F. Lecomte

La tuberculose péritonéale : une entité toujours présente. À propos de quatre observations

INTRODUCTION Tuberculous peritonitis, a major problem in developing country, occurs preferentially in immigrant population and in patients with acquired immune deficiency syndrome (AIDS). Although rare in France, it did not disappear and epidemiological, clinical and therapeutic approach deserve to be reminded. EXEGESIS We reported 4 patients (immigrants in two cases), occurred in caucasian and African persons (one with AIDS). Disease was characterized by fever, abdominal pain, anorexia, weight loss and ascites. Biological and radiological were unconclusive. Cell count analysis from ascitic fluid show a lymphocytic predominance with negative direct smear for Ziehl-Neelsen strain. Tuberculous peritonitis was established with combined visual and histological diagnosic laparoscopic examination. CONCLUSION These observations have the interest to underline that tuberculous peritonitis must be evoked in case of lymphocytic ascitis. We believe an aggressive diagnostic approach, particulary with peritoneal biopsy, is warranted for the diagnosis of tuberculous peritonitis. Validity of PCR amplification is ascitic fluid still needs to be established.

Communication brèveLa tuberculose péritonéale : une entité toujours présente. À propos de quatre observationsTuberculosis peritonitis: an always present disease. About 4 new cases

INTRODUCTION Tuberculous peritonitis, a major problem in developing country, occurs preferentially in immigrant population and in patients with acquired immune deficiency syndrome (AIDS). Although rare in France, it did not disappear and epidemiological, clinical and therapeutic approach deserve to be reminded. EXEGESIS We reported 4 patients (immigrants in two cases), occurred in caucasian and African persons (one with AIDS). Disease was characterized by fever, abdominal pain, anorexia, weight loss and ascites. Biological and radiological were unconclusive. Cell count analysis from ascitic fluid show a lymphocytic predominance with negative direct smear for Ziehl-Neelsen strain. Tuberculous peritonitis was established with combined visual and histological diagnosic laparoscopic examination. CONCLUSION These observations have the interest to underline that tuberculous peritonitis must be evoked in case of lymphocytic ascitis. We believe an aggressive diagnostic approach, particulary with peritoneal biopsy, is warranted for the diagnosis of tuberculous peritonitis. Validity of PCR amplification is ascitic fluid still needs to be established.

Granulomatous Nephritis as the First Manifestation of Whipple Disease

TO THE EDITOR: Renal involvement is rare in Whipple disease (1, 2). It occurs late during the course of the disease (1, 2). We observed a patient who developed renal failure related to granulomatous nephritis. Polymerase chain reaction (PCR) analysis of 16S ribosomal RNA (rRNA) was positive in renal biopsy specimens, demonstrating the presence of Tropheryma whippelii (3). A 60-year-old man presented in April 1996 with an 8-month history of polyarthralgias involving the shoulders, elbows, wrists, and knees. The following laboratory findings were abnormal: erythrocyte sedimentation rate, 56 mm/h; protein level, 60 g/L; albumin level, 26 g/L; serum creatinine concentration, 239 mol/L; creatinine clearance, 46 mL/min; 24-hour urinary protein excretion, 2 g; and microscopic hematuria. Autoantibody screening (rheumatoid factors, antinuclear antibodies, complement profile, anticardiolipin and antiphospholipid antibodies, antineutrophil cytoplasmic antibodies, cryoglobulin, and angiotensin-converting enzyme levels) was negative. Bacterial (Chlamydia pneumoniae, Mycoplasma pneumoniae, brucella, and Lyme disease) and viral (hepatitis, parvovirus, and HIV) serologic tests also had negative results. Renal histology demonstrated diffuse interstitial and granulomatous nephritis; direct immunofluorescence and Ziehl-Nielsen stains were negative. Other investigations, including abdominal and thoracic computed tomography, echocardiography, biopsies of accessory salivary glands and duodenum, were normal. Steroid therapy, 0.8 mg/kg per day, resulted in partial improvement of joint manifestations. Biochemical renal abnormalities remained unchanged. In January 1999, the patient presented with a 10-kg weight loss. Physical examination revealed cachexia and a tender abdomen on palpation. Laboratory tests showed serum creatinine concentration of 326 mol/L; creatinine clearance of 21 mL/min; 24-hour urinary protein excretion of 1.7 g; and malabsorption syndrome with steatorrhea, 6 g/d. Results of duodenal biopsy were normal. Colonoscopy was done, and ileum biopsy revealed the diagnosis of Whipple disease. Although trimethoprim-sulfamethoxazole treatment was started, the patient died 3 weeks later. Retrospective PCR analysis of renal biopsy specimens (performed in April 1996) was positive for 16S rRNA. Because renal manifestations may precede other systemic signs of Whipple disease, we suggest that when unexplained granulomatous nephritis is noted, evaluation for an underlying Whipple disease should be done. In 1996, PCR using 16S rRNA was not yet available in our university hospital; this test would have demonstrated Tropheryma whippelii in our patients renal histology. Our data indicate that PCR analysis may be helpful in identifying suspicious cases of Whipple disease before the appearance of the usual clinical manifestations and foamy macrophages that are positive on periodic acid-Schiff staining.