Time to liquid culture positivity can substitute for colony counting on agar plates in early bactericidal activity studies of antituberculosis agents

The measurement of early bactericidal activity (EBA) is the first step in the clinical investigation of antituberculosis agents. EBA is determined by quantifying the viable sputum mycobacterial load on consecutive days of treatment. To investigate whether time to positivity (TTP) in mycobacterial liquid culture can substitute for colony forming unit (CFU) counting on agar plates we compared the error variation of TTP and CFU in 2115 pooled sputum samples collected overnight from 250 individuals included in five EBA studies. We found that the technical variation between duplicate laboratory measurements and the within-subject or day-to-day variation were similar for TTP (8.5% and 27.4% of total variation, respectively) and CFU (6.7% and 29.3% of total variation). The ability of the measurements to separate the EBA of 22 treatment arms was determined with group rank correlation of means and one-way analysis of variance. Except for the EBA over 0-2 days, individual and group EBAs measured with TTP and CFU were highly correlated. Treatment group means rank correlation coefficients were r = 0.472, r = 0.910 and r = 0.818, respectively, for EBA 0-2 days, EBA 0-7 days and EBA 0-14 days. Analysis of variance significantly favoured TTP over CFU for discrimination between groups with F values of 6.58 and 1.87, 7.77 and 4.58, and 8.71 and 3.56, respectively. We conclude that TTP is an acceptable alternative to CFU counting for the determination of the viable sputum mycobacterial load in EBA studies of up to 14 days duration.