RT Journal
A1 Pattakos G, Koch CG, Blackstone EH
T1 JEhovah's witnesses may not have identical outcomes with nontransfused non-witnesses after cardiac surgery—reply
JF JAMA Internal Medicine
JO JAMA Internal Medicine
YR 2013
FD February 11
VO 173
IS 3
SP 248
OP 249
DO 10.1001/jamainternmed.2013.2139
UL http://dx.doi.org/10.1001/jamainternmed.2013.2139
AB 
Indeed, our original analysis included a comparison of Witness patients propensity-score matched to 322 non-Witness patients who were not transfused. The journal requested that we remove this part of the analysis from the printed article, but some of the material is included in the supplementary online-only content (see eAppendices 1 and 2 and eFigures 1-3 in our original investigation1). Reoperation for bleeding was less (0.3% vs 3.7%; P = .003) in non-Witness nontransfused patients compared with Witness patients (Table). Intensive care unit and postoperative lengths of stay were also slightly shorter and less variable (P = .02). However, other complications and hospital mortality were similar (Table), as were predischarge hematocrit values (Witness vs non-Witness nontransfused patients, 31% [4.6%] vs 31% [4.6%] [P = .70]). Survival of Witness patients was 86%, 69%, 51%, and 34% at 5, 10, 15, and 20 years after surgery, respectively, vs 86%, 69%, 52%, and 38% among matched non-Witness nontransfused patients, respectively (P > . 40).