Relationship between hypothermia and blood loss in adult patients undergoing open lumbar spine surgery

Intraoperative blood loss during open lumbar spine surgery is associated with adverse events and is a contributor to higher medical costs. Intraoperative hypothermia has been shown to increase blood loss and postoperative allogeneic blood transfusion rates in other realms of orthopedic surgery, but it has not been studied extensively in patients undergoing spine surgery.

To determine whether a clinically relevant association exists between intraoperative core body temperature and blood loss or transfusion rates in adult patients undergoing open lumbar spine surgery.

In this retrospective medical record review, the surgical records of 174 adult patients who underwent open, nonmicroscopically assisted lumbar spine surgery performed by a single surgeon at a single institution were evaluated. Maximum, minimum, and average temperature, hypothermic temperature, and temperature range parameters were compared with intraoperative, total, and net blood loss and blood transfusion parameters. Additional patient demographic and perioperative characteristics were compared with blood loss and transfusion parameters to determine potential confounders. Analysis of variance, Spearman rank correlation, and generalized multiple linear regression analysis were performed to test for an association between temperature and blood loss or allogeneic transfusion rates. Statistical significance was set at P≤.05.

After implementation of exclusion criteria, 160 patient records and 168 surgical procedures were included in the analysis. For patients whose temperature decreased to a hypothermic level at some point during the procedure, hypothermic maximum temperature was protective against blood loss on bivariate analysis (P≤.02), but this finding lost significance after multivariate regression analysis (P>.09). Temperature range was associated with increased blood loss on bivariate analyses (P<.001) but also lost significance after adjusting for covariates in regression analysis (P≥.65). Surgery type (P≤.001) and operative time (P≤.001) were the most robust predictors of increased blood loss (P=.005) and were significantly associated with temperature (P<.001).

No effect was found on perioperative blood loss from any temperature parameter or hypothermia in adult patients who underwent lumbar spine surgery once covariates were controlled for with multivariate analysis. One possible interpretation of these results is that the effect of temperature on blood loss can be explained by its strong relationship to the confounders of operative time and surgery type.