Background: There have been several reports about target blood glucose levels in the perioperative period, and recently, glycemic variability was reported to be associated with mortality in surgical patients in the intensive care unit (ICU). Meanwhile, evaluation of protracted postoperative wound healing has been considered important for determining the outcomes in surgical patients. Infection, poor nutritional status, anemia, jaundice, and chronic renal failure have been identified as risk factors for protracted wound healing. In addition to these risk factors, glycemic variability may be associated with protracted wound healing or poor prognosis in non-critically ill surgical patients with diabetes. However, studies on the relationship between glycemic variability and protracted wound healing in these patients are limited.
Aims: To investigate the relationship between protracted wound healing and glycemic variability as well as known risk factors in non-critically ill surgical patients with diabetes who underwent interventions for glycemic control in the perioperative period.
Method: We retrospectively analyzed 372 non-critically ill surgical patients with diabetes who were hospitalized and whose glucose levels were monitored >3 times per day from April 2014 to March 2016 in two centers. Non-critically ill surgical patients were defined as surgical patients who did not require longer time of care in the ICU. Protracted wound healing was evaluated by the number of days from surgery to removal of stitches on a cutaneous wound (days of wound with stitches). White blood cell counts; levels of C-reactive protein, albumin, hemoglobin, and total bilirubin; estimated glomerular filtration rate; and body temperature were adopted as risk factors for protracted wound healing (known risk factors). Here, we analyzed the relationship between days of wound with stitches and perioperative changes in these factors (the day of stitch removal – preoperative) as well as preoperative these factors using linear regression analysis. Furthermore, hypoglycemia was defined as a blood glucose level <70 mg/dL on any test performed in the hospital. Hypoglycemia, glycemic variability (standard deviation [SD], coefficient of variation [CV], and average daily risk range [ADRR]), and mean glucose concentrations were determined from all the glycemic data collected during the period of days the wound had stitches. We analyzed the association between days of wound with stitches and these factors using linear regression analysis as well. We also analyzed the association of mortality with known risk factors on day 1 of hospital stay (preoperative) and change in these factors (day of discharge from hospital − day 1 of hospital stay [preoperative]) using logistic regression analysis. Lastly, we analyzed the association between mortality and those glycemic variability indexes that were determined from all the glycemic data collected during the entire hospital stay using logistic regression analysis.
Results: According to the univariate analysis, none of the preoperative known risk factors and none of the perioperative changes in these factors were significantly associated with days of wound with stitches and mortality. In contrast, hypoglycemia, ADRR, CV, SD, and mean glucose concentrations were significantly associated with days of wound with stitches and mortality. After stepwise multivariate adjustment, high ADRR, high CV, and high SD were independently significantly associated with long days of wound with stitches, with standardized partial regression coefficients (β) of 0.23, 0.49, and 0.41, respectively (p < 0.0001 for all three results). Moreover, hypoglycemia, high ADRR, high CV, high SD, and high mean glucose concentrations were independently significantly associated with increased mortality according to the following measures in respective order: (1) odds ratio (OR), 16.83; 95% confidence interval (CI), 6.26–45.25; p < 0.0001; (2) OR, 1.10; 95% CI, 1.06–1.15; p < 0.0001; (3) OR, 1.10; 95% CI, 1.04–1.16; p = 0.001; (4) OR, 1.05; 95% CI, 1.03–1.07; p < 0.0001; and (5) OR, 1.03; 95% CI, 1.02–1.05; p < 0.0001.
Discussion: Glycemic variability, rather than known risk factors, was found to be associated with protracted wound healing in non-critically ill surgical patients with diabetes. This is supported by the fact that increased glycemic variability causes oxidative stress, which in turn causes circulatory disturbance of microvessels that nourish the wound, resulting in protracted wound healing.

Glycemic variability affects protracted postoperative wound healing in patients with diabetes ~multicenter study~