Tuberculosis in Kenya: A national sampling survey of drug resistance and other factors: An East African and British medical research council co-operative investigation

Abstract

A survey was performed in a random sample of 11 of 30 administrative districts in Kenya in 1964, to obtain information on:— 1. ( a) the proportions of patients with pulmonary and with extra-pulmonary tuberculosis; 2. ( b) the prevalence of bacteriologically-positive pulmonary tuberculosis: 3. ( c) the prevalence of initial and acquired drug resistance; and 4. ( d) the radiographic extent and type of pulmonary disease. Of 1,296 patients, 89·8% had pulmonary tuberculosis only, 2·5% had extra-pulmonary and pulmonary tuberculosis and 7·7% had extra-pulmonary tuberculosis only. Of the 132 patients with extra-pulmonary tuberculosis, 48% had lymph gland involvement, 21% bone and joint disease and 15% pleural effusion. Of the 1,196 patients with pulmonary tuberculosis (with or without extra-pulmonary tuberculosis), 93% gave no history of previous chemotherapy, 1% gave a history of possible and 6% of definite previous chemotherapy. (The proportions for the patients with extra-pulmonary tuberculosis only were 90%, 3% and 7% respectively.) A bacteriological specimen (nearly always sputum) was produced by 1,080 of the patients with pulmonary tuberculosis; of the 1,037 with bacteriological results available the smear was positive in 53% and the culture in 71%. The positivity rates were higher in adults than in children. Four (0·5 %) of the 739 patients with sensitivity tests available were excreting anonymous mycobacteria. Of 632 patients with no history of previous chemotherapy and a sensitivity test available, 14·7% had a strain resistant to one or more drugs, 11·1% being resistant to isoniazid and/or streptomycin, 10·3% to isoniazid, 2·2% to streptomycin and 6·0% to PAS. The proportion resistant to one drug only was 12·0%. The prevalence of resistance in children below the age of 15 was 6·2%, compared with 16·2% for adults (P = 0·02). Of the patients with a history of definite previous chemotherapy, 2% of 42 with a history of less than one month had a resistant strain as had 48% of 21 with a history of one to three months and 90% of 21 with a history of three months or more. Of 86 patients with no history of previous chemotherapy but a history of contact with patients who had been under treatment for tuberculosis, 23% had a resistant strain, compared with 13% of the 546 patients with no history of contact (P < 0·05). Of 939 patients with chest radiographs available and a diagnosis in Kenya of pulmonary tuberculosis, assessed independently in London, a lung lesion was reported in 848. Gross, extensive or moderate disease was present in 72% of the patients and cavitation was present in 68% of them. The type of radiographic disease was assessed as acute in 38%, mixed-acute in 4%, mixed in 28% and chronic in 31%. The disease in children was usually less extensive, less often cavitated but more acute than in adults. The positivity rates on smear and culture were higher in both adults and children in those with extensive disease and cavitation, and the prevalence of drug resistance was higher in those with extensive cavitation or with chronic disease.