Letters to the Editor: Letters & Announcements
Bijker and Kalkman1 note that low blood pressure, 1 of the 3 significant predictors of 1-yr mortality in the original study by Monk et al.,2 was not among the covariates in our study.3 They suggest that low intraoperative blood pressure and low bispectral index (BIS) could be interrelated, meaning that low blood pressure could, in fact, be the reason for impaired long-term survival.
Rather than let the statistical relationship between time during which BIS <45 and 1-yr mortality risk being prematurely interpreted as a call for change in anesthesia practice, we extended the statistical analysis.3 However, in doing so, we were not clear enough in our publication, thereby leading Bijker and Kalkman to overlook our most important finding. In other words, after including a very trivial covariate in the Cox proportional hazard model, malignancy status at the time for surgery, time during which BIS <45 was no longer significantly related to 1- or 2-yr mortality.3
It was previously known, and from our analyses it is obviously critical, which covariates are included in an explanatory statistical model.4 Even if malignancy status logically seems to be a relevant (and from our results seems relevant) covariate in the Cox model, the entire story can only be definitely resolved by a randomized study. From our data (all details not published), it seems that patients having surgery for intraabdominal malignancies with a duration of anesthesia exceeding 3 h randomized to 2 different BIS levels would be a suitable design for study to limit sample size. Even this study is unfortunately too large for our institution.
Maj-Lis Lindholm, RN
Rolf Sandin, MD, PhD
Department of Physiology and Pharmacology
Section Anesthesiology and Intensive Care
1.Bijker JB, Kalkman CJ. The role of intraoperative blood pressure in the association between low bispectral index values and mortality within 2 years after surgery. Anesth Analg 2010;110:639
2.Monk TG, Saini V, Weldon BC, Sigl JC. Anesthetic management and one-year mortality after noncardiac surgery. Anesth Analg 2005;100:4–10
3.Lindholm ML, Träff S, Granath F, Greenwald SD, Ekbom A, Lennmarken C, Sandin RH. Mortality within 2 years after surgery in relation to low intraoperative bispectral index values and preexisting malignant disease. Anesth Analg 2009;108:508–12
4.Pace NL. Independent predictors from stepwise logistic regression may be nothing more than publishable P values. Anesth Analg 2008;107:1775–8