Background: Surgical site infections (SSIs) are associated with significant morbidity, mortality, and resource utilization and are potentially preventable. Peri-operative hyperglycaemia has been associated with increased SSIs and previous recommendations have been to treat glucose levels above 200 mg/dL. However, recent studies have questioned the optimal glycaemic control regimen to prevent SSIs. Whether the benefits of strict or intensive glycaemic control with insulin infusion as compared to conventional management outweigh the risks remains controversial.
Objectives: To summarise the evidence for the impact of glycaemic control in the peri-operative period on the incidence of surgical site infections, hypoglycaemia, level of glycaemic control, all-cause and infection-related mortality, and hospital length of stay and to investigate for differences of effect between different levels of glycaemic control.
Search strategy: A search strategy was developed to search the following databases: Cochrane Wounds Group Specialised Register (searched 25 March 2009), The Cochrane Central Register of Controlled Trials, The Cochrane Library 2009, Issue 1; Ovid MEDLINE (1950 to March Week 2 2009); Ovid EMBASE (1980 to 2009 Week 12) and EBSCO CINAHL (1982 to March Week 3 2009). The search was not limited by language or publication status.
Selection criteria: Randomised controlled trials (RCTs) were eligible for inclusion if they evaluated two (or more) glycaemic control regimens in the peri-operative period (within one week pre-, intra-, and/or post-operative) and reported surgical site infections as an outcome.
Data collection and analysis: The standard method for conducting a systematic review in accordance with the Cochrane Wounds Group was used. Two review authors independently reviewed the results from the database searches and identified relevant studies. Two review authors extracted study data and outcomes from each study and reviewed each study for methodological quality. Any disagreement was resolved by discussion or by referral to a third review author.
Main results: Five RCTs met the pre-specified inclusion criteria for this review. No trials evaluated strict glycaemic control in the immediate pre-operative period or outside the intensive care unit. Due to heterogeneity in patient populations, peri-operative period, glycaemic target, route of insulin administration, and definitions of outcome measures, combination of the results of the five included trials into a meta-analysis was not appropriate. The methodological quality of the trials was variable. In terms of outcomes, only one trial demonstrated a significant reduction in SSIs with strict glycaemic control, but the quality of this trial was difficult to assess as a result of poor reporting; furthermore the baseline rate of SSIs was high (30%). The other trials were either underpowered to detect a difference in SSIs, due to a low baseline rate (less than or equal to 5%), or did not report SSIs as a single outcome but as part of a composite. Of the three trials reporting hypoglycaemia (which was not consistently defined) all had a higher rate in the strict glycaemic control group but none attributed significant morbidity to the hypoglycaemia. Adequacy of glucose control between groups was measured differently among studies. Studies could not be compared due to differences in target ranges, and were susceptible to measurement bias due to differences in frequency of measurement and lack of blinding by the providers following the glycaemic protocols. Infection-related mortality was not reported in any of the trials, and no trials demonstrated a significant difference in all-cause mortality. Length of hospital stay was significantly reduced in the strict glycaemic control groups in only one trial.
Authors' conclusions: There is insufficient evidence to support strict glycaemic control versus conventional management (maintenance of glucose < 200 mg/dL) for the prevention of SSIs. No trials were found that evaluated strict glycaemic control in the immediate pre-operative period or outside the setting of an intensive care unit. The trials were limited by small sample size, inconsistencies in the definitions of the outcome measures and methodological quality. Further large randomised trials are required to address this question and may be most appropriately performed in patients at high risk for SSIs.