Laboratory diagnosis of pulmonary tuberculosis is based on the presence of tubercle bacilli in sputum by direct microscopy, culture and/or animal inoculation. Culture examination, followed by tests for identifying the bacilli, is recognized as the most accurate and reliable method. Its efficacy depends on the laboratory techniques employed and its use in different practical situations such as epidemiological surveys, active community Case-finding, organization of diagnostic services and evaluation of diagnosis and treatment in tuberculosis control programmes. But the practicability of culture method in developing countries must be studied. The present paper deals with a systematic study of data from four investigations designed to elucidate the influence of certain operational factors on the utility of the culture method.

STUDY I: is a longitudinal survey in a randomly selected population in 134 villages in the three sub-divisions of Bangalore district. The analysis is based on the material from the first round, when two samples of sputum, (spot and overnight) were collected at intervals of 24-48 hours from persons aged 5 years and above having abnormal x ray shadows. The specimens were collected in house to house visits, stored after collection in insulated box with ice container and transported to the main laboratory at the National Tuberculosis Institute (NTI). The interval between collection of specimens in the field and culture in the laboratory was 1-7 days. A smear was stained and examined first by fluorescence microscopy and then by Ziehl-Neelsen (ZN) method. Each specimen was cultured on two slopes of Lowenstein-Jensen medium. All positive cultures were submitted to further identification tests; i.e., growth at room temperature, rate of growth at 37%C, pigment production in the dark and exposure to light, catalase and peroxidase reactions, niacin production, and sensitivity to INH, SM and PAS. STUDY II: relates to a mass Case-finding programme in Tumkur district when two specimens (spot and overnight) were collected from individuals aged 20 years and above with symptoms suggestive of pulmonary tuberculosis and from positive tuberculin reactors below 20 years voluntarily reporting with symptoms. The specimens were then treated in the same way as in Study I. STUDY III: pertains to the technical assessment of microscopy using Ziehl-Neelsen method performed by the auxiliary health staff of Peripheral Health Institutions in Bangalore district. A spot specimen was collected daily by auxiliary staff at each health facility from patients who were symptomatics. All smears were examined by ZN method at each centre and the corresponding sputum specimens were transported to NTI laboratory twice weekly. Duplicate smears were made and reexamined and culture was also done at NTI. All positive cultures were identified as in Study I. No refrigeration facilities were available in these centres and specimens were not transported in an insulated box. Rest of the procedures were followed as in previous studies. STUDY IV: is connected with operational and technical assessment of the District Tuberculosis Programme in Anantapur district one year after its commencement. A sample was taken from all patients who started treatment during a particular period but did not collect their drugs. Spot specimens were collected in the field, stored without any refrigeration and transported to NTI laboratory, thereafter the same procedure was followed as above.

An analysis of these four studies brought out certain operational factors affecting the culture method. (1) The results showed that an interval of 7 days between collection of sputum in the field and its processing in the laboratory did not affect the yield of positive cultures, even though the specimens were stored and transported under field conditions. (2) A higher proportion of positive cases were detected by culture than by direct microscopy but the magnitude of additional yield was dependant upon the procedure of selecting persons for sputum examination. (3) In service programmes restricted to persons with symptoms who attend diagnostic centres, the increase in yield is too small, to justify the introduction of culture examination.