Mycobacterial PD peritonitis

One of the most feared complications of peritoneal dialysis is that of PD peritonitis. Infection with gram positive (50-70%) organisms accounts for the vast majority of cases. The typical presentation is with abdominal pain, fever and cloudy dialysate fluid. To make the diagnosis of PD peritonitis, a sample of the PD fluid must be sent for cell count, differential and culture + sensitivity. A PD fluid white cell count of >100/cc is generally accepted as enough for diagnosis.

Empiric therapy can usually be given intra-peritoneally and consists of broad spectrum gram positive and gram negative coverage until the culture data is back. If it is the case where a patient comes in the middle of the night and there is no-one available to give IP antibiotics, then they should be given intravenously.

The cell count differential can sometimes give a clue that pathogens other than bacteria may be lurking around. A predominant lymphocytosis in the PD fluid should raise the suspicion of mycobacterial organisms.

Also, the presence of prolonged symptoms despite treatment of bacterial pathogens, or relapsing peritonitis with negative bacterial cultures, should ring alarms bells and start the search for alternative organisms.

A recent review of tuberculous peritonitis associated with peritoneal dialysis, reported that 89% presented with abdominal pain, 81% with fever and 77% with cloudy dialysate. 30% had a lymphocyte predominant dialysate, but 65% still had a neutrophil predominance. When suspicion arises there are various ways to make the diagnosis – PCR, mycobacterial staining, culture and sometimes a peritoneal biopsy is necessary.

The International Society for Peritoneal Dialysis recommends treatment with isoniazid, rifampin, pyrazinamide and ofloxacin. Three months for pyrazinamide and ofloxacin; with at least 12 months of isoniazid and rifampin. Pyridoxine should be given with isoniazid to reduce the risk of peripheral neuropathy. There is no strong consensus on whether the PD catheter needs to be removed in the case of tuberculous peritonitis – if the patient looks sick, the safer thing to do is remove it, particularly if it is an atypical mycobacterium. In one small study of ten patients, 6 were still on PD nine months after their infection.

TB needs to be kept in the differential for our patients, especially those who are immunosuppressed, come from endemic areas or may have occupational exposures.