Skip to main content

Table 3 Sample size calculation for non-inferiority safety hypothesis of tranexamic acid vs. placebo

From: Rationale and design of the PeriOperative ISchemic Evaluation-3 (POISE-3): a randomized controlled trial evaluating tranexamic acid and a strategy to minimize hypotension in noncardiac surgery

Placebo event rate NI margin, absolute risk difference TXA vs. placebob Power Sample sizec
  0.90 8,868
10%a 1.18% 0.95 10,963
  0.97 12,441
0.90 8,078
11%a 1.29% 0.95 9,990
  0.97 11,340
0.90 7,390
12%a 1.39% 0.95 9,136
  0.97 10,367
  1. NI non-inferiority margin, TXA tranexamic acid
  2. aThe estimates are based on the event rates observed in the POISE-2 trial (aspirin-placebo group) [5] and in a simulated subpopulation of the VISION cohort [1, 2] matching the POISE-3 inclusion criteria
  3. bThese are the NI margins expressed as absolute risk differences, corresponding to a NI margin expressed as hazard ratio of 1.125, when the placebo event rate is 10%, 11%, or 12%
  4. cThese sample sizes are obtained from non-inferiority test for two survival curves using Cox’s proportional hazard model, with a hazard ratio of 1.125 as non-inferiority margin, and a hazard ratio of 0.9 as actual hazard ratio (one-sided test, alpha of 0.025)