Multiple-puncture tuberculin testing: reason for variable response to tine test.

1040 'Tine Test' (TT) discs from 3 production batches were examined under a dissecting microscope. The dried tuberculin coatings of the 4 metal tines of the test discs were found to vary from being thin and smooth to being thick with droplet formation. Variations in coating were identified in the following proportions: 47% thin, 37% medium, and 16% thick (N = 4160 tines), these proportions varying significantly between the 3 batches (p much less than 0.001). Pairs of thin and thick coated TT discs were selected and applied to 103 tuberculin positive volunteers. When thickly coated TT were applied the sum of the diameters of induration was substantially smaller (p less than 0.001) than when thinly coated TT were used. At a greater than or equal to 2 mm limit for a positive reaction the thick TT failed to detect 16% of subjects positive to a 'Monotest' (MT) or a Mantoux (MX) 10 IU and the thin TT failed to detect 5% (p less than 0.05); at a greater than or equal to 5 mm limit the figures were 77% and 59% (p less than 0.01) respectively. In moderately sensitized populations, thick coated TT could lead to an excess of false negative results. The TT is of limited value as a screening test unless thin and uniform coating of tuberculin can be guaranteed.