RT Journal
A1 Conway W, Gagandeep S
T1 OPerative blood loss and survival in pancreatic cancer
JF Archives of Surgery
JO Archives of Surgery
YR 2009
FD June 1
VO 144
IS 6
SP 591
OP 595
DO 10.1001/archsurg.2009.101
UL http://dx.doi.org/10.1001/archsurg.2009.101
AB 
Kazanjian et al1 recently reported determinants of survival in 182 patients who underwent pancreaticoduodenectomy for adenocarcinoma of the pancreatic head. Through multivariate analysis, the authors found that operative estimated blood loss (EBL) and time of surgery (1987-1995 or 1996-2005) were significantly related to survival, whereas perioperative blood transfusion was not. We have 3 comments on this important study. First, the significance of EBL should be interpreted with caution, because this variable is highly subjective and notoriously inaccurate.2- 4 Simulations have found that operative personnel (including surgeons) visually underestimate EBL. Because surgeons have an obvious bias, which is amplified when EBL becomes a performance indicator or the end point of a study, it is not surprising that the most accurate estimations of EBL are from anesthesiologists.2