Successful Treatment of Tuberculous Peritonitis While Maintaining Patient on CAPD

Dave Tan, Paul A. Fein, Arletha Jorden, Morrell M. A vram

Although conventional wisdom advises removal of the Tenckhoff catheter as part of the therapy for tuberculous peritonitis, there are a few recent reports of cases successfully treated while maintaining the patients on CAPD. We wish to report three cases treated without interrupting CAPD. In two of the patients. cultures were positive for Mycobacterium tuberculosis and in the third case, although the cultures were negative, the patient improved on anti-Tb medications. Smear for AFB was positive in one patient; and two had a positive PPD. All had predominance of lymphocytes and monocytes in effluent. The total WBC count was 160-300 and two patients hadfever. All had abdominal pain.

One patient was treated with INH and ethambutol; one with INH and rifampin and one (who was suspected of being HIV+ ) also received pyrazinamide (PZA) until culture was available. Cultures grew in 4-6 weeks. All were started on therapy prior to having the culture results, and all showed clinical improvement within two weeks. One patient had his catheter replaced two months later because of pseudomonas peritonitis, continued on CAPD for an additionalfive months, then changed to HD because of recurrent bacterial peritonitis. One patient died of complications of diabetic vascular disease three months later with no evidence of peritonitis. One patient has remained on anti-Tb treatmentfor seven months and is doing well on CAPD.

Key words

Tb -tuberculosis, peritonitis, CAPD


the Long Island College Hospital, Brooklyn, New York 11201, USA.


Bacteria are the most common causes of peritonitis among peritoneal dialysis patients. However, tuberculous peritonitis is being increasingly recognized in this group (1-3). This complication of CAPD is mainly treated by removing the Tenckhoff catheter and by anti Tb therapy. We report three cases of Tb peritonitis treated with anti Tb medications while the Tenckhoff catheter remained in place.

Case 1

A 37-year-old black male with a past history of intravenous drug abuse, hypertension and alcoholism was admitted because of worsening azotemia in August, 1988. After being treated a few times on hemodialysis, a Tenckhoff catheter was placed and the patient began on CAPD. The patient developed peritonitis three months later and was treated with antibiotics. Staphylococcus epidermidis was cultured from the PD effluent at one point. His abdominal pain, fever and malaise persisted despite antibiotics. Work-up which included a chest CT showed significant subcarinal, anterior mediastinal lymph nodes and a left hilar mass. Sputum and bronchoscopy sampling were negative for AFB . Four months after the Tenckhoff catheter surgery , PD fluid lymphocytosis (67 -86% ) was noted in two specimens containing from 200 to 400 WBC's. The patient was then started on INH, rifampin and PZA. Nine days later, his abdominal pain resolved and the PD fluid became clear. Total PD fluid WBC count was 40 with 45% lymphocytes and 50% monocytes 21 days after starting the anti TB regimen. One out of three PD fluid specimens subsequently grew out Mycobacterium tuberculosis. The patient remained free of Tb peritonitis for nine months when the Tenckhoff catheter was removed because of enterobacter peritonitis which failed to respond to medical management.

Case 2

A 58-year-old black female with ESRD from longstanding diabetes mellitus was admitted in February , 1988, for Tenckhoff catheter placement. Ten months later, she developed a purulent discharge at the catheter exit site and was treated successfully with vancomycin and gentamicin. However, she soon developed a vague "gassy" abdominal discomfort. She also complained of malaise and tired easily. Except for fibrin, the PD fluid was initially clear .The PD fluid total WBC count was 81 with 64% PMN's and 24% lymphocytes; fluid bacterial cultures were negative. She was treated with vancomycin and gentamicin but two weeks later, the patient developed a fever of 103F and the fluid remained hazy. The fluid WBC count increased to 136 with 69% PMN's and 25% monocytes. Her tuberculin skin test was positive. She was started on INH and rifampin a few days later. The patient felt better five days after the anti Tb medications were begun. Due to persistent vomiting from rifampin, ethambutol was given instead. Six weeks after the INH and rifampin/ ethambutol were started, the PD fluid cleared and remained clear until her death a few weeks later from gangrene of the right leg. A PD fluid specimen submitted for culture subsequently grew out Mycobacterium tuberculosis.

Case 3

A 77-year-old white male with a history of atherosclerotic heart disease, hypertension and congestive heart failure developed ESRD and was started on CAPD in March, 1990. He developed a bacterial peritonitis during the first few weeks on CAPD. This soon resolved with intraperitoneal antibiotics. However, on 5/22/90, he had an acute episode of abdominal pain. The PD effluent was clear. Total fluid WBC count was 54 with 57% lymphocytes and 39% monocytes. On 5/31/90, the PD fluid became hazy and the total WBC count increased to 216 with 8% lymphocytes and 92% monocytes. PD fluid bacterial cultures were negative. He still had abdominal pain but was afebrile. The PPD was ( + ) on 6/5/90. He was then started on INH and rifampin. Six days later the abdominal pain markedly improved and the PD fluid became clear .PD effluent specimens sent for AFB smears and cultures were all negative.


Compared to the general population, ESRD patients have a 10 fold increase in their risk of infection with Mycobacterium tuberculosis (4). Tuberculosis in this group usually manifests non-specifically and may often have extrapulmonary involvement.

Prior to their treatment with anti-tuberculous drugs, all three of our patients had been treated with antibiotics intraperitoneally: for peritonitis in two patients and an exit site infection in the third. A significant lymphocytosis in the PD fluid was seen in two of the patients. Mallat et at. , in their review of Tb peritonitis among PD patients, noted the wide variation of lymphocyte counts in the PD fluid and the paucity of a positive AFB smear (5). Two of our patients who underwent tuberculin skin testing had positive results. As most ESRD patients have impaired cellular immunity, the tuberculin skin test may be useful only if positive. The third patient who had a culture-negative peritonitis and a high percentage of lymphocytes in the PO fluid benefitted from a trial of anti-tuberculous drugs. Some investigators recommend a limited laparotomy, laparoscopy or peritoneal needle biopsy to provide a histologic diagnosis (6, 7).

In recent years, Tb peritonitis among CAPD patients has been treated with anti-tuberculous medications without removal of the Tenckhoff catheter (3,5,8). A nine-month course of INH and rifampin was reported to be very successful in treating extrapulmonary tuberculosis among a large number of patients (9) .In a CAPD patient with prolonged fever and cloudy effluent, switching from oral to intraperitoneal rifampin resulted in defervescence and the disappearance of leukocytes from the effluent (3). Because of the possibility of HIV in our first patient and of his having acquired an INH-resistant form of Mycobacterium tuberculosis, we added a third drug, PZA. Hence, we feel that most CAPD patients who develop Tb peritonitis may be reasonably treated with a nine-month course of INH and rifampin with the Tenckhoff catheter remaining in place. Indications for removal of the Tenckhoff catheter other than Tb peritonitis may arise. These include the appearance of a different microorganism which may be difficult to eradicate with antibiotics alone or changing the patient to hemodialysis due to a decreasing peritoneal clear ance and ultrafiltration rate (3). The latter probably results from the sequelae of prolonged inflammation of Tb peritonitis.


  1. Holly HP, Tucker T, Moffatt TL, Dodds KA, Dodds HM. Tuberculous peritonitis in patients undergoing chronic home peritoneal dialysis. Am I Kidney Dis 1982; I: 222-26.
  2. McKerrow KI, Neale TI. Tuberculous peritonitis in chronic renal failure managed by CAPD. Aust NZ IMed 1983; 13: 343-47.
  3. Cheng IKP, Chan PCK, Man KC. Tuberculous peritonitis complicating long-term peritoneal dialysis. Am I Nephro11989; 9: 155-61.
  4. Belcon MC, Smith EKM, Kahana LM, Shimizu AG. Tuberculosis in dialysis patients. Clin Nephrology 1986; 17: 14-18.
  5. Mallat SG, Brensilver IM. Tuberculous peritonitis in a CAPD patient cured without catheter removal. Case report, review of the literature and guidelines for treatment and diagnosis. Am I Kidney Dis 1989; 13: 57.
  6. Borhanmanesh F, Hekmat K, Yaezzadeh K et al. Tuberculous peritonitis. Prospective study of 32 cases in Iran. Ann Intern Med 1972; 78: 567 72.
  7. Levine H. Needle biopsy diagnosis of tuberculous peritonitis. Am Rev Resp Dis 1968; 97: 889-94.
  8. Baumgartner DD, Arterbery YE, Hale AI, Gupta RK, Bradley SF. Peritoneal dialysis-associated tuberculous peritonitis in an intravenous drug user with AIDS. Am I Kidney Dis 1989; 14: 154-57.
  9. Dutt AK, Moers D, Stead WW. Short-course chemotherapy for extra-pulmonary tuberculosis. Ann Intern Med 1986; 104: 7-12.
Correspondence to:
Dave Tan, M.D., The Long Island College Hospital, Atlantic Avenue and Hicks Street, Brooklyn, NY 11201.