Low BMI Associated With High 30-Day Postsurgical Mortality

Surgical patients with a body mass index at the lower end of the normal range were more likely to die within 30 days of the procedure than those in the moderately overweight range, researchers found.


Compared with patients with a BMI of 26.3 to 29.6, those with a value below 23.1 had a significantly higher risk of death (adjusted OR 1.40, 95% CI 1.25 to 1.58), according to an analysis of 189,533 surgeries performed in 2005 and 2006 and recorded in the National Surgical Quality Improvement Program (NSQIP) database.


Patients with higher BMI values above 23.1 -- including the morbidly obese -- had about the same risk of 30-day mortality as the moderately overweight, George J. Stukenborg, PhD, of the University of Virginia in Charlottesville, and colleagues, reported online in Archives of Surgery.


But for some individual types of surgeries, obesity was associated with increased mortality, the researchers found.


"These individual types of procedures include procedures with which the general surgeon should have definite experience: colorectal resection, colostomy formation, cholecystectomy, hernia repair, mastectomy, and wound debridement," Stukenborg and colleagues wrote.


NSQIP data in the study were extracted from medical records at 183 participating hospitals. Those with low volumes reported all cases performed each year whereas high-volume hospitals reported the first 40 consecutive cases for 42 eight-day cycles each year.


Patients in the study were categorized into BMI quintiles, with values of less than 23.1 being the lowest and those above 35.2 being the highest. For the overall 30-day mortality risk calculation, the middle quintile -- 26.3 to 29.6 -- served as the reference.


BMI values of 20 to 25 are considered normal. A value of 30 is the standard threshold separating overweight from frank obesity.


Odds ratios for 30-day mortality in the two lowest and two highest quintiles were adjusted for procedure type and baseline mortality risk. The latter is a standard part of the NSQIP data and is calculated from more than 30 patient variables including sociodemographic factors, comorbidities, and preoperative laboratory values such as serum albumin and white blood cell count.


Data reported for the years 2005 and 2006 were analyzed from 183 sites included in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database of general and vascular surgery. Of the189,533 patients reviewed in the evaluation, 3245 (1.7%) died within 30 days of surgery.


Measures of patient BMI were divided into nonstandard quintiles specific to the NSQIP study population. Baseline differences in overall existing mortality risk were evaluated for each patient, using the NSQIP probability of 30-day mortality risk score, which draws on more than 30 demographic variables as well as comorbidity factors and preoperative lab values. In addition, data were assessed to calculate the effect on patient mortality risk contributed by each of 45 independently evaluated categories of principal operating procedure.


The latter analysis found that the top 3 procedural categories associated with high 30-day mortality (compared with laparoscopy as a middle-range reference procedure responsible for 2% of the patient deaths) were exploratory laparotomy (13.9% of patient deaths), lower-extremity amputations (8.1%), and small bowel resection (7.9%). On the safer end of the procedural scale, 19 types of surgery each were associated with 1% or fewer of the patient deaths.


The authors also found notable interactions between BMI and some procedural categories, indicating that the link between BMI and mortality was statistically different for those categories than for patients who underwent laparoscopy. These findings suggest that the interaction between BMI and mortality plays a role within certain categories of surgery that may be classified as high risk or low risk, relative to laparoscopy.


Compared with the reference procedure, statistically significant (P < .05) differences in mortality odds were found for colostomies (AOR, 1.09; P = .009), wound debridement (AOR, 1.68; P = .002), ileostomy (AOR, 1.56; P = .008), musculoskeletal procedures (AOR, 0.49; P = .03), endarterectomy of head/neck vasculature (AOR, 0.55; P = .002), upper gastrointestinal procedures (AOR, 0.42; P < .001), cholecystectomy (AOR, 0.27; P = .04), and mastectomy (AOR, 0.05; P = .001).


"There is increasing evidence and supportive literature that obesity affects the practice of surgery in the United States and worldwide," explain the authors. "Reports to date have shown that, at the very least, complication rates and hospital resources have been affected by procedures on obese patients." The American College of Surgeons NSQIP database, used in the current study, "allows for the evaluation of a larger number of patients and for the examination of more specific types of procedures." The researchers conclude that "BMI is a significant predictor of mortality within 30 days of surgery, even after adjusting for the contribution to mortality risk made by type of surgery and for a specific patient's overall expected risk of death. Patients with a BMI of less than 23.1 demonstrated a significant increased risk of death.