 |
|
|
 |
Angry Fat To the Editor: A 23-year-old Maori man presented in January 2004, with a 2-week history of exercise intolerance secondary to painful, swollen legs. He reported sweats and weight loss (10 kg). He was febrile (39.3°C) and had a nodular rash over the lower legs consistent with erythema nodosum. Blood tests showed anemia and neutropenia: hemoglobin 106 g/l, white blood cell count 2.6 ´ 109/l (neutrophils 62%, lymphocytes 35%), platelets 223 ´ 109/l; increased C-reactive protein (56 mg/l), and mildly increased transaminases. Serology demonstrated previous contact with hepatitis B, Epstein-Barr, and cytomegaloviruses. Mantoux test, midstream urine, serum creatine kinase, a1-antitrypsin, amylase, and computerized tomography (CT) of the chest were normal. After a 2-week course of indomethacin his rash resolved, but he was readmitted one month later with ongoing leg pain and fever. CT of the abdomen and pelvis showed widespread increased signal in the subcutaneous and intraperitoneal fat consistent with generalized panniculitis (Figure 1A). Vasculitis serology, bone marrow examination, blood cultures, echocardiography, and urine microscopy were negative. Over the next 2 months his symptoms progressed; he became immobile and lost more weight (10 kg). The plantar aspects of the feet were severely tender and there was moderate weakness and wasting of the hip and shoulder girdles. His deterioration was complicated by a lobar pneumonia and acute respiratory failure for which he required intensive care treatment. Magnetic resonance imaging (MRI) showed circumferential subcutaneous panniculitis involving the thighs and lower abdominal wall with no muscle involvement (Figure 1B). A full-thickness biopsy (MRI-guided) from normal-looking thigh skin showed diffuse infiltration of the subcutaneous lobules by small T lymphocytes (CD3, CD56 negative) without obvious atypia, accompanied by histiocytes (CD68 positive) and small numbers of B lymphocytes (CD20 positive) (Figure 1C). The lymphocytes rimmed intact fat cells, a feature of lymphoma, but there were no cytophagic features. T cell gene rearrangement analysis of abnormal tissue taken from the paraffin block showed no evidence of monoclonality. A muscle biopsy from vastus lateralis showed myofiber necrosis with mild inflammatory changes around the microvasculature of the surrounding connective tissue. In summary, the findings were lobular lymphocytic panniculitis in fat and ischemic necrosis in muscle. The patient was started on pulsed methylprednisone followed by high dose prednisone therapy and he rapidly improved. Over the next 12 months, on a combination of methotrexate and tapered prednisone, his symptoms resolved and his weight normalized. On at least 2 occasions he relapsed when he omitted his medications.
Panniculitis has never been regarded as a primary condition, with the exception of Weber-Christian disease, which is of uncertain etiology and now only of historical interest. It is usually a manifestation of another illness1. For example, erythema nodosum (the commonest form of panniculitis) secondary to streptococcal infection. The pathological classification has always been descriptive, using a mixture of geographic features, (septal or lobular or both); the infiltrating cell type; (neutrophils, lymphocytes, histiocytes, or mixed) and other characteristics such as vasculitis, calcification, granulomata, hemorrhage, cytophagia, or necrosis. Some of these features are specific for certain etiologies, for example acute neutrophilic panniculitis and a1 antitrypsin defficiency. However, lobular lymphocytic panniculitis is not one of these and can be associated with a diverse range of conditions such as tuberculosis (erythema induratum when fully developed), chronic viral hepatitis, systemic lupus erythematosus, dermatomyositis, severe weight loss, cold injury, and lymphoma2-4. This result is less helpful to the clinician, but soft tissue imaging may be useful for showing the extent of inflammation (and selecting an appropriate biopsy site), when the overlying skin is normal. We have no evidence that the T lymphocytes infiltrating the fat are malignant and so for diagnostic purposes we assume they are reactive. Therefore we are proposing that both the panniculitis and the systemic effects of this illness are secondary to cytokines secreted by reactive T cells, which in turn can be modulated by steroid therapy. DAVID L. JARDINE, FRACP, Department of General Medicine, E-mail: david.jardine@cdhb.govt.nz; JACQUI GARDNER, FRCPA, Department of Pathology, Christchurch Hospital; RICHARD STEELE, FRACP, Clinical Immunologist, Wellington Hospital, Wellington, New Zealand. ACKNOWLEDGMENT The authors are grateful for the advice of Dr. John O'Donnell and Dr. Ken Macdonald who reviewed the manuscript. Dr. Wayne Bailey prepared the images for the figure. REFERENCES 1. Weedon D. Skin Pathology. 2nd edition. New York: Churchill Livingstone; 2002:521-41. 2. Sutra-Loubet C, Carlotti A, Guillemette J, Wallach D. Neutrophylic panniculitis. J Am Acad Dermatol 2004;50:280-5. 3. Solans R, Cortes J, Selva A, et al. Panniculitis: a cutaneous manifestation of dermatomyositis. J Am Acad Dermatol 2002;46:S148-S150. 4. Craig AJ, Cualing H, Thomas G, Lamerson C, Smith R. Cytophagic histiocytic panniculitis — a syndrome associated with benign and malignant panniculitis: case comparison and review of the literature. J Am Acad Dermatol 1998;39:721-36. 5. Solomon AR, Kantak AG, Ramirez JE, Rajarman S, Raimer SS, Goldman AS. Suppressor-cytotoxic T-lymphocyte panniculitis. Pediatr Dermatol 1986;3:295-9. |
