LETTER

The isolation of intestinal spirochetes from the blood of a series of six French patients with a range of serious clinical problems was reported recently (3). These organisms and another from the blood of a patient with AIDS in the United States were subsequently identified as Brachyspira (Serpulina) pilosicoli (7).

Intestinal carriage of B. pilosicoli occurs in Australia and is particularly common in Aboriginals (6), homosexual males (5), and immigrants who have recently arrived from developing countries (2). A recent study has shown that routine incubation protocols used for blood cultures are too short to allow detection of B. pilosicoli (1). The purpose of the study described here was to determine if spirochetemia due to this organism may be common but going undiagnosed in Australia.

To investigate the above-mentioned possibility, blood samples (n = 1,063) submitted for routine culture between August 1998 and February 2000 were selected from patients whose clinical symptoms resembled those of the French patients or whose symptoms had been described previously as having occurred in individuals with intestinal spirochete carriage (4, 7). These symptoms and conditions included malignancy (n = 108), human immunodeficiency virus infection (n = 9), other immune deficiencies (n = 109), stroke (n = 17), alcohol intoxication (n = 5), liver disease (n = 29), and gastrointestinal pain or diarrhea (n = 205). Patients with fever (n = 456) or septic shock (n = 125) were also chosen, and a proportion of those who were suffering from one of the previously mentioned conditions also had fever (256 of 482; 53.1%) or sepsis (44 of 482; 9.1%) recorded. Patients (n = 801) were from urban (n = 300) and rural Western Australia (n = 323) and from near Darwin (n = 178) in the Northern Territory of Australia. Patients were Aboriginal Australians (n = 302; 37.7%), non-Aboriginal individuals (n = 241; 30.8%), or of unknown ethnic background (n = 352; 43.9%).

The BACTEC 9240 automated blood culture system using Anaerobic/F medium was used for the study. To comply with our findings relating to the use of this equipment for detection of B. pilosicoli (1), the incubation protocol of urban samples was extended to 14 days after no positive signal was achieved at 3 days. Rural samples that signaled negative after 5 days of incubation by the standard procedure were sent from the laboratory of origin overnight to the Western Australian Centre for Pathology and Medical Research (PathCentre), Perth, Australia, and were reincubated in the BACTEC machinery for 21 days. Terminal subculture onto blood agar with 14 days of anaerobic incubation was carried out after this time, and a blood film was examined by phase contrast microscopy for the presence of spirochetes.

Intestinal spirochetes were neither cultured nor observed by these methods. Although no data were obtained on intestinal carriage of B. pilosicoli in these patients, the populations investigated included those reported to have a high prevalence of colonization. The cultural conditions have been shown to be the most appropriate for B. pilosicoli detection and growth in blood (1). Thus, we conclude that spirochetemia by intestinal spirochetes is not common in these populations. Despite these results, B. pilosicoli should not be completely disregarded as a possible cause of bacteremia in certain patient groups, particularly those with a high risk of intestinal carriage with the organism, since the condition has been described in both France and the United States.