[Tuberculosis and HIV in Bangui, Central African Republic: strong prevalence and management difficulties]. / Tuberculose et VIH à Bangui, République centrafricaine: forte prévalence et difficultés de prise en charge.

The increasing incidence of tuberculosis in relation with the HIV-AIDS epidemic poses a major public health problem in sub-Saharan Africa. The purpose of this retrospective study was to analyze the prevalence of HIV-1 infection, clinical presentation, and bacteriological findings in patients treated for tuberculosis in a hospital department in Bangui, Central African Republic between January 1996 and December 1998. Among the 1142 patients who benefited for HIV serology, HIV-1 prevalence of was 82% (IC95%: 79-85%). Most patients (92%), had not undergone HIV serology before hospitalization. Mean age was 34 years. Sex ratio F/M was 1.21. Diagnosis of tuberculosis was based mainly on clinical and radiological data. Positive sputum smears were available for only 52% of the patients. The most frequent site of tuberculosis was the lungs with no significant difference between the HIV-positive and HIV-negative groups (83% versus 79% respectively). Sputum-smear examination was positive in 42% of the patients with no significant difference between the HIV-positive and HIV-negative groups (43% versus 37%). Upon admission patients usually presented advanced disease, with 11% dying within a week after hospitalization. There was a steep increase in the prevalence of HIV in tuberculosis-infected patients in Banqui, from 32% in 1988 to 62% in 1994. In spite of the existence of a National Tuberculosis Control Program, diagnostic facilities remain limited and diagnosis of tuberculosis and HIV-infection is often delayed. Outpatient care must be improved.

Day-case transfer of inpatients for diagnostic and therapeutic endoscopic retrograde cholangiopancreatography.

BACKGROUND: To use facilities flexibly and efficiently, a diagnostic and therapeutic endoscopic retrograde cholangiopancreatography (ERCP) service was offered to inpatients from referring hospitals on a day-case basis. METHODS: Patients were transferred by ambulance with a nurse escort and returned to the parent hospital after a short period of recovery. The activity of this service was audited. RESULTS: A total of 188 patients (70 men and 118 women of mean age 63.5 (range 22-94) years) were referred, 55 by physicians or geriatricians and 133 by surgeons from 19 hospitals up to 100 miles distant. Some 109 (58 per cent) had jaundice, 12 (6 per cent) cholangitis, seven (4 per cent) acute pancreatitis and 41 (22 per cent) abdominal pain. All patients arrived between 08:30 and 12:00 hours, all but 16 between 09:30 and 11:00 hours, and all but one with an escort; 129 patients travelled less than 10 miles. The mean time spent in the authors' hospital was 5.9 (range 2.2-9.5) h. The mean time spent away from the parent hospital by patient and escort was 7.0 (range 3.5-10) h. Cannulation failed in 12 patients (6 per cent), two with duodenal tumours. Sphincterotomy was performed in 70 cases, mechanical lithotripsy in seven and stents were inserted in 67. Nine patients were admitted, four for percutaneous stent insertion after failed ERCP and three for abdominal pain (one acute pancreatitis); two patients with sepsis and malignancy died after admission to this hospital. After return to the parent hospital, a further patient died following laparotomy for malignancy and five patients developed abdominal pain (three acute pancreatitis). CONCLUSION: These findings show that short-stay ERCP can be undertaken on transferred inpatients safely and efficiently without adverse effect on success or complication rates.