Non-O:1 Vibrio Cholerae Bacteremia and Peritonitis in a Patient With Nephrotic Syndrome

To The Editors: Nonagglutinating Vibrio cholerae organisms are structurally and biochemically similar to the classical V. cholerae but are not agglutinated by the O subgroup 1 antiserum. These organisms cause diarrheal illness1, 2 but rarely cause more invasive disease including septicemia, cholecystitis, meningitis and peritonitis.3-5 Previous reports of non-O:1 V. cholerae septicemia,4, 5 including three cases from our hospital, have mainly been in patients with chronic liver disease or hematologic malignancies, although septicemia has been reported in a previously healthy patient.4 We report a case of bacteremia and primary peritonitis caused by non-O:1 V. cholerae in a child with nephrotic syndrome. Case report. A 4-year-old boy with frequently relapsing nephrotic syndrome was referred to this hospital for further management. He was being treated with daily low dosage prednisolone for the prior 13 months. After preliminary investigations, including a renal biopsy, revealed that he had minimal change disease, he was treated with prednisolone as per the International Study on Kidney Diseases in Children regimen.6 Two days after commencing treatment the patient developed watery diarrhea and vomiting. This was followed by the onset of fever, abdominal pain and distension requiring hospitalization. On examination he had anasarca, was febrile and appeared toxic. Pulse rate was 108/min and blood pressure was 130/86 mm Hg. His abdomen was distended with generalized tenderness and guarding. Free fluid was present in the peritoneal cavity and bowel sounds were absent. The rest of the examination was unremarkable. Laboratory investigations at admission revealed a total leukocyte count of 14 700/mm3 with 4% band forms, 75% neutrophils, 1% eosinophils, 19% lymphocytes and 1% monocytes. The hemoglobin was 11.6 g/dl and the platelet count was 280 000/mm3. The 24-h urine protein was 55 mg in 420 ml. The serum total protein and albumin were 2.7 and 0.8 g/dl, respectively. Blood urea was 34 mg/dl and serum creatinine was 0.4 mg/dl. No pathogens were isolated in stool culture. A plain radiograph of the abdomen revealed multiple air-fluid levels, and an ultrasound of the abdomen showed free fluid in the abdomen with no focal abscess formation. A peritoneal tap revealed turbid peritoneal fluid with a total leukocyte count of 8900/mm3 and a differential of 93% neutrophils and 7% monocytes and 200 units of lactate dehydrogenase. V. cholerae that did not agglutinate with O:1 or O:139 antisera was isolated in culture from the peritoneal fluid as well as from the blood culture. The isolate was susceptible to cefotaxime, ceftazidime, chloramphenicol, gentamicin, amikacin, ciprofloxacin and trimethoprim-sulfamethoxazole. The patient was treated intravenously with cefotaxime. In addition he was maintained on nasogastric drainage and intravenous fluids with no oral feedings. Albumin infusion and spironolactone were also given to reduce the edema. Four days after starting antibiotic treatment the patient was no longer toxic and he became afebrile after 8 days of antibiotics. Repeat blood culture at that time was sterile. Antibiotics were continued for 14 days. At discharge 18 days after starting treatment, he was afebrile and had no edema, abdominal distension or tenderness. He was tolerating oral feeding and had normal bowel sounds. Discussion. Peritonitis is a common complication in patients with nephrotic syndrome. The predominant organism is Streptococcus pneumoniae with Gram-negative enteric organisms being the next most frequent.7 Transmural migration of enteric organisms through the gastrointestinal tract with hematogenous dissemination in children with nephrotic syndrome who have low serum immunoglobulin values and decreased opsonizing capacity of the ascitic fluid has been postulated to be one of the pathogenetic mechanisms for bacteremia and peritonitis in these patients. However, some authors have thought this an unlikely theory because of the paucity of Gram-negative organisms causing peritonitis in their series of patients with nephrotic syndrome.8 Our patient had diarrhea preceding the onset of peritonitis and septicemia suggesting an enteric infection. Although V. cholerae was not isolated from the feces at the time of admission, it was most likely the causative agent of diarrhea; the organism was isolated in only one of five previously reported patients with septicemia who had preceding diarrhea.4 This would support the theory that invasive disease caused by V. cholerae in this patient was a result of transmural migration of the organism. Moreover the two predominant risk factors for invasive disease with non O:1 V. cholerae in previously reported cases are chronic liver disease with portosystemic shunting of blood and hematologic malignancies.4, 5 In both conditions transmural migration of enteric bacteria is thought to play an important role in the pathogenesis of invasive disease. Although S. pneumoniae is the predominant cause of septicemia and peritonitis in patients with nephrotic syndrome, a number of other organisms including unusual pathogens such as non-O:1 V. cholerae may be responsible for the infection. Maya Thomas, D.C.H.; Thomas Cherian, M.D.; P. Raghupathy, M.D. Department of Child Health Christian Medical College and Hospital Vellore 632004, India