Tuberculous bowel obstruction at a university teaching hospital in Northwestern Tanzania: a surgical experience with 118 cases.

backgroundBowel obstruction resulting from intestinal tuberculosis has been reported to be more prevalent in developing countries including Tanzania. This study was undertaken to describe the clinicopathological profile, surgical management and outcome of tuberculous intestinal obstruction in our local setting and to identify factors responsible for poor outcome among these patients.

methodsThis was a prospective descriptive study of patients operated for tuberculous intestinal obstruction at Bugando Medical Centre (BMC) in northwestern Tanzania from April 2008 to March 2012. Ethical approval to conduct the study was obtained from relevant authorities. Statistical data analysis was performed using SPSS version 17.0.

resultsA total of 118 patients with tuberculous intestinal obstruction were studied. The male to female ratio was 1.8: 1. The median age was 26 years (range 11-67 years). The modal age group was 21-30 years. Thirty-one (26.3%) patients had associated pulmonary tuberculosis and 25 (21.2%) patients were HIV positive with a median CD4+ count of 225 cells /μl. Small bowel strictures were the most common operative findings accounting for 72.9% of cases. The ileo-caecal region was the commonest area of involvement in 68 (57.6%) patients. The right hemicolectomy with ileo-transverse anastomosis was the most frequent surgical procedure performed in 66 (55.9%) patients. Postoperatively all the patients received antituberculous drugs for a period of one year. Postoperative complication rate was 37.3% and surgical site infection (SSI) was the most frequent complication in 42.8% of cases. HIV positivity and low CD4+ count were the main predictors of SSI (p < 0.001). The overall median length of hospital stay was 24 days. Patients who had postoperative complications stayed longer in the hospital and this was statistically significant (p = 0.011). Mortality rate was 28.8% and it was significantly associated with co-existing medical illness, delayed presentation, HIV positivity, low CD 4 count (<200 cells/μl), ASA class and presence of complications (p < 0.001). The follow up of patients was generally poor as more than fifty percent of patients were lost to follow up.

conclusionTuberculous bowel obstruction remains rampant in our environment and contributes significantly to high morbidity and mortality. The majority of patients present late when the disease becomes complicated. A high index of suspicion, proper evaluation and therapeutic trial in suspected patients is essential for an early diagnosis and timely definitive treatment, in order to decrease the morbidity and mortality associated with this disease.