, a majority with varus deformities, but did not report any isolated proximal fibular stress fractures. Manish and colleagues² reported the only documented case of an isolated proximal fibular stress fracture in a patient with osteoarthritic varus deformity. The patient presented initially with pain and paresthesias of the lower thigh and leg consistent with an L5 radiculopathy. They believed that the varus deformity and the repetitive contraction of the lateral knee muscles put increased shear forces on the fibula leading to the stress fracture. Our patients did not present with any radicular symptoms, a history of acute worsening pain, or an increased activity concerning for a stress fracture. Instead, our patients presented with progressively worsening knee pain typical of severe OA and incidental findings on imaging of tension-side fibular stress fractures. An MRI with and without gadolinium confirmed the diagnosis of a healed fracture in our first patient and a healing fracture in our second patient.

CONCLUSION

Although exceedingly rare in osteoarthritic varus knees, we presented 2 cases of MRI-confirmed proximal fibular stress fractures in this report. As demonstrated, patients may present with symptoms of OA or radicular symptoms as described by Manish and colleagues.² Presentation may also include an expansile lesion on imaging, prompting a differential diagnosis that includes a neoplasm. If present in the setting of an osteoarthritic varus knee, stress fractures of the proximal fibula should heal with conservative treatment and not affect the plan or outcome of TKA.

ABSTRACT

Stress fractures are often missed, especially in unusual clinical settings. We report on 2 patients who presented to our orthopedic surgery clinic with incidental findings of asymptomatic proximal fibular tension side stress fractures in severe longstanding varus osteoarthritic knees. Initial plain films demonstrated an expansile deformity of the proximal fibular shaft, and differential diagnosis included a healed or healing fracture versus possible neoplasm. Magnetic resonance imaging with and without gadolinium was utilized to rule out the latter prior to planned total knee arthroplasty.

Continue to: The proximal fibula...

 

The proximal fibula is a rare site for stress fractures, with most of these fractures occurring in military recruits.¹ To the authors’ knowledge, there has been only 1 documented case of a proximal fibular stress fracture in patients with severe osteoarthritis (OA) and fixed varus deformity, which mimicked L5 radiculopathy.² We are not aware of any reports of asymptomatic tension-side fibular stress fractures in varus knees. In our 2 cases, the patients were indicated for total knee arthroplasty (TKA) for varus degenerative joint disease after failing nonoperative treatment; however, further work-up was justified to rule out neoplasm after plain films revealed expansile deformities of the proximal fibular shaft. Each patient subsequently underwent magnetic resonance imaging (MRI) with and without gadolinium contrast, which demonstrated a healed and healing proximal fibular stress fracture. Magnetic resonance imaging is rarely indicated in the evaluation of degenerative joint disease, and stress fractures about a varus knee generally occur on the compression side of the tibia and are symptomatic.^3-7 The patients provided informed written consent for print and electronic publication of this case report.

CASE REPORT

The first patient was a 77-year-old male who presented with longstanding knee pain, left greater than right, exacerbated by weight-bearing activities. The patient had no improvement with physical therapy or anti-inflammatory medication. He denied any history of trauma, weakness, paresthesias, or a recent increase in activity. The patient also denied any fevers, chills, night sweats, or other constitutional symptoms. On physical examination, the patient had an antalgic gait and limited range of motion bilaterally. Examination of his right lower extremity demonstrated a fixed 5° varus deformity. No distinct point tenderness was noted.

Radiographs of the right knee demonstrated varus deformity and tricompartmental degenerative changes with severe medial joint space narrowing. An expansile deformity of the proximal right fibular shaft was also noted (Figure 1), which was not present on the films 2 years earlier (Figure 2). The absence of this deformity on previous imaging raised the suspicion of a tumor. An MRI with and without gadolinium, which was obtained to rule out a neoplastic process, showed an old, healed proximal fibular shaft fracture with chronic periosteal reaction (Figure 3). There was no marrow edema to suggest acute injury and no neoplastic lesion. He was reassured regarding the benign findings and was scheduled for a left TKA, as his pain was more severe on the left knee. The patient’s stress fracture healed without complications, and he underwent a successful left TKA. He returned approximately 6 months after his procedure with worsening right knee pain and underwent a successful TKA on the right knee as well.

The second patient was a 67-year-old male with longstanding bilateral knee pain, right greater than left, with no antecedent trauma. He denied a history of increased activity, or weakness or paresthesias. He denied any fevers, chills, night sweats, or other constitutional symptoms. One year prior to presentation at our clinic, he had received corticosteroid injections and hyaluronic acid, without relief. The patient also had a history with another surgeon of arthroscopy 1 year earlier and subchondroplasty 3 years before presentation to our clinic. On physical examination, the patient’s right knee displayed a fixed 7° varus deformity with decreased range of motion, effusion, and diffuse crepitus. Further examination revealed tenderness to palpation of the proximal fibula.

Radiographs of the right knee showed degenerative joint disease with varus deformity and medial compartment joint space narrowing. They also demonstrated an expansile deformity of mixed lucency and sclerosis involving the proximal right fibular shaft (Figure 4). Although these findings appeared to be consistent with a stress fracture, their appearance was also suspicious for a neoplasm. To rule out malignancy, an MRI with and without gadolinium was obtained that revealed a healing stress fracture of the proximal fibula (Figure 5). The patient was reassured, and plans were made to proceed with a TKA. The patient’s stress fracture healed without complications, and he underwent successful right TKA. Radiographs from the patient’s 8-week follow-up showed a healed fibular stress fracture (Figure 6).

Continue to: DISCUSSION

 

 

DISCUSSION

To our knowledge, this is the first report of incidental tension-side stress fractures in varus osteoarthritic knees. Stress fractures have been classified into 2 groups, fatigue fractures and insufficiency fractures. Fatigue fractures occur when abnormal stress is applied to normal bones, and insufficiency fractures result when normal stress is applied to abnormal bones.⁸ Stress fractures can also be classified into risk categories based on which bone is involved and the loading of the bone.⁹ Sites loaded in tension have increased risk of nonunion, progression to complete fracture, and reoccurrence compared with sites loaded in compression.⁹ Stress fractures of the fibula occur rarely, and when present, they are more commonly observed in the distal fibula in athletes and military recruits.1 Stress fractures occur rarely in patients with primary OA, and when present in this setting, obesity and malalignment are the contributing factors.³ Neither patient was obese in our case (body mass index of 27 and 28, respectively), but significant varus deformity was present in both patients. Stress fractures occurring near the knee in the setting of a varus deformity generally occur on the compression side of the tibia and are symptomatic.^3-7

Regarding malalignment, Cheung and colleagues¹⁰ reported about a case of an elderly female with OA of the knee with valgus deformity that initially developed a proximal fibular stress fracture followed by a proximal tibial stress fracture. However, both of our patients had varus deformities. Mullaji and Shetty³ documented stress fractures in 34 patients with OA, a majority with varus deformities, but did not report any isolated proximal fibular stress fractures. Manish and colleagues² reported the only documented case of an isolated proximal fibular stress fracture in a patient with osteoarthritic varus deformity. The patient presented initially with pain and paresthesias of the lower thigh and leg consistent with an L5 radiculopathy. They believed that the varus deformity and the repetitive contraction of the lateral knee muscles put increased shear forces on the fibula leading to the stress fracture. Our patients did not present with any radicular symptoms, a history of acute worsening pain, or an increased activity concerning for a stress fracture. Instead, our patients presented with progressively worsening knee pain typical of severe OA and incidental findings on imaging of tension-side fibular stress fractures. An MRI with and without gadolinium confirmed the diagnosis of a healed fracture in our first patient and a healing fracture in our second patient.

CONCLUSION

Although exceedingly rare in osteoarthritic varus knees, we presented 2 cases of MRI-confirmed proximal fibular stress fractures in this report. As demonstrated, patients may present with symptoms of OA or radicular symptoms as described by Manish and colleagues.² Presentation may also include an expansile lesion on imaging, prompting a differential diagnosis that includes a neoplasm. If present in the setting of an osteoarthritic varus knee, stress fractures of the proximal fibula should heal with conservative treatment and not affect the plan or outcome of TKA.

Perniosis, or chilblain, is characterized by localized, tender, erythematous skin lesions that occur as an abnormal reaction to exposure to cold and damp conditions. Although the lesions favor the distal extremities, perniosis may present anywhere on the body. Lesions can develop within hours to days following exposure to temperature less than 10°C or damp environments with greater than 60% humidity.¹ Acute cases may lead to pruritus and tenderness, whereas chronic cases may involve lesions that blister or ulcerate and can take weeks to heal. We report an unusual case of erythematous plaques arising on the buttocks of a 73-year-old woman using ice pack treatments for chronic low back pain.

Case Report

A 73-year-old woman presented with recurrent tender lesions on the buttocks of 5 years’ duration. Her medical history was remarkable for hypertension, hypothyroidism, and lumbar spinal fusion surgery 5 years prior. Physical examination revealed indurated erythematous plaques with areas of erosions on the left buttock with some involvement of the right buttock (Figure 1).

After a trial of oral valacyclovir for presumed herpes simplex infection provided no relief, a punch biopsy of the left buttock was performed, which revealed a cell-poor interface dermatitis with superficial and deep perivascular and periadnexal lymphocytic infiltrates (Figure 2). Perieccrine lymphocytes were present in a small portion of the reticular dermis (Figure 3). The patient revealed she had been sitting on ice packs for several hours daily since the lumbar spinal fusion surgery 5 years prior to alleviate chronic low back pain.

Based on the clinicopathologic correlation, a diagnosis of perniosis secondary to ice pack therapy was made. An evaluation for concomitant or underlying connective tissue disease (CTD) including a complete blood cell count with sedimentation rate, antinuclear antibodies (ANAs), serum protein electrophoresis, and serum levels of cryoglobulins and complement components was unremarkable. Our patient was treated with simple analgesia and was encouraged to avoid direct contact with ice packs for extended periods of time. Because of her low back pain, she continued to use ice packs but readjusted them sporadically and decreased frequency of use. She had complete resolution of the lesions at 6-month follow-up.

 

Comment

Perniosis is a self-limited condition, manifesting as erythematous plaques or nodules following exposure to cold and damp conditions. It was first reported in 1902 by Hochsinger² as tender submental plaques occurring in children after exposure to cold weather. Since then, reports of perniosis have been described in equestrians and long-distance cyclists as well as in the context of other outdoor activities.^3-5 In all cases, patients developed perniosis at sites of exposure to cold or damp conditions.

Perniosis arising in patients using ice pack therapy is a rare and recent phenomenon, with only 3 other known reported cases.^6,7 In all cases, including ours, patients reported treating chronic low back pain with ice packs for more than 2 hours per day. Clinical presentations included erythematous to purpuric plaques with ulceration on the lower back or buttocks that reoccurred with subsequent use of ice packs. No concomitant CTD was reported.⁶

Much controversy exists as to whether idiopathic perniosis (IP) increases susceptibility to acquiring an autoimmune disease or if IP is a form of CTD that follows a more indolent course.⁸ In a prospective study of 33 patients with underlying IP, no patients developed lupus erythematosus (LE), with a median follow-up of 38 months.⁹ A study by Crowson and Magro⁸ revealed that 18 of 39 patients with perniotic lesions had an associated systemic disease including LE, human immunodeficiency virus, viral hepatitis, rheumatoid arthritis, cryofibrinogenemia, hypergammaglobulinemia, iritis, or Crohn disease. Of the 21 other patients who had no underlying CTD or systemic disease, 10 had a positive ANA test but no systemic symptoms; therefore, all 21 of these patients were classified as cases of IP.⁸

Cutaneous biopsy to distinguish between IP and autoimmune perniosis remains controversial; perniotic lesions and discoid LE share histopathologic features,⁹ as was evident with our case, which demonstrated overlapping findings of vacuolar change with superficial and deep perivascular and periadnexal lymphoid infiltrates. Typical features of IP include thrombosed capillaries in the papillary dermis and lymphocytic exocytosis localized to the acrosyringia, whereas secondary perniosis has superficial and deep perivascular and perieccrine lymphocytic infiltrates with vascular thrombosis in the reticular dermis. Vascular ectasia, dermal mucinosis, basement membrane zone thickening, and erythrocyte extravasation are not reliable and may be seen in both cases.⁸ One study revealed the only significant difference between both entities was the perieccrine distribution of lymphocytic infiltrate in cases of IP (P=.007), whereas an absence of perieccrine involvement was noted in autoimmune cases.⁹

Direct immunofluorescence (DIF) may help differentiate IP from autoimmune perniosis. In a prospective study by Viguier et al,⁹ 6 of 9 patients with IP had negative DIF and 3 had slight nonspecific C3 immunoreactivity of dermal vessels. Conversely, in patients with autoimmune perniosis, positive DIF with the lupus band test was seen in 3 of 7 patients, all who had a positive ANA test⁹; however, positive ANA levels also were reported in patients with autoimmune perniosis but negative DIF, suggesting that DIF lacks specificity in diagnosing autoimmune perniosis.

Although histopathologic findings bear similarities to LE, there are no guidelines to suggest for or against laboratory testing for CTD in patients presenting with perniosis. Some investigators have suggested that any patient with clinical features suggestive of perniosis should undergo laboratory evaluation including a complete blood cell count and assessment for antibodies to Ro, ANA, rheumatoid factor, cryofibrinogens, and antiphospholipid antibodies.⁹ Serum protein electrophoresis and immunofixation electrophoresis may be done to exclude monoclonal gammopathy.

For idiopathic cases, treatment is aimed at limiting or removing cold exposure. Patients should be advised regarding the use of long-term ice pack use and the potential development of perniosis. For chronic perniosis lasting beyond several weeks, a combination of a slow taper of oral prednisone, hydroxychloroquine, and quinacrine has been successful in patients with persistent lesions despite making environmental modifications.³ Intralesional triamcinolone acetonide and nifedipine also have been effective in perniotic hand lesions.¹⁰

Conclusion

We report a rare case of perniosis on the buttocks that arose in a patient who utilized ice packs for treatment of chronic low back pain. Ice pack–induced perniosis may be an underreported entity. Histopathologic examination is nondescript, as overlapping features of perniosis and LE have been observed with no underlying CTD present. Correlation with patient history and clinical examination is paramount in diagnosis and management.

Perniosis, or chilblain, is characterized by localized, tender, erythematous skin lesions that occur as an abnormal reaction to exposure to cold and damp conditions. Although the lesions favor the distal extremities, perniosis may present anywhere on the body. Lesions can develop within hours to days following exposure to temperature less than 10°C or damp environments with greater than 60% humidity.¹ Acute cases may lead to pruritus and tenderness, whereas chronic cases may involve lesions that blister or ulcerate and can take weeks to heal. We report an unusual case of erythematous plaques arising on the buttocks of a 73-year-old woman using ice pack treatments for chronic low back pain.

Case Report

A 73-year-old woman presented with recurrent tender lesions on the buttocks of 5 years’ duration. Her medical history was remarkable for hypertension, hypothyroidism, and lumbar spinal fusion surgery 5 years prior. Physical examination revealed indurated erythematous plaques with areas of erosions on the left buttock with some involvement of the right buttock (Figure 1).

After a trial of oral valacyclovir for presumed herpes simplex infection provided no relief, a punch biopsy of the left buttock was performed, which revealed a cell-poor interface dermatitis with superficial and deep perivascular and periadnexal lymphocytic infiltrates (Figure 2). Perieccrine lymphocytes were present in a small portion of the reticular dermis (Figure 3). The patient revealed she had been sitting on ice packs for several hours daily since the lumbar spinal fusion surgery 5 years prior to alleviate chronic low back pain.

Based on the clinicopathologic correlation, a diagnosis of perniosis secondary to ice pack therapy was made. An evaluation for concomitant or underlying connective tissue disease (CTD) including a complete blood cell count with sedimentation rate, antinuclear antibodies (ANAs), serum protein electrophoresis, and serum levels of cryoglobulins and complement components was unremarkable. Our patient was treated with simple analgesia and was encouraged to avoid direct contact with ice packs for extended periods of time. Because of her low back pain, she continued to use ice packs but readjusted them sporadically and decreased frequency of use. She had complete resolution of the lesions at 6-month follow-up.

 

Comment

Perniosis is a self-limited condition, manifesting as erythematous plaques or nodules following exposure to cold and damp conditions. It was first reported in 1902 by Hochsinger² as tender submental plaques occurring in children after exposure to cold weather. Since then, reports of perniosis have been described in equestrians and long-distance cyclists as well as in the context of other outdoor activities.^3-5 In all cases, patients developed perniosis at sites of exposure to cold or damp conditions.

Perniosis arising in patients using ice pack therapy is a rare and recent phenomenon, with only 3 other known reported cases.^6,7 In all cases, including ours, patients reported treating chronic low back pain with ice packs for more than 2 hours per day. Clinical presentations included erythematous to purpuric plaques with ulceration on the lower back or buttocks that reoccurred with subsequent use of ice packs. No concomitant CTD was reported.⁶

Much controversy exists as to whether idiopathic perniosis (IP) increases susceptibility to acquiring an autoimmune disease or if IP is a form of CTD that follows a more indolent course.⁸ In a prospective study of 33 patients with underlying IP, no patients developed lupus erythematosus (LE), with a median follow-up of 38 months.⁹ A study by Crowson and Magro⁸ revealed that 18 of 39 patients with perniotic lesions had an associated systemic disease including LE, human immunodeficiency virus, viral hepatitis, rheumatoid arthritis, cryofibrinogenemia, hypergammaglobulinemia, iritis, or Crohn disease. Of the 21 other patients who had no underlying CTD or systemic disease, 10 had a positive ANA test but no systemic symptoms; therefore, all 21 of these patients were classified as cases of IP.⁸

Cutaneous biopsy to distinguish between IP and autoimmune perniosis remains controversial; perniotic lesions and discoid LE share histopathologic features,⁹ as was evident with our case, which demonstrated overlapping findings of vacuolar change with superficial and deep perivascular and periadnexal lymphoid infiltrates. Typical features of IP include thrombosed capillaries in the papillary dermis and lymphocytic exocytosis localized to the acrosyringia, whereas secondary perniosis has superficial and deep perivascular and perieccrine lymphocytic infiltrates with vascular thrombosis in the reticular dermis. Vascular ectasia, dermal mucinosis, basement membrane zone thickening, and erythrocyte extravasation are not reliable and may be seen in both cases.⁸ One study revealed the only significant difference between both entities was the perieccrine distribution of lymphocytic infiltrate in cases of IP (P=.007), whereas an absence of perieccrine involvement was noted in autoimmune cases.⁹

Direct immunofluorescence (DIF) may help differentiate IP from autoimmune perniosis. In a prospective study by Viguier et al,⁹ 6 of 9 patients with IP had negative DIF and 3 had slight nonspecific C3 immunoreactivity of dermal vessels. Conversely, in patients with autoimmune perniosis, positive DIF with the lupus band test was seen in 3 of 7 patients, all who had a positive ANA test⁹; however, positive ANA levels also were reported in patients with autoimmune perniosis but negative DIF, suggesting that DIF lacks specificity in diagnosing autoimmune perniosis.

Although histopathologic findings bear similarities to LE, there are no guidelines to suggest for or against laboratory testing for CTD in patients presenting with perniosis. Some investigators have suggested that any patient with clinical features suggestive of perniosis should undergo laboratory evaluation including a complete blood cell count and assessment for antibodies to Ro, ANA, rheumatoid factor, cryofibrinogens, and antiphospholipid antibodies.⁹ Serum protein electrophoresis and immunofixation electrophoresis may be done to exclude monoclonal gammopathy.

For idiopathic cases, treatment is aimed at limiting or removing cold exposure. Patients should be advised regarding the use of long-term ice pack use and the potential development of perniosis. For chronic perniosis lasting beyond several weeks, a combination of a slow taper of oral prednisone, hydroxychloroquine, and quinacrine has been successful in patients with persistent lesions despite making environmental modifications.³ Intralesional triamcinolone acetonide and nifedipine also have been effective in perniotic hand lesions.¹⁰

Conclusion

We report a rare case of perniosis on the buttocks that arose in a patient who utilized ice packs for treatment of chronic low back pain. Ice pack–induced perniosis may be an underreported entity. Histopathologic examination is nondescript, as overlapping features of perniosis and LE have been observed with no underlying CTD present. Correlation with patient history and clinical examination is paramount in diagnosis and management.

Perniosis, or chilblain, is characterized by localized, tender, erythematous skin lesions that occur as an abnormal reaction to exposure to cold and damp conditions. Although the lesions favor the distal extremities, perniosis may present anywhere on the body. Lesions can develop within hours to days following exposure to temperature less than 10°C or damp environments with greater than 60% humidity.¹ Acute cases may lead to pruritus and tenderness, whereas chronic cases may involve lesions that blister or ulcerate and can take weeks to heal. We report an unusual case of erythematous plaques arising on the buttocks of a 73-year-old woman using ice pack treatments for chronic low back pain.

Case Report

A 73-year-old woman presented with recurrent tender lesions on the buttocks of 5 years’ duration. Her medical history was remarkable for hypertension, hypothyroidism, and lumbar spinal fusion surgery 5 years prior. Physical examination revealed indurated erythematous plaques with areas of erosions on the left buttock with some involvement of the right buttock (Figure 1).

After a trial of oral valacyclovir for presumed herpes simplex infection provided no relief, a punch biopsy of the left buttock was performed, which revealed a cell-poor interface dermatitis with superficial and deep perivascular and periadnexal lymphocytic infiltrates (Figure 2). Perieccrine lymphocytes were present in a small portion of the reticular dermis (Figure 3). The patient revealed she had been sitting on ice packs for several hours daily since the lumbar spinal fusion surgery 5 years prior to alleviate chronic low back pain.

Based on the clinicopathologic correlation, a diagnosis of perniosis secondary to ice pack therapy was made. An evaluation for concomitant or underlying connective tissue disease (CTD) including a complete blood cell count with sedimentation rate, antinuclear antibodies (ANAs), serum protein electrophoresis, and serum levels of cryoglobulins and complement components was unremarkable. Our patient was treated with simple analgesia and was encouraged to avoid direct contact with ice packs for extended periods of time. Because of her low back pain, she continued to use ice packs but readjusted them sporadically and decreased frequency of use. She had complete resolution of the lesions at 6-month follow-up.

 

Comment

Perniosis is a self-limited condition, manifesting as erythematous plaques or nodules following exposure to cold and damp conditions. It was first reported in 1902 by Hochsinger² as tender submental plaques occurring in children after exposure to cold weather. Since then, reports of perniosis have been described in equestrians and long-distance cyclists as well as in the context of other outdoor activities.^3-5 In all cases, patients developed perniosis at sites of exposure to cold or damp conditions.

Perniosis arising in patients using ice pack therapy is a rare and recent phenomenon, with only 3 other known reported cases.^6,7 In all cases, including ours, patients reported treating chronic low back pain with ice packs for more than 2 hours per day. Clinical presentations included erythematous to purpuric plaques with ulceration on the lower back or buttocks that reoccurred with subsequent use of ice packs. No concomitant CTD was reported.⁶

Much controversy exists as to whether idiopathic perniosis (IP) increases susceptibility to acquiring an autoimmune disease or if IP is a form of CTD that follows a more indolent course.⁸ In a prospective study of 33 patients with underlying IP, no patients developed lupus erythematosus (LE), with a median follow-up of 38 months.⁹ A study by Crowson and Magro⁸ revealed that 18 of 39 patients with perniotic lesions had an associated systemic disease including LE, human immunodeficiency virus, viral hepatitis, rheumatoid arthritis, cryofibrinogenemia, hypergammaglobulinemia, iritis, or Crohn disease. Of the 21 other patients who had no underlying CTD or systemic disease, 10 had a positive ANA test but no systemic symptoms; therefore, all 21 of these patients were classified as cases of IP.⁸

Cutaneous biopsy to distinguish between IP and autoimmune perniosis remains controversial; perniotic lesions and discoid LE share histopathologic features,⁹ as was evident with our case, which demonstrated overlapping findings of vacuolar change with superficial and deep perivascular and periadnexal lymphoid infiltrates. Typical features of IP include thrombosed capillaries in the papillary dermis and lymphocytic exocytosis localized to the acrosyringia, whereas secondary perniosis has superficial and deep perivascular and perieccrine lymphocytic infiltrates with vascular thrombosis in the reticular dermis. Vascular ectasia, dermal mucinosis, basement membrane zone thickening, and erythrocyte extravasation are not reliable and may be seen in both cases.⁸ One study revealed the only significant difference between both entities was the perieccrine distribution of lymphocytic infiltrate in cases of IP (P=.007), whereas an absence of perieccrine involvement was noted in autoimmune cases.⁹

Direct immunofluorescence (DIF) may help differentiate IP from autoimmune perniosis. In a prospective study by Viguier et al,⁹ 6 of 9 patients with IP had negative DIF and 3 had slight nonspecific C3 immunoreactivity of dermal vessels. Conversely, in patients with autoimmune perniosis, positive DIF with the lupus band test was seen in 3 of 7 patients, all who had a positive ANA test⁹; however, positive ANA levels also were reported in patients with autoimmune perniosis but negative DIF, suggesting that DIF lacks specificity in diagnosing autoimmune perniosis.

Although histopathologic findings bear similarities to LE, there are no guidelines to suggest for or against laboratory testing for CTD in patients presenting with perniosis. Some investigators have suggested that any patient with clinical features suggestive of perniosis should undergo laboratory evaluation including a complete blood cell count and assessment for antibodies to Ro, ANA, rheumatoid factor, cryofibrinogens, and antiphospholipid antibodies.⁹ Serum protein electrophoresis and immunofixation electrophoresis may be done to exclude monoclonal gammopathy.

For idiopathic cases, treatment is aimed at limiting or removing cold exposure. Patients should be advised regarding the use of long-term ice pack use and the potential development of perniosis. For chronic perniosis lasting beyond several weeks, a combination of a slow taper of oral prednisone, hydroxychloroquine, and quinacrine has been successful in patients with persistent lesions despite making environmental modifications.³ Intralesional triamcinolone acetonide and nifedipine also have been effective in perniotic hand lesions.¹⁰

Conclusion

We report a rare case of perniosis on the buttocks that arose in a patient who utilized ice packs for treatment of chronic low back pain. Ice pack–induced perniosis may be an underreported entity. Histopathologic examination is nondescript, as overlapping features of perniosis and LE have been observed with no underlying CTD present. Correlation with patient history and clinical examination is paramount in diagnosis and management.

THE CASE

A 33-year-old Caucasian woman presented to the emergency department with a 6-day history of fever (103°-104°F) and right groin pain and swelling. Associated symptoms included headache, diarrhea, malaise, weakness, nausea, cough, and anorexia. Upon presentation, she admitted to a recent hike on a bubonic plague–endemic trail in Colorado.

Her vital signs were unremarkable, and the physical examination demonstrated normal findings except for tender, erythematous, nonfluctuant right inguinal lymphadenopathy. The patient was admitted for intractable pain and fever and started on intravenous cefoxitin 2 g IV every 8 hours and oral doxycycline 100 mg every 12 hours for pelvic inflammatory disease vs tick- or flea-borne illness. Due to the patient’s recent trip to a plague-infested area, our suspicion for Yersinia pestis infection was high.

The patient’s work-up included a negative pregnancy test and urinalysis. A complete blood count demonstrated a white blood cell count of 8.6 (4.3-10.5) × 10³/UL with a 3+ left shift and a platelet count of 112 (180-500) × 10³/UL. A complete metabolic panel showed hypokalemia and hyponatremia (potassium 2.8 [3.5-5.1] mmol/L and sodium 134 [137-145] mmol/L). Blood cultures were negative for any bacterial or fungal growth after 48 hours; stool cultures were negative for Salmonella, Shigella, Campylobacter, Giardia, generalized Yersinia, and Escherichia coli O157:H7. Swabs for Gardnerella vaginalis, Trichomonas vaginalis, Candida, Chlamydia trachomatis, and Neisseria gonorrhea also were negative. Lyme, Bartonella henselae, and heterophile antibodies were also negative. Francisella tularensis was not cultured due to low suspicion.

Imaging included a normal chest x-ray and a computed tomography scan of the abdomen and pelvis that showed enlarged right inguinal lymph nodes with fatty stranding, a thicker distal right iliopsoas, hepatosplenomegaly, and an enlarged right adnexa (FIGURE 1). Initial ultrasound of the bubo showed 2 enlarged suprapubic lymph nodes, the largest measuring 3.5 × 1.4 × 2.4 cm³ (FIGURE 2), and 8 enlarged inguinal nodes.

The patient continued to have a low-grade fever, diarrhea, and inguinal lymphadenopathy throughout her first 2 hospitalized days. The cefoxitin was discontinued by Day 3, and the consulting infectious disease physician started oral metronidazole 500 mg every 12 hours due to the patient’s failure to improve. Later that night, the patient experienced increasing erythema and pain in her right inguinal region. A repeat ultrasound showed increased inguinal lymphadenopathy with the largest nodes measuring 2.9 × 1.5 × 2.5 cm³ and 2.7 × 1.3 × 2 cm³ (FIGURE 3).

Although doxycycline is considered an acceptable regimen for Y pestis infection, the infectious disease physician added oral ciprofloxacin 750 mg every 12 hours the following morning, as the patient had not improved.

THE DIAGNOSIS

Although the initial gram stain was negative for Yersinia, clinical suspicion pointed to a diagnosis of bubonic plague. Serology was considered; however, it was not available through the hospital. A definitive diagnosis required bubo aspiration and culture, which was performed but required 48 hours before results would be available.

Continue to: By Day 5, the patient was clinically improved and...

 

By Day 5, the patient was clinically improved and deemed safe for discharge on empiric treatment with ciprofloxacin 750 mg twice daily and doxycycline 100 mg twice daily to complete a 14-day course of antibiotic therapy for bubonic plague. The bubo culture subsequently grew Y pestis, confirming the diagnosis. The patient made a full recovery and was greatly improved when seen in the outpatient setting by the treating infectious disease physician. Outpatient ultrasound repeated 3 weeks after discharge showed borderline lymphadenopathy, no greater than 1 cm.

DISCUSSION

Between 2000 and 2009, there were 57 cases of Y pestis in the United States; in early 2015, 11 cases were found in 6 Western states.¹ The plague presents in the bubonic form 80% to 95% of the time, and it has never been reported in Michigan (where we treated this patient); however, there was a laboratory case in Illinois. Although rats were traditionally the host for Y pestis, the prairie dog, Cynomys gunnisoni, is a host in the United States.² Rodents are the most important hosts, but more than 200 mammalian species, including domestic pets, have had reported infections. Transmission is primarily via flea bites, but Y pestis also may be transmitted via respiratory secretion, inhalation, or direct handling of contaminated animal tissues. Due to the risk of respiratory spread, the Centers for Disease Control and Prevention must be notified of a diagnosis.^3,4

Y pestis travels from the site of the flea bite to regional lymph nodes, where it reproduces, and the resultant inflammatory reaction creates buboes. The bacteria then circulate in the blood to other organs, although Y pestis bacteria are primarily removed by the liver and spleen. Patients often develop symptoms such as headache, fevers, chills, and gastrointestinal distress. Diagnosis is reached by bubo culture or rapid testing for the F1 antigen. Early intervention with antibiotics is crucial as untreated bubonic plague has a mortality rate of 50% to 90%.^3,4

The differential diagnosis for unilateral inguinal lymphadenopathy with associated constitutional symptoms was broad, in this case, and included pelvic inflammatory disease, bubonic plague, iliopsoas abscess, lymphogranuloma venereum, bartonellosis, infectious mononucleosis, and tick-borne diseases, such as ehrlichiosis, tularemia, Lyme disease, Rocky Mountain spotted fever, and Colorado tick fever.

Although the initial gram stain demonstrated no organisms, a subsequent bubo aspiration and culture grew Y pestis, confirming the diagnosis.

Treatment. Food and Drug Administration–approved treatments include streptomycin (gentamicin 5 mg/kg/day IM or IV for 14 days is more widely utilized), doxycycline 200 mg PO once daily for 10 to 14 days, and fluoroquinolones (ciprofloxacin 500-750 mg every 12 hours for 10-14 days). Trimethoprim-sulfamethoxazole may be used as an alternative, but limitations include potentially incomplete or slowed responses.

Continue to: THE TAKEAWAY

 

 

THE TAKEAWAY

This case points to the importance of a complete, systematic approach to each patient. While bubonic plague is not a diagnosis that would immediately come to mind in a patient visiting an emergency department in Michigan, a thorough history revealed a recent trip to a bubonic plague–endemic area. A thorough physical exam demonstrated unilateral painful inguinal adenopathy—which, when paired with the patient’s history—was consistent with the uncommon diagnosis of bubonic plague.

ACKNOWLEDGEMENTS
The authors thank Brian Waite, MD, and James Addison, MD, for critically revising this report for important intellectual content.

CORRESPONDENCE
Katherine Lazet, DO, 3838 N First Avenue, Evansville, IN 47710; [email protected]

THE CASE

A 33-year-old Caucasian woman presented to the emergency department with a 6-day history of fever (103°-104°F) and right groin pain and swelling. Associated symptoms included headache, diarrhea, malaise, weakness, nausea, cough, and anorexia. Upon presentation, she admitted to a recent hike on a bubonic plague–endemic trail in Colorado.

Her vital signs were unremarkable, and the physical examination demonstrated normal findings except for tender, erythematous, nonfluctuant right inguinal lymphadenopathy. The patient was admitted for intractable pain and fever and started on intravenous cefoxitin 2 g IV every 8 hours and oral doxycycline 100 mg every 12 hours for pelvic inflammatory disease vs tick- or flea-borne illness. Due to the patient’s recent trip to a plague-infested area, our suspicion for Yersinia pestis infection was high.

The patient’s work-up included a negative pregnancy test and urinalysis. A complete blood count demonstrated a white blood cell count of 8.6 (4.3-10.5) × 10³/UL with a 3+ left shift and a platelet count of 112 (180-500) × 10³/UL. A complete metabolic panel showed hypokalemia and hyponatremia (potassium 2.8 [3.5-5.1] mmol/L and sodium 134 [137-145] mmol/L). Blood cultures were negative for any bacterial or fungal growth after 48 hours; stool cultures were negative for Salmonella, Shigella, Campylobacter, Giardia, generalized Yersinia, and Escherichia coli O157:H7. Swabs for Gardnerella vaginalis, Trichomonas vaginalis, Candida, Chlamydia trachomatis, and Neisseria gonorrhea also were negative. Lyme, Bartonella henselae, and heterophile antibodies were also negative. Francisella tularensis was not cultured due to low suspicion.

Imaging included a normal chest x-ray and a computed tomography scan of the abdomen and pelvis that showed enlarged right inguinal lymph nodes with fatty stranding, a thicker distal right iliopsoas, hepatosplenomegaly, and an enlarged right adnexa (FIGURE 1). Initial ultrasound of the bubo showed 2 enlarged suprapubic lymph nodes, the largest measuring 3.5 × 1.4 × 2.4 cm³ (FIGURE 2), and 8 enlarged inguinal nodes.

The patient continued to have a low-grade fever, diarrhea, and inguinal lymphadenopathy throughout her first 2 hospitalized days. The cefoxitin was discontinued by Day 3, and the consulting infectious disease physician started oral metronidazole 500 mg every 12 hours due to the patient’s failure to improve. Later that night, the patient experienced increasing erythema and pain in her right inguinal region. A repeat ultrasound showed increased inguinal lymphadenopathy with the largest nodes measuring 2.9 × 1.5 × 2.5 cm³ and 2.7 × 1.3 × 2 cm³ (FIGURE 3).

Although doxycycline is considered an acceptable regimen for Y pestis infection, the infectious disease physician added oral ciprofloxacin 750 mg every 12 hours the following morning, as the patient had not improved.

THE DIAGNOSIS

Although the initial gram stain was negative for Yersinia, clinical suspicion pointed to a diagnosis of bubonic plague. Serology was considered; however, it was not available through the hospital. A definitive diagnosis required bubo aspiration and culture, which was performed but required 48 hours before results would be available.

Continue to: By Day 5, the patient was clinically improved and...

 

By Day 5, the patient was clinically improved and deemed safe for discharge on empiric treatment with ciprofloxacin 750 mg twice daily and doxycycline 100 mg twice daily to complete a 14-day course of antibiotic therapy for bubonic plague. The bubo culture subsequently grew Y pestis, confirming the diagnosis. The patient made a full recovery and was greatly improved when seen in the outpatient setting by the treating infectious disease physician. Outpatient ultrasound repeated 3 weeks after discharge showed borderline lymphadenopathy, no greater than 1 cm.

DISCUSSION

Between 2000 and 2009, there were 57 cases of Y pestis in the United States; in early 2015, 11 cases were found in 6 Western states.¹ The plague presents in the bubonic form 80% to 95% of the time, and it has never been reported in Michigan (where we treated this patient); however, there was a laboratory case in Illinois. Although rats were traditionally the host for Y pestis, the prairie dog, Cynomys gunnisoni, is a host in the United States.² Rodents are the most important hosts, but more than 200 mammalian species, including domestic pets, have had reported infections. Transmission is primarily via flea bites, but Y pestis also may be transmitted via respiratory secretion, inhalation, or direct handling of contaminated animal tissues. Due to the risk of respiratory spread, the Centers for Disease Control and Prevention must be notified of a diagnosis.^3,4

Y pestis travels from the site of the flea bite to regional lymph nodes, where it reproduces, and the resultant inflammatory reaction creates buboes. The bacteria then circulate in the blood to other organs, although Y pestis bacteria are primarily removed by the liver and spleen. Patients often develop symptoms such as headache, fevers, chills, and gastrointestinal distress. Diagnosis is reached by bubo culture or rapid testing for the F1 antigen. Early intervention with antibiotics is crucial as untreated bubonic plague has a mortality rate of 50% to 90%.^3,4

The differential diagnosis for unilateral inguinal lymphadenopathy with associated constitutional symptoms was broad, in this case, and included pelvic inflammatory disease, bubonic plague, iliopsoas abscess, lymphogranuloma venereum, bartonellosis, infectious mononucleosis, and tick-borne diseases, such as ehrlichiosis, tularemia, Lyme disease, Rocky Mountain spotted fever, and Colorado tick fever.

Although the initial gram stain demonstrated no organisms, a subsequent bubo aspiration and culture grew Y pestis, confirming the diagnosis.

Treatment. Food and Drug Administration–approved treatments include streptomycin (gentamicin 5 mg/kg/day IM or IV for 14 days is more widely utilized), doxycycline 200 mg PO once daily for 10 to 14 days, and fluoroquinolones (ciprofloxacin 500-750 mg every 12 hours for 10-14 days). Trimethoprim-sulfamethoxazole may be used as an alternative, but limitations include potentially incomplete or slowed responses.

Continue to: THE TAKEAWAY

 

 

THE TAKEAWAY

This case points to the importance of a complete, systematic approach to each patient. While bubonic plague is not a diagnosis that would immediately come to mind in a patient visiting an emergency department in Michigan, a thorough history revealed a recent trip to a bubonic plague–endemic area. A thorough physical exam demonstrated unilateral painful inguinal adenopathy—which, when paired with the patient’s history—was consistent with the uncommon diagnosis of bubonic plague.

ACKNOWLEDGEMENTS
The authors thank Brian Waite, MD, and James Addison, MD, for critically revising this report for important intellectual content.

CORRESPONDENCE
Katherine Lazet, DO, 3838 N First Avenue, Evansville, IN 47710; [email protected]

THE CASE

A 33-year-old Caucasian woman presented to the emergency department with a 6-day history of fever (103°-104°F) and right groin pain and swelling. Associated symptoms included headache, diarrhea, malaise, weakness, nausea, cough, and anorexia. Upon presentation, she admitted to a recent hike on a bubonic plague–endemic trail in Colorado.

Her vital signs were unremarkable, and the physical examination demonstrated normal findings except for tender, erythematous, nonfluctuant right inguinal lymphadenopathy. The patient was admitted for intractable pain and fever and started on intravenous cefoxitin 2 g IV every 8 hours and oral doxycycline 100 mg every 12 hours for pelvic inflammatory disease vs tick- or flea-borne illness. Due to the patient’s recent trip to a plague-infested area, our suspicion for Yersinia pestis infection was high.

The patient’s work-up included a negative pregnancy test and urinalysis. A complete blood count demonstrated a white blood cell count of 8.6 (4.3-10.5) × 10³/UL with a 3+ left shift and a platelet count of 112 (180-500) × 10³/UL. A complete metabolic panel showed hypokalemia and hyponatremia (potassium 2.8 [3.5-5.1] mmol/L and sodium 134 [137-145] mmol/L). Blood cultures were negative for any bacterial or fungal growth after 48 hours; stool cultures were negative for Salmonella, Shigella, Campylobacter, Giardia, generalized Yersinia, and Escherichia coli O157:H7. Swabs for Gardnerella vaginalis, Trichomonas vaginalis, Candida, Chlamydia trachomatis, and Neisseria gonorrhea also were negative. Lyme, Bartonella henselae, and heterophile antibodies were also negative. Francisella tularensis was not cultured due to low suspicion.

Imaging included a normal chest x-ray and a computed tomography scan of the abdomen and pelvis that showed enlarged right inguinal lymph nodes with fatty stranding, a thicker distal right iliopsoas, hepatosplenomegaly, and an enlarged right adnexa (FIGURE 1). Initial ultrasound of the bubo showed 2 enlarged suprapubic lymph nodes, the largest measuring 3.5 × 1.4 × 2.4 cm³ (FIGURE 2), and 8 enlarged inguinal nodes.

The patient continued to have a low-grade fever, diarrhea, and inguinal lymphadenopathy throughout her first 2 hospitalized days. The cefoxitin was discontinued by Day 3, and the consulting infectious disease physician started oral metronidazole 500 mg every 12 hours due to the patient’s failure to improve. Later that night, the patient experienced increasing erythema and pain in her right inguinal region. A repeat ultrasound showed increased inguinal lymphadenopathy with the largest nodes measuring 2.9 × 1.5 × 2.5 cm³ and 2.7 × 1.3 × 2 cm³ (FIGURE 3).

Although doxycycline is considered an acceptable regimen for Y pestis infection, the infectious disease physician added oral ciprofloxacin 750 mg every 12 hours the following morning, as the patient had not improved.

THE DIAGNOSIS

Although the initial gram stain was negative for Yersinia, clinical suspicion pointed to a diagnosis of bubonic plague. Serology was considered; however, it was not available through the hospital. A definitive diagnosis required bubo aspiration and culture, which was performed but required 48 hours before results would be available.

Continue to: By Day 5, the patient was clinically improved and...

 

By Day 5, the patient was clinically improved and deemed safe for discharge on empiric treatment with ciprofloxacin 750 mg twice daily and doxycycline 100 mg twice daily to complete a 14-day course of antibiotic therapy for bubonic plague. The bubo culture subsequently grew Y pestis, confirming the diagnosis. The patient made a full recovery and was greatly improved when seen in the outpatient setting by the treating infectious disease physician. Outpatient ultrasound repeated 3 weeks after discharge showed borderline lymphadenopathy, no greater than 1 cm.

DISCUSSION

Between 2000 and 2009, there were 57 cases of Y pestis in the United States; in early 2015, 11 cases were found in 6 Western states.¹ The plague presents in the bubonic form 80% to 95% of the time, and it has never been reported in Michigan (where we treated this patient); however, there was a laboratory case in Illinois. Although rats were traditionally the host for Y pestis, the prairie dog, Cynomys gunnisoni, is a host in the United States.² Rodents are the most important hosts, but more than 200 mammalian species, including domestic pets, have had reported infections. Transmission is primarily via flea bites, but Y pestis also may be transmitted via respiratory secretion, inhalation, or direct handling of contaminated animal tissues. Due to the risk of respiratory spread, the Centers for Disease Control and Prevention must be notified of a diagnosis.^3,4

Y pestis travels from the site of the flea bite to regional lymph nodes, where it reproduces, and the resultant inflammatory reaction creates buboes. The bacteria then circulate in the blood to other organs, although Y pestis bacteria are primarily removed by the liver and spleen. Patients often develop symptoms such as headache, fevers, chills, and gastrointestinal distress. Diagnosis is reached by bubo culture or rapid testing for the F1 antigen. Early intervention with antibiotics is crucial as untreated bubonic plague has a mortality rate of 50% to 90%.^3,4

The differential diagnosis for unilateral inguinal lymphadenopathy with associated constitutional symptoms was broad, in this case, and included pelvic inflammatory disease, bubonic plague, iliopsoas abscess, lymphogranuloma venereum, bartonellosis, infectious mononucleosis, and tick-borne diseases, such as ehrlichiosis, tularemia, Lyme disease, Rocky Mountain spotted fever, and Colorado tick fever.

Although the initial gram stain demonstrated no organisms, a subsequent bubo aspiration and culture grew Y pestis, confirming the diagnosis.

Treatment. Food and Drug Administration–approved treatments include streptomycin (gentamicin 5 mg/kg/day IM or IV for 14 days is more widely utilized), doxycycline 200 mg PO once daily for 10 to 14 days, and fluoroquinolones (ciprofloxacin 500-750 mg every 12 hours for 10-14 days). Trimethoprim-sulfamethoxazole may be used as an alternative, but limitations include potentially incomplete or slowed responses.

Continue to: THE TAKEAWAY

 

 

THE TAKEAWAY

This case points to the importance of a complete, systematic approach to each patient. While bubonic plague is not a diagnosis that would immediately come to mind in a patient visiting an emergency department in Michigan, a thorough history revealed a recent trip to a bubonic plague–endemic area. A thorough physical exam demonstrated unilateral painful inguinal adenopathy—which, when paired with the patient’s history—was consistent with the uncommon diagnosis of bubonic plague.

ACKNOWLEDGEMENTS
The authors thank Brian Waite, MD, and James Addison, MD, for critically revising this report for important intellectual content.

CORRESPONDENCE
Katherine Lazet, DO, 3838 N First Avenue, Evansville, IN 47710; [email protected]

Small bowel obstruction (SBO) accounts for 2% of all cases of abdominal pain presenting to the ED and 15% of abdominal pain admissions to surgical units from the ED.^1,2 SBO can be a difficult diagnosis; the most common symptoms include nausea, vomiting, abdominal pain, obstipation, and constipation. The symptomatology depends on multiple factors: the area of the blockage, length of obstruction, and degree of the obstruction (either partial or complete).³ An upper gastrointestinal (GI) blockage classically presents with nausea and vomiting, while a lower GI blockage often presents with abdominal pain, constipation, and obstipation. Complications of obstruction range from significant morbidity—such as bowel strangulation (23%) and sepsis (31%)—to mortality (9%).⁴ ED POCUS allows for rapid and accurate diagnosis of SBO.

CASE

A 60-year-old female with a past medical history of peptic ulcer disease and multiple abdominal surgeries, including umbilical hernia repair, appendectomy, and total abdominal hysterectomy, presented to the ED with an 8-hour history of nausea and vomiting. She reported that her abdomen felt bloated. She had experienced non-bloody, watery stools for the prior 3 weeks. She also reported three to four weeks of epigastric abdominal pain similar to her previous “ulcer pain.” Of note, she was evaluated in GI clinic one day prior to her ED visit for dysphagia, abdominal distention, and diarrhea and was scheduled for an outpatient upper endoscopy. Initial vitals were significant for a heart rate of 100 beats/min. Physical exam was significant for a mildly distended abdomen, tender to palpation at epigastrium without rebound or guarding. Labs showed a white blood cell count of 11.8 K/uL and otherwise unremarkable complete blood count, basic metabolic panel, liver function tests, and lactate measurement. Given the patient’s history of multiple abdominal surgeries and clinical presentation, POCUS was performed to evaluate for SBO. Dilated loops of small bowel were visualized in the lower abdomen gas, suggestive of SBO.

Since the small bowel encompasses a large portion of the abdomen, to fully evaluate for SBO, multiple views are necessary. These include the epigastrium, bilateral colic gutters, and suprapubic regions.5 Use the low-frequency curvilinear transducer to obtain these views, scanning in the transverse and sagittal planes (see Figures 1 and 2). Scan while moving the transducer in columns (ie, “mowing the lawn”), making sure to cover the entire abdomen. To assure that you are evaluating the small bowel, and not the large bowel, look for the characteristic plicae circularis of the small bowel (shown in Figure 3). In children and very slender adults, the high-frequency linear probe may provide enough depth to obtain adequate views.

 

A fluid-filled small intestinal segment >2.5 cm is consistent with a diagnosis of SBO. Measuring the diameter of the small bowel is both the most sensitive and specific sign; a measurement of greater than 2.5 cm is diagnostic, with a sensitivity of 97% and specificity of 91% (see Figure 4).⁶ This can be somewhat difficult to visualize, as bowel loops are multidirectional and diameters can mistakenly be taken on an indirect cut; to avoid over- or underestimation of bowel diameter, you may want to measure in the short axis using a transverse cross-sectional view of the bowel.

Lack of peristalsis is suggestive of a closed-loop obstruction. However, this finding may be more difficult to visualize, as it requires several continuous minutes of scanning or repeated exams to truly establish absent peristalsis. In prolonged courses of SBO, the bowel wall can measure >3 mm, which suggests necrosis, warranting accelerated surgical intervention. In addition, the detection of extraluminal peritoneal fluid can help determine the severity of the SBO, and small versus large fluid amounts can help determine whether medical or surgical management is warranted (see Figure 5).⁷

DISCUSSION

Increased time to diagnosis of SBO can lead to prolonged patient suffering and greater complication rates. The gold standard for diagnosing SBO—CT with intravenous and oral contrast—can take hours, requiring patients, who are often nauseated, to ingest and tolerate oral contrast. In the past, an “obstructive series” of x-rays would have been used early in the work-up of possible SBO.⁶

Recent literature suggests that POCUS is not only faster, more cost effective, and advantageous (involving no ionizing radiation), but also more accurate than x-rays. Specifically, a meta-analysis by Taylor et al showed pooled estimates for obstructive series x-rays have a sensitivity (Sn) of 75%, a specificity (Sp) of 66%, a positive likelihood ratio (+LR) of 1.6, and a negative likelihood ratio (-LR) of 0.43.¹ On the other hand, pooled results from ED studies of emergency medicine (EM) residents performing POCUS in patients with signs and symptoms suspicious for SBO showed POCUS had a Sn of 97%, Sp of 90%, +LR of 9.5, and a -LR of 0.04.^1,5,8  While detractors point to the operator-dependent nature of POCUS, literature suggests that with EM residents novice to POCUS for SBO (defined as less than 5 previous scans for SBO) were given a 10-minute didactic session and yielded Sn 94%, Sp 81%, +LR 5.0, -LR 0.07.⁵ Unluer et al trained novice EM residents for 6 hours and found them to yield Sn 98%, Sp 95%, +LR 19.5, and -LR 0.02.⁸ Thus, while it is no surprise that those with more training attain better results, both studies show it does not take much time for EM providers to surpass the accuracy of x-rays with POCUS.

CASE CONCLUSION

The findings on POCUS highly suggested the diagnosis of an SBO. A CT scan of the abdomen and pelvis with intravenous and oral contrast was ordered to further evaluate obstruction, transition point, and possible complications, including signs of ischemia per surgical request. CT demonstrated dilated loops of small bowel with transition point in the right lower quadrant, with a small amount of mesenteric fluid consistent with SBO with possible early bowel compromise due to ischemia. General surgery admitted the patient; conservative treatment with serial abdominal exams, nasogastric tube, NPO and bowel rest was ordered. The patient’s diet was gradually advanced, and she was discharged on the eleventh day of hospitalization.

SUMMARY

POCUS is a useful non-invasive tool that can accurately diagnose SBO. POCUS has increased sensitivity and specificity when compared to abdominal X-rays. This bedside imaging will not only give the ED provider rapid diagnostic information but also lead to expedited surgical intervention.

Small bowel obstruction (SBO) accounts for 2% of all cases of abdominal pain presenting to the ED and 15% of abdominal pain admissions to surgical units from the ED.^1,2 SBO can be a difficult diagnosis; the most common symptoms include nausea, vomiting, abdominal pain, obstipation, and constipation. The symptomatology depends on multiple factors: the area of the blockage, length of obstruction, and degree of the obstruction (either partial or complete).³ An upper gastrointestinal (GI) blockage classically presents with nausea and vomiting, while a lower GI blockage often presents with abdominal pain, constipation, and obstipation. Complications of obstruction range from significant morbidity—such as bowel strangulation (23%) and sepsis (31%)—to mortality (9%).⁴ ED POCUS allows for rapid and accurate diagnosis of SBO.

CASE

A 60-year-old female with a past medical history of peptic ulcer disease and multiple abdominal surgeries, including umbilical hernia repair, appendectomy, and total abdominal hysterectomy, presented to the ED with an 8-hour history of nausea and vomiting. She reported that her abdomen felt bloated. She had experienced non-bloody, watery stools for the prior 3 weeks. She also reported three to four weeks of epigastric abdominal pain similar to her previous “ulcer pain.” Of note, she was evaluated in GI clinic one day prior to her ED visit for dysphagia, abdominal distention, and diarrhea and was scheduled for an outpatient upper endoscopy. Initial vitals were significant for a heart rate of 100 beats/min. Physical exam was significant for a mildly distended abdomen, tender to palpation at epigastrium without rebound or guarding. Labs showed a white blood cell count of 11.8 K/uL and otherwise unremarkable complete blood count, basic metabolic panel, liver function tests, and lactate measurement. Given the patient’s history of multiple abdominal surgeries and clinical presentation, POCUS was performed to evaluate for SBO. Dilated loops of small bowel were visualized in the lower abdomen gas, suggestive of SBO.

Since the small bowel encompasses a large portion of the abdomen, to fully evaluate for SBO, multiple views are necessary. These include the epigastrium, bilateral colic gutters, and suprapubic regions.5 Use the low-frequency curvilinear transducer to obtain these views, scanning in the transverse and sagittal planes (see Figures 1 and 2). Scan while moving the transducer in columns (ie, “mowing the lawn”), making sure to cover the entire abdomen. To assure that you are evaluating the small bowel, and not the large bowel, look for the characteristic plicae circularis of the small bowel (shown in Figure 3). In children and very slender adults, the high-frequency linear probe may provide enough depth to obtain adequate views.

 

A fluid-filled small intestinal segment >2.5 cm is consistent with a diagnosis of SBO. Measuring the diameter of the small bowel is both the most sensitive and specific sign; a measurement of greater than 2.5 cm is diagnostic, with a sensitivity of 97% and specificity of 91% (see Figure 4).⁶ This can be somewhat difficult to visualize, as bowel loops are multidirectional and diameters can mistakenly be taken on an indirect cut; to avoid over- or underestimation of bowel diameter, you may want to measure in the short axis using a transverse cross-sectional view of the bowel.

Lack of peristalsis is suggestive of a closed-loop obstruction. However, this finding may be more difficult to visualize, as it requires several continuous minutes of scanning or repeated exams to truly establish absent peristalsis. In prolonged courses of SBO, the bowel wall can measure >3 mm, which suggests necrosis, warranting accelerated surgical intervention. In addition, the detection of extraluminal peritoneal fluid can help determine the severity of the SBO, and small versus large fluid amounts can help determine whether medical or surgical management is warranted (see Figure 5).⁷

DISCUSSION

Increased time to diagnosis of SBO can lead to prolonged patient suffering and greater complication rates. The gold standard for diagnosing SBO—CT with intravenous and oral contrast—can take hours, requiring patients, who are often nauseated, to ingest and tolerate oral contrast. In the past, an “obstructive series” of x-rays would have been used early in the work-up of possible SBO.⁶

Recent literature suggests that POCUS is not only faster, more cost effective, and advantageous (involving no ionizing radiation), but also more accurate than x-rays. Specifically, a meta-analysis by Taylor et al showed pooled estimates for obstructive series x-rays have a sensitivity (Sn) of 75%, a specificity (Sp) of 66%, a positive likelihood ratio (+LR) of 1.6, and a negative likelihood ratio (-LR) of 0.43.¹ On the other hand, pooled results from ED studies of emergency medicine (EM) residents performing POCUS in patients with signs and symptoms suspicious for SBO showed POCUS had a Sn of 97%, Sp of 90%, +LR of 9.5, and a -LR of 0.04.^1,5,8  While detractors point to the operator-dependent nature of POCUS, literature suggests that with EM residents novice to POCUS for SBO (defined as less than 5 previous scans for SBO) were given a 10-minute didactic session and yielded Sn 94%, Sp 81%, +LR 5.0, -LR 0.07.⁵ Unluer et al trained novice EM residents for 6 hours and found them to yield Sn 98%, Sp 95%, +LR 19.5, and -LR 0.02.⁸ Thus, while it is no surprise that those with more training attain better results, both studies show it does not take much time for EM providers to surpass the accuracy of x-rays with POCUS.

CASE CONCLUSION

The findings on POCUS highly suggested the diagnosis of an SBO. A CT scan of the abdomen and pelvis with intravenous and oral contrast was ordered to further evaluate obstruction, transition point, and possible complications, including signs of ischemia per surgical request. CT demonstrated dilated loops of small bowel with transition point in the right lower quadrant, with a small amount of mesenteric fluid consistent with SBO with possible early bowel compromise due to ischemia. General surgery admitted the patient; conservative treatment with serial abdominal exams, nasogastric tube, NPO and bowel rest was ordered. The patient’s diet was gradually advanced, and she was discharged on the eleventh day of hospitalization.

SUMMARY

POCUS is a useful non-invasive tool that can accurately diagnose SBO. POCUS has increased sensitivity and specificity when compared to abdominal X-rays. This bedside imaging will not only give the ED provider rapid diagnostic information but also lead to expedited surgical intervention.

Small bowel obstruction (SBO) accounts for 2% of all cases of abdominal pain presenting to the ED and 15% of abdominal pain admissions to surgical units from the ED.^1,2 SBO can be a difficult diagnosis; the most common symptoms include nausea, vomiting, abdominal pain, obstipation, and constipation. The symptomatology depends on multiple factors: the area of the blockage, length of obstruction, and degree of the obstruction (either partial or complete).³ An upper gastrointestinal (GI) blockage classically presents with nausea and vomiting, while a lower GI blockage often presents with abdominal pain, constipation, and obstipation. Complications of obstruction range from significant morbidity—such as bowel strangulation (23%) and sepsis (31%)—to mortality (9%).⁴ ED POCUS allows for rapid and accurate diagnosis of SBO.

CASE

A 60-year-old female with a past medical history of peptic ulcer disease and multiple abdominal surgeries, including umbilical hernia repair, appendectomy, and total abdominal hysterectomy, presented to the ED with an 8-hour history of nausea and vomiting. She reported that her abdomen felt bloated. She had experienced non-bloody, watery stools for the prior 3 weeks. She also reported three to four weeks of epigastric abdominal pain similar to her previous “ulcer pain.” Of note, she was evaluated in GI clinic one day prior to her ED visit for dysphagia, abdominal distention, and diarrhea and was scheduled for an outpatient upper endoscopy. Initial vitals were significant for a heart rate of 100 beats/min. Physical exam was significant for a mildly distended abdomen, tender to palpation at epigastrium without rebound or guarding. Labs showed a white blood cell count of 11.8 K/uL and otherwise unremarkable complete blood count, basic metabolic panel, liver function tests, and lactate measurement. Given the patient’s history of multiple abdominal surgeries and clinical presentation, POCUS was performed to evaluate for SBO. Dilated loops of small bowel were visualized in the lower abdomen gas, suggestive of SBO.

Since the small bowel encompasses a large portion of the abdomen, to fully evaluate for SBO, multiple views are necessary. These include the epigastrium, bilateral colic gutters, and suprapubic regions.5 Use the low-frequency curvilinear transducer to obtain these views, scanning in the transverse and sagittal planes (see Figures 1 and 2). Scan while moving the transducer in columns (ie, “mowing the lawn”), making sure to cover the entire abdomen. To assure that you are evaluating the small bowel, and not the large bowel, look for the characteristic plicae circularis of the small bowel (shown in Figure 3). In children and very slender adults, the high-frequency linear probe may provide enough depth to obtain adequate views.

 

A fluid-filled small intestinal segment >2.5 cm is consistent with a diagnosis of SBO. Measuring the diameter of the small bowel is both the most sensitive and specific sign; a measurement of greater than 2.5 cm is diagnostic, with a sensitivity of 97% and specificity of 91% (see Figure 4).⁶ This can be somewhat difficult to visualize, as bowel loops are multidirectional and diameters can mistakenly be taken on an indirect cut; to avoid over- or underestimation of bowel diameter, you may want to measure in the short axis using a transverse cross-sectional view of the bowel.

Lack of peristalsis is suggestive of a closed-loop obstruction. However, this finding may be more difficult to visualize, as it requires several continuous minutes of scanning or repeated exams to truly establish absent peristalsis. In prolonged courses of SBO, the bowel wall can measure >3 mm, which suggests necrosis, warranting accelerated surgical intervention. In addition, the detection of extraluminal peritoneal fluid can help determine the severity of the SBO, and small versus large fluid amounts can help determine whether medical or surgical management is warranted (see Figure 5).⁷

DISCUSSION

Increased time to diagnosis of SBO can lead to prolonged patient suffering and greater complication rates. The gold standard for diagnosing SBO—CT with intravenous and oral contrast—can take hours, requiring patients, who are often nauseated, to ingest and tolerate oral contrast. In the past, an “obstructive series” of x-rays would have been used early in the work-up of possible SBO.⁶

Recent literature suggests that POCUS is not only faster, more cost effective, and advantageous (involving no ionizing radiation), but also more accurate than x-rays. Specifically, a meta-analysis by Taylor et al showed pooled estimates for obstructive series x-rays have a sensitivity (Sn) of 75%, a specificity (Sp) of 66%, a positive likelihood ratio (+LR) of 1.6, and a negative likelihood ratio (-LR) of 0.43.¹ On the other hand, pooled results from ED studies of emergency medicine (EM) residents performing POCUS in patients with signs and symptoms suspicious for SBO showed POCUS had a Sn of 97%, Sp of 90%, +LR of 9.5, and a -LR of 0.04.^1,5,8  While detractors point to the operator-dependent nature of POCUS, literature suggests that with EM residents novice to POCUS for SBO (defined as less than 5 previous scans for SBO) were given a 10-minute didactic session and yielded Sn 94%, Sp 81%, +LR 5.0, -LR 0.07.⁵ Unluer et al trained novice EM residents for 6 hours and found them to yield Sn 98%, Sp 95%, +LR 19.5, and -LR 0.02.⁸ Thus, while it is no surprise that those with more training attain better results, both studies show it does not take much time for EM providers to surpass the accuracy of x-rays with POCUS.

CASE CONCLUSION

The findings on POCUS highly suggested the diagnosis of an SBO. A CT scan of the abdomen and pelvis with intravenous and oral contrast was ordered to further evaluate obstruction, transition point, and possible complications, including signs of ischemia per surgical request. CT demonstrated dilated loops of small bowel with transition point in the right lower quadrant, with a small amount of mesenteric fluid consistent with SBO with possible early bowel compromise due to ischemia. General surgery admitted the patient; conservative treatment with serial abdominal exams, nasogastric tube, NPO and bowel rest was ordered. The patient’s diet was gradually advanced, and she was discharged on the eleventh day of hospitalization.

SUMMARY

POCUS is a useful non-invasive tool that can accurately diagnose SBO. POCUS has increased sensitivity and specificity when compared to abdominal X-rays. This bedside imaging will not only give the ED provider rapid diagnostic information but also lead to expedited surgical intervention.

Multiple myeloma is the most common plasma cell neoplasm, with an estimated 24,000 cases occurring annually.¹ Symptomatic multiple myeloma most commonly presents with one or more of the cardinal CRAB phenomena of hypercalcemia, renal dysfunction, anemia, or lytic bone lesions.² Less commonly, patients may present with plasmacytomas (focal lesions of malignant plasma cells), which may involve bony or soft tissues.¹

Plasma cell neoplasms occasionally involve the joints, including the elbows, typically as plasmacytomas. The elbow is an unusual but reported location of plasmacytomas.^3,4 A case of multiple myeloma and amyloid light-chain (AL) amyloidosis has been reported, with manifestations including pseudomyopathy, bone marrow plasmacytosis, and bilateral trochanteric bursitis.⁵Bursitis is defined as inflammation of the synovial-fluid–containing sacs that lubricate joints. The olecranon bursa is commonly affected. Etiologies include infection, inflammatory disease, trauma, and malignancy. Furthermore, there is an association between bursitis and immunosuppression.^6,7 The most common modes of therapy used to treat bursitis are nonsteroidal anti-inflammatory drugs, corticosteroid injections, and surgical management.

Trochanteric bursitis has been attributed to multiple myeloma in one previous case report, but we are not aware of any previous cases of olecranon bursitis caused by multiple myeloma. Here, we present the case of a 46-year-old man with heavily pretreated multiple myeloma and amyloidosis who developed left olecranon bursitis contemporaneously with disease relapse; flow cytometric analysis of the bursal fluid demonstrated an abnormal plasma cell population, establishing the etiology.

Case presentation and summary

A 46-year-old man with a longstanding history of multiple myeloma developed swelling of the left elbow that was initially painless in September 2016. He had been diagnosed with IgA kappa multiple myeloma and AL deposition in 2011. Over the course of his disease, he was treated with the following sequence of therapies: cyclophosphamide, bortezomib, and dexamethasone, followed by melphalan-conditioned autologous peripheral blood stem cell transplant; lenalidomide and dexamethasone; carfilzomib and dexamethasone; pomalidomide, bortezomib, and dexamethasone; and bortezomib, lenalidomide, dexamethasone, doxorubicin, cyclophosphamide, and etoposide, followed by second melphalan-conditioned autologous peripheral blood stem cell transplant. In addition to treatment with numerous novel and chemotherapeutic agents, his disease course was notable for amyloid deposition in the liver, bone marrow, and kidneys, which resulted in dialysis dependence.

After the second autologous transplant, he achieved a very good partial response and experienced about 9 months of remission, after which laboratory evaluation indicated recurrence of IgA kappa monoclonal protein and free kappa light-chains, which increased slowly over several months without focal symptoms, cytopenias, or decline in organ function (Figure 1).

Twelve months after his second transplant, he presented in September 2016 with 4 weeks of left elbow swelling, with the appearance suggesting a fluid collection over the left olecranon process (Figure 2). The fluid collection was not painful unless bumped or pushed. The maximum pain level was 1-2 on a scale of 0-10. His daughter drained the fluid collection on 2 occasions, but it reaccumulated over 2 to 3 days. He reported no fevers, chills, or sweats. He did not have any redness at the site. He did not report any systemic symptoms.

Physical examination of the left elbow demonstrated a ballotable fluid collection associated with the olecranon, with no associated warmth, tenderness, or erythema. Bursal fluid was sampled, yielding orange-colored serous fluid with bland characteristics (Figure 3). Microbiologic studies were negative (Table 1). We did not suspect a malignant cause initially.

The fluid collection persisted despite treatment with nonsteroidal anti-inflammatory drugs and serial drainage procedures approximately twice per week. It became more erythematous and uncomfortable. We repeated diagnostic sampling at 13 months post-transplant. Cytospin revealed scant plasma cells. A multiparametric 8-color flow cytometric analysis was performed on the bursal fluid. It demonstrated the presence of a small abnormal population of plasma cells (0.04%). The abnormal plasma cells showed expression of CD138 and bright CD38 with aberrant expression of CD56, dim CD45, and loss of CD19, CD81 and CD27. They did not express CD117 or CD20 (Figure 4).

Because of the patient’s discomfort and his history of multidrug-refractory multiple myeloma, we obtained computed tomography imaging of the axial and appendicular skeleton, which demonstrated diffuse small lytic lesions, none larger than 3 mm, including the left elbow joint. The patient began systemic treatment with ixazomib, pomalidomide, and dexamethasone and then received radiation therapy of 20 Gy in 4 fractions to the left olecranon area. The bursal fluid collection remained stable in size but required periodic, though less frequent, drainage procedures. Unfortunately, the patient only tolerated 2 cycles of systemic therapy before experiencing hypercalcemia, exacerbation of hepatic amyloidosis, and a decline in performance status. He died 17 months after the transplant.

 

Discussion

Our patient experienced left olecranon bursitis simultaneously with relapse of multiple myeloma and AL amyloidosis. Evaluation for infectious causes was negative, and the bursal fluid did not have strongly inflammatory characteristics. Furthermore, a small plasma cell population was isolated from the fluid. Imaging did not reveal an underlying dominant lytic lesion. Although we do not have direct pathologic confirmation, the clinical scenario and flow cytometry findings support our interpretation that the patient’s bursitis was caused by or at least related to underlying multiple myeloma. While reactive plasma cells are also CD38 positive and CD138 positive, they maintain the expression of CD19 and CD45 without aberrant expression of CD56 or CD117 and do not show loss of expression of CD81 or CD27. In this situation, we suspect that either a plasmacytoma involving the soft tissue of the bursa or amyloid infiltration of the synovium may have occurred. Anti-myeloma therapies and radiation therapy did not result in control of the bursitis, though it should be noted that the patient’s highly refractory disease progressed despite treatment with a combination of later-generation immunomodulatory imide and proteasome inhibitor therapies.

Cases of malignant bursitis have been reported several times in the literature, though nearly all of the instances involved connective tissue or metastatic tumors. Tumor histologies include osteochondroma,^8,9 malignant fibrous histiocytoma,¹⁰ synovial sarcoma,¹¹ and metastatic breast cancer.¹²

Hematologic malignancies are more rare causes of bursitis; our literature search identified a report of 2 cases of non-Hodgkin lymphoma mimicking rheumatoid arthritis. The joints were the knee and elbow. Synovial fluid from one case was clear and yellow, with leukocytosis with a neutrophilic predominance (similar to our case). In both cases, pathology confirmed lymphomatous infiltration of the synovium.¹³ Notably, we identified a case of a previously healthy 35-year-old woman with bilateral trochanteric bursitis. Biopsy of tissue from the right trochanteric bursa demonstrated positive birefringence, diagnostic of AL amyloidosis. The patient also had a biclonal paraprotein accompanied by calvarial lytic lesions. She was treated with a corticosteroid pulse and bisphosphonates, followed by autologous hematopoietic stem cell transplant. ⁵ Our case shares features with the above case, including the relatively young age of the patient and the presence of AL amyloidosis.

Our patient wished to avoid a surgical biopsy procedure, and therefore we utilized flow cytometry of the bursal fluid to establish that the etiology of fluid collection was consistent with his concurrent relapse of multiple myeloma. We believe that we are reporting the second case of multiple myeloma-associated bursitis and the first case associated with multiple myeloma relapse; to our knowledge, it is the first to be diagnosed with the aid of flow cytometry.

Because of our patient’s reliance on hemodialysis beginning one year prior to his presentation with olecranon bursitis, we entertain “dialysis elbow” within the differential diagnosis. Dialysis elbow is a relatively uncommon complication of dialysis, in which patients develop olecranon bursitis on the same side as the hemodialysis access after a prolonged (months to years) duration of hemodialysis. Serositis and mechanical forces are the hypothesized etiologies¹⁴; infectious and rheumatologic causes were excluded from the reported cases. Nevertheless, we favor a malignant cause based upon the flow cytometry findings indicating involvement by immunophenotypically abnormal plasma cells.

Our patient was treated initially with serial drainage and nonsteroidals, which had little impact. After diagnosis of a plasma cell population in the fluid, we offered local treatment with radiation and systemic treatment of multiple myeloma, which offered better but suboptimal control. Possible treatments for olecranon bursitis include surgery, corticosteroid injections, anti-inflammatories, and serial drainage. Nonsurgical management may be more effective than surgical management, and corticosteroid injection carries significant risks. On the other hand, serial drainage does not confer additional infection risk in cases with aseptic etiology.¹⁵ We combined conservative measures as well as treatment of the underlying disease, but we believe that our patient did not derive significant benefit because of the refractory nature of his disease; he also expressed a preference to avoid surgical intervention.
 

 

Conclusion

Bursitis is a rare but thought-provoking potential manifestation of multiple myeloma and AL amyloidosis; we believe that our patient’s bursitis was related to plasma cell neoplasia based upon co-occurrence with disease relapse. His bursitis turned out to be an early indicator of impending systemic relapse. In this particular case, in which the patient wished to avoid surgical intervention, flow cytometry was of great value, and we believe that our case is the first report of malignant bursitis being diagnosed by flow cytometry. Our patient’s case shares similarities with other biopsy-confirmed cases of malignant bursitis, but we were able to avoid the need for surgical biopsy or bursal stripping.

The authors thank Jennifer Wilham MT (ASCP), Pat Byrd MT (ASCP), and Darlene Mann MT (ASCP) for their technical support.

Multiple myeloma is the most common plasma cell neoplasm, with an estimated 24,000 cases occurring annually.¹ Symptomatic multiple myeloma most commonly presents with one or more of the cardinal CRAB phenomena of hypercalcemia, renal dysfunction, anemia, or lytic bone lesions.² Less commonly, patients may present with plasmacytomas (focal lesions of malignant plasma cells), which may involve bony or soft tissues.¹

Plasma cell neoplasms occasionally involve the joints, including the elbows, typically as plasmacytomas. The elbow is an unusual but reported location of plasmacytomas.^3,4 A case of multiple myeloma and amyloid light-chain (AL) amyloidosis has been reported, with manifestations including pseudomyopathy, bone marrow plasmacytosis, and bilateral trochanteric bursitis.⁵Bursitis is defined as inflammation of the synovial-fluid–containing sacs that lubricate joints. The olecranon bursa is commonly affected. Etiologies include infection, inflammatory disease, trauma, and malignancy. Furthermore, there is an association between bursitis and immunosuppression.^6,7 The most common modes of therapy used to treat bursitis are nonsteroidal anti-inflammatory drugs, corticosteroid injections, and surgical management.

Trochanteric bursitis has been attributed to multiple myeloma in one previous case report, but we are not aware of any previous cases of olecranon bursitis caused by multiple myeloma. Here, we present the case of a 46-year-old man with heavily pretreated multiple myeloma and amyloidosis who developed left olecranon bursitis contemporaneously with disease relapse; flow cytometric analysis of the bursal fluid demonstrated an abnormal plasma cell population, establishing the etiology.

Case presentation and summary

A 46-year-old man with a longstanding history of multiple myeloma developed swelling of the left elbow that was initially painless in September 2016. He had been diagnosed with IgA kappa multiple myeloma and AL deposition in 2011. Over the course of his disease, he was treated with the following sequence of therapies: cyclophosphamide, bortezomib, and dexamethasone, followed by melphalan-conditioned autologous peripheral blood stem cell transplant; lenalidomide and dexamethasone; carfilzomib and dexamethasone; pomalidomide, bortezomib, and dexamethasone; and bortezomib, lenalidomide, dexamethasone, doxorubicin, cyclophosphamide, and etoposide, followed by second melphalan-conditioned autologous peripheral blood stem cell transplant. In addition to treatment with numerous novel and chemotherapeutic agents, his disease course was notable for amyloid deposition in the liver, bone marrow, and kidneys, which resulted in dialysis dependence.

After the second autologous transplant, he achieved a very good partial response and experienced about 9 months of remission, after which laboratory evaluation indicated recurrence of IgA kappa monoclonal protein and free kappa light-chains, which increased slowly over several months without focal symptoms, cytopenias, or decline in organ function (Figure 1).

Twelve months after his second transplant, he presented in September 2016 with 4 weeks of left elbow swelling, with the appearance suggesting a fluid collection over the left olecranon process (Figure 2). The fluid collection was not painful unless bumped or pushed. The maximum pain level was 1-2 on a scale of 0-10. His daughter drained the fluid collection on 2 occasions, but it reaccumulated over 2 to 3 days. He reported no fevers, chills, or sweats. He did not have any redness at the site. He did not report any systemic symptoms.

Physical examination of the left elbow demonstrated a ballotable fluid collection associated with the olecranon, with no associated warmth, tenderness, or erythema. Bursal fluid was sampled, yielding orange-colored serous fluid with bland characteristics (Figure 3). Microbiologic studies were negative (Table 1). We did not suspect a malignant cause initially.

The fluid collection persisted despite treatment with nonsteroidal anti-inflammatory drugs and serial drainage procedures approximately twice per week. It became more erythematous and uncomfortable. We repeated diagnostic sampling at 13 months post-transplant. Cytospin revealed scant plasma cells. A multiparametric 8-color flow cytometric analysis was performed on the bursal fluid. It demonstrated the presence of a small abnormal population of plasma cells (0.04%). The abnormal plasma cells showed expression of CD138 and bright CD38 with aberrant expression of CD56, dim CD45, and loss of CD19, CD81 and CD27. They did not express CD117 or CD20 (Figure 4).

Because of the patient’s discomfort and his history of multidrug-refractory multiple myeloma, we obtained computed tomography imaging of the axial and appendicular skeleton, which demonstrated diffuse small lytic lesions, none larger than 3 mm, including the left elbow joint. The patient began systemic treatment with ixazomib, pomalidomide, and dexamethasone and then received radiation therapy of 20 Gy in 4 fractions to the left olecranon area. The bursal fluid collection remained stable in size but required periodic, though less frequent, drainage procedures. Unfortunately, the patient only tolerated 2 cycles of systemic therapy before experiencing hypercalcemia, exacerbation of hepatic amyloidosis, and a decline in performance status. He died 17 months after the transplant.

 

Discussion

Our patient experienced left olecranon bursitis simultaneously with relapse of multiple myeloma and AL amyloidosis. Evaluation for infectious causes was negative, and the bursal fluid did not have strongly inflammatory characteristics. Furthermore, a small plasma cell population was isolated from the fluid. Imaging did not reveal an underlying dominant lytic lesion. Although we do not have direct pathologic confirmation, the clinical scenario and flow cytometry findings support our interpretation that the patient’s bursitis was caused by or at least related to underlying multiple myeloma. While reactive plasma cells are also CD38 positive and CD138 positive, they maintain the expression of CD19 and CD45 without aberrant expression of CD56 or CD117 and do not show loss of expression of CD81 or CD27. In this situation, we suspect that either a plasmacytoma involving the soft tissue of the bursa or amyloid infiltration of the synovium may have occurred. Anti-myeloma therapies and radiation therapy did not result in control of the bursitis, though it should be noted that the patient’s highly refractory disease progressed despite treatment with a combination of later-generation immunomodulatory imide and proteasome inhibitor therapies.

Cases of malignant bursitis have been reported several times in the literature, though nearly all of the instances involved connective tissue or metastatic tumors. Tumor histologies include osteochondroma,^8,9 malignant fibrous histiocytoma,¹⁰ synovial sarcoma,¹¹ and metastatic breast cancer.¹²

Hematologic malignancies are more rare causes of bursitis; our literature search identified a report of 2 cases of non-Hodgkin lymphoma mimicking rheumatoid arthritis. The joints were the knee and elbow. Synovial fluid from one case was clear and yellow, with leukocytosis with a neutrophilic predominance (similar to our case). In both cases, pathology confirmed lymphomatous infiltration of the synovium.¹³ Notably, we identified a case of a previously healthy 35-year-old woman with bilateral trochanteric bursitis. Biopsy of tissue from the right trochanteric bursa demonstrated positive birefringence, diagnostic of AL amyloidosis. The patient also had a biclonal paraprotein accompanied by calvarial lytic lesions. She was treated with a corticosteroid pulse and bisphosphonates, followed by autologous hematopoietic stem cell transplant. ⁵ Our case shares features with the above case, including the relatively young age of the patient and the presence of AL amyloidosis.

Our patient wished to avoid a surgical biopsy procedure, and therefore we utilized flow cytometry of the bursal fluid to establish that the etiology of fluid collection was consistent with his concurrent relapse of multiple myeloma. We believe that we are reporting the second case of multiple myeloma-associated bursitis and the first case associated with multiple myeloma relapse; to our knowledge, it is the first to be diagnosed with the aid of flow cytometry.

Because of our patient’s reliance on hemodialysis beginning one year prior to his presentation with olecranon bursitis, we entertain “dialysis elbow” within the differential diagnosis. Dialysis elbow is a relatively uncommon complication of dialysis, in which patients develop olecranon bursitis on the same side as the hemodialysis access after a prolonged (months to years) duration of hemodialysis. Serositis and mechanical forces are the hypothesized etiologies¹⁴; infectious and rheumatologic causes were excluded from the reported cases. Nevertheless, we favor a malignant cause based upon the flow cytometry findings indicating involvement by immunophenotypically abnormal plasma cells.

Our patient was treated initially with serial drainage and nonsteroidals, which had little impact. After diagnosis of a plasma cell population in the fluid, we offered local treatment with radiation and systemic treatment of multiple myeloma, which offered better but suboptimal control. Possible treatments for olecranon bursitis include surgery, corticosteroid injections, anti-inflammatories, and serial drainage. Nonsurgical management may be more effective than surgical management, and corticosteroid injection carries significant risks. On the other hand, serial drainage does not confer additional infection risk in cases with aseptic etiology.¹⁵ We combined conservative measures as well as treatment of the underlying disease, but we believe that our patient did not derive significant benefit because of the refractory nature of his disease; he also expressed a preference to avoid surgical intervention.
 

 

Conclusion

Bursitis is a rare but thought-provoking potential manifestation of multiple myeloma and AL amyloidosis; we believe that our patient’s bursitis was related to plasma cell neoplasia based upon co-occurrence with disease relapse. His bursitis turned out to be an early indicator of impending systemic relapse. In this particular case, in which the patient wished to avoid surgical intervention, flow cytometry was of great value, and we believe that our case is the first report of malignant bursitis being diagnosed by flow cytometry. Our patient’s case shares similarities with other biopsy-confirmed cases of malignant bursitis, but we were able to avoid the need for surgical biopsy or bursal stripping.

The authors thank Jennifer Wilham MT (ASCP), Pat Byrd MT (ASCP), and Darlene Mann MT (ASCP) for their technical support.

Multiple myeloma is the most common plasma cell neoplasm, with an estimated 24,000 cases occurring annually.¹ Symptomatic multiple myeloma most commonly presents with one or more of the cardinal CRAB phenomena of hypercalcemia, renal dysfunction, anemia, or lytic bone lesions.² Less commonly, patients may present with plasmacytomas (focal lesions of malignant plasma cells), which may involve bony or soft tissues.¹

Plasma cell neoplasms occasionally involve the joints, including the elbows, typically as plasmacytomas. The elbow is an unusual but reported location of plasmacytomas.^3,4 A case of multiple myeloma and amyloid light-chain (AL) amyloidosis has been reported, with manifestations including pseudomyopathy, bone marrow plasmacytosis, and bilateral trochanteric bursitis.⁵Bursitis is defined as inflammation of the synovial-fluid–containing sacs that lubricate joints. The olecranon bursa is commonly affected. Etiologies include infection, inflammatory disease, trauma, and malignancy. Furthermore, there is an association between bursitis and immunosuppression.^6,7 The most common modes of therapy used to treat bursitis are nonsteroidal anti-inflammatory drugs, corticosteroid injections, and surgical management.

Trochanteric bursitis has been attributed to multiple myeloma in one previous case report, but we are not aware of any previous cases of olecranon bursitis caused by multiple myeloma. Here, we present the case of a 46-year-old man with heavily pretreated multiple myeloma and amyloidosis who developed left olecranon bursitis contemporaneously with disease relapse; flow cytometric analysis of the bursal fluid demonstrated an abnormal plasma cell population, establishing the etiology.

Case presentation and summary

A 46-year-old man with a longstanding history of multiple myeloma developed swelling of the left elbow that was initially painless in September 2016. He had been diagnosed with IgA kappa multiple myeloma and AL deposition in 2011. Over the course of his disease, he was treated with the following sequence of therapies: cyclophosphamide, bortezomib, and dexamethasone, followed by melphalan-conditioned autologous peripheral blood stem cell transplant; lenalidomide and dexamethasone; carfilzomib and dexamethasone; pomalidomide, bortezomib, and dexamethasone; and bortezomib, lenalidomide, dexamethasone, doxorubicin, cyclophosphamide, and etoposide, followed by second melphalan-conditioned autologous peripheral blood stem cell transplant. In addition to treatment with numerous novel and chemotherapeutic agents, his disease course was notable for amyloid deposition in the liver, bone marrow, and kidneys, which resulted in dialysis dependence.

After the second autologous transplant, he achieved a very good partial response and experienced about 9 months of remission, after which laboratory evaluation indicated recurrence of IgA kappa monoclonal protein and free kappa light-chains, which increased slowly over several months without focal symptoms, cytopenias, or decline in organ function (Figure 1).

Twelve months after his second transplant, he presented in September 2016 with 4 weeks of left elbow swelling, with the appearance suggesting a fluid collection over the left olecranon process (Figure 2). The fluid collection was not painful unless bumped or pushed. The maximum pain level was 1-2 on a scale of 0-10. His daughter drained the fluid collection on 2 occasions, but it reaccumulated over 2 to 3 days. He reported no fevers, chills, or sweats. He did not have any redness at the site. He did not report any systemic symptoms.

Physical examination of the left elbow demonstrated a ballotable fluid collection associated with the olecranon, with no associated warmth, tenderness, or erythema. Bursal fluid was sampled, yielding orange-colored serous fluid with bland characteristics (Figure 3). Microbiologic studies were negative (Table 1). We did not suspect a malignant cause initially.

The fluid collection persisted despite treatment with nonsteroidal anti-inflammatory drugs and serial drainage procedures approximately twice per week. It became more erythematous and uncomfortable. We repeated diagnostic sampling at 13 months post-transplant. Cytospin revealed scant plasma cells. A multiparametric 8-color flow cytometric analysis was performed on the bursal fluid. It demonstrated the presence of a small abnormal population of plasma cells (0.04%). The abnormal plasma cells showed expression of CD138 and bright CD38 with aberrant expression of CD56, dim CD45, and loss of CD19, CD81 and CD27. They did not express CD117 or CD20 (Figure 4).

Because of the patient’s discomfort and his history of multidrug-refractory multiple myeloma, we obtained computed tomography imaging of the axial and appendicular skeleton, which demonstrated diffuse small lytic lesions, none larger than 3 mm, including the left elbow joint. The patient began systemic treatment with ixazomib, pomalidomide, and dexamethasone and then received radiation therapy of 20 Gy in 4 fractions to the left olecranon area. The bursal fluid collection remained stable in size but required periodic, though less frequent, drainage procedures. Unfortunately, the patient only tolerated 2 cycles of systemic therapy before experiencing hypercalcemia, exacerbation of hepatic amyloidosis, and a decline in performance status. He died 17 months after the transplant.

 

Discussion

Our patient experienced left olecranon bursitis simultaneously with relapse of multiple myeloma and AL amyloidosis. Evaluation for infectious causes was negative, and the bursal fluid did not have strongly inflammatory characteristics. Furthermore, a small plasma cell population was isolated from the fluid. Imaging did not reveal an underlying dominant lytic lesion. Although we do not have direct pathologic confirmation, the clinical scenario and flow cytometry findings support our interpretation that the patient’s bursitis was caused by or at least related to underlying multiple myeloma. While reactive plasma cells are also CD38 positive and CD138 positive, they maintain the expression of CD19 and CD45 without aberrant expression of CD56 or CD117 and do not show loss of expression of CD81 or CD27. In this situation, we suspect that either a plasmacytoma involving the soft tissue of the bursa or amyloid infiltration of the synovium may have occurred. Anti-myeloma therapies and radiation therapy did not result in control of the bursitis, though it should be noted that the patient’s highly refractory disease progressed despite treatment with a combination of later-generation immunomodulatory imide and proteasome inhibitor therapies.

Cases of malignant bursitis have been reported several times in the literature, though nearly all of the instances involved connective tissue or metastatic tumors. Tumor histologies include osteochondroma,^8,9 malignant fibrous histiocytoma,¹⁰ synovial sarcoma,¹¹ and metastatic breast cancer.¹²

Hematologic malignancies are more rare causes of bursitis; our literature search identified a report of 2 cases of non-Hodgkin lymphoma mimicking rheumatoid arthritis. The joints were the knee and elbow. Synovial fluid from one case was clear and yellow, with leukocytosis with a neutrophilic predominance (similar to our case). In both cases, pathology confirmed lymphomatous infiltration of the synovium.¹³ Notably, we identified a case of a previously healthy 35-year-old woman with bilateral trochanteric bursitis. Biopsy of tissue from the right trochanteric bursa demonstrated positive birefringence, diagnostic of AL amyloidosis. The patient also had a biclonal paraprotein accompanied by calvarial lytic lesions. She was treated with a corticosteroid pulse and bisphosphonates, followed by autologous hematopoietic stem cell transplant. ⁵ Our case shares features with the above case, including the relatively young age of the patient and the presence of AL amyloidosis.

Our patient wished to avoid a surgical biopsy procedure, and therefore we utilized flow cytometry of the bursal fluid to establish that the etiology of fluid collection was consistent with his concurrent relapse of multiple myeloma. We believe that we are reporting the second case of multiple myeloma-associated bursitis and the first case associated with multiple myeloma relapse; to our knowledge, it is the first to be diagnosed with the aid of flow cytometry.

Because of our patient’s reliance on hemodialysis beginning one year prior to his presentation with olecranon bursitis, we entertain “dialysis elbow” within the differential diagnosis. Dialysis elbow is a relatively uncommon complication of dialysis, in which patients develop olecranon bursitis on the same side as the hemodialysis access after a prolonged (months to years) duration of hemodialysis. Serositis and mechanical forces are the hypothesized etiologies¹⁴; infectious and rheumatologic causes were excluded from the reported cases. Nevertheless, we favor a malignant cause based upon the flow cytometry findings indicating involvement by immunophenotypically abnormal plasma cells.

Our patient was treated initially with serial drainage and nonsteroidals, which had little impact. After diagnosis of a plasma cell population in the fluid, we offered local treatment with radiation and systemic treatment of multiple myeloma, which offered better but suboptimal control. Possible treatments for olecranon bursitis include surgery, corticosteroid injections, anti-inflammatories, and serial drainage. Nonsurgical management may be more effective than surgical management, and corticosteroid injection carries significant risks. On the other hand, serial drainage does not confer additional infection risk in cases with aseptic etiology.¹⁵ We combined conservative measures as well as treatment of the underlying disease, but we believe that our patient did not derive significant benefit because of the refractory nature of his disease; he also expressed a preference to avoid surgical intervention.
 

 

Conclusion

Bursitis is a rare but thought-provoking potential manifestation of multiple myeloma and AL amyloidosis; we believe that our patient’s bursitis was related to plasma cell neoplasia based upon co-occurrence with disease relapse. His bursitis turned out to be an early indicator of impending systemic relapse. In this particular case, in which the patient wished to avoid surgical intervention, flow cytometry was of great value, and we believe that our case is the first report of malignant bursitis being diagnosed by flow cytometry. Our patient’s case shares similarities with other biopsy-confirmed cases of malignant bursitis, but we were able to avoid the need for surgical biopsy or bursal stripping.

The authors thank Jennifer Wilham MT (ASCP), Pat Byrd MT (ASCP), and Darlene Mann MT (ASCP) for their technical support.

Cutaneous B-cell lymphomas represent a group of lymphomas derived from B lymphocytes in various stages of differentiation. The skin can be the site of primary or secondary involvement of any of the B-cell lymphomas. Primary cutaneous B-cell lymphomas present in the skin without evidence of extracutaneous disease at the time of diagnosis.¹ The World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues recognizes 5 distinct primary cutaneous B-cell lymphoma subtypes: primary cutaneous follicle center lymphoma; primary cutaneous marginal zone lymphoma; primary cutaneous diffuse large B-cell lymphoma (DLBCL), leg type; DLBCL, not otherwise specified; and intravascular DLBCL.^1-3 The DLBCL, not otherwise specified, category includes less common provisional entities with insufficient evidence to be recognized as distinct diseases. Epstein-Barr virus (EBV)–positive DLBCL is a rare subtype in this group.⁴

This article reviews the different clinicopathologic subtypes of primary cutaneous B-cell lymphoma. It also serves to help dermatologists recognize primary cutaneous EBV-positive DLBCL as a rare and aggressive form of this disease.

Case Report

An 84-year-old white man presented with a pruritic eruption on the arms, legs, back, neck, and face of 5 months’ duration. His medical history was notable for prostate cancer that was successfully treated with radiation therapy 6 years prior. The patient denied any constitutional symptoms such as fever, chills, night sweats, or weight loss, and review of systems was negative. The patient was taking prednisone, which alleviated the pruritus, but the lesions persisted.

Physical examination revealed multiple pink to erythematous papules and subcutaneous nodules involving the face, neck, back, arms, and legs (Figure 1). No scale, crust, or ulceration was present. Palpation of the cervical, supraclavicular, axillary, and inguinal lymph nodes was negative for lymphadenopathy.

Punch biopsies of representative lesions on the upper back and right arm revealed diffuse and nodular infiltrates of large atypical lymphoid cells with scattered centroblasts and immunoblasts (Figures 2 and 3). Immunohistochemical staining demonstrated CD79, MUM-1, and EBV-encoded RNA positivity among the neoplastic cells. The Ki-67 proliferative index was greater than 90%. The neoplastic cells were negative for CD5, CD10, CD20, CD21, CD30, CD56, CD123, CD138, PAX5, C-MYC, BCL-2, BCL-6, cyclin D1, TCL-1A, and terminal deoxynucleotidyl transferase. Polymerase chain reaction showed a clonal B-cell population.

A peripheral blood smear did not show evidence of a B-cell lymphoproliferative process. A bone marrow biopsy was performed and did not show evidence of B-cell lymphoid neoplasia but did show reactive lymphoid aggregates composed of CD4⁺ and CD10⁺ T cells. Peripheral blood T-cell rearrangement and JAK2 were negative.

Based on clinical and histologic findings, the patient was diagnosed with primary cutaneous EBV-positive DLBCL. The patient was started on CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) chemotherapy for treatment of this aggressive cutaneous lymphoma, which initially resulted in clinical improvement of the lesions and complete involution of the subcutaneous nodules. After the sixth cycle of CHOP, he developed faintly erythematous indurated papules on the upper arms, chest, and back. Biopsy confirmed recurrence of the EBV-positive cutaneous lymphoma, and he started salvage chemotherapy with gemcitabine, oxaliplatin, and rituximab every 2 weeks; however, 4 months later (9 months after the initial presentation) he died from complications of the disease.

 

Comment

Etiology
Epstein-Barr virus–positive DLBCL, also called EBV-positive DLBCL of the elderly, was initially described in 2003 by Oyama et al⁵ and was included as a provisional entity in the 2008 World Health Organization classification system as a rare subtype of the DLBCL, not otherwise specified, category.² It is defined as an EBV-positive monoclonal large B-cell proliferation that occurs in immunocompetent patients older than 50 years.⁶ Epstein-Barr virus is a human herpesvirus that demonstrates tropism for lymphocytes and survives in human hosts by establishing latency in B cells. Under normal immune conditions, the proliferation of EBV-infected B cells is prevented by cytotoxic T cells.⁷ It is important to recognize that patients with EBV-positive DLBCL do not have a known immunodeficiency state; therefore, it has been postulated that EBV-positive DLBCL might be caused by age-related senescence of the immune system.^4,8

Epidemiology and Clinical Features
Epstein-Barr virus–positive DLBCL is more common in Asian countries than in Western countries, and there is a slight male predominance.⁶ A majority of patients present with extranodal disease at the time of diagnosis, and the skin is the most common extranodal site of involvement.^6,9 Rare cases of primary cutaneous involvement also have been described.^7,9,10 Cutaneous manifestations include erythematous papules and subcutaneous nodules. Other sites of extranodal involvement include the lungs, oral cavity, pharynx, gastrointestinal tract, and bone marrow.^8,9 However, EBV-positive DLBCL is an aggressive lymphoma and prognosis is poor irrespective of the primary site of involvement.

Histopathology
Two morphologic subtypes can be seen on histology. The polymorphic pattern is characterized by a broad range of B-cell maturation with admixed reactive cells (eg, lymphocytes, histiocytes, plasma cells). The monomorphic or large-cell pattern is characterized by monotonous sheets of large transformed B cells.^4,11 Many cases show both histologic patterns, and these morphologic variants do not impart any clinical or prognostic significance. Regardless of the histologic subtype, the neoplastic cells express pan B-cell antigens (eg, CD19, CD20, CD79a, PAX5), as well as MUM-1, BCL-2, and EBV-encoded RNA.⁴ Cases with plasmablastic features, as in our patient, may show weak or absent CD20 staining.¹² Detection of EBV by in situ hybridization is required for the diagnosis.

Diagnosis
Workup for a suspected cutaneous lymphoma should include a complete history and physical examination; laboratory studies; and relevant imaging evaluation such as computed tomography of the chest, abdomen, and pelvis with or without whole-body positron emission tomography. A bone marrow biopsy and aspirate also should be performed in all cutaneous lymphomas with intermediate to aggressive clinical behavior. Accurate staging evaluation is integral to confirm the absence of extracutaneous involvement and to provide prognostic and anatomic information for the appropriate selection of treatment.¹³

Prognosis and Management
Primary cutaneous lymphomas tend to have different clinical behaviors and prognoses compared to histologically similar systemic lymphomas; therefore, different therapeutic strategies are warranted.¹⁴ Epstein-Barr virus–positive DLBCL has an aggressive clinical course with a median survival of 2 years.⁸ Patients with EBV-positive DLBCL have a poorer overall survival and treatment response when compared to patients with EBV-negative DLBCLs.⁴ Primary cutaneous B-cell lymphomas with indolent behavior, such as primary cutaneous marginal zone lymphoma and primary cutaneous follicle center lymphoma, can be treated with surgical excision, radiation therapy, or observation.¹⁵ No standard treatment exists for EBV-positive DLBCL, but R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone), which is the standard treatment of primary cutaneous DLBCL, leg type, may provide a survival benefit.^13,15 Further studies are required to determine optimal treatment strategies.

Conclusion

Although rare, EBV-positive DLBCL is an important entity to consider when evaluating a patient with a suspected primary cutaneous lymphoma. Workup to rule out an underlying systemic lymphoma with relevant laboratory evaluation, imaging studies, and bone marrow biopsy is critical. Prognosis is poor and treatment is difficult, as standard treatment protocols have yet to be determined.

Cutaneous B-cell lymphomas represent a group of lymphomas derived from B lymphocytes in various stages of differentiation. The skin can be the site of primary or secondary involvement of any of the B-cell lymphomas. Primary cutaneous B-cell lymphomas present in the skin without evidence of extracutaneous disease at the time of diagnosis.¹ The World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues recognizes 5 distinct primary cutaneous B-cell lymphoma subtypes: primary cutaneous follicle center lymphoma; primary cutaneous marginal zone lymphoma; primary cutaneous diffuse large B-cell lymphoma (DLBCL), leg type; DLBCL, not otherwise specified; and intravascular DLBCL.^1-3 The DLBCL, not otherwise specified, category includes less common provisional entities with insufficient evidence to be recognized as distinct diseases. Epstein-Barr virus (EBV)–positive DLBCL is a rare subtype in this group.⁴

This article reviews the different clinicopathologic subtypes of primary cutaneous B-cell lymphoma. It also serves to help dermatologists recognize primary cutaneous EBV-positive DLBCL as a rare and aggressive form of this disease.

Case Report

An 84-year-old white man presented with a pruritic eruption on the arms, legs, back, neck, and face of 5 months’ duration. His medical history was notable for prostate cancer that was successfully treated with radiation therapy 6 years prior. The patient denied any constitutional symptoms such as fever, chills, night sweats, or weight loss, and review of systems was negative. The patient was taking prednisone, which alleviated the pruritus, but the lesions persisted.

Physical examination revealed multiple pink to erythematous papules and subcutaneous nodules involving the face, neck, back, arms, and legs (Figure 1). No scale, crust, or ulceration was present. Palpation of the cervical, supraclavicular, axillary, and inguinal lymph nodes was negative for lymphadenopathy.

Punch biopsies of representative lesions on the upper back and right arm revealed diffuse and nodular infiltrates of large atypical lymphoid cells with scattered centroblasts and immunoblasts (Figures 2 and 3). Immunohistochemical staining demonstrated CD79, MUM-1, and EBV-encoded RNA positivity among the neoplastic cells. The Ki-67 proliferative index was greater than 90%. The neoplastic cells were negative for CD5, CD10, CD20, CD21, CD30, CD56, CD123, CD138, PAX5, C-MYC, BCL-2, BCL-6, cyclin D1, TCL-1A, and terminal deoxynucleotidyl transferase. Polymerase chain reaction showed a clonal B-cell population.

A peripheral blood smear did not show evidence of a B-cell lymphoproliferative process. A bone marrow biopsy was performed and did not show evidence of B-cell lymphoid neoplasia but did show reactive lymphoid aggregates composed of CD4⁺ and CD10⁺ T cells. Peripheral blood T-cell rearrangement and JAK2 were negative.

Based on clinical and histologic findings, the patient was diagnosed with primary cutaneous EBV-positive DLBCL. The patient was started on CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) chemotherapy for treatment of this aggressive cutaneous lymphoma, which initially resulted in clinical improvement of the lesions and complete involution of the subcutaneous nodules. After the sixth cycle of CHOP, he developed faintly erythematous indurated papules on the upper arms, chest, and back. Biopsy confirmed recurrence of the EBV-positive cutaneous lymphoma, and he started salvage chemotherapy with gemcitabine, oxaliplatin, and rituximab every 2 weeks; however, 4 months later (9 months after the initial presentation) he died from complications of the disease.

 

Comment

Etiology
Epstein-Barr virus–positive DLBCL, also called EBV-positive DLBCL of the elderly, was initially described in 2003 by Oyama et al⁵ and was included as a provisional entity in the 2008 World Health Organization classification system as a rare subtype of the DLBCL, not otherwise specified, category.² It is defined as an EBV-positive monoclonal large B-cell proliferation that occurs in immunocompetent patients older than 50 years.⁶ Epstein-Barr virus is a human herpesvirus that demonstrates tropism for lymphocytes and survives in human hosts by establishing latency in B cells. Under normal immune conditions, the proliferation of EBV-infected B cells is prevented by cytotoxic T cells.⁷ It is important to recognize that patients with EBV-positive DLBCL do not have a known immunodeficiency state; therefore, it has been postulated that EBV-positive DLBCL might be caused by age-related senescence of the immune system.^4,8

Epidemiology and Clinical Features
Epstein-Barr virus–positive DLBCL is more common in Asian countries than in Western countries, and there is a slight male predominance.⁶ A majority of patients present with extranodal disease at the time of diagnosis, and the skin is the most common extranodal site of involvement.^6,9 Rare cases of primary cutaneous involvement also have been described.^7,9,10 Cutaneous manifestations include erythematous papules and subcutaneous nodules. Other sites of extranodal involvement include the lungs, oral cavity, pharynx, gastrointestinal tract, and bone marrow.^8,9 However, EBV-positive DLBCL is an aggressive lymphoma and prognosis is poor irrespective of the primary site of involvement.

Histopathology
Two morphologic subtypes can be seen on histology. The polymorphic pattern is characterized by a broad range of B-cell maturation with admixed reactive cells (eg, lymphocytes, histiocytes, plasma cells). The monomorphic or large-cell pattern is characterized by monotonous sheets of large transformed B cells.^4,11 Many cases show both histologic patterns, and these morphologic variants do not impart any clinical or prognostic significance. Regardless of the histologic subtype, the neoplastic cells express pan B-cell antigens (eg, CD19, CD20, CD79a, PAX5), as well as MUM-1, BCL-2, and EBV-encoded RNA.⁴ Cases with plasmablastic features, as in our patient, may show weak or absent CD20 staining.¹² Detection of EBV by in situ hybridization is required for the diagnosis.

Diagnosis
Workup for a suspected cutaneous lymphoma should include a complete history and physical examination; laboratory studies; and relevant imaging evaluation such as computed tomography of the chest, abdomen, and pelvis with or without whole-body positron emission tomography. A bone marrow biopsy and aspirate also should be performed in all cutaneous lymphomas with intermediate to aggressive clinical behavior. Accurate staging evaluation is integral to confirm the absence of extracutaneous involvement and to provide prognostic and anatomic information for the appropriate selection of treatment.¹³

Prognosis and Management
Primary cutaneous lymphomas tend to have different clinical behaviors and prognoses compared to histologically similar systemic lymphomas; therefore, different therapeutic strategies are warranted.¹⁴ Epstein-Barr virus–positive DLBCL has an aggressive clinical course with a median survival of 2 years.⁸ Patients with EBV-positive DLBCL have a poorer overall survival and treatment response when compared to patients with EBV-negative DLBCLs.⁴ Primary cutaneous B-cell lymphomas with indolent behavior, such as primary cutaneous marginal zone lymphoma and primary cutaneous follicle center lymphoma, can be treated with surgical excision, radiation therapy, or observation.¹⁵ No standard treatment exists for EBV-positive DLBCL, but R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone), which is the standard treatment of primary cutaneous DLBCL, leg type, may provide a survival benefit.^13,15 Further studies are required to determine optimal treatment strategies.

Conclusion

Although rare, EBV-positive DLBCL is an important entity to consider when evaluating a patient with a suspected primary cutaneous lymphoma. Workup to rule out an underlying systemic lymphoma with relevant laboratory evaluation, imaging studies, and bone marrow biopsy is critical. Prognosis is poor and treatment is difficult, as standard treatment protocols have yet to be determined.

Cutaneous B-cell lymphomas represent a group of lymphomas derived from B lymphocytes in various stages of differentiation. The skin can be the site of primary or secondary involvement of any of the B-cell lymphomas. Primary cutaneous B-cell lymphomas present in the skin without evidence of extracutaneous disease at the time of diagnosis.¹ The World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues recognizes 5 distinct primary cutaneous B-cell lymphoma subtypes: primary cutaneous follicle center lymphoma; primary cutaneous marginal zone lymphoma; primary cutaneous diffuse large B-cell lymphoma (DLBCL), leg type; DLBCL, not otherwise specified; and intravascular DLBCL.^1-3 The DLBCL, not otherwise specified, category includes less common provisional entities with insufficient evidence to be recognized as distinct diseases. Epstein-Barr virus (EBV)–positive DLBCL is a rare subtype in this group.⁴

This article reviews the different clinicopathologic subtypes of primary cutaneous B-cell lymphoma. It also serves to help dermatologists recognize primary cutaneous EBV-positive DLBCL as a rare and aggressive form of this disease.

Case Report

An 84-year-old white man presented with a pruritic eruption on the arms, legs, back, neck, and face of 5 months’ duration. His medical history was notable for prostate cancer that was successfully treated with radiation therapy 6 years prior. The patient denied any constitutional symptoms such as fever, chills, night sweats, or weight loss, and review of systems was negative. The patient was taking prednisone, which alleviated the pruritus, but the lesions persisted.

Physical examination revealed multiple pink to erythematous papules and subcutaneous nodules involving the face, neck, back, arms, and legs (Figure 1). No scale, crust, or ulceration was present. Palpation of the cervical, supraclavicular, axillary, and inguinal lymph nodes was negative for lymphadenopathy.

Punch biopsies of representative lesions on the upper back and right arm revealed diffuse and nodular infiltrates of large atypical lymphoid cells with scattered centroblasts and immunoblasts (Figures 2 and 3). Immunohistochemical staining demonstrated CD79, MUM-1, and EBV-encoded RNA positivity among the neoplastic cells. The Ki-67 proliferative index was greater than 90%. The neoplastic cells were negative for CD5, CD10, CD20, CD21, CD30, CD56, CD123, CD138, PAX5, C-MYC, BCL-2, BCL-6, cyclin D1, TCL-1A, and terminal deoxynucleotidyl transferase. Polymerase chain reaction showed a clonal B-cell population.

A peripheral blood smear did not show evidence of a B-cell lymphoproliferative process. A bone marrow biopsy was performed and did not show evidence of B-cell lymphoid neoplasia but did show reactive lymphoid aggregates composed of CD4⁺ and CD10⁺ T cells. Peripheral blood T-cell rearrangement and JAK2 were negative.

Based on clinical and histologic findings, the patient was diagnosed with primary cutaneous EBV-positive DLBCL. The patient was started on CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) chemotherapy for treatment of this aggressive cutaneous lymphoma, which initially resulted in clinical improvement of the lesions and complete involution of the subcutaneous nodules. After the sixth cycle of CHOP, he developed faintly erythematous indurated papules on the upper arms, chest, and back. Biopsy confirmed recurrence of the EBV-positive cutaneous lymphoma, and he started salvage chemotherapy with gemcitabine, oxaliplatin, and rituximab every 2 weeks; however, 4 months later (9 months after the initial presentation) he died from complications of the disease.

 

Comment

Etiology
Epstein-Barr virus–positive DLBCL, also called EBV-positive DLBCL of the elderly, was initially described in 2003 by Oyama et al⁵ and was included as a provisional entity in the 2008 World Health Organization classification system as a rare subtype of the DLBCL, not otherwise specified, category.² It is defined as an EBV-positive monoclonal large B-cell proliferation that occurs in immunocompetent patients older than 50 years.⁶ Epstein-Barr virus is a human herpesvirus that demonstrates tropism for lymphocytes and survives in human hosts by establishing latency in B cells. Under normal immune conditions, the proliferation of EBV-infected B cells is prevented by cytotoxic T cells.⁷ It is important to recognize that patients with EBV-positive DLBCL do not have a known immunodeficiency state; therefore, it has been postulated that EBV-positive DLBCL might be caused by age-related senescence of the immune system.^4,8

Epidemiology and Clinical Features
Epstein-Barr virus–positive DLBCL is more common in Asian countries than in Western countries, and there is a slight male predominance.⁶ A majority of patients present with extranodal disease at the time of diagnosis, and the skin is the most common extranodal site of involvement.^6,9 Rare cases of primary cutaneous involvement also have been described.^7,9,10 Cutaneous manifestations include erythematous papules and subcutaneous nodules. Other sites of extranodal involvement include the lungs, oral cavity, pharynx, gastrointestinal tract, and bone marrow.^8,9 However, EBV-positive DLBCL is an aggressive lymphoma and prognosis is poor irrespective of the primary site of involvement.

Histopathology
Two morphologic subtypes can be seen on histology. The polymorphic pattern is characterized by a broad range of B-cell maturation with admixed reactive cells (eg, lymphocytes, histiocytes, plasma cells). The monomorphic or large-cell pattern is characterized by monotonous sheets of large transformed B cells.^4,11 Many cases show both histologic patterns, and these morphologic variants do not impart any clinical or prognostic significance. Regardless of the histologic subtype, the neoplastic cells express pan B-cell antigens (eg, CD19, CD20, CD79a, PAX5), as well as MUM-1, BCL-2, and EBV-encoded RNA.⁴ Cases with plasmablastic features, as in our patient, may show weak or absent CD20 staining.¹² Detection of EBV by in situ hybridization is required for the diagnosis.

Diagnosis
Workup for a suspected cutaneous lymphoma should include a complete history and physical examination; laboratory studies; and relevant imaging evaluation such as computed tomography of the chest, abdomen, and pelvis with or without whole-body positron emission tomography. A bone marrow biopsy and aspirate also should be performed in all cutaneous lymphomas with intermediate to aggressive clinical behavior. Accurate staging evaluation is integral to confirm the absence of extracutaneous involvement and to provide prognostic and anatomic information for the appropriate selection of treatment.¹³

Prognosis and Management
Primary cutaneous lymphomas tend to have different clinical behaviors and prognoses compared to histologically similar systemic lymphomas; therefore, different therapeutic strategies are warranted.¹⁴ Epstein-Barr virus–positive DLBCL has an aggressive clinical course with a median survival of 2 years.⁸ Patients with EBV-positive DLBCL have a poorer overall survival and treatment response when compared to patients with EBV-negative DLBCLs.⁴ Primary cutaneous B-cell lymphomas with indolent behavior, such as primary cutaneous marginal zone lymphoma and primary cutaneous follicle center lymphoma, can be treated with surgical excision, radiation therapy, or observation.¹⁵ No standard treatment exists for EBV-positive DLBCL, but R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone), which is the standard treatment of primary cutaneous DLBCL, leg type, may provide a survival benefit.^13,15 Further studies are required to determine optimal treatment strategies.

Conclusion

Although rare, EBV-positive DLBCL is an important entity to consider when evaluating a patient with a suspected primary cutaneous lymphoma. Workup to rule out an underlying systemic lymphoma with relevant laboratory evaluation, imaging studies, and bone marrow biopsy is critical. Prognosis is poor and treatment is difficult, as standard treatment protocols have yet to be determined.

Sarcoidosis is a multiorgan, systemic, granulomatous disease that most commonly affects the cutaneous, pulmonary, ocular, and cardiac organ systems. Cutaneous involvement occurs in approximately 20% to 35% of patients, with approximately 25% of patients demonstrating only dermatologic findings.¹ Cutaneous sarcoidosis can have a highly variable presentation. Ichthyosiform sarcoidosis (IS) is a rare form of this disease that has been described as presenting as polygonal adherent scales.² It often is associated with internal organ involvement. We present a case of IS without any organ system involvement at the time of diagnosis. A review of the English-language literature was performed to ascertain the internal organ associations most commonly reported with IS.

Case Report

A 66-year-old black woman presented to dermatology with dark scaly patches noted by her primary care physician to be present on both of the lower extremities. The patient believed they were present for at least 4 years. She described dark spots confined to the lower legs that had gradually increased in size. Review of systems was negative for fever, chills, night sweats, weight loss, vision changes, cough, dyspnea, and joint pains, and there was no history of either personal or familial cutaneous diseases.

Physical examination revealed cutaneous patches of thin white scale with a sharp edge in arciform patterns on the lower extremities. Several of these patches were hyperpigmented and xerotic in appearance (Figure 1). The patches were limited to the lower legs, with no other lesions noted.

A punch biopsy of the skin on the right lower leg was performed. Histopathologic analysis showed epidermal compact hyperkeratosis with deep granulomatous infiltration into the subcutaneous tissue (Figures 2A and 2B). At high power, these granulomas were noted to be noncaseating naked granulomas composed of epithelioid histiocytes surrounded by sparse lymphocytic inflammation (Figure 2C). Special stains including acid-fast bacilli, Fite, and periodic acid–Schiff were negative. The diagnosis of IS was made based on clinical presentation and primarily by histopathologic analysis.

The patient’s cutaneous lesions were treated with fluocinonide ointment 0.05% twice daily. Although she did not notice a dramatic improvement in the plaques, they stabilized in size. Her primary care physician was notified and advised to begin a workup for involvement of other organ systems by sarcoidosis. Her initial evaluation, which included a chest radiograph and electrocardiogram, were unremarkable. Despite multiple attempts to persuade the patient to return for further follow-up, neither dermatology nor her primary care physician were able to complete a full workup.

 

Comment

Etiology
Although there are several theories regarding the etiology of sarcoidosis, the exact cause remains unknown. The body’s immune response, infectious agents, genetics, and the environment have all been thought to play a role. It has been well established that helper T cell (T_H1) production of interferon and increased levels of tumor necrosis factor propagate the inflammatory response seen in sarcoidosis.³ More recently, T_H17 cells have been found in cutaneous lesions, bronchoalveolar lavage samples, and the blood of patients with sarcoidosis, especially in those with active disease progression.³ Infectious agents such as mycobacteria and propionibacteria DNA or RNA also have been found in sarcoid samples.⁴ Several HLA-DRB1 variants have been associated with an increased incidence of sarcoidosis.⁵

Presentation
Characteristic dermatologic findings of sarcoidosis include macules, papules, nodules, and plaques located on the face, especially the nose, cheeks, and ears, and on the shins or ankles, as well as similar lesions around tattoos or scars. Sarcoid lesions also have been described as angiolupoid, lichenoid, annular, verrucous, ulcerative, and psoriasiform. Here we present an example of the uncommon type, ichthyosiform. Ichthyosiform sarcoidosis is a rare variant described primarily in dark-skinned individuals, a finding supported by both our case and prior reports. Most reported cases have described IS lesions as having a pasted-on appearance, with adherent centers on the extensor surfaces of the lower extremities, head, and/or neck.⁶ Our case follows this descriptive pattern previously reported with adherent patches limited to the lower extremities.

Histopathology
The key histopathologic finding is the presence of noncaseating granulomas on biopsy. Sarcoid “specific” lesions rest on the identification of the noncaseating granulomas, while “nonspecific” lesions such as erythema nodosum fail to demonstrate this finding.¹

Systemic Involvement
The IS type is believed to be an excellent marker for systemic disease, with approximately 95% of reported cases having some form of systemic illness.⁶ Acquired ichthyosis should warrant further investigation for systemic disease. Early recognition could be beneficial for the patient because the ichthyosiform type is believed to precede the diagnosis of systemic disease in most cases by a median of 3 months.⁶

The most common site of internal sarcoid involvement is the lungs, but the lymph nodes, eyes, liver, spleen, heart, and central nervous system also can be involved. Patients can present with nonspecific symptoms such as erythema nodosum in the skin, dyspnea, cough, chest pain, vision changes, enlarged lymph nodes, headaches, joint pain, fever, fatigue, weight loss, and malaise. According to a PubMed search of articles indexed for MEDLINE using the term ichthyosiform sarcoidosis, 16 cases have been reported in the English-language literature (eTable).^1,6-14 Of these 16 cases, 3 involved men and 13 involved women. The median age of a patient diagnosed with IS was 37 years. The respiratory system was found to be the most common organ system involved (14 of 16 patients), with hilar adenopathy and restrictive lung disease being the most common findings. Neurologic findings and hepatic involvement also were seen in 3 and 3 patients, respectively. Eight of 16 cases had an elevated serum angiotensin-converting enzyme level. Details of systemic involvement in other cases of IS are listed in the eTable.

 

Management
Most patients are given topical corticosteroids for their cutaneous lesions, but patients with systemic involvement will likely need some type of systemic immunosuppressive therapy to control their disease. Systemic therapy often is warranted in IS because of reports of rapid progression. Our case differs from these prior reports in the relative stability of the disease at the last patient encounter. Systemic treatment commonly includes oral corticosteroids such as prednisone. Other options, such as hydroxychloroquine, methotrexate, azathioprine, pentoxifylline, thalidomide, cyclophosphamide, cyclosporine, and infliximab, can be considered if other treatments fail.¹³ Ichthyosiform sarcoidosis patients should continue to have regular follow-up to monitor for disease progression.

Differential
When evaluating an acquired ichthyosis, dermatologists can consider other associations such as Hodgkin disease, hypothyroidism, multiple myeloma, carcinomatosis, and chronic malnutrition.¹ Skin biopsy demonstrating granuloma formation also is not specific for sarcoidosis. Other etiologies, such as autoimmune diseases, immunodeficiency disorders, infections, foreign body granulomas, neoplasms, and drug reactions, should be considered.¹⁵ All patients with acquired ichthyosis should undergo a thorough evaluation for internal involvement.

Conclusion

We presented a case of IS, a rare type of sarcoidosis commonly associated with further internal involvement of the respiratory, nervous, or hepatic organ systems. Recognition of an acquired form of ichthyosis and its potential disease associations, including sarcoidosis, is important to improve early detection of any internal disease, allowing prompt initiation of treatment.

Sarcoidosis is a multiorgan, systemic, granulomatous disease that most commonly affects the cutaneous, pulmonary, ocular, and cardiac organ systems. Cutaneous involvement occurs in approximately 20% to 35% of patients, with approximately 25% of patients demonstrating only dermatologic findings.¹ Cutaneous sarcoidosis can have a highly variable presentation. Ichthyosiform sarcoidosis (IS) is a rare form of this disease that has been described as presenting as polygonal adherent scales.² It often is associated with internal organ involvement. We present a case of IS without any organ system involvement at the time of diagnosis. A review of the English-language literature was performed to ascertain the internal organ associations most commonly reported with IS.

Case Report

A 66-year-old black woman presented to dermatology with dark scaly patches noted by her primary care physician to be present on both of the lower extremities. The patient believed they were present for at least 4 years. She described dark spots confined to the lower legs that had gradually increased in size. Review of systems was negative for fever, chills, night sweats, weight loss, vision changes, cough, dyspnea, and joint pains, and there was no history of either personal or familial cutaneous diseases.

Physical examination revealed cutaneous patches of thin white scale with a sharp edge in arciform patterns on the lower extremities. Several of these patches were hyperpigmented and xerotic in appearance (Figure 1). The patches were limited to the lower legs, with no other lesions noted.

A punch biopsy of the skin on the right lower leg was performed. Histopathologic analysis showed epidermal compact hyperkeratosis with deep granulomatous infiltration into the subcutaneous tissue (Figures 2A and 2B). At high power, these granulomas were noted to be noncaseating naked granulomas composed of epithelioid histiocytes surrounded by sparse lymphocytic inflammation (Figure 2C). Special stains including acid-fast bacilli, Fite, and periodic acid–Schiff were negative. The diagnosis of IS was made based on clinical presentation and primarily by histopathologic analysis.

The patient’s cutaneous lesions were treated with fluocinonide ointment 0.05% twice daily. Although she did not notice a dramatic improvement in the plaques, they stabilized in size. Her primary care physician was notified and advised to begin a workup for involvement of other organ systems by sarcoidosis. Her initial evaluation, which included a chest radiograph and electrocardiogram, were unremarkable. Despite multiple attempts to persuade the patient to return for further follow-up, neither dermatology nor her primary care physician were able to complete a full workup.

 

Comment

Etiology
Although there are several theories regarding the etiology of sarcoidosis, the exact cause remains unknown. The body’s immune response, infectious agents, genetics, and the environment have all been thought to play a role. It has been well established that helper T cell (T_H1) production of interferon and increased levels of tumor necrosis factor propagate the inflammatory response seen in sarcoidosis.³ More recently, T_H17 cells have been found in cutaneous lesions, bronchoalveolar lavage samples, and the blood of patients with sarcoidosis, especially in those with active disease progression.³ Infectious agents such as mycobacteria and propionibacteria DNA or RNA also have been found in sarcoid samples.⁴ Several HLA-DRB1 variants have been associated with an increased incidence of sarcoidosis.⁵

Presentation
Characteristic dermatologic findings of sarcoidosis include macules, papules, nodules, and plaques located on the face, especially the nose, cheeks, and ears, and on the shins or ankles, as well as similar lesions around tattoos or scars. Sarcoid lesions also have been described as angiolupoid, lichenoid, annular, verrucous, ulcerative, and psoriasiform. Here we present an example of the uncommon type, ichthyosiform. Ichthyosiform sarcoidosis is a rare variant described primarily in dark-skinned individuals, a finding supported by both our case and prior reports. Most reported cases have described IS lesions as having a pasted-on appearance, with adherent centers on the extensor surfaces of the lower extremities, head, and/or neck.⁶ Our case follows this descriptive pattern previously reported with adherent patches limited to the lower extremities.

Histopathology
The key histopathologic finding is the presence of noncaseating granulomas on biopsy. Sarcoid “specific” lesions rest on the identification of the noncaseating granulomas, while “nonspecific” lesions such as erythema nodosum fail to demonstrate this finding.¹

Systemic Involvement
The IS type is believed to be an excellent marker for systemic disease, with approximately 95% of reported cases having some form of systemic illness.⁶ Acquired ichthyosis should warrant further investigation for systemic disease. Early recognition could be beneficial for the patient because the ichthyosiform type is believed to precede the diagnosis of systemic disease in most cases by a median of 3 months.⁶

The most common site of internal sarcoid involvement is the lungs, but the lymph nodes, eyes, liver, spleen, heart, and central nervous system also can be involved. Patients can present with nonspecific symptoms such as erythema nodosum in the skin, dyspnea, cough, chest pain, vision changes, enlarged lymph nodes, headaches, joint pain, fever, fatigue, weight loss, and malaise. According to a PubMed search of articles indexed for MEDLINE using the term ichthyosiform sarcoidosis, 16 cases have been reported in the English-language literature (eTable).^1,6-14 Of these 16 cases, 3 involved men and 13 involved women. The median age of a patient diagnosed with IS was 37 years. The respiratory system was found to be the most common organ system involved (14 of 16 patients), with hilar adenopathy and restrictive lung disease being the most common findings. Neurologic findings and hepatic involvement also were seen in 3 and 3 patients, respectively. Eight of 16 cases had an elevated serum angiotensin-converting enzyme level. Details of systemic involvement in other cases of IS are listed in the eTable.

 

Management
Most patients are given topical corticosteroids for their cutaneous lesions, but patients with systemic involvement will likely need some type of systemic immunosuppressive therapy to control their disease. Systemic therapy often is warranted in IS because of reports of rapid progression. Our case differs from these prior reports in the relative stability of the disease at the last patient encounter. Systemic treatment commonly includes oral corticosteroids such as prednisone. Other options, such as hydroxychloroquine, methotrexate, azathioprine, pentoxifylline, thalidomide, cyclophosphamide, cyclosporine, and infliximab, can be considered if other treatments fail.¹³ Ichthyosiform sarcoidosis patients should continue to have regular follow-up to monitor for disease progression.

Differential
When evaluating an acquired ichthyosis, dermatologists can consider other associations such as Hodgkin disease, hypothyroidism, multiple myeloma, carcinomatosis, and chronic malnutrition.¹ Skin biopsy demonstrating granuloma formation also is not specific for sarcoidosis. Other etiologies, such as autoimmune diseases, immunodeficiency disorders, infections, foreign body granulomas, neoplasms, and drug reactions, should be considered.¹⁵ All patients with acquired ichthyosis should undergo a thorough evaluation for internal involvement.

Conclusion

We presented a case of IS, a rare type of sarcoidosis commonly associated with further internal involvement of the respiratory, nervous, or hepatic organ systems. Recognition of an acquired form of ichthyosis and its potential disease associations, including sarcoidosis, is important to improve early detection of any internal disease, allowing prompt initiation of treatment.

Sarcoidosis is a multiorgan, systemic, granulomatous disease that most commonly affects the cutaneous, pulmonary, ocular, and cardiac organ systems. Cutaneous involvement occurs in approximately 20% to 35% of patients, with approximately 25% of patients demonstrating only dermatologic findings.¹ Cutaneous sarcoidosis can have a highly variable presentation. Ichthyosiform sarcoidosis (IS) is a rare form of this disease that has been described as presenting as polygonal adherent scales.² It often is associated with internal organ involvement. We present a case of IS without any organ system involvement at the time of diagnosis. A review of the English-language literature was performed to ascertain the internal organ associations most commonly reported with IS.

Case Report

A 66-year-old black woman presented to dermatology with dark scaly patches noted by her primary care physician to be present on both of the lower extremities. The patient believed they were present for at least 4 years. She described dark spots confined to the lower legs that had gradually increased in size. Review of systems was negative for fever, chills, night sweats, weight loss, vision changes, cough, dyspnea, and joint pains, and there was no history of either personal or familial cutaneous diseases.

Physical examination revealed cutaneous patches of thin white scale with a sharp edge in arciform patterns on the lower extremities. Several of these patches were hyperpigmented and xerotic in appearance (Figure 1). The patches were limited to the lower legs, with no other lesions noted.

A punch biopsy of the skin on the right lower leg was performed. Histopathologic analysis showed epidermal compact hyperkeratosis with deep granulomatous infiltration into the subcutaneous tissue (Figures 2A and 2B). At high power, these granulomas were noted to be noncaseating naked granulomas composed of epithelioid histiocytes surrounded by sparse lymphocytic inflammation (Figure 2C). Special stains including acid-fast bacilli, Fite, and periodic acid–Schiff were negative. The diagnosis of IS was made based on clinical presentation and primarily by histopathologic analysis.

The patient’s cutaneous lesions were treated with fluocinonide ointment 0.05% twice daily. Although she did not notice a dramatic improvement in the plaques, they stabilized in size. Her primary care physician was notified and advised to begin a workup for involvement of other organ systems by sarcoidosis. Her initial evaluation, which included a chest radiograph and electrocardiogram, were unremarkable. Despite multiple attempts to persuade the patient to return for further follow-up, neither dermatology nor her primary care physician were able to complete a full workup.

 

Comment

Etiology
Although there are several theories regarding the etiology of sarcoidosis, the exact cause remains unknown. The body’s immune response, infectious agents, genetics, and the environment have all been thought to play a role. It has been well established that helper T cell (T_H1) production of interferon and increased levels of tumor necrosis factor propagate the inflammatory response seen in sarcoidosis.³ More recently, T_H17 cells have been found in cutaneous lesions, bronchoalveolar lavage samples, and the blood of patients with sarcoidosis, especially in those with active disease progression.³ Infectious agents such as mycobacteria and propionibacteria DNA or RNA also have been found in sarcoid samples.⁴ Several HLA-DRB1 variants have been associated with an increased incidence of sarcoidosis.⁵

Presentation
Characteristic dermatologic findings of sarcoidosis include macules, papules, nodules, and plaques located on the face, especially the nose, cheeks, and ears, and on the shins or ankles, as well as similar lesions around tattoos or scars. Sarcoid lesions also have been described as angiolupoid, lichenoid, annular, verrucous, ulcerative, and psoriasiform. Here we present an example of the uncommon type, ichthyosiform. Ichthyosiform sarcoidosis is a rare variant described primarily in dark-skinned individuals, a finding supported by both our case and prior reports. Most reported cases have described IS lesions as having a pasted-on appearance, with adherent centers on the extensor surfaces of the lower extremities, head, and/or neck.⁶ Our case follows this descriptive pattern previously reported with adherent patches limited to the lower extremities.

Histopathology
The key histopathologic finding is the presence of noncaseating granulomas on biopsy. Sarcoid “specific” lesions rest on the identification of the noncaseating granulomas, while “nonspecific” lesions such as erythema nodosum fail to demonstrate this finding.¹

Systemic Involvement
The IS type is believed to be an excellent marker for systemic disease, with approximately 95% of reported cases having some form of systemic illness.⁶ Acquired ichthyosis should warrant further investigation for systemic disease. Early recognition could be beneficial for the patient because the ichthyosiform type is believed to precede the diagnosis of systemic disease in most cases by a median of 3 months.⁶

The most common site of internal sarcoid involvement is the lungs, but the lymph nodes, eyes, liver, spleen, heart, and central nervous system also can be involved. Patients can present with nonspecific symptoms such as erythema nodosum in the skin, dyspnea, cough, chest pain, vision changes, enlarged lymph nodes, headaches, joint pain, fever, fatigue, weight loss, and malaise. According to a PubMed search of articles indexed for MEDLINE using the term ichthyosiform sarcoidosis, 16 cases have been reported in the English-language literature (eTable).^1,6-14 Of these 16 cases, 3 involved men and 13 involved women. The median age of a patient diagnosed with IS was 37 years. The respiratory system was found to be the most common organ system involved (14 of 16 patients), with hilar adenopathy and restrictive lung disease being the most common findings. Neurologic findings and hepatic involvement also were seen in 3 and 3 patients, respectively. Eight of 16 cases had an elevated serum angiotensin-converting enzyme level. Details of systemic involvement in other cases of IS are listed in the eTable.

 

Management
Most patients are given topical corticosteroids for their cutaneous lesions, but patients with systemic involvement will likely need some type of systemic immunosuppressive therapy to control their disease. Systemic therapy often is warranted in IS because of reports of rapid progression. Our case differs from these prior reports in the relative stability of the disease at the last patient encounter. Systemic treatment commonly includes oral corticosteroids such as prednisone. Other options, such as hydroxychloroquine, methotrexate, azathioprine, pentoxifylline, thalidomide, cyclophosphamide, cyclosporine, and infliximab, can be considered if other treatments fail.¹³ Ichthyosiform sarcoidosis patients should continue to have regular follow-up to monitor for disease progression.

Differential
When evaluating an acquired ichthyosis, dermatologists can consider other associations such as Hodgkin disease, hypothyroidism, multiple myeloma, carcinomatosis, and chronic malnutrition.¹ Skin biopsy demonstrating granuloma formation also is not specific for sarcoidosis. Other etiologies, such as autoimmune diseases, immunodeficiency disorders, infections, foreign body granulomas, neoplasms, and drug reactions, should be considered.¹⁵ All patients with acquired ichthyosis should undergo a thorough evaluation for internal involvement.

Conclusion

We presented a case of IS, a rare type of sarcoidosis commonly associated with further internal involvement of the respiratory, nervous, or hepatic organ systems. Recognition of an acquired form of ichthyosis and its potential disease associations, including sarcoidosis, is important to improve early detection of any internal disease, allowing prompt initiation of treatment.

Crizotinib is a multitargeted tyrosine kinase inhibitor that blocks anaplastic lymphoma kinase (ALK), hepatocyte growth factor receptor (c-Met), and their oncogenic variants ALK fusion proteins or c-Met/hepatocyte growth factor receptor mutant variants.¹ Additionally, crizotinib was approved by the US Food and Drug Administration in 2011 for the treatment of patients with non–small cell lung cancer (NSCLC) whose tumors are echinoderm microtubule-associated proteinlike 4 (EML4)/ALK or ROS1 positive.^2,3 Among unselected populations of patients with NSCLC, the frequency of EML4/ALK rearrangements ranges from 1.5% to 6.7%.¹ Crizotinib is superior to standard chemotherapy in patients with ALK-positive NSCLC.²

In clinical trials, adverse reactions (grades 1 to 4) to crizotinib occurring in at least 25% of patients included visual disturbances, gastrointestinal tract disorders, fatigue, and pitting edema.^1,2,4 Adverse reactions (grades 3 and 4) occurring in more than 5% of patients included elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, dyspnea, pneumonia, and neutropenia.^1,4 Although the incidence of dermatologic adverse reactions is approximately 11%, substantial progression of drug eruptions rarely has been reported.^2,5 We describe a case of lichenoid drug eruption (LDE) that appeared 4 weeks after initiation of crizotinib treatment in a patient with ALK-positive metastatic lung adenocarcinoma.

Case Report

A 61-year-old man presented with a history of ALK-positive NSCLC with lung-to-lung metastasis and pleural seeding treated with a right lower lobectomy and chemotherapy 9 years prior. Chemotherapy was reattempted 5 years later. Targeted therapy with gefitinib was initiated following the lobectomy and 5 years later with erlotinib. The NSCLC was stable, as indicated by computed tomography performed once every 3 or 6 months. After 5 years of treatment, follow-up computed tomography showed slowly growing nodular shadows in the right middle and lower lung fields. Due to this disease progression, treatment with crizotinib (250 mg twice daily) was initiated. Four weeks after the initiation of crizotinib therapy, mild itchy skin eruptions developed on all extremities and the lower lip. He also reported that the skin lesions became more itchy and red with sun exposure. He had no history of drug allergies and denied taking any other medications.

Physical examination revealed multiple brown to violaceous, slightly scaly, flat-topped polygonal papules or plaques on both lower legs (Figure 1A), dorsal hands (Figure 1B), and extensor sites of the elbows, as well as lacelike fine white lines on the lower lip (Figure 1C). There were no nail lesions. The patient’s dermatologic history was unremarkable, except for a few vitiligo lesions on the dorsal hands, extensor sites of the elbows, and mouth angles diagnosed 20 years earlier.

A skin biopsy from the right dorsal hand revealed a lichenoid infiltrate in the superficial dermis composed of lymphocytes, histiocytes and scattered eosinophils, focal parakeratosis, focal hypergranulosis, mild acanthosis, and basal vacuolization (Figure 2A). In addition, some dyskeratotic keratinocytes in the stratum spinosum and granulosum were identified (Figure 2B). The histopathology was consistent with the diagnosis of an LDE. Direct immunofluorescence revealed no globular or cytoid body–like deposits of immunoglobulin, with IgM, IgA, IgG, or C3 in the epidermis, dermis, and basement membrane zone. Routine laboratory studies revealed elevated liver enzymes, including an ALT level of 115 U/L (reference range, 0–40 U/L) and AST level of 60 U/L (reference range, 5–45 U/L). Negative results for the serum hepatitis B surface antigen and anti– hepatitis C virus tests were recorded. The patient had no medical history of alcohol consumption or abnormal liver function tests. The skin lesions were treated with diflucortolone valerate fatty ointment 0.1% twice daily and abnormal liver functions were treated with silymarin (150 mg per cap twice daily). He experienced some improvement.

A causality assessment was performed using the Naranjo Adverse Drug Reaction Probability Scale,^6,7 and we concluded that crizotinib was the possible cause (Naranjo score, 4) of this adverse drug reaction (Table). Because the skin reaction was tolerable and liver enzymes were mildly elevated (ALT, 50 U/L; AST, 48 U/L), the offending drug was continued to benefit the underlying disease. His NSCLC was stable on computed tomography 3 months later.

 

Comment

The number of indicated uses of crizotinib, an oral small-molecule ALK tyrosine kinase inhibitor for the treatment of NSCLC, has gradually increased, but only a few cases of cutaneous adverse reactions, such as erythema multiforme and severe photosensitivity dermatitis, have been reported.^2,5 Skin toxicity is a common and well-known side effect of other small-molecule tyrosine kinase inhibitors, particularly epidermal growth factor receptor inhibitors.⁸ However, LDE is not commonly associated with small-molecule tyrosine kinase inhibitors, though it has been described in a few patients taking imatinib for chronic myelogenous leukemia and gastrointestinal tract stromal tumors.^9,10

The clinical morphology of LDE may resemble lichen planus, but certain features, such as larger skin lesions, the absence of Wickham striae, and photodistribution, help to differentiate between the two.¹⁰ Histologically, some findings are more common in LDE, including focal parakeratosis, cytoid bodies in the cornified and granular layers, and the presence of eosinophils.¹¹

Our patient developed lichenoid rashes after 1 month of crizotinib therapy. The latency period for developing a medication-induced LDE varies from months to 1 year and is dependent on the dosage, host response, prior exposure, and concomitant drug administration. No additional medications had been added to our patient’s regimen after initiating crizotinib therapy, and he did not take any other known medications. Ultimately, based on the time-event relationship, morphology, distribution, and histopathologic findings, we concluded that our patient developed an LDE due to crizotinib.

Our patient also had a history of vitiligo affecting the hands, elbows, and mouth angles for 20 years. Although there are limited reports of a possible causal link between lichen planus or drug-induced lichen planus eruption and vitiligo,^12-14 we do not think these conditions were associated in our case because the patient’s vitiligo lesions persisted for many years, did not progress, and remained inactive and stable, and there was a lack of co-localization of LDE and vitiligo.

Our patient reported that the skin eruptions worsened after sun exposure. Oser and Janne⁵ also reported a patient with ALK-positive metastatic lung adenocarcinoma who developed severe crizotinib-induced photosensitive rashes. Further accumulation of similar cases and pathophysiological studies will be necessary to clarify whether this photosensitivity dermatitis is caused by ALK inhibition itself or mediated through host-immune mechanisms.⁵

Conclusion

As crizotinib prescriptions for patients with NSCLC are increasing, clinicians should be aware of the possibility of cutaneous LDEs occurring as an adverse effect. Additionally, physicians should treat appropriately to avoid unnecessarily discontinuing a potentially life-saving medication and to improve quality of life for patients with NSCLC who are treated with crizotinib.

Crizotinib is a multitargeted tyrosine kinase inhibitor that blocks anaplastic lymphoma kinase (ALK), hepatocyte growth factor receptor (c-Met), and their oncogenic variants ALK fusion proteins or c-Met/hepatocyte growth factor receptor mutant variants.¹ Additionally, crizotinib was approved by the US Food and Drug Administration in 2011 for the treatment of patients with non–small cell lung cancer (NSCLC) whose tumors are echinoderm microtubule-associated proteinlike 4 (EML4)/ALK or ROS1 positive.^2,3 Among unselected populations of patients with NSCLC, the frequency of EML4/ALK rearrangements ranges from 1.5% to 6.7%.¹ Crizotinib is superior to standard chemotherapy in patients with ALK-positive NSCLC.²

In clinical trials, adverse reactions (grades 1 to 4) to crizotinib occurring in at least 25% of patients included visual disturbances, gastrointestinal tract disorders, fatigue, and pitting edema.^1,2,4 Adverse reactions (grades 3 and 4) occurring in more than 5% of patients included elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, dyspnea, pneumonia, and neutropenia.^1,4 Although the incidence of dermatologic adverse reactions is approximately 11%, substantial progression of drug eruptions rarely has been reported.^2,5 We describe a case of lichenoid drug eruption (LDE) that appeared 4 weeks after initiation of crizotinib treatment in a patient with ALK-positive metastatic lung adenocarcinoma.

Case Report

A 61-year-old man presented with a history of ALK-positive NSCLC with lung-to-lung metastasis and pleural seeding treated with a right lower lobectomy and chemotherapy 9 years prior. Chemotherapy was reattempted 5 years later. Targeted therapy with gefitinib was initiated following the lobectomy and 5 years later with erlotinib. The NSCLC was stable, as indicated by computed tomography performed once every 3 or 6 months. After 5 years of treatment, follow-up computed tomography showed slowly growing nodular shadows in the right middle and lower lung fields. Due to this disease progression, treatment with crizotinib (250 mg twice daily) was initiated. Four weeks after the initiation of crizotinib therapy, mild itchy skin eruptions developed on all extremities and the lower lip. He also reported that the skin lesions became more itchy and red with sun exposure. He had no history of drug allergies and denied taking any other medications.

Physical examination revealed multiple brown to violaceous, slightly scaly, flat-topped polygonal papules or plaques on both lower legs (Figure 1A), dorsal hands (Figure 1B), and extensor sites of the elbows, as well as lacelike fine white lines on the lower lip (Figure 1C). There were no nail lesions. The patient’s dermatologic history was unremarkable, except for a few vitiligo lesions on the dorsal hands, extensor sites of the elbows, and mouth angles diagnosed 20 years earlier.

A skin biopsy from the right dorsal hand revealed a lichenoid infiltrate in the superficial dermis composed of lymphocytes, histiocytes and scattered eosinophils, focal parakeratosis, focal hypergranulosis, mild acanthosis, and basal vacuolization (Figure 2A). In addition, some dyskeratotic keratinocytes in the stratum spinosum and granulosum were identified (Figure 2B). The histopathology was consistent with the diagnosis of an LDE. Direct immunofluorescence revealed no globular or cytoid body–like deposits of immunoglobulin, with IgM, IgA, IgG, or C3 in the epidermis, dermis, and basement membrane zone. Routine laboratory studies revealed elevated liver enzymes, including an ALT level of 115 U/L (reference range, 0–40 U/L) and AST level of 60 U/L (reference range, 5–45 U/L). Negative results for the serum hepatitis B surface antigen and anti– hepatitis C virus tests were recorded. The patient had no medical history of alcohol consumption or abnormal liver function tests. The skin lesions were treated with diflucortolone valerate fatty ointment 0.1% twice daily and abnormal liver functions were treated with silymarin (150 mg per cap twice daily). He experienced some improvement.

A causality assessment was performed using the Naranjo Adverse Drug Reaction Probability Scale,^6,7 and we concluded that crizotinib was the possible cause (Naranjo score, 4) of this adverse drug reaction (Table). Because the skin reaction was tolerable and liver enzymes were mildly elevated (ALT, 50 U/L; AST, 48 U/L), the offending drug was continued to benefit the underlying disease. His NSCLC was stable on computed tomography 3 months later.

 

Comment

The number of indicated uses of crizotinib, an oral small-molecule ALK tyrosine kinase inhibitor for the treatment of NSCLC, has gradually increased, but only a few cases of cutaneous adverse reactions, such as erythema multiforme and severe photosensitivity dermatitis, have been reported.^2,5 Skin toxicity is a common and well-known side effect of other small-molecule tyrosine kinase inhibitors, particularly epidermal growth factor receptor inhibitors.⁸ However, LDE is not commonly associated with small-molecule tyrosine kinase inhibitors, though it has been described in a few patients taking imatinib for chronic myelogenous leukemia and gastrointestinal tract stromal tumors.^9,10

The clinical morphology of LDE may resemble lichen planus, but certain features, such as larger skin lesions, the absence of Wickham striae, and photodistribution, help to differentiate between the two.¹⁰ Histologically, some findings are more common in LDE, including focal parakeratosis, cytoid bodies in the cornified and granular layers, and the presence of eosinophils.¹¹

Our patient developed lichenoid rashes after 1 month of crizotinib therapy. The latency period for developing a medication-induced LDE varies from months to 1 year and is dependent on the dosage, host response, prior exposure, and concomitant drug administration. No additional medications had been added to our patient’s regimen after initiating crizotinib therapy, and he did not take any other known medications. Ultimately, based on the time-event relationship, morphology, distribution, and histopathologic findings, we concluded that our patient developed an LDE due to crizotinib.

Our patient also had a history of vitiligo affecting the hands, elbows, and mouth angles for 20 years. Although there are limited reports of a possible causal link between lichen planus or drug-induced lichen planus eruption and vitiligo,^12-14 we do not think these conditions were associated in our case because the patient’s vitiligo lesions persisted for many years, did not progress, and remained inactive and stable, and there was a lack of co-localization of LDE and vitiligo.

Our patient reported that the skin eruptions worsened after sun exposure. Oser and Janne⁵ also reported a patient with ALK-positive metastatic lung adenocarcinoma who developed severe crizotinib-induced photosensitive rashes. Further accumulation of similar cases and pathophysiological studies will be necessary to clarify whether this photosensitivity dermatitis is caused by ALK inhibition itself or mediated through host-immune mechanisms.⁵

Conclusion

As crizotinib prescriptions for patients with NSCLC are increasing, clinicians should be aware of the possibility of cutaneous LDEs occurring as an adverse effect. Additionally, physicians should treat appropriately to avoid unnecessarily discontinuing a potentially life-saving medication and to improve quality of life for patients with NSCLC who are treated with crizotinib.

Crizotinib is a multitargeted tyrosine kinase inhibitor that blocks anaplastic lymphoma kinase (ALK), hepatocyte growth factor receptor (c-Met), and their oncogenic variants ALK fusion proteins or c-Met/hepatocyte growth factor receptor mutant variants.¹ Additionally, crizotinib was approved by the US Food and Drug Administration in 2011 for the treatment of patients with non–small cell lung cancer (NSCLC) whose tumors are echinoderm microtubule-associated proteinlike 4 (EML4)/ALK or ROS1 positive.^2,3 Among unselected populations of patients with NSCLC, the frequency of EML4/ALK rearrangements ranges from 1.5% to 6.7%.¹ Crizotinib is superior to standard chemotherapy in patients with ALK-positive NSCLC.²

In clinical trials, adverse reactions (grades 1 to 4) to crizotinib occurring in at least 25% of patients included visual disturbances, gastrointestinal tract disorders, fatigue, and pitting edema.^1,2,4 Adverse reactions (grades 3 and 4) occurring in more than 5% of patients included elevated alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, dyspnea, pneumonia, and neutropenia.^1,4 Although the incidence of dermatologic adverse reactions is approximately 11%, substantial progression of drug eruptions rarely has been reported.^2,5 We describe a case of lichenoid drug eruption (LDE) that appeared 4 weeks after initiation of crizotinib treatment in a patient with ALK-positive metastatic lung adenocarcinoma.

Case Report

A 61-year-old man presented with a history of ALK-positive NSCLC with lung-to-lung metastasis and pleural seeding treated with a right lower lobectomy and chemotherapy 9 years prior. Chemotherapy was reattempted 5 years later. Targeted therapy with gefitinib was initiated following the lobectomy and 5 years later with erlotinib. The NSCLC was stable, as indicated by computed tomography performed once every 3 or 6 months. After 5 years of treatment, follow-up computed tomography showed slowly growing nodular shadows in the right middle and lower lung fields. Due to this disease progression, treatment with crizotinib (250 mg twice daily) was initiated. Four weeks after the initiation of crizotinib therapy, mild itchy skin eruptions developed on all extremities and the lower lip. He also reported that the skin lesions became more itchy and red with sun exposure. He had no history of drug allergies and denied taking any other medications.

Physical examination revealed multiple brown to violaceous, slightly scaly, flat-topped polygonal papules or plaques on both lower legs (Figure 1A), dorsal hands (Figure 1B), and extensor sites of the elbows, as well as lacelike fine white lines on the lower lip (Figure 1C). There were no nail lesions. The patient’s dermatologic history was unremarkable, except for a few vitiligo lesions on the dorsal hands, extensor sites of the elbows, and mouth angles diagnosed 20 years earlier.

A skin biopsy from the right dorsal hand revealed a lichenoid infiltrate in the superficial dermis composed of lymphocytes, histiocytes and scattered eosinophils, focal parakeratosis, focal hypergranulosis, mild acanthosis, and basal vacuolization (Figure 2A). In addition, some dyskeratotic keratinocytes in the stratum spinosum and granulosum were identified (Figure 2B). The histopathology was consistent with the diagnosis of an LDE. Direct immunofluorescence revealed no globular or cytoid body–like deposits of immunoglobulin, with IgM, IgA, IgG, or C3 in the epidermis, dermis, and basement membrane zone. Routine laboratory studies revealed elevated liver enzymes, including an ALT level of 115 U/L (reference range, 0–40 U/L) and AST level of 60 U/L (reference range, 5–45 U/L). Negative results for the serum hepatitis B surface antigen and anti– hepatitis C virus tests were recorded. The patient had no medical history of alcohol consumption or abnormal liver function tests. The skin lesions were treated with diflucortolone valerate fatty ointment 0.1% twice daily and abnormal liver functions were treated with silymarin (150 mg per cap twice daily). He experienced some improvement.

A causality assessment was performed using the Naranjo Adverse Drug Reaction Probability Scale,^6,7 and we concluded that crizotinib was the possible cause (Naranjo score, 4) of this adverse drug reaction (Table). Because the skin reaction was tolerable and liver enzymes were mildly elevated (ALT, 50 U/L; AST, 48 U/L), the offending drug was continued to benefit the underlying disease. His NSCLC was stable on computed tomography 3 months later.

 

Comment

The number of indicated uses of crizotinib, an oral small-molecule ALK tyrosine kinase inhibitor for the treatment of NSCLC, has gradually increased, but only a few cases of cutaneous adverse reactions, such as erythema multiforme and severe photosensitivity dermatitis, have been reported.^2,5 Skin toxicity is a common and well-known side effect of other small-molecule tyrosine kinase inhibitors, particularly epidermal growth factor receptor inhibitors.⁸ However, LDE is not commonly associated with small-molecule tyrosine kinase inhibitors, though it has been described in a few patients taking imatinib for chronic myelogenous leukemia and gastrointestinal tract stromal tumors.^9,10

The clinical morphology of LDE may resemble lichen planus, but certain features, such as larger skin lesions, the absence of Wickham striae, and photodistribution, help to differentiate between the two.¹⁰ Histologically, some findings are more common in LDE, including focal parakeratosis, cytoid bodies in the cornified and granular layers, and the presence of eosinophils.¹¹

Our patient developed lichenoid rashes after 1 month of crizotinib therapy. The latency period for developing a medication-induced LDE varies from months to 1 year and is dependent on the dosage, host response, prior exposure, and concomitant drug administration. No additional medications had been added to our patient’s regimen after initiating crizotinib therapy, and he did not take any other known medications. Ultimately, based on the time-event relationship, morphology, distribution, and histopathologic findings, we concluded that our patient developed an LDE due to crizotinib.

Our patient also had a history of vitiligo affecting the hands, elbows, and mouth angles for 20 years. Although there are limited reports of a possible causal link between lichen planus or drug-induced lichen planus eruption and vitiligo,^12-14 we do not think these conditions were associated in our case because the patient’s vitiligo lesions persisted for many years, did not progress, and remained inactive and stable, and there was a lack of co-localization of LDE and vitiligo.

Our patient reported that the skin eruptions worsened after sun exposure. Oser and Janne⁵ also reported a patient with ALK-positive metastatic lung adenocarcinoma who developed severe crizotinib-induced photosensitive rashes. Further accumulation of similar cases and pathophysiological studies will be necessary to clarify whether this photosensitivity dermatitis is caused by ALK inhibition itself or mediated through host-immune mechanisms.⁵

Conclusion

As crizotinib prescriptions for patients with NSCLC are increasing, clinicians should be aware of the possibility of cutaneous LDEs occurring as an adverse effect. Additionally, physicians should treat appropriately to avoid unnecessarily discontinuing a potentially life-saving medication and to improve quality of life for patients with NSCLC who are treated with crizotinib.