The International Journal of Tuberculosis and Lung Disease, 2009; 13 (6):737–743SETTING: Resource-limited settings in sub-Saharan African countries. OBJECTIVE: To utilise African giant pouched rats (Cricetomys gambianus) for the diagnosis of tuberculosis (TB) in human sputum. DESIGN: A specially designed cage with 10 sniffi ng holes and cassette-carrier was used. The sputum samples were put in the sample cassette, containing 10 samples in line, placed under matching sniffi ng holes. Rats were trained to sniff each consecutive sample, and indicate TB positives by fi xing their nose for 5 seconds at the sniffi ng hole. This behaviour was maintained by food reinforcement upon correct indications. A total of 3416 samples were used. RESULTS: Of the 20 trained rats, 18 were able to discriminate positive from negative sputum samples, with average daily sensitivities ranging from 72% to 100%, and average daily false-positives ranging from 0.7% to 8.1%. The use of multiple rats signifi cantly increased sensitivity and negative predictive value. CONCLUSION: Utilising trained sniffer rats for TB detection is a potentially faster screening method and is at least as sensitive as smear microscopy. This method could therefore be suitable for active case fi nding, especially where large numbers of samples are to be analysed in resource-limited settings, to complement existing diagnostic techniques.