Letters Section Editor: Stephen J. Lurie,
MD, PhD, Senior Editor.
To the Editor: Given the low frequency of SSIs
in the study of Dr Pryor and colleagues,1 we
are concerned that their sample size was too small to obtain reliable results.
The authors stated that 300 patients provided an 80% power to rule out a 40%
difference between the treatment groups (the specified null hypothesis) at
an α of .05. Using an 11% infection rate in patients given 35% oxygen,
as reported by Greif et al2 and observed
by Pryor et al, we compute that an 80% power to detect a 40% reduction in
the infection rate from 11% to 6.6% would require 651 patients; 930 patients
would be required to detect a 40% increase from 11% to 15.4%. The study thus
appears to have been underpowered from the start—and then was stopped
after only 160 patients were randomized. For a study anticipated to have 300
patients, 160 patients (53.3%) is a curious a priori stopping point.
Greif R, Sessler DI. Supplemental Oxygen and Risk of Surgical Site Infection. JAMA. 2004;291(16):1956. doi:10.1001/jama.291.16.1957-a