DETAILS FOR THIS REVIEW

Study Population: 20 randomized controlled trials, with 2,670 adults (average age of 56 to 73 years) with diabetes mellitus; allocated to an intensive (n = 1,320) or a conventional (n = 1,350) glycemic control target in a perioperative hospital setting

End Points: Primary outcomes: all-cause mortality, hypoglycemic episodes, severe hypoglycemic episodes, infectious complications; Secondary outcomes: cardiovascular events, renal failure, length of intensive care unit stay, length of hospital stay, health-related quality of life

Narrative: Approximately 11.6% of U.S. adults and 10.5% of adults worldwide have diabetes^1,2; these individuals are at increased risk of perioperative complications.^3,4 Previous limited data suggested that patients without diabetes benefit from intensive perioperative glycemic control, but minimal evidence was found targeting outcomes for patients with diabetes.⁵ Further, a 2012 Cochrane review found no clear differences for most outcomes when comparing intensive with conventional perioperative glycemic control in people with diabetes, but a post hoc analysis suggested increased hypoglycemic events with intensive control.⁶ An update of the 2012 original, this Cochrane review includes several new trials.⁷

In this review, one-half of the randomized controlled trials were conducted in the United States (10 of 20 included studies), with three studies conducted in China, two in multiple-country study centers, and one each in Colombia, Lebanon, Egypt, Brazil, and Canada. All studies focused on tight glycemic control via insulin infusion, with the intensive group focusing on a lower glycemic target than the conventional group. Study designs were heterogeneous, with some studies using glucose infusions and others adding subcutaneous insulin boluses.

For the primary outcomes studied, low-certainty evidence demonstrated that intensive glycemic control likely increased severe hypoglycemic events (risk ratio = 4.73; 95% CI, 2.12 to 10.55; eight studies; 1,896 participants; absolute risk difference = 2.3%; number needed to harm = 43).⁸

For secondary outcomes, low-certainty evidence demonstrated that intensive glycemic control may result in decreased cardiovascular events (risk ratio = 0.73; 95% CI, 0.55 to 0.97; 11 studies; 1,454 participants; absolute risk difference = 6.6%; number needed to treat = 16). These events were defined as “any incidents that may cause damage to the cardiovascular system.”

When comparing intensive glycemic control with conventional glycemic control in people with diabetes who are undergoing surgery, there is likely no difference in perioperative all-cause mortality from 28 days to 1 year (high-certainty evidence), infectious complications (low-certainty evidence), renal failure events (low-certainty evidence), length of intensive care unit stay (low-certainty evidence), or length of hospital stay (very low-certainty evidence).

Given heterogeneity between studies and scarce outcomes data, this meta-analysis could not draw conclusions on blood pressure effects, health-related quality of life, or other socioeconomic effects.

Caveats: This systematic review demonstrated high-certainty evidence of no difference between intensive and conventional perioperative glucose control in all-cause mortality; however, the heterogeneity of the study design and outcome measurements affected the conclusions. Individual randomized controlled trials examined different types of surgeries, and unique surgeries affect glycemic demands differently; yet, all surgeries were lumped together in the analysis. Additionally, in the studies the terms “intensive glycemic control” and “conventional glycemic control” included a disparate range of blood glucose targets. Therefore, participants in the experimental intensive therapy group in one trial may have received different or even no interventions if they were in the experimental group in another trial with different glycemic targets. Partly because of this variance, no recommendations about optimal perioperative glycemic range could be made.

Individual study outcomes for cardiovascular events varied across studies but did include clinically relevant outcomes, such as rates of ischemic events, myocardial infarction, cerebrovascular accident, ventricular arrythmias, and major adverse cardiovascular events.

Although the data in this study can be generalizable to adults with diabetes who are undergoing surgery, less than 50% of the studies (9 out of 20) reported a preprocedural A1C (range = 6.5% to 8.4%). Many of the included studies failed to define the duration or severity of diabetes in each group, raising questions about individual tolerability of more stringent glycemic control. Additionally, it is reasonable to consider that individuals with poor baseline diabetes control would not be undergoing nonemergent surgery, further limiting generalizability.

Finally, the study excluded multiple patients with comorbidities that are prevalent in primary care, including kidney disease, liver disease, and obesity.

Conclusions: When comparing intensive vs. usual glycemic control in patients with diabetes who are undergoing surgery, this review demonstrated no significant benefit in all-cause mortality. When a difference occurred between groups, low-certainty evidence supported a likely reduction in cardiovascular events and an increase in severe hypoglycemic episodes with intensive control. Given these mixed findings and the significant heterogeneity of the studies and outcome reporting, we have assigned a color recommendation of yellow (more information needed).

Of note, two ongoing trials that otherwise met criteria for this review may add value to future updates. It would likely be beneficial for future studies to focus on more homogenous study models, particularly with a more aligned definition of a glycemic target for intensive glycemic control. Future studies may consider targeting outcomes separated by surgical procedure type, socioeconomic effect, and health-related quality of life because this information is absent in current literature.

Editor’s Note: The views expressed are those of the authors and do not necessarily reflect the official policy or position of the U.S. Navy, U.S. Department of Defense, or U.S. government.