Prediabetics and Post-op Risk: Real Increase or Artifact?

Cardiac surgery patients with impaired glucose tolerance or prediabetes have long been thought to suffer more major postoperative adverse events than those with normal glucose tolerance. New evidence suggests that may not be the case.

Researchers in Boston have found that the two groups of patients—prediabetics and nondiabetics—experience similar rates of myocardial infarction, renal failure, stroke and death, at least when a cutoff of hemoglobin A_1c (HbA_1c) of 6% or greater is used to define impaired glucose tolerance.

“Patients with impaired glucose tolerance are a special group,” said Balachundhar Subramaniam, MPH, MD, assistant professor of anaesthesia at Harvard Medical School and Beth Israel Deaconess Medical Center, both in Boston. “They’re not yet diagnosed as diabetics, and [yet] may be prone to more risk than patients who are already diagnosed and well controlled.”

To determine the relative adverse event rate for nondiabetics versus prediabetics, Dr. Subramaniam and colleagues tested 927 cardiac surgery patients for HbA_1c within two weeks of their surgery; 638 (69%) were not diagnosed with diabetes. An HbA_1c level below 6% was used to further distinguish true nondiabetics (n=472) from those with impaired glucose tolerance (HbA_1c≥6%; n=166).

The patients with impaired glucose tolerance had HbA_1c values as high as 7.8% (mean, 6.2%). They also had a higher incidence of significant comorbidities.

Nevertheless, postoperative outcomes did not differ significantly between the true nondiabetics and those with impaired glucose tolerance (Table 1).

Table 1. Glucose Status and Outcomes After Cardiac Surgery Nondiabetics (n=472) Impaired Glucose Tolerance (n=166) P Value Permanent stroke, n (%) 11 (2%) 5 (3%) 0.576 Sternal infection, n (%) 4 (1%) 1 (0.5%) 1.000 Renal failure, n (%) 9 (2%) 5 (3%) 0.372 Myocardial infarction, n (%) 3 (0.6%) 2 (1.2%) 0.609 Death, n (%) 14 (3%) 6 (3.6%) 0.616

As Dr. Subramaniam reported at the 2010 annual meeting of the Society of Cardiovascular Anesthesiologists (abstract SCA67), the results may hinge on the investigators’ definition of impaired glucose tolerance. He cited a recent article in The New England Journal of Medicine (2010;362:800-811) in which HbA_1c levels above 5.5% were used to determine patients with impaired glucose tolerance.

“It could be that there is no actual difference in major perioperative adverse events between true nondiabetics and those with impaired glucose tolerance,” Dr. Subramaniam told Anesthesiology News. It also is possible that impaired glucose tolerance may not be defined with an HbA_1c cutoff of 6%, which is why the researchers plan to reanalyze the data using 5.5% as a benchmark.

However, in work they plan to present at the 2010 annual meeting of the American Society of Anesthesiologists (ASA; abstract A1143), Dr. Subramaniam and his colleagues found that using the lower threshold for HbA_1c does not seem to predict a patient’s risk for major complications of cardiac surgery such as death, postoperative myocardial infarction, stroke and kidney failure (Table 2).

Table 2. Major Complications After Cardiac Surgery by Glucose Cutpoint HbA1C <5.5 HbA1C ≥5.5 P Value Death, n (%) 3 (2) 20 (3) 0.78 Myocardial infarction, n (%) 1 (1) 4 (0.6) 1.00 Renal failure, n (%) 6 (4) 16 (2.3) 0.26 Deep sternal wound, n (%) 2 (1.3) 4 (6) 0.29 Permanent stroke, n (%) 2 (1.3) 15 (2.2) 0.75 Combined outcome (%) 8 6.4 0.47

In an unrelated study also to be presented at the ASA meeting (abstract A720), Cleveland Clinic researchers analyzed the effect of blood sugar levels on postoperative mortality in a pool of more than 61,500 patients who underwent noncardiac surgery. They did not find a link between in-hospital death and preoperative blood glucose. However, the researchers found that patients with elevated blood glucose before surgery were more likely to be dead a year after their procedure, and that the risk for death rose with the patient’s blood sugar.

Boris Mraovic, MD, associate professor of anesthesiology at Thomas Jefferson University, in Philadelphia, said preoperative glucose control is not well studied, despite the fact that diabetes increases perioperative morbidity and mortality. “There are only few retrospective studies with conflicting results,” Dr. Mraovic said. “This could be explained by the fact that HbA_1c reflects only a mean glucose value and not glucose variability, which may play an important role in increasing risks.”

Dr. Mraovic also questioned whether the cardiac study was sufficiently powered to illustrate outcome differences between nondiabetic patients and those with impaired glucose tolerance. “The other way to analyze the data would be to find threshold HbA_1cvalues for a variety of complications,” he said.