Reason for variable response to tine test.

Reason for variable response to tine test In comparative studies with the Mantoux test the tine method has shown false-negative rates of between 23% and 64%,1-though studies on highly sensitised tuberculous patients have tended to show a closer correlation.4 This inconsistency has remained unexplained by studies of the duration of application, skin temperature, and release of skin tension before removing the tine. I have looked at variability of the tuberculin coating, a factor not previously considered, as a cause of false-negative and diminished responses. Microscopic inspection of tines shows that there are two characteristic types of tuberculin coating. Either the tip is covered uniformly to a depth of 125 mm, or a globule forms some 1-5-1-75 mm from the tip, leaving the tip itself bare or with diminished coating (figure). All four tines of a disc usually have one or other characteristic coating. Globular tine units ranged from six to 18 in any one container of 25. 2mm Variable tuberculin coating of tines. Left tine is smooth coated with good tip cover, whereas tuberculin has retracted towvards base of right tine, a globule has formed, and tip is poorly covered. A series of tests were carried out on volunteers. Smooth-coated tines were selected from batch No 528-267, and one applied on each arm of 20 persons positive to a Mantoux test (5 tuberculin units of purified protein derivative) and read after 48 hours. All reactions were positive and similar between arms on each individual. The sum of the diameters of induration on each arm were 108 mm and 105 mm. In the second test, smooth-and globular-coated tines were separated from the same batch and applied to opposite arms of volunteers known to be tuberculin positive. Again 20 paired tests were carried out and read after 48 hours. In seven persons the arms tested with globular-coated tines had a negative reaction, whereas the opposite arms tested with smooth-coated tines were positive. Where globular-coated units produced reactions, these were, with one exception, less than the opposite smooth-coated reaction. The sum of the diameters of induration from 20 smooth-coated tines was 148 mm and from the 20 globular-coated tines 82 mm. Analysis by the Wilcoxon signed-rank test indicates that dissimilar reactions of this order would arise less than one in 100 times by chance. In the third test, globular-and smooth-coated tines were selected from batch No 528-267 and distributed randomly in pairs before …