Effect of HbA1c and Perioperative Glucose on Postoperative Mortality
HbA1c and 30-day mortality were not associated when controlling for glucose in noncardiac or cardiac procedures.
Hemoglobin A1c (HbA1c) level may be positively associated with perioperative glucose level but not with 30-day mortality, and perioperative glucose may predict 30-day mortality linearly and nonlinearly for noncardiac and cardiac procedures, respectively, according to a study published in Diabetes Care.
Researchers performed a retrospective analysis of data from 13,077 individuals who were at least 18 years of age and who were undergoing noncardiac (n=6684) and cardiac (n=6393) procedures to determine average perioperative blood glucose levels (average glucose from day of surgery to postoperative day 3) and 30-day postoperative mortality.
The purpose of the study was to determine whether there was an association between preoperative HbA1c level and perioperative glucose level and 30-day mortality in individuals undergoing either a cardiac or noncardiac procedure.
Study results showed a significant positive yet nonlinear association between preoperative HbA1c levels and perioperative glucose levels (P <.001), with levels virtually linear for HbA1c between 6% and 7.5%, followed by a flattening of the line for higher HbA1c values. Preoperative HbA1c levels and subsequent perioperative glucose levels were found to have a stronger association, with greater variation in individuals undergoing noncardiac procedures compared with cardiac procedures.
After controlling for other predictors, an increase in 30-day mortality was found in women (odds ratio 1.8 [95% CI, 1.2-2.7]; P =.004) and in those with very low body mass index (BMI), lower HbA1c level, increased age, and increased perioperative glucose level (P =.001, P =.01, P <.001, P =.04, respectively) undergoing noncardiac surgeries.
Those who underwent cardiac procedures were noted to have an increased mortality for increased age, BMI, and both low and high average glucose levels (a U-shaped curve), with stronger 30-day mortality association for perioperative glucose levels outside the 120 to 160 mg/dL range (4.5% at 100 mg/dL, 1.5% at 140 mg/dL, 6.9% at 200 mg/dL). Interestingly, preoperative HbA1c level was not found to be a predictor of postoperative mortality for those who underwent cardiac procedures despite the strong separate association noted between perioperative HbA1c and perioperative glucose levels.
A subsequent analysis controlling for age, glucose level, BMI, and sex found a negative association between noncardiac procedures and 30-day mortality; however, as time from surgery increased, the association of HbA1c with mortality became more positive. When not controlling for perioperative glucose, the trend of mortality vs HbA1c level remained virtually the same for individuals undergoing noncardiac surgeries yet the association became more positive for individuals undergoing cardiac procedures.
Researchers concluded that perioperative HbA1c levels predicted perioperative glucose levels, whereas the average perioperative glucose levels predicted 30-day mortality. HbA1c was not strongly associated with 30-day mortality for individuals undergoing noncardiac and cardiac procedures, and a negative association was found for noncardiac surgeries. In addition, after controlling for age, BMI, sex, and perioperative glucose level, elevated HbA1c was not associated with increased 30-day mortality. For those undergoing noncardiac surgeries, a stronger association was found between average perioperative blood glucose levels and 30-day mortality, and a U-shaped curve was found in individuals who underwent cardiac surgeries, emphasizing an increase in 30-day mortality in individuals with perioperative blood glucose levels of <120 mg/dL and >160 mg/dL.
Therefore, clinicians should monitor perioperative glucose in individuals undergoing both noncardiac and cardiac procedures carefully, with an increased vigilance in keeping perioperative glucose in individuals undergoing cardiac procedures between >120 and <160 mg/dL as this was associated with a lower mortality rate.
van den Boom W, Schroeder RA, Manning MW, Setji TL, Fiestan GO, Dunson DB. Effect of A1C and glucose on postoperative mortality in noncardiac and cardiac surgeries [published online February 13, 2018]. Diabetes Care. doi: 10.2337/dc17-2232