At any given time, a patient’s blood-glucose level is easy to measure. Beyond the standard pre/postoperative lab values, there are finger sticks, transdermal meters, and other modalities that make taking a patient’s glucose “snapshot” pretty straightforward. So why don’t we surgeons keep track of it more frequently before and after joint replacement, when, according to the prognostic study by Shohat et al. in the July 5, 2018 issue of JBJS, fluctuating glucose levels can have a critical impact on outcomes?
By retrospectively studying more than 5,000 patients who had undergone either total hip or total knee arthroplasty, the authors found that increased variability of glucose levels (measured by a coefficient of variation) was associated with increased risks of 90-day mortality, surgical-site infection, and periprosthetic joint infection. Specifically, the authors demonstrated that for every 10-percentage-point increase in the glycemic coefficient of variation, the risk of 90-day mortality increased by 26%, and the risk of periprosthetic or surgical-site infection increased by 20%. These are remarkable increases in extremely important outcome measures, and the associations held regardless of the patient’s mean glucose values prior to or after the surgery. In fact, some of the highest levels of glucose variability were found in patients who had well-controlled glucose levels preoperatively. Furthermore, as Charles Cornell, MD points out in a commentary on this study, “Glucose variability appears to affect surgical prognosis more than chronic hyperglycemia.”
These findings were surprising and a bit concerning. I don’t tend to order routine blood-glucose measurements postoperatively on patients who appear to be euglycemic based on preoperative testing. Yet, according to these data, maybe I should. Findings of high glucose variability postoperatively might now prompt me to consult with endocrine or perioperative medicine specialists or at least consider informing patients with fluctuating glucose levels that they may be at increased risk of serious postoperative complications.
Measuring a patient’s blood sugar is neither challenging nor prohibitively expensive. So why don’t we monitor it more closely? Probably because, until now, we have not had a compelling reason to do so with “low-risk” patients. What this study suggests is that our definition of a “low-risk” patient from a glycemic-control standpoint may be misinformed. And while further research needs to be performed to corroborate these findings, that is a pretty scary thought to digest.
Chad A. Krueger, MD
JBJS Deputy Editor for Social Media