A single surgical unit's experience with abdominal tuberculosis in the HIV/AIDS era

Human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) has resulted in a resurgence of abdominal tuberculosis in South Africa, and these patients often present to general surgeons. We describe a single-hospital experience in a region of high HIV prevalence.

A prospective database of all patients with suspected abdominal tuberculosis was maintained from January 2003 until July 2005.

There were 67 patients (20 men, 47 women) with an average age of 32 years (range 27-61 years). The erythrocyte sedimentation rate was universally elevated (105 +/- 23). Altogether, 23 patients were HIV-positive and 7 were HIV-negative. The status was unknown in the remainder. Chest radiographs demonstrated an abnormality in 17 patients (22%). Abdominal ultrasonography was performed in 59 patients and computed tomography in 12. Twelve laparotomies were performed, seven as emergencies. None in the elective laparotomy group died, whereas the mortality rate in the emergency group was 60%. Laparoscopy was insufficient for a variety of reasons. Two patients underwent appendectomy and two excision of a perianal fistula. Two patients underwent biopsy of a palpable subcutaneous node, which confirmed the diagnosis in both cases. A definitive diagnosis was achieved in all 12 patients subjected to laparotomy and at colonoscopic biopsy in 2, lymph node biopsy in 2, appendectomy in 2, perianal fistulectomy in 2, and percutaneous drainage in 2. In the remaining 47 patients the diagnosis was made on the basis of the clinical presentation and radiologic imaging. The patients were commenced on antituberculous therapy. The in-hospital mortality in this group was 10%. Therapy was continued at a centralized tuberculosis facility independent of the hospital. Surgical follow-up was poor, with only five (7%) patients completing the 6-month review at a surgical clinic.

A resurgence in tuberculosis during the HIV era produces a new spectrum of presentations for the surgeon. Emergency surgery is associated with high mortality. Bacterial and histologic evidence of infection are difficult to obtain, and indirect clinical and imaging evidence are used to commence a trial of therapy. A short-term clinical response is regarded as proof of disease. Lack of follow-up means that the efficacy of this strategy is unproven. Health policy changes are needed to enable appropriate surgical follow-up to determine the most effective management algorithm.