Glycemic Control and Prevention of Surgical Site Infection

Abstract

Background:

Stress hyperglycemia is associated with increased risk of surgical site infections (SSIs). Use of strict or tight glycemic control with intensive insulin therapy to prevent SSIs is controversial.

Methods:

Review of pertinent English-language literature.

Results:

There is a large body of literature supporting an association between stress hyperglycemia and SSIs. The quality of evidence from randomized controlled trials and meta-analyses that strict glycemic control reduces SSIs or any infections is low, and the strength of recommendation for strict glycemic control is weak due to the associated increase in moderate and severe hypoglycemia.

Conclusion:

Current recommendations for glycemic control in surgical patients are informed primarily by trials using intensive insulin therapy in critically ill patients. Further research is necessary to ascertain the optimal glycemic target for non-critically ill patients, to determine if subsets of patients may benefit from strict glycemic control, and to identify alternative methods for treating stress hyperglycemia and explaining the mechanisms by which it increases infectious risk.

Surgical site infections (SSIs) are associated with increased morbidity, mortality, re-admissions, hospital length of stay, resource utilization, and cost [1–2]. An estimated 55% of SSIs are preventable with evidence-based tactics such as antibiotic prophylaxis [3]. However, despite increased compliance with antibiotic prophylaxis guidelines, SSIs continue to be a substantial problem, suggesting that additional interventions are needed [4 –6].

Stress hyperglycemia is defined as acute, transient elevations in blood glucose concentrations that occur during illness [7]. Although described typically in non-diabetic patients, it may occur in patients with undiagnosed diabetes mellitus or impaired glucose tolerance and in diabetics with transient deterioration in glycemic control. Stress hyperglycemia has multiple consequences, including immunosuppression, effects on the cardiovascular system, pro-thrombotic effects, stimulation of pro-inflammatory mediators, endothelial cell dysfunction, and effects on the central nervous system [8]. Stress hyperglycemia has been associated with increased infectious complications and mortality in surgical patients including but not limited to trauma [9,10], burn [11], and post-operative cardiac surgery patients [12].

The 1999 U.S. Centers for Disease Control and Prevention (CDC) guidelines recommend that serum blood glucose concentrations should be controlled in diabetics and that hyperglycemia (>200 mg/dL) should be avoided peri-operatively [13]. In 2001, this recommendation was challenged by a single-center (Leuven) trial that reported that intensive insulin therapy (IIT; 80–110 mg/dL) reduced in-hospital mortality and morbidity significantly as compared to conventional therapy (180–200 mg/dL) in cardiac surgery patients [14]. Subsequently, IIT (also known as strict or tight glycemic control) became adopted widely. Since then, additional studies have questioned tight glycemic control in critically ill patients, resulting in updated guidelines and recommendations for glycemic control in the peri-operative period and in the intensive care unit (ICU). The goals of this review are: 1) To evaluate the evidence for a causal relationship between stress hyperglycemia and SSIs; 2) to compare the effectiveness of tight versus conventional glycemic control in preventing SSIs; 3) to describe current recommendations for peri-operative glycemic control; and 4) to pose questions for future research in this area.

Does stress hyperglycemia cause surgical site infections?

There are multiple observational studies demonstrating an association between stress hyperglycemia and SSIs (Table 1) [15 –27]. However, association does not always represent causality. Stress hyperglycemia may instead be a surrogate for increased severity of disease or a consequence rather than a cause of infection. The Bradford Hill criteria are a set of nine conditions that support a causal relationship between an exposure and an outcome [28]. These criteria can be used to examine the relationship between hyperglycemia and SSIs.

Table 1.

Examples of Studies (2008–2013) Demonstrating Strength and Consistency of Association between Hyperglycemia and SSIs in Adult Surgery Patients

Study	Number of patients	Diabetes status	Timing of glucose measurement	Glucose threshold (mg/dL)	Adjusted OR	95% CI
Breast surgery
Vilar-Compte et al. 2008¹⁵	260	Diabetic and non-diabetic	Post-op	150	2.9	1.2–6.2
Colorectal surgery
Jackson et al. 2012¹⁶	7,576	Diabetic and non-diabetic	Pre-op	160–200	1.4	1.1–1.9
Ho et al. 2011¹⁷	650	Diabetic and non-diabetic	Post-op	200	3.0^*	1.4–6.5^*
Serra-Aracil et al. 2011¹⁸	383 (colon)	Diabetic and non-diabetic	Post-op	200	4.5^*	1.1–18.9^*
Serra-Aracil et al. 2011¹⁸	228 (rectal)	Diabetic and non-diabetic	Post-op	200	9.4^*	1.0–86.0^*
Ata et al. 2010¹⁹	226	Diabetic and non-diabetic	Post-op	140	3.2	1.4–7.2
McConnell et al. 2009²⁰	149	Diabetic	Post-op	200	3.6	1.3–9.9
Colorectal and bariatric surgery
Kwon et al. 2013²¹	11,633	Diabetic and non-diabetic	Post-op	180	3.1	1.7–5.6
General and vascular surgery
Ramos et al. 2008²²	995	Diabetic and non-diabetic	Post-op	200	1.8	1.4–2.5
Hepatopancreaticobiliary surgery
Ambiru et al. 2008²³	265	Diabetic and non-diabetic	Post-op	200	6.6	3.5–12.5
Transplant surgery
Park et al. 2009²⁴	680	Diabetic and non-diabetic	Intra-op	200	2.3	1.3–4.0
Orthopedic surgery
Richards et al. 2012²⁵	790	Non-diabetic	Not specified	200	2.7	1.1–6.7
Jamsen et al. 2010²⁶	1,565	Diabetic and non-diabetic	Pre-op	126	4.4	1.3–14.8
Olsen et al. 2008²⁷	2,316	Diabetic and non-diabetic	Pre-op	125	3.3	1.4–7.5

Organ space surgical site infection only.

Pre-op=pre-operative; Post-op=post-operative; OR=odds ratio; CI=confidence interval.

Strength, biologic gradient, and consistency

There is a strong correlation between stress hyperglycemia and SSIs (strength) observed across different surgical populations and regardless of diabetic status (consistency; Table 1 [15 –27]). Several studies have also demonstrated a correlation between higher glucose concentrations and increased risk for SSIs (biologic gradient). A single-center study showed that among general surgery patients, SSI rates increased with each stratum of post-operative glucose concentrations (Table 2 [19]). Furthermore, post-operative glucose concentration was the only independent predictor of SSI. Another study identified a trend toward increased SSIs as glucose concentrations increased beyond 140 mg/dL (Table 2 [29]). The results were only statistically significant by uni-variate analysis, but the study may have been underpowered due to the low SSI rate (220/13,800 or 1.8%). A multi-center study of 11,633 bariatric and colorectal surgery patients demonstrated that the odds of infection increased depending on whether hyperglycemia (>180 mg/dL) occurred on the day of surgery alone (odds ratio [OR], 1.7; 95% confidence interval [CI] 0.98–2.94), only on post-operative day 1 or 2 (OR, 2.08; 95% CI 1.43–3.02), or on both post-operative days 1 and 2 (OR, 3.1; 95% CI 1.72–5.59) [21]. Although this study did not measure SSIs as a separate outcome, it provides additional support for a dose–response relationship between hyperglycemia and infection.

Table 2.

Biological Gradient between Pre- and Post-Operative Glucose Levels and SSIs

	Uni-variate analysis		Multi-variable analysis
Glucose level	OR	95% CI	OR	95% CI
Ata et al 2010¹⁹
Pre-operative glucose (mg/dL)
<110	1 [Reference]
111–140	1.24	0.10–2.53
141–180	1.08	0.38–3.12
181–220	5.87	2.38–14.44
>220	3.51	1.16–10.63
Post-operative glucose (mg/dL)
<110	1 [Reference]		1 [Reference]
111–140	3.61	1.22–10.67	3.61	1.22–10.67
141–180	6.26	2.17–18.17	6.26	2.17–18.17
181–220	5.92	1.73–20.22	5.92	1.73–20.22
>220	17.1	3.71–39.64	17.1	3.71–39.64

Jeon et al 2011²⁹
Pre-operative glucose (mg/dL)
<80	1.14	0.52–2.53	0.99	0.44–2.23
80–109	1 [Reference]		1 [Reference]
110–139	0.81	0.54–1.22	0.76	0.50–1.16
140–180	1.23	0.80–1.89	1.06	0.67–1.67
≥180	1.71	1.15–2.55	1.21	0.77–1.91
Post-operative glucose (mg/dL)
<80	1.53	0.35–6.75	1.41	0.31–6.31
80–109	1 [Reference]		1 [Reference]
110–139	1.19	0.66–2.11	1.28	0.71–2.30
140–180	1.27	0.73–2.20	1.30	0.73–2.31
≥180	2.53	1.50–4.27	1.72	0.84–1.13

SSIs=surgical site infections.

Temporality

The issue of temporality or whether hyperglycemia precedes SSIs has been addressed in studies of trauma patients. Admission hyperglycemia greater than 200 mg/dL in injured patients is associated with increased risk of infection [9,30,31]. In a retrospective study of critically ill trauma patients not treated with a standardized glycemic control protocol, there was an independent association (after adjustment for age and injury severity score) between higher glucose concentrations in the first week after trauma admission and subsequent skin and wound infections [32]. A later study that occurred after institution of a tight glycemic control protocol reported that acute glucose elevations could predict the presence of an infection [33]. The acute glucose elevations occurred on average two days prior to the diagnosis of infection. Although hyperglycemia preceding SSIs may still be a surrogate for physiologic status (and therefore represent patients who are at a higher risk for infection), these studies support a temporal relationship between hyperglycemia and SSIs.

Specificity

The existence of a one-to-one effect, or specificity, is the most difficult of the Bradford Hill criteria to prove for hyperglycemia and SSIs. There are many risk factors that can contribute to SSIs [34] and stress hyperglycemia is associated with a multitude of adverse effects other than infection [7]. However, epidemiologists have argued that the utility of the specificity criterion is uncertain when discussing complex causal systems [35]. Thus, the lack of specificity between hyperglycemia and SSIs does not necessarily exclude a causal relationship.

Plausibility

Stress hyperglycemia may contribute to the development of infectious complications and SSIs by impairing immunologic function. First, hyperglycemia affects neutrophil function, including respiratory burst, endothelial adherence, chemotaxis, phagocytosis, and bactericidal activity [36,37]. These deficiencies have been linked to an increased incidence of nosocomial infections in critically ill patients [38,39]. Furthermore, studies suggest that insulin administration may partially restore neutrophil function; insulin therapy has been demonstrated to improve neutrophil production of reactive oxygen species [40], bactericidal activity [41], and phagocytic activity [42,43]. Second, short-term hyperglycemia has been associated with non-enzymatic glycosylation of circulating immunoglobulin, which results in its inactivation [44]. Third, hyperglycemia may result in diminished T-cell-mediated cellular immunity; short-term hyperglycemia induces a decrease in white blood cell count and peripheral lymphocytes [45]. Thus, there is a plausible biologic explanation for a causal relationship between hyperglycemia and SSIs.

Coherence

The association between hyperglycemia and SSI is consistent with existing knowledge and theory. As described above, laboratory and animal data support the theory that hyperglycemia results in immune dysfunction, which increases risk of infection. Observational studies have demonstrated the association between hyperglycemia and SSIs, and trials have demonstrated reduction in SSIs with glycemic control (described below). Lastly, the association is consistent with the larger framework of modifiable risk factors that contribute to SSIs.

Analogy

One alternative explanation is that glycemic variability or fluctuations in glucose concentrations causes SSIs. Glycemic variability has been demonstrated to correlate with mortality in multiple studies, measured using different methods including standard deviation [46 –50], range [51 –54], and changes in glucose concentrations over time [49,55]. A study in critically ill trauma patients identified that highly variable glucose patterns within the first week after admission predicted infection [32]. However, a highly variable glucose pattern was not defined explicitly and SSIs were not a separate outcome measure. Another study reported a relationship between the coefficient of variation of glucose concentrations and mortality but not SSIs in surgical patients [29]. Currently, there is inadequate evidence to suggest that glycemic variability and not hyperglycemia causes SSIs, but further research is required.

Another potential confounder is that insulin has independent effects that could reduce infectious risk. For example, insulin has direct anti-inflammatory effects through inhibition of natural factor kappa β [56]. Studies in endotoxemic rats and thermally injured children with normal glucose concentrations showed that insulin administration resulted in decreased pro-inflammatory and increased anti-inflammatory cytokine expression [57]. A secondary analysis of the Leuven trial addressed the question of whether glycemic control or insulin resulted in the improved outcomes in cardiac surgery patients; the beneficial effects of intensive insulin therapy were correlated with the lowest glucose range rather than the total insulin dose [58]. The study concluded that the beneficial effects were a consequence of the metabolic control rather than of the properties of insulin.

Experiment

Prospective observational and randomized controlled trials have suggested that treatment of hyperglycemia reduces the risk of SSI. The Portland Diabetic Project was a longitudinal observational cohort study of diabetic patients undergoing coronary artery bypass grafting that began in 1987 [12,59]. Institution of a continuous insulin infusion protocol to maintain post-operative blood glucose concentrations below 150 mg/dL for three days reduced the risk of deep incisional sternal SSIs by 66% (p<0.001) [12]. Additional randomized trials and meta-analyses demonstrating a reduction in SSIs associated with tight glycemic control are described in the next section.

In summary, stress hyperglycemia is associated with SSIs and other infections. Many but not all of the Bradford Hill criteria are met; thus, causality cannot be proved definitively. Ultimately, the question most relevant to clinical practice is whether aggressive treatment of stress hyperglycemia reduces the risk of SSI.

Does tight glycemic control reduce surgical site infections?

In 2009, a Cochrane review compared strict and conventional glycemic control regimens for the prevention of SSIs in adults [60]. At that time, six randomized trials were identified that met the original inclusion and exclusion criteria. However, a meta-analysis was not possible due to substantial clinical heterogeneity between patient populations, target ranges, and differences in outcome assessment. The review concluded that there was insufficient evidence for strict glycemic control versus conventional management (<200 mg/dL) to prevent SSIs.

A subsequent meta-analysis of IIT in adult cardiac surgery patients included five randomized trials [61], and showed a reduced risk of infection (relative risk [RR], 0.50; 95% CI 0.29–0.84; I2=2%). There are several limitations to this meta-analysis. First, the primary outcome of infection was not defined explicitly. Inspection of the individual trials reveals that four of the trials evaluated SSIs exclusively, although the definitions were not standardized [62 –65]. One trial evaluated pneumonia and SSIs as a composite outcome [66]. Second, two of the five trials were from the same institution and had overlapping time frames for patient recruitment suggesting that patients may have been double-counted [62,63]. Third, one trial did not have adequate sequence generation for randomization, two did not have adequate allocation concealment, and none had any blinding [61].

In another meta-analysis that included both surgical and medical critically ill patients, IIT reduced the risk of sepsis (RR, 0.79; 95% CI 0.62–1.00; I2=53%) [67]. The risk of other infections including SSI, urinary tract, and pneumonia also was reduced (RR, 0.78; 95% CI 0.62–0.97; I2=51%). However, there was substantial clinical and methodological heterogeneity between studies, and the I2 values also suggest that there was heterogeneity between studies. Sensitivity analysis demonstrated that the heterogeneity was primarily due to the Leuven trial, which reported a significant reduction in sepsis. The Leuven trial has been criticized for the liberal use of parenteral nutrition, the lack of a published insulin protocol, and lack of reproducibility in other centers. Exclusion of the Leuven trial resulted in no sepsis benefit being identified (RR, 0.89; 95% CI 0.74–1.09; I2=29%) [67]. In addition, the largest randomized trial of tight glycemic control in both medical and surgical critically ill patients, the Normoglycemia in Intensive Care Evaluation-Survival Using Glucose Algorithm Regulation (NICE-SUGAR) trial, reported no difference in sepsis (RR, 1.04; 95% CI 0.91–1.19) [68].

In summary, the quality of the evidence for a reduction in SSIs, sepsis, or infection from tight glycemic control with IIT is poor. Furthermore, in determining the strength of recommendation for an intervention, other factors such as the balance between all of the risks and benefits must be considered [69].

What are the current recommendations for peri-operative glycemic control?

Systematic reviews evaluating the effects of strict glycemic control on mortality and hypoglycemia rates have concluded that there is no clear benefit and potential harm [67,70]. The NICE-SUGAR trial of more than 6,000 patients reported that IIT to keep blood glucose concentrations between 81 to 108 mg/dL increased mortality (OR, 1.14; 95% CI 1.02–1.28; p=0.02) [68]. In subgroup analyses, intensive control in patients undergoing an operation (OR, 1.31; 95% CI 1.07–1.61) was not statistically significant.

Intensive insulin therapy increased hypoglycemia (<40 mg/dL) significantly in the NICE-SUGAR trial (OR, 14.7; 95% CI 9.0–25.9; p<0.001). A follow-up study demonstrated that IIT was associated with moderate hypoglycemia (41 to 70 mg/dL) as well as severe hypoglycemia (<40 mg/dL), and that both were associated with increased risk of death [71]. These results are consistent with a meta-analysis, which suggests that IIT increases the risk of hypoglycemia significantly (RR, 6.00; 95% CI 4.06–9.87; I2=57%) [67]. Although a single hypoglycemic episode may seem inconsequential, multiple studies have reported that even one hypoglycemic episode is associated with an increased risk of mortality [72 –76].

The only current guidelines that include a target for intra-operative glycemic control are from the Society of Thoracic Surgeons [77]. Although the guidelines do not reference NICE-SUGAR, the recommendations (<180 mg/dL) are consistent with the trial's conclusions [68]. The American Association of Clinical Endocrinologists and American Diabetes Association consensus statement recommends a glucose range of 140 to 180 mg/dL in critically ill patients based on NICE-SUGAR [78] They recommend <140 mg/dL premeal and <180 mg/dL random glucose concentrations in non-critically ill patients treated with insulin, based on clinical experience and judgment [78]. Lastly, the American College of Physicians recommends the most liberal targets of 140 to 200 mg/dL in critically ill surgical patients and no less than 140 mg/dL in non-critically ill patients [79].

In 2012, the American College of Critical Care Medicine (ACCM) published guidelines for the use of an insulin infusion for the management of hyperglycemia in surgical and medical critically ill patients [80]. The recommendations are to institute protocols to achieve: 1) target glucose concentrations <150 mg/dL with an absolute maximum of 180 mg/dL; and 2) low rates of hypoglycemia (<70 mg/dL). This recommendation did not vary among subsets of patients including cardiac surgery and traumatic injury patients. The guidelines also provide recommendations for processes of care for insulin therapy and for quality improvement of glycemic control programs.

Thus, despite the passage of over a decade and the publication of numerous additional studies, the target glucose concentration has not changed significantly since the 1999 CDC recommendations for peri-operative glycemic control. Furthermore, the quality of evidence for the current recommendations remains low. Well-designed, prospective trials that incorporate the lessons learned to date are necessary to address remaining gaps in knowledge.

What are the future areas for research regarding glycemic control and prevention of SSIs?

1. What is the optimal glycemic target (defined as effective in reducing SSIs and other morbidity without significantly increasing hypoglycemia) for non-critically ill surgical patients? What is the optimal glycemic target in pediatric surgery patients? Given the concerns about hypoglycemia in ICU patients where glucose monitoring is recommended every 1–2 h among patients receiving an insulin infusion [80], studies implementing strict glycemic control on surgical wards have been few. In a randomized trial using sliding scale regular insulin to achieve a target of 100 to 140 mg/dL in non-critically ill general surgery patients, hypoglycemia <70 mg/dL occurred in almost 5% of patients [81]. In children, studies have been performed primarily in pediatric cardiac surgery patients because they are at high risk for hyperglycemia and therefore considered to be most likely to benefit from tight glycemic control. However, observational studies are conflicting as to whether stress hyperglycemia correlates to outcomes in this patient population [82 –84]. Furthermore, a recent two-center randomized trial found no difference in infection rate, mortality, length of stay, or organ failure in pediatric cardiac surgery patients receiving tight glycemic control (80 to 110 mg/dL) versus conventional care [85].

2. Are there subgroups of patients for whom strict glycemic control is recommended to prevent SSIs or to improve outcomes? The Leuven trial demonstrated the greatest morbidity and mortality benefit from IIT, one reason being that cardiac surgery patients may be different systematically from other patients. For example, cardiopulmonary bypass and hypothermia suppress insulin secretion, which may increase hyperglycemia [86]. As mentioned previously, a meta-analysis of cardiac surgery patients showed that IIT reduces infections although the analysis has substantive methodological flaws [61]. Based on the NICE-SUGAR trial, other subgroups that may receive a mortality benefit from tight glycemic control include critically ill trauma patients and patients receiving corticosteroids [68]. Future trials should focus on specific populations in whom a benefit may exist, or stratify randomization and plan a priori for subgroup analyses.

3. How should glycemic variability be measured? Does glycemic variability impact outcome including SSIs? What mechanisms account for the effects of glycemic variability? How can glycemic variability be minimized to improve outcomes? As previously mentioned, regardless of the measurement method, glycemic variability has been correlated with mortality [87]. In fact, one explanation put forth for the lack of mortality benefit in the NICE-SUGAR trial is that there was no significant difference in glycemic variability (standard deviation) between the treatment arms, but the intensive insulin group had a higher rate of hypoglycemia [68,87]. The ACCM guidelines for insulin infusion recommend that standard deviation and coefficient of variation (standard deviation/mean) be reported as measures of glycemic variability in all published studies [80]. Glycemic variability has been demonstrated to increase oxidative stress, leading to endothelial dysfunction and endothelial damage [88]. Multiple factors can influence glycemic variability including both patient and provider-level characteristics, such as severity of illness and intensity of care. Future research should investigate the effects of interventions, such as use of computerized algorithms to minimize glycemic variability from insulin infusions [89].

4. What is the effectiveness and safety of alternative methods to intravenous insulin infusion to achieve glycemic control and improve outcomes? Although randomized trials in critically ill patients have primarily studied the effectiveness of intravenous insulin infusions, other types of insulin have been evaluated. In the RABBIT-2 Surgery study, a basal-bolus insulin regimen with a long-acting (glargine) and short-acting (glulisine) insulin analogue was compared to sliding scale regular insulin in ward surgical patients [81]. Both treatment arms targeted fasting and premeal glucose levels between 100 and 140 mg/dL. Patients in the sliding scale insulin group had higher mean glucose levels and increased SSIs but a lower rate of serum glucose concentrations <70 mg/dL. Non-insulin oral hypoglycemic medications have a limited role in the management of inpatient hyperglycemia, particularly in a non-stable or fasting patient [78]. Nonetheless, there is limited evidence that metformin reduces stress-induced hyperglycemia in burn and non-diabetic critically ill patients but further study is required to quantify both potential benefits and harms [90,91].

In conclusion, there is good evidence that hyperglycemia is a risk factor for SSIs. Based on the Bradford Hill criteria, there may be a cause-effect relationship. There is low quality evidence that treatment of hyperglycemia with insulin reduces the risk of SSIs or infection in surgical patients, and there is good evidence that the risk of hypoglycemia is increased. Given the risks and benefits, the optimal upper threshold for glycemic control to prevent SSIs and other infectious complications appears to lie between the traditional target of 200 mg/dL and the strict target of 110 mg/dL. Future research should be aimed toward identifying subgroups in whom stricter control may be indicated; developing methods for reducing glycemic variability and the risk of hypoglycemia; and assessing the feasibility, effectiveness, and safety of achieving glycemic control in non-intensive care unit settings.

Author Disclosure Statement

No competing financial interests exist.

References

Kirkland

, Briggs

, Trivette

et al. The impact of surgical-site infections in the 1990s: Attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol, 1999; 20:725–730.

de Lissovoy

, Fraeman

, Hutchins

et al. Surgical site infection: Incidence and impact on hospital utilization and treatment costs. Am J Infect Control, 2009; 37:387–397.

Umscheid

, Mitchell

, Doshi

et al. Estimating the proportion of healthcare-associated infections that are reasonably preventable and the related mortality and costs. Infect Control Hosp Epidemiol, 2011; 32:101–114.

Stulberg

, Delaney

, Neuhauser

et al. Adherence to surgical care improvement project measures and the association with postoperative infections. JAMA, 2010; 303:2479–2485.

Hawn

, Vick

, Richman

et al. Surgical site infection prevention: time to move beyond the surgical care improvement program. Ann Surg, 2011; 254:494–501.

Ingraham

, Cohen

, Bilimoria

et al. Association of surgical care improvement project infection-related process measure compliance with risk-adjusted outcomes: implications for quality measurement. J Am Coll Surg, 2010; 211:705–714.

Dungan

, Braithwaite

, Preiser

. Stress hyperglycaemia. Lancet, 2009; 373:1798–807.

Clement

, Braithwaite

, Magee

et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care, 2004; 27:553–591.

Yendamuri

, Fulda

, Tinkoff

. Admission hyperglycemia as a prognostic indicator in trauma. J Trauma, 2003; 55:33–38.

10.

Bochicchio

, Salzano

, Joshi

et al. Admission preoperative glucose is predictive of morbidity and mortality in trauma patients who require immediate operative intervention. Am Surg, 2005; 171–174.

11.

Gore

, Chinkes

, Heggers

, Herndon

, Wolf

, Desai

. Association of hyperglycemia with increased mortality after severe burn injury. J Trauma, 2001; 51:540–544.

12.

Furnary

, Wu

, Bookin

. Effect of hyperglycemia and continuous intravenous insulin infusions on outcomes of cardiac surgical procedures: The Portland Diabetic Project. Endocr Pract, 2004; 10,Suppl 2:21–33.

13.

Mangram

, Horan

, Pearson

et al. Guideline for prevention of surgical site Infection, 1999. Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee. Am J Infect Control, 1999; 27:97–132.

14.

van den Berghe

, Wouters

, Weekers

et al. Intensive insulin therapy in the critically ill patients. N Engl J Med, 2001; 345:1359–1367.

15.

Vilar-Compte

, Alvarez de Iturbe

, Martin-Onraet

et al. Hyperglycemia as a risk factor for surgical site infections in patients undergoing mastectomy. Am J Infect Control, 2008; 36:192–198.

16.

Jackson

, Amdur

, White

, Macsata

. Hyperglycemia is associated with increased risk of morbidity and mortality after colectomy for cancer. J Am Coll Surg, 2012; 214:68–80.

17.

, Stein

, Trencheva

et al. Differing risk factors for incisional and organ/space surgical site infections following abdominal colorectal surgery. Dis Colon Rectum, 2011; 54:818–825.

18.

Serra-Aracil

, Garcia-Domingo

, Pares

et al. Surgical site infection in elective operations for colorectal cancer after the application of preventive measures. Arch Surg, 2011; 146:606–612.

19.

Ata

, Lee

, Bestle

et al. Postoperative hyperglycemia and surgical site infection in general surgery patients. Arch Surg, 2010; 145:858–864.

20.

McConnell

, Johnson

, Porter

. Surgical site infections following colorectal surgery in patients with diabetes: Association with postoperative hyperglycemia. J Gastrointest Surg, 2009; 13:508–515.

21.

Kwon

, Thompson

, Dellinger

et al. Importance of perioperative glycemic control in general surgery: A report from the surgical care and outcomes assessment program. Ann Surg, 2013; 257:8–14.

22.

Ramos

, Khalpey

, Lipsitz

et al. Relationship of perioperative hyperglycemia and postoperative infections in patients who undergo general and vascular surgery. Ann Surg, 2008; 248:585–591.

23.

Ambiru

, Kato

, Kimura

et al. Poor postoperative blood glucose control increases surgical site infections after surgery for hepato-biliary-pancreatic cancer: A prospective study in a high-volume institute in Japan. J Hosp Infect, 2008; 68:230–233.

24.

Park

, Hsu

, Neelakanta

et al. Severe intraoperative hyperglycemia is independently associated with surgical site infection after liver transplantation. Transplantation, 2009; 87:1031–1036.

25.

Richards

, Kauffmann

, Zuckerman

, Obremskey

, May

. Relationship of hyperglycemia and surgical-site infection in orthopaedic surgery. J Bone Joint Surg Am, 2012; 94:1181–1186.

26.

Jamsen

, Nevalainen

, Kalliovalkama

, Moilanen

. Preoperative hyperglycemia predicts infected total knee replacement. Eur J Intern Med, 2010; 21:196–201.

27.

Olsen

, Nepple

, Riew

et al. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Joint Surg Am, 2008; 90:62–69.

28.

Hill

. The environment and disease: Association or causation? Proc R Soc Med, 1965; 58:295–300.

29.

Jeon

, Furuya

, Berman

, Larson

. The role of pre-operative and post-operative glucose control in surgical-site infections and mortality. PLoS ONE, 2012; 7:e45616.

30.

Laird

, Miller

, Kilgo

et al. Relationship of early hyperglycemia to mortality in trauma patients. J Trauma, 2004; 56:1058–1062.

31.

Sung

, Bochicchio

, Joshi

et al. Admission hyperglycemia is predictive of outcome in critically ill trauma patients. J Trauma, 2005; 59:80–83.

32.

Bochicchio

, Sung

, Joshi

et al. Persistent hyperglycemia is predictive of outcome in critically ill trauma patients. J Trauma, 2005; 58:921–924.

33.

Bochicchio

, Bochicchio

, Joshi

et al. Acute glucose elevation is highly predictive of infection and outcome in critically injured trauma patients. Ann Surg, 2010; 252:597–602.

34.

Fry

. Fifty ways to cause surgical site infections. Surg Infect (Larchmt), 2011; 12:497–500.

35.

Hofler

. The Bradford Hill considerations on causality: a counterfactual perspective. Emerg Themes Epidemiol, 2005; 2:11.

36.

Delamaire

, Maugendre

, Moreno

et al. Impaired leucocyte functions in diabetic patients. Diabet Med, 1997; 14:29–34.

37.

Kwoun

, Ling

, Lydon

et al. Immunologic effects of acute hyperglycemia in nondiabetic rats. JPEN, 1997; 21:91–95.

38.

Simms

, D'Amico

. Posttraumatic auto-oxidative polymorphonuclear neutrophil receptor injury predicts the development of nosocomial infection. Arch Surg, 1997; 132:171–177.

39.

Stephan

, Yang

, Tankovic

et al. Impairment of polymorphonuclear neutrophil functions precedes nosocomial infections in critically ill patients. Crit Care Med, 2002; 30:315–322.

40.

Walrand

, Guillet

, Boirie

, Vasson

. In vivo evidences that insulin regulates human polymorphonuclear neutrophil functions. J Leukoc Biol, 2004.

41.

Gallacher

, Thomson

, Fraser

, Fisher

, Gemmell

, MacCuish

. Neutrophil bactericidal function in diabetes mellitus: Evidence for association with blood glucose control. Diabet Med, 1995; 12:916–920.

42.

Rassias

, Givan

, Marrin

et al. Insulin increases neutrophil count and phagocytic capacity after cardiac surgery. Anesth Analg, 2002; 94:1113–1119.

43.

Rassias

, Marrin

, Arruda

et al. Insulin infusion improves neutrophil function in diabetic cardiac surgery patients. Anesth Analg, 1999; 88:1011–1016.

44.

Black

, Hennessey

, Andrassy

. Short-term hyperglycemia depresses immunity through nonenzymatic glycosylation of circulating immunoglobulin. J Trauma, 1990; 30:830–832.

45.

von Kanel

, Mills

, Dimsdale

. Short-term hyperglycemia induces lymphopenia and lymphocyte subset redistribution. Life Sci, 2001; 69:255–262.

46.

Egi

, Bellomo

, Stachowski

, French

, Hart

. Variability of blood glucose concentration and short-term mortality in critically ill patients. Anesthesiology, 2006; 105:244–252.

47.

Krinsley

. Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med, 2008; 36:3008–3013.

48.

Meynaar

, Eslami

, Abu-Hanna

et al. Blood glucose amplitude variability as predictor for mortality in surgical and medical intensive care unit patients: a multicenter cohort study. J Crit Care, 2012; 27:119–124.

49.

Hermanides

, Vriesendorp

, Bosman

et al. Glucose variability is associated with intensive care unit mortality. Crit Care Med, 2010; 38:838–842.

50.

Ali

, O'Brien

Jr. , Dungan

et al. Glucose variability and mortality in patients with sepsis. Crit Care Med, 2008; 36:2316–2321.

51.

Hirshberg

, Larsen

, Van Duker

. Alterations in glucose homeostasis in the pediatric intensive care unit: Hyperglycemia and glucose variability are associated with increased mortality and morbidity. Pediatr Crit Care Med, 2008; 9:361–366.

52.

Eslami

, Taherzadeh

, Schultz

, Abu-Hanna

. Glucose variability measures and their effect on mortality: a systematic review. Intensive Care Med, 2011; 37:583–593.

53.

Bagshaw

, Bellomo

, Jacka

et al. The impact of early hypoglycemia and blood glucose variability on outcome in critical illness. Crit Care, 2009; 13:R91.

54.

Jacka

, Torok-Both

, Bagshaw

. Blood glucose control among critically ill patients with brain injury. Can J Neurol Sci, 2009; 36:436–442.

55.

Wintergerst

, Buckingham

, Gandrud

et al. Association of hypoglycemia, hyperglycemia, and glucose variability with morbidity and death in the pediatric intensive care unit. Pediatrics, 2006; 118:173–179.

56.

Hirsch

. Effect of insulin therapy on nonglycemic variables during acute illness. Endocr Pract, 2004; 10,Suppl 2:63–70.

57.

Jeschke

, Klein

, Herndon

. Insulin treatment improves the systemic inflammatory reaction to severe trauma. Ann Surg, 2004; 239:553–560.

58.

Van den Berghe

, Wouters

, Bouillon

et al. Outcome benefit of intensive insulin therapy in the critically ill: Insulin dose versus glycemic control. Crit Care Med, 2003; 31:359–366.

59.

Furnary

, Wu

. Eliminating the diabetic disadvantage: The Portland Diabetic Project. Semin Thorac Cardiovasc Surg, 2006; 302–308.

60.

Kao

, Meeks

, Moyer

, Lally

. Peri-operative glycaemic control regimens for preventing surgical site infections in adults. Cochrane Database Syst Rev, 2009; CD006806.

61.

Hua

, Chen

, Li

et al. Intensive intraoperative insulin therapy versus conventional insulin therapy during cardiac surgery: A meta-analysis. J Cardiothorac Vasc Anesth, 2012; 26:829–834.

62.

Albacker

, Carvalho

, Schricker

, Lachapelle

. High-dose insulin therapy attenuates systemic inflammatory response in coronary artery bypass grafting patients. Annals Thorac Surg, 2008; 86:20–27.

63.

Albacker

, Carvalho

, Schricker

, Lachapelle

. Myocardial protection during elective coronary artery bypass grafting using high-dose insulin therapy. Annals Thorac Surg, 2007; 84:1920–1927.

64.

Gandhi

, Nuttall

, Abel

et al. Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Ann Intern Med, 2007; 146:233–243.

65.

Chan

, Galas

, Hajjar

et al. Intensive perioperative glucose control does not improve outcomes of patients submitted to open-heart surgery: A randomized controlled trial. Clinics (Sao Paulo), 2009; 64:51–60.

66.

Lazar

, Chipkin

, Fitzgerald

et al. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events[see comment].2004; 1497–1502.

67.

Kansagara

, Fu

, Freeman

et al. Intensive insulin therapy in hospitalized patients: A systematic review. Ann Intern Med, 2011; 154:268–282.

68.

Finfer

, Chittock

, Su

et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med, 2009; 360:1283–1297.

69.

Guyatt

, Oxman

, Vist

et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ, 2008; 336:924–926.

70.

Griesdale

, de Souza

, van Dam

et al. Intensive insulin therapy and mortality among critically ill patients: A meta-analysis including NICE-SUGAR study data. CMAJ, 2009; 180:821–827.

71.

Finfer

, Liu

, Chittock

et al. Hypoglycemia and risk of death in critically ill patients. N Engl J Med, 2012; 367:1108–1118.

72.

Arabi

, Tamim

, Rishu

. Hypoglycemia with intensive insulin therapy in critically ill patients: Predisposing factors and association with mortality. Crit Care Med, 2009; 37:2536–2544.

73.

Krinsley

, Grover

. Severe hypoglycemia in critically ill patients: Risk factors and outcomes. Crit Care Med, 2007; 35:2262–2267.

74.

Hermanides

, Bosman

, Vriesendorp

et al. Hypoglycemia is associated with intensive care unit mortality. Crit Care Med, 2010; 38:1430–1434.

75.

Vriesendorp

, DeVries

, Rosendaal

, Hoekstra

. Severe hypoglycemia in critically ill: risk and outcomes. Crit Care Med, 2008; 36:1390.

76.

Vriesendorp

, DeVries

, van Santen

et al. Evaluation of short-term consequences of hypoglycemia in an intensive care unit. Crit Care Med, 2006; 34:2714–2718.

77.

Lazar

, McDonnell

, Chipkin

et al. The Society of Thoracic Surgeons practice guideline series: Blood glucose management during adult cardiac surgery. Ann Thorac Surg, 2009; 87:663–669.

78.

Moghissi

, Korytkowski

, DiNardo

et al. American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Endocr Pract, 2009; 15:353–369.

79.

Qaseem

, Humphrey

, Chou

et al. Use of intensive insulin therapy for the management of glycemic control in hospitalized patients: A clinical practice guideline from the American College of Physicians. Ann Intern Med, 2011; 154:260–267.

80.

Jacobi

, Bircher

, Krinsley

et al. Guidelines for the use of an insulin infusion for the management of hyperglycemia in critically ill patients. Crit Care Med, 2012; 40:3251–376.

81.

Umpierrez

, Smiley

, Jacobs

et al.

Randomized study of basal-bolus insulin therapy in the inpatient management of patients with type 2 diabetes undergoing general surgery (RABBIT 2 surgery)

Diabetes Care, 2011; 34:256–261.

82.

Ulate

, Raj

, Rotta

. Critical illness hyperglycemia in pediatric cardiac surgery. J Diabetes Sci Technol, 2012; 6:29–36.

83.

DeCampli

, Olsen

, Munro

, Felix

. Perioperative hyperglycemia: effect on outcome after infant congenital heart surgery. Ann Thorac Surg, 2010; 89:181–185.

84.

Yates

, Dyke

2nd , Taeed

et al. Hyperglycemia is a marker for poor outcome in the postoperative pediatric cardiac patient. Pediatr Crit Care Med, 2006; 7:351–355.

85.

Agus

, Steil

, Wypij

et al. Tight glycemic control versus standard care after pediatric cardiac surgery. N Engl J Med, 2012; 367:1208–1219.

86.

Kawahito

, Kitahata

, Kitagawa

, Oshita

. Intensive insulin therapy during cardiovascular surgery. J Med Invest, 2010; 57:191–204.

87.

Egi

, Bellomo

, Reade

. Is reducing variability of blood glucose the real but hidden target of intensive insulin therapy? Crit Care, 2009; 13:302.

88.

Monnier

, Mas

, Ginet

et al. Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. JAMA, 2006; 295:1681–1687.

89.

Horibe

, Nair

, Yurina

, Neradilek

, Rozet

. A novel computerized fading memory algorithm for glycemic control in postoperative surgical patients. Anesth Analg, 2012; 115:580–587.

90.

Gore

, Wolf

, Herndon

, Wolfe

. Metformin blunts stress-induced hyperglycemia after thermal injury. J Trauma, 2003; 54:555–561.

91.

Mojtahedzadeh

, Jafarieh

, Najafi

et al. Comparison of metformin and insulin in the control of hyperglycaemia in non-diabetic critically ill patients. Endokrynol Pol, 2012; 63:206–211.