ABSTRACT
The bleeding time is a readily and easily performed clinical test with immediate results, but there is a degree of subjectivity in its performance and interpretation. We performed a study on 27 volunteers designed to determine the normal range, interobserver reproducibility, and biological variability of the test. Bleeding times in these normal subjects ranged from as low as 129 seconds to as high as 803 seconds. The interobserver variability was 106 seconds (2 standard deviations of the mean of the differences of paired results of repeated measurements), and the coefficient of variation was 18%. For bleeding times taken on the same subjects 6 weeks apart, when the same nurse performed the test at both visits, the difference was 150 seconds (2 standard deviations of the mean of the differences of paired samples) and the coefficient of variation was 27%, and they were essentially the same if a different nurse performed the tests at each visit. There is a wide range in the bleeding times among subjects. However, within individuals there is little biological variability, and most of the difference over time is due to interobserver variability. This suggests that changes in bleeding time are clinically useful in predicting platelet responsiveness in individual patients.
