Best Practice & Research Clinical Anaesthesiology Vol. 22, No. 3, pp.

519535, 2008
doi:10.1016/j.bpa.2008.06.005 available online at http://www.sciencedirect.com

7 Diabetes, Hyperglycemia, and Infections*

Ashley M. Shilling * Jacob Raphael
MD MD

Assistant Professor of Anesthesiology

Assistant Professor of Anesthesiology University of Virginia Medical Center, PO Box 800710, Charlottesville, VA 22908, USA

Postoperative infection is not only a major source of morbidity and mortality in patients undergoing surgery, but also an important cause of increased hospital stay and resource utilization. Diabetes has been shown in multiple studies to increase the risk of post-surgical infection. More recently, hyperglycemia has been investigated as an independent risk factor for postoperative infection. This paper will review the effects of intra-operative, postoperative, and longterm glycemic control on postoperative infection rates. The mechanisms by which surgery causes hyperglycemia will be reviewed, as well as the immunologic and humeral effects of hyperglycemia. Key words: diabetes; hyperglycemia; surgical infection; anesthesia; infectious complications.

Postoperative infections are a devastating complication for patients undergoing surgery. They are the second most common cause of nosocomial infection and account for approximately 17% of all hospital-acquired infections.1 These infections lead to longer hospital and intensive care unit stays, substantially increased mortality, and contribute signicantly to healthcare costs.2 For centuries, scientists have searched for infectious risk factors, as well as for means of infection modication and prevention. It has been well established that patients with Diabetes Mellitus (DM) are at increased risk for both surgical and nosocomial infections.35 Infection rates have been quoted to be two to ve times more prevalent in diabetics than in the nondiabetic population.6 A large, albeit retrospective cohort study, showed that nearly half of all diabetics have at least one yearly hospitalization or physician claim for an infection.7 The relative risk for diabetic infectious disease-related hospitalization was 2.17. Further, death attributed to infection was signicantly higher in diabetics. A large,
* Disclosures: This review was funded with departmental and institutional support. * Corresponding author. Tel.: 1 434 924 2283; Fax: 1 434 982 0019. E-mail addresses: abm5f@virginia.edu (A.M. Shilling), jr5ef@virginia.edu (J. Raphael).

1521-6896/$ - see front matter 2008 Elsevier Ltd. All rights reserved.

520 A. M. Shilling and J. Raphael

12-month prospective study revealed that patients with DM had increased risk for lower respiratory tract, urinary tract, skin, and mucous membrane infections.8 One still-unanswered question, however, is whether this increased risk is due to hyperglycemia itself, or other associated features found in the diabetic disease state such as microvascular changes or immunologic dysfunction. A large number of studies have looked to answer this question. If perioperative hyperglycemia is an independent risk factor for post-surgical infection, anesthesiologists and surgeons may be empowered to signicantly affect infectious outcomes without changing the more difcult-to-control variables of diabetic status or long-term glycemic control. Due to the consistency among studies and compelling evidence that diabetes leads to increased infection risk, major focus has been placed on whether hyperglycemia, as an independent risk factor, is associated with increased infection. Even in nondiabetics, hyperglycemia is associated with an increased risk of morbidity and mortality.9 The overwhelming majority of studies examining hyperglycemia and surgical infection have focused on the cardiac population, likely because 1628% of patients undergoing cardiac surgery are diabetic.1012 Furthermore, deep sternal wound infections, though rare, remain a devastating and often lethal complication following cardiothoracic surgery, with mortality as high as 14%.13,14 Increased hospitalization, resource utilization, and morbidity are only a few of the consequences of sternal wound infections. A landmark study, published in 2001, by van der Berghe and coworkers, cast light on the signicance of postoperative hyperglycemia.15 This study included ICU patients, the majority of which had undergone cardiac surgery. They showed that treating hyperglycemia aggressively with a target blood glucose level between 80110 g/dL in the surgical intensive care unit signicantly reduced morbidity and mortality, both in the intensive care unit and during hospital stay. Patients treated aggressively achieved not only lower glucose levels with increased in-hospital survival and decreased infections, but also experienced long-term effects of increased one-year survival. One-year mortality was decreased from 8.0% to 4.6%. The effect was greatest in patients with multi-organ failure in the face of sepsis. The results of this trial instigated a urry of studies and aroused interest in the deleterious effects of hyperglycemia, with focus on the cardiac surgical population. POSTOPERATIVE HYPERGLYCEMIA IN THE CARDIAC SURGERY PATIENT In a retrospective study, Zerr and colleagues at the Portland Diabetic Project established that increased mean glucose levels for the rst two days following cardiac surgery is an independent risk factor for deep sternal wound infection in diabetics.16 This group found that a continuous intravenous insulin protocol was superior to standard subcutaneous insulin for glycemic control. Improved glycemic control decreased deep sternal wound infections from 2.4% to 1.5%. The same group conducted a prospective study of 2,467 diabetic patients undergoing cardiac surgery.17 They looked at both the efcacy of subcutaneous insulin injections versus continuous intravenous insulin infusions in maintaining glucose less than 200 mg/dL, as well as deep sternal wound infection rates between these two groups. Their results indicated that, indeed, postoperative hyperglycemia was associated with deep sternal wound infection in diabetics. Furthermore, a continuous insulin infusion, which they also found to be superior to subcutaneous insulin in reaching target glucose levels, resulted in a 66% reduction in deep sternal wound infection. They concluded that hyperglycemia in the postoperative setting may be

Diabetes, Hyperglycemia, and Infections 521

a causative factor for deep sternal wound infections. Of note, this group looked solely at the diabetic population and postoperative glucose control. A multitude of studies prior to, and following the Portland group ndings, have conrmed the deleterious effects of postoperative hyperglycemia and infection in the diabetic cardiac surgery patient population. A prospective study of 761 cardiac surgery patients showed not only that diabetics were at an increased risk for wound infections, but also that strict glucose control using an insulin infusion to maintain glucose between 120160 mg/dL signicantly reduced the risk of wound infection in diabetics.6 A chart review of 411 diabetic patients undergoing coronary artery bypass grafting (CABG) demonstrated a signicant increase in risk of developing infections including leg and sternal wound, urinary tract infection (UTI), and pneumonia when serum glucose levels (measured up to 36 hours after surgery) were higher than 200 mg/dL.18 Yet another retrospective study of diabetics undergoing CABG found not only poor glycemic control in their population, but also an increase in adverse outcomes including sepsis and infection in patients with elevated blood glucose within the rst 24 hours postoperatively.19 In addition to insulin treatment, Lazar and colleagues used a glucose, insulin, potassium (GIK) protocol in diabetics undergoing CABG and found improved 2 year survival and fewer wound infections (1% versus 10%) in patients randomized to GIK therapy with target glucose of 125200 mg/dL versus standard therapy.20 This group hypothesized that GIK, and specically the insulin component, improved endothelial function, decreased vascular inammation, and reduced thrombogenicity. The number and consistency of these studies indicates that diabetics undergoing cardiac surgery who experience postoperative hyperglycemia are likely at increased risk for infectious complications. Of note, the target glucose levels for these studies ranged from 120200 mg/dL, while the van der Berghe target was in the lower range of 80110 mg/dL. Postoperative hyperglycemia may also increase infection in the nondiabetic cardiac patient. One prospective study of diabetic and nondiabetic patients undergoing CABG found that those nondiabetics with postoperative hyperglycemia had an increased incidence of mediastinitis.21 Thus, tight postoperative glucose control may not only benet the diabetic cardiac patient, but also the nondiabetic patient. There has been little data refuting the correlation between postoperative hyperglycemia and infection in cardiac patients, however one large retrospective study in 2003 of 1574 patients, (35% with diabetes) undergoing CABG did not nd a statistically signicant increase in surgical or nosocomial infection rates in patients with postoperative hyperglycemia.22 The authors conrmed that diabetics have higher infection rates and found that each 50 mg/dL increase in blood glucose was associated with longer postoperative hospitalization and higher hospitalization cost. One important feature of this study is the large percentage of nondiabetic patients. Whether the non-diabetic patient benets equally from tight postoperative glycemic control is less certain. Despite this study, the majority of investigations have shown improved infectious outcomes with aggressive postoperative glucose control in the setting of cardiac surgery, with data being stronger in the diabetic population. INTRAOPERATIVE HYPERGLYCEMIA IN THE CARDIAC SURGERY PATIENT While the majority of studies demonstrate increased infection in cardiac patients with elevated postoperative glucose levels, the evidence for improved outcomes with

522 A. M. Shilling and J. Raphael

tighter intra-operative glucose control is less compelling. A retrospective study by Gandhi in 2005 found that increased mean introperative glucose in cardiac surgery patients was not a predictor of infection.23 A more recent prospective, randomized study by the same group examined 400 diabetic and nondiabetic cardiac surgery patients. They investigated tight intraoperative glucose control (80100 mg/dL) versus treatment of intraoperative glucose when levels exceeded 200 mg/dl.24 Postoperatively, all patients received tight glucose control to maintain normoglycemia. Though not adequately powered to detect differences in deep sternal wound infections, aggressive intraoperative insulin therapy did not reduce the incidence of sternal wound infection. Furthermore, they found a statistically signicant increased risk of stroke in the intensive glucose control groups and a non-statistically signicant trend towards increased death. Notably, the glucose targets for the Gandhi paper most closely resembles that from van der Berghe15, while the majority of other studies had higher glucose targets. Another study casting doubt on strict intraoperative glucose control was a prospective, observational study by Ouattara looking at intraoperative glucose control in 200 diabetic cardiac surgery patients.25 This group showed that infection was not signicantly more frequent in patients with poor glycemic control, though other outcomes of morbidity including cardiovascular, neurologic, and renal were associated with hyperglycemia. Thus, the data for infection and intra-operative glucose control is less clear and Gandhis study24 actually raises concerns for aggressive intra-operative glucose control. While severe postoperative hyperglycemia may be detrimental, overzealous intra-operative glucose control may lead to worsened neurologic outcomes. POSTOPERATIVE HYPERGLYCEMIA IN THE NONCARDIAC PATIENT While data from the cardiac surgery population is often retrospective, the studies of hyperglycemia and infection in the noncardiac population are almost nonexistent. A retrospective study looking at patients undergoing infrainguinal vascular surgery investigated hyperglycemia within 48 hours following surgery.26 Thirty-one percent of patients developed a postoperative infection requiring antibiotic treatment within 30 days of surgery. Infections included surgical wound infections, graft infections, pneumonia, and urinary tract infections. The majority of infections were wound-related (23%). Using univariate logistic regression analysis, postoperative hyperglycemia was associated with increased postoperative infections (Figure 1). Interestingly, after

* **

***

Figure 1. Percentage of patients undergoing peripheral vascular surgery who experienced at least one infection per glucose quartile; * P0.006; **P0.28; ***P-0.66. Taken from source26 with permission.

Diabetes, Hyperglycemia, and Infections 523

correcting for postoperative glucose levels, they found Type II diabetes to be associated with fewer postoperative infections. One explanation suggested by the authors was that insulin may have a benecial, independent effect on infection and the diabetic patients treated for hyperglycemia experienced the immunologic and anti-inammatory benets of insulin. Vriesendorp and colleagues also performed a restrospective study of 151 ASA I and II patients undergoing oesophagectomy.27 They examined the relationship between 48 hour post-operative glucose levels and infection. Infections included pneumonia, wound and urinary tract infection, and sepsis. Ninety-seven percent of their patients developed hyperglycemia (glucose > 6.1 mmol/L) within the rst 48 postoperative hours. Thirty-six percent of patients had at least one infectious complication and 9.9% had greater than one infection. Using univariate and multivariate regression analyses, no association was found between postoperative glucose levels and infection or insulin administration and infection, despite the common incidence of hyperglycemia in this patient population. Though retrospective in nature, these patients were enrolled in a prospective, randomized study of different outcomes. Reasons for this conicting result compared to the majority of other studies may be due to the patient population studied: little cardiovascular disease was present due to ASA 1 or 2 enrollment requirements. The authors speculate that in a population with minimal or no cardiac and vascular disease, increased glucose is not associated with infection risk. Thus, the major benets of treating hyperglycemia may be gained only from patients with cardiovascular compromise or diabetes. A prospective, randomized study of 78 patients, undergoing intracerebral aneurysm clipping after subarachnoid hemorrhage, examined intensive (blood glucose levels 80 120 mg/dL) versus conventional insulin therapy (blood glucose levels 80220 mg/dL) begun intra-op and continued up to 14 days postoperatively or until ICU discharge.28 The authors found infection rates of 42% in conventional treatment group versus 27% in the intensive insulin treatment group. Pneumonia was the most commonly-occurring infection followed by UTI, wound, and sepsis. Interestingly, incidence of vasospasm, neurologic outcomes, and mortality were not statistically signicant between the two groups. Another group, led by Pomposelli29, examined postoperative glucose levels and the rate of nosocomial infections. One hundred diabetic patients without existing infection who were undergoing elective abdominal or cardiac surgery were monitored for perioperative glucose control and postoperative infection. At least one serum glucose level greater than 220 mg/dL on POD 1 was 87.5% sensitive but nonspecic (33.3%) in predicting a nosocomial infection. In patients with hyperglycemia, the infection rate was 2.7 times higher than those with all serum glucose values less than 220 mg/dL. The authors concluded that patients with early postoperative hyperglycemia may have an increased incidence of infection; however, these ndings must be interpreted carefully as association does not necessarily mean causation. Another small, prospective, randomized study of general surgery ICU patients reconrmed those ndings as well and found a reduced incidence of nosocomial infection in strict versus standard glucose control protocols.30 PREOPERATIVE GLUCOSE CONTROL IN THE DIABETIC AND NONDIABETIC SURGICAL POPULATION While the majority of studies have looked at the more-easily controlled and measured variables of intraoperative or postoperative glucose, there have been several cardiac

524 A. M. Shilling and J. Raphael

and noncardiac studies investigating long-term preoperative glycemic control and postoperative infections. When looking at long-term glucose control, HbA1C is often used as a surrogate measure. It is accepted that HbA1C reects average glucose levels over a time period of 90 days. In a small, prospective study in 1992, Bishop and colleagues examined 90 patients undergoing penile prosthesis surgery.31 Approximately one third of the group was diabetic. They found that all periprosthetic infections occurred in diabetic patients and the majority of these (80%) were in patients with poorly-controlled diabetes (HbA1C greater than 11.5%). They concluded that a preoperative elevation in HbA1C is a strong indicator of postoperative infection and recommended denying surgery to patients until better glucose control is obtained. Despite the small numbers, this became the standard of care for patients undergoing penile prosthesis surgery but was later challenged by a larger, prospective study. Wilson and colleagues enrolled 389 patients undergoing elective penile prosthesis, the majority of which were not diabetic (n 275).32 The investigators used the previously cited 11.5% HbA1C as their cut-off for poor glucose control. Unlike Bishop, they found the incidence of surgical site infection to be signicantly lower, and found no association between elevated HbA1C levels and surgical site infections. They concluded that HbA1C levels may not be a clinical predictor of postoperative infection in penile implant surgery. Despite the conicting results of these two studies, it is well dened that a HbA1C of 11.5 correlates with a blood glucose level of more than 310 mg/dL. Using this value in the aforementioned studies essentially deems an average glucose below 310 mg/dL in the diabetic and nondiabetic population acceptable prior to undergoing surgery. Thus, results from these studies may be signicantly different when using a more acceptable HbA1C level. A prospective study by Latham investigated the role of long-term glucose control versus postoperative glucose control on infection in the cardiac surgery population.33 They examined diabetic status, postoperative glucose, and HbA1C levels above or below 8 mg/dL, and found that the rate of surgical site infection correlated with the degree of hyperglycemia during the postoperative period. Hyperglycemia, regardless of diabetic status, was an independent risk factor for surgical site infection. Postoperative glucose greater than 200 increased surgical site infection two-fold. They also found that a history of diabetes was associated with a 2.7-fold increase in the risk of developing a wound infection. Interestingly, poor chronic glucose control, as evidenced by HbA1C greater than 8, was not associated with an increase in infection rate. Despite this, the authors found that patients with a HbA1C level less than 8% had signicantly less postoperative hyperglycemia than those with higher levels. Another study of patients undergoing CABG did not nd a correlation between HbA1C levels preoperatively and sternal wound infections.21 Dronge and colleagues retrospectively examined 490 diabetic patients undergoing noncardiac surgery and their preoperative HbA1C levels.34 Their primary outcome was any form of infectionsurgical or nosocomial. Notably, they used a more universally-accepted denition of acceptable glucose control (HbA1C less than 7%) and found that diabetic patients with a HbA1C level less than 7% were signicantly less likely to experience a postoperative infectious complications. This study chose the most commonly accepted measure of acceptable long-term glycemic control and is more applicable to current practices. Trick and colleagues14 found that diabetes is an independent risk factor for deep sternal wound infection only when preoperative glucose is greater than 200 mg/dL. Guvener35, in a retrospective cohort study, conrmed that diabetics undergoing CABG have a higher incidence of infectious complications than nondiabetics. They

Diabetes, Hyperglycemia, and Infections 525

also found preoperative glucose to be a predictor of post-operative infection after CABG in diabetics. Reasons why better preoperative glucose control may offer improved postoperative infectious outcomes is likely multifactorial. First, better preoperative control of glucose may confer better glucose regimens and postoperative control. Indeed, Latham showed that patients with HbA1C less than 8% had less hyperglycemia postoperatively.33 Secondly, patients with better long-term glucose control may have better general health and an improved metabolic milieu that may be protective from infection.34 Indeed, the evidence seems compelling that not only diabetes, but also postoperative hyperglycemia, increases the risk of developing postoperative infection, particularly in the cardiac surgical population. The data suggests, but is less convincing, for intra-operative and long-term glucose control. Further, there is evidence that overzealous intra-operative glucose control may be neurologically detrimental. Because anesthesiologists often do not have the opportunity to play a major role in longterm maximization of patients prior to surgery, the evidence supporting improved outcomes with postoperative and potentially intra-operative glucose control is encouraging. These variables are obviously more modiable and could offer signicant benets both immediately in the postoperative period, and also in the long-term. What still remains unclear is the glucose threshold in which patient risk increases. It then becomes an important question as to why postoperative glucose control reduces the incidence of surgical infections, and whether improved long-term control or intraoperative control has benets. One further important question is whether glucose variability is an important factor for infection. SURGICAL STRESS RESPONSE Even in the 1800s, the serious challenges posed by diabetes were acknowledged. Treves wrote, Diabetes offers a serious bar to any kind of operation, and injuries involving open wounds, haemorrhage, or damage to the blood vessels are exceeding grave in subjects of this disease. A wound in the diabetic patient will probably not heal while the tissues appear to offer the most favourable soil for the development of putrefaction and pyogenic bacteria. The wound gapes, suppurates, and sloughs. Gangrene readily follows an injury in diabetics, and such patients show terrible proneness to the low form of erysipelas, and cellulites.36 One hundred fty years ago, Reyboso observed glucosuria, a condition induced by ether anesthesia, in which glucose is discharged in the urine, and in 1877 Claude Bernard described hyperglycemia during hemorrhagic shock.37 Today, it is well known that any type of acute stress or injury results in insulin resistance, glucose intolerance, and hyperglycemia; a constellation termed diabetes of injury.38,39 Regardless of diabetic status, surgery qualies as a signicant stress on the body and the degree of hyperglycemia perioperatively depends on the type, severity, and extent of surgery (Figure 2). Supercial surgeries such as on the eye and ear may cause a small increase in glucose, while intraabdominal and cardiac surgeries tend to cause a more pronounced response.40 While blood glucose levels typically increase to 126180 mg/dL in elective intraperitoneal procedures41, nondiabetics undergoing coronary artery bypass surgery trend toward glucose levels reaching or exceeding 270 mg/dL42,43 Surgical stress activates neuro-endocrine pathways, stimulating both the adrenergic system and hypothalamic-pituitary-adrenal axis. The hypothalamus, in response to stressful events, releases Corticotropin Releasing Hormone (CRH) which causes a resultant release of

526 A. M. Shilling and J. Raphael

Figure 2. Relationship between the increment of post-operative endogenous glucose production and the duration of the exposure of the peritoneum during abdominal hysterectomy. Taken from source40 with permission.

Adrenocorticotropic Hormone (ACTH) from the pituitary, resulting in release of glucocorticoids by the adrenal cortex. Concomitantly, the adrenergic response to stress includes release of catecholamines such as epinephrine and norepinephrine. The resultant endocrine milieu of the stress response is characterized by increased counterregulatory hormones and catecholamines, cortisol, and glucagon. A rise in the level of Growth Hormone (GH) and cytokines such as IL-1, TNF, IL-6 is also observed. It is now clearly understood that hyperglycemia frequently results from the stress response. Furthermore, in the diabetic patient, impaired glucose control leads to an exaggerated hyperglycemic response. There are several mechanisms causing the significant rise in glucose. First, increases in hepatic gluconeogenesis and glycogenolysis occur due to counter-regulatory hormones. The release of epinephrine and norepinephrine stimulate hepatic glycogenolysis and gluconeogensis. This is the major cause for increased glucose levels, though decreased glucose uptake and clearance due to insulin resistance is another reason for increasing glucose levels during stress. Most insulin resistance occurs in skeletal muscle, but adipose, liver, and cardiac tissues may also contribute to this. In addition, there is a large release of proinammatory cytokines and acute phase reactive proteins such as TNF-alpha, IL-1, IL-6, and C-Reactive Protein (CRP) that also lead to peripheral insulin resistance. Furthermore, failure of glucose to exhibit its normal negative feedback on gluconeogenesis also occurs during stress. With high levels of catecholamines, the feedback leading to insulin secretion and glucagon inhibition is diminished. Yet another mechanism in which stress can cause glucose surges is that, hyperglycemia, itself, is pro-inammatory and perpetuates the glycemic response. At the cellular level, hyperglycemia increases pro-inammatory transcription factors which, in turn, induce transcription of proinammatory cytokines TNF-alpha, IL-1, IL-2, IL-6, IL-8 and IL-18. This further leads to worsened hyperglycemia.44,45 Importantly, surgical stress-induced hyperglycemia leads to inammation and

Diabetes, Hyperglycemia, and Infections 527

further hyperglycemia. Additionally, studies have shown that glucose uctuations may lead to a greater oxidative stress and increased cytokine production.45,46 EFFECTS OF HYPERGLYCEMIA ON IMMUNE-SYSTEM Since it is well-documented that surgical stress can lead to profound hyperglycemia, even in the nondiabetic patient, the next logical question is: why does hyperglycemia lead to susceptibility to infection? Hyperglycemia and diabetes cause a number of deleterious effects on immune defense mechanismsboth cellular and humeral (Figure 3). Included are changes in leukocyte function, altered microvascular response, and changes in the complement cascade, cytokine network, and chemokine formation. Some of the earliest research on the harmful effects of hyperglycemia was focused on leukocyte function. In vitro, hyperglycemic conditions may affect leukocyte function in several ways, such as a decrease in chemotaxis, phagocytosis, adherence, and bacteriocidal activity. The actual degree of leukocyte function and responsiveness in diabetic subjects has been shown to be inversely related to the extent of hyperglycemia.47 It was shown almost 40 years ago that chemotaxis may be inhibited by hyperglycemia and the effects may actually be reversed with insulin.4850 Like chemotaxis, phagocytosis has been shown in multiple studies to be impaired by diabetes and hyperglycemia.50,51 Even glucose levels of 200 mg/dL may impair phagocytic function in vitro or in the clinical setting. Encouragingly, decreased phagocytosis may improve, but not completely normalize, after 36 hours of normoglycemia.52 In animal models, hyperglycemia caused impairment of phagocytosis in both monocytes and granulocytes and the immune dysfunction was partially reversed by insulin.53 An animal study of post-burn rabbits and maintenance of normoglycemia with exogenous insulin, showed improved monocyte phagocytosis by more than a mean of 150% when animals were treated with insulin. These animals demonstrated improved oxidative-mediated killing and suppressed growth hormone secretion.54 Within the last decade, it has been shown that neutrophil dysfunction in the setting of hyperglycemia can be improved in vivo as well. A study by Rassias and colleagues55 looked at intraoperative and postoperative hyperglycemia in diabetic patients undergoing cardiac surgery. They found that nonspecic phagocytic activity of neutrophils diminished one hour after separation from cardio-pulmonary bypass with hyperglycemia, but improved with aggressive glucose control. Though this study was not designed to compare infection rates, three patients in the standard insulin treatment group had infectious complications (septic mediastinitis, nosocomial pneumonia, and urinary tract infection), while none of the aggressively-treated patients developed any infections. In addition to its effects on chemotaxis and phagocytosis, there is evidence that hyperglycemia may affect polymorphoneutrophils (PMN) adherence and apoptosis. In a small study involving diabetic patients, it was shown that PMN adherence to a synthetic nylon column was decreased in the presence of hyperglycemia, but this effect was reversed after restoring normoglycemia.56,57 Several studies have shown that hyperglycemia may increase WBC apoptosis but this has also been questioned by other investigators.58 Six studies have evaluated glycemic threshold for dysfunction of neutrophils and the median glycemic threshold was found to be 200 mg/dL with a range between 130275 mg/dL.18 Another means in which hyperglycemia may affect leukocyte function is through antigen presentation by monocytes. Antigen presentation and clearance, an important componenent of infection prevention, may be diminished by hyperglycemia. Turina et al

528 A. M. Shilling and J. Raphael

Figure 3. The effects of hyperglycemia on the immune system. Taken with permission from Mauermann WJ, Nermergut EC. Anesthesiologists role in the prevention of surgical site infection. Anesthesiology 2006; 105(2):413-421.

showed that monocyte expression of human leukocyte antigen (HLA)-DR was significantly depressed in a hyperglycemic medium (400 mg/dl glucose) after only 24 hours compared to monocytes in a normoglycemic medium (100 mg/dl). Interestingly, the addition of insulin was associated with a greater depression in monocyte expression of surface antigen. Decreased levels of monocyte HLA-DR have been shown to

Diabetes, Hyperglycemia, and Infections 529

correlate with infections in critically ill patients.5961 Most studies have concluded that the bactericidal activity of neutrophils is decreased in both diabetics and in hyperglycemic environments. This may vary, however, with the organisms being investigated.62,63 One study conducted to specically elucidate immune-suppression as a result of the diabetic state versus hyperglycemia examined 37 nondiabetic patients given a 75 gram glucose solution. This resulted in signicant hyperglycemia as expected.64 Within two hours, there was a signicant lymphopenia and lymphocyte subset redistribution. The authors speculate that leukocyte sequestration to the vascular endothelium by cellular adhesion molecules (CAM) is partly responsible for the lymphopenia. Indeed, it has been shown that hyperglycemia increases expression of both intracellular adhesion molecules and E-selectins.65 Data has suggested that PMN respiratory oxidative burst is inhibited by high glucose concentrations in vitro. This, in part, explains the decreased intracellular pathogen lysis that occurs during hyperglycemia. Phagocytosis is also impaired by decreased oxidative burst generation in monocytes.53,66 Both the diabetic state and hyperglycemia are proinammatory. Type I and II diabetics have increased levels of TNF-alpha, IL-6, and IL-8.67,68 As mentioned previously, increased levels of IL-6 IL-18, and TNF-alpha have also been found in hyperglycemic environments.45 While inammatory responses are important in eradication of infectious agents, the resulting edema can lead to hypoxia as well as microvascular and macrovascular dysfunction. Also, a relatively small dose of 75 grams of glucose in healthy subjects causes inammatory changes and oxidative stress.69 Thus, additional detrimental effects of hyperglycemia include increased reactive oxidative species (ROS), increased oxidative stress, and overproduction of free radicals.44 Hyperglycemia can lead to nonenzymatic glycosylation of proteins as well as the addition of sugar molecules to lysine residues on extracellular proteins. These glycosylation processes result in protein inactivation and dysfunction of immunoglobins. It has been shown that glycosylation of the C3 element of the complement system prevents it from invading bacterial surfaces.70 Glycosolation of collagen as well as increased collagenase activity results from hyperglycemia in animals, thus decreasing wound collagen levels, impairing wound healing, and potentially increasing susceptibility to infection. In the surgical setting, a decrease in tensile strength of intestinal anastomosis was shown in hyperglycemic animals, but was reversed to normal strength in diabetic animals when glucose was normalized.71 Hyperglycemia may exert some of its detrimental effects on the endothelium as well. Endothelium-dependent vasodilation is attenuated by hyperglycemia in both in vitro and in vivo studies.72,73 It has also been shown that in diabetes, nitric oxide (NO) production is decreased and the vascular response to NO may be blunted. The resultant failure of vasodilation may hinder phagocytes from reaching their infectious target. A study by Williams et al. looked at the effects of hyperglycemia on endothelium-dependent vasodilation in humans.74 Using the brachial artery, they found that acute hyperglycemia impairs endothelium-dependent vasodilation in healthy humans. Proposed mechanisms for the endothelial dysfunction include hyperglycemiamediated formation of oxygen-derived free radicals which inactivate nitric oxide, formation of glycosylation end products, activation of protein kinase C which may lead to increased generation of vasoconstrictor prostanoids75, and phosphorylation of endothelial cell muscarinic receptors.76 An enzyme inhibiting nitric oxide synthetase, Asymmetric Dimethylarginine (ADMS), has been shown to be increased in a subset of patients with hyperglycemia77 and may be an independent predictor of mortality.

530 A. M. Shilling and J. Raphael

Further, hyperglycemia can cause small vessel vasculopathy, thus leading to tissue hypoxia, ischemia, and impaired antibiotic prophylaxis.78 The hyperglycemic environment may enhance the virulence of various microorganisms. Candida albicans shows competitive binding and inhibition of complementmediated phagocytosis in a hyperglycemic environment.70 Glucosuria enhances Escherichia coli growth and may be a reason for the increased incidence of urinary tract infections in diabetics.79 Enzymes important in DNA repair, inammation, and programmed cell death may also be affected by hyperglycemia.80 As we are learning more about the mechanisms by which hyperglycemia affects immunity, it becomes clear that glucose levels need to be monitored and controlled, particularly in the perioperative setting when patients are prone to infectious complications. There is mounting evidence that postoperative hyperglycemia, in both the diabetic patient population as well as the nondiabetic population, may increase infectious complications. The strongest data is in the cardiac surgical population, particularly in the postoperative setting, though there is suggestion that generalized perioperative hyperglycemia may be deleterious. Large, prospective, randomized trials investigating perioperative glucose control in the operating setting do not exist and there is an immense need for future studies, particularly in the noncardiac surgery patient population. In the preoperative setting, it is vital to question patients and screen appropriate patients for diabetes. In cardiac surgery patients, strict postoperative glycemic control is of paramount importance. Still to be determined is the glucose concentration in which infection risks increase. The original target of less than 220 mg/dL has been shown to be inadequate and more recent data suggests lower values may be benecial. Unfortunately, studies show improved benet when glucose targets are in the broad range of 80200 mg/dL. The optimal target glucose level is still undetermined. One further question is whether glucose uctuations play an important role in the detrimental effects of hyperglycemia. It is prudent to maintain normoglycemia and limit glucose uctuations as much as possible. Many studies have demonstrated that correction of stress-induced hyperglycemia with insulin decreases infection and other morbidities. Insulin has a multitude of benecial effects including enhanced neutrophil phagocytosis and chemotaxis. Insulin has anti-inammatory properties such as reducing inammatory cytokines and suppressing proinammatory transcription factors. Insulin also reduces superoxide ion generation, thus reducing ROS-induced tissue injury. The anti-inammatory and antioxidant effect of insulin can be seen within two hours of administration and 2 units an hour of insulin is similar to the effects of 100 mg of intravenous hydrocortisone.81 Insulin may also decrease CRP concentration which is an important component in the acute phase response. Insulin has also been shown to have a benecial effect on lipid metabolism which may alter infection by scavenging endotoxin in the circulation.82 Finally, insulin may lower triglyceride and free fatty acid concentrations.83 An insulin infusion protocol has proven superior to intermittent subcutaneous boluses in improving glucose control, though concerns for hypoglycemia have been raised. In the Portland study, 0.04% of diabetic patients had symptomatic hypoglycemia, though several studies have been stopped due to safety concerns for hypoglycemia.44 One study of nondiabetic cardiopulmonary bypass surgery patients found that 40% of patients treated with insulin developed hypoglycemia requiring treatment.43 When using an insulin infusion, glucose levels should be aggressively checked. We recommend checking glucose hourly, especially when changes in management are being instigated. As we learn more about hyperglycemia, it is important that anesthesiologists be aware of the potential deleterious effects of hyperglycemia and means in which

Diabetes, Hyperglycemia, and Infections 531

we may modify patient risks. Even in the nondiabetic patient population, surgical stress often leads to hyperglycemia and should be monitored and treated. This is particularly true in the cardiac surgery patient population. The optimal glucose level is still undetermined at this time, though the original thought of keeping glucose less than 200 mg/dL may be indadequate. More recent data suggests lower targets of 80120 mg/dL, but is the most conclusive in the post-operative cardiac surgery patient. Aggressive insulin therapy is certainly the mainstay for glycemic control, though various anesthetic techniques such as neuraxial and regional anesthesia may improve glycemic control by modication of the stress response.40 Pharmacologic modulation of the stress response is also promising and warrants further investigation. Practice points All patients undergoing surgery should be questioned preoperatively about diabetes and screened for diabetes if deemed necessary. Diabetic patients should be counseled preoperatively regarding the need to maintain good glucose control, as well as their increased risk for infection development. Even in nondiabetic patients, surgical stress often leads to insulin resistance and hyperglycemia and should be monitored and treated. This is particularly true in the cardiac surgery patient population. Tight postoperative glycemic control should be practiced, especially in patients that have undergone cardiac surgery. The physician should aim for normoglycemia and try to avoid uctuations.

Future research The role of perioperative glycemic control in infection prevention is still not clearly elucidated, particularly in the noncardiac surgery patient. The signicance of long-term glucose control and surgical infection risk needs further evaluation. Further research needs to be conducted on the optimal target glucose levels intra-operatively and postoperatively.. Means of modifying the surgical stress response such as neuraxial anesthesia and pharmacologic agents need to be further investigated.

REFERENCES
1. National Nosocomial Infections Surveillance (NNIS) report, data summary from October 1986-April 1996, issued May 1996. A report from the National Nosocomial Infections Surveillance (NNIS) System. American Journal of Infection Control 1996; 24: 380388. 2. Kirkland KB, Briggs JP, Trivette SL et al. The impact of surgical-site infections in the 1990s: attributable mortality, excess length of hospitalization, and extra costs. Infection Control and Hospital Epidemiology 1999; 20: 725730.

532 A. M. Shilling and J. Raphael 3. Gadaleta D, Risucci DA, Nelson RL et al. Effects of morbid obesity and diabetes mellitus on risk of coronary artery bypass grafting. The American Journal of Cardiology 1992; 70: 16131614. 4. Lilienfeld DE, Vlahov D, Tenney JH & McLaughlin JS. Obesity and diabetes as risk factors for postoperative wound infections after cardiac surgery. American Journal of Infection Control 1988; 16: 36. 5. Hoogwerf BJ, Sheeler LR & Licata AA. Endocrine management of the open heart surgical patient. Seminars in Thoracic and Cardiovascular Surgery 1991; 3: 7580. 6. Hruska LA, Smith JM, Hendy MP et al. Continuous insulin infusion reduces infectious complications in diabetics following coronary surgery. Journal of Cardiac Surgery 2005; 20: 403407. 7. Shah BR & Hux JE. Quantifying the risk of infectious diseases for people with diabetes. Diabetes Care 2003; 26: 510513. 8. Muller LM, Gorter KJ, Hak E et al. Increased risk of common infections in patients with type 1 and type 2 diabetes mellitus. Clinical Infectious Diseases 2005; 41: 281288. 9. Capes SE, Hunt D, Malmberg K & Gerstein HC. Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet 2000; 355: 773778. 10. Slaughter MS, Olson MM, Lee Jr. JT & Ward HB. A fteen-year wound surveillance study after coronary artery bypass. The Annals of Thoracic Surgery 1993; 56: 10631068. 11. Edwards FH, Grover FL, Shroyer AL et al. The Society of Thoracic Surgeons National Cardiac Surgery Database: current risk assessment. The Annals of Thoracic Surgery 1997; 63: 903908. 12. Ghali WA, Quan H & Brant R. Coronary artery bypass grafting in Canada: national and provincial mortality trends, 1992-1995. CMAJ : Canadian Medical Association journal 1998; 159: 2531. 13. Loop FD, Lytle BW, Cosgrove DM et al. J. Maxwell Chamberlain memorial paper. Sternal wound complications after isolated coronary artery bypass grafting: early and late mortality, morbidity, and cost of care. The Annals of Thoracic Surgery 1990; 49: 179186 [discussion: 186187]. 14. Trick WE, Scheckler WE, Tokars JI et al. Modiable risk factors associated with deep sternal site infection after coronary artery bypass grafting. The Journal of Thoracic and Cardiovascular Surgery 2000; 119: 108114. *15. van den Berghe G, Wouters P, Weekers F et al. Intensive insulin therapy in the critically ill patients. The New England Journal of Medicine 2001; 345: 13591367. *16. Zerr KJ, Furnary AP, Grunkemeier GL et al. Glucose control lowers the risk of wound infection in diabetics after open heart operations. The Annals of Thoracic Surgery 1997; 63: 356361. *17. Furnary AP, Zerr KJ, Grunkemeier GL & Starr A. Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. The Annals of Thoracic Surgery 1999; 67: 352360 [discussion: 360362]. 18. Golden SH, Peart-Vigilance C, Kao WH & Brancati FL. Perioperative glycemic control and the risk of infectious complications in a cohort of adults with diabetes. Diabetes Care 1999; 22: 14081414. 19. McAlister FA, Man J, Bistritz L et al. Diabetes and coronary artery bypass surgery: an examination of perioperative glycemic control and outcomes. Diabetes Care 2003; 26: 15181524. *20. Lazar HL, Chipkin SR, Fitzgerald CA et al. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004; 109: 14971502. 21. Swenne CL, Lindholm C, Borowiec J et al. Peri-operative glucose control and development of surgical wound infections in patients undergoing coronary artery bypass graft. The Journal of Hospital Infection 2005; 61: 201212. 22. Estrada CA, Young JA, Nifong LW & Chitwood Jr. WR. Outcomes and perioperative hyperglycemia in patients with or without diabetes mellitus undergoing coronary artery bypass grafting. The Annals of Thoracic Surgery 2003; 75: 13921399. 23. Gandhi GY, Nuttall GA, Abel MD et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clinic Proceedings 2005; 80: 862866. *24. Gandhi GY, Nuttall GA, Abel MD et al. Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Annals of Internal Medicine 2007; 146: 233243. 25. Ouattara A, Lecomte P, Le Manach Y et al. Poor intraoperative blood glucose control is associated with a worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005; 103: 687694.

Diabetes, Hyperglycemia, and Infections 533 26. Vriesendorp TM, Morelis QJ, Devries JH et al. Early post-operative glucose levels are an independent risk factor for infection after peripheral vascular surgery. A retrospective study. European Journal of Vascular and Endovascular Surgery 2004; 28: 520525. 27. Vriesendorp TM, DeVries JH, Hulscher JB et al. Early postoperative hyperglycaemia is not a risk factor for infectious complications and prolonged in-hospital stay in patients undergoing oesophagectomy: a retrospective analysis of a prospective trial. Critical Care 2004; 8: R437R442. 28. Bilotta F, Spinelli A, Giovannini F et al. The effect of intensive insulin therapy on infection rate, vasospasm, neurologic outcome, and mortality in neurointensive care unit after intracranial aneurysm clipping in patients with acute subarachnoid hemorrhage: a randomized prospective pilot trial. Journal of Neurosurgical Anesthesiology 2007; 19: 156160. 29. Pomposelli JJ, Baxter 3rd JK, Babineau TJ et al. Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. JPEN. Journal of Parenteral and Enteral Nutrition 1998; 22: 7781. 30. Grey NJ & Perdrizet GA. Reduction of nosocomial infections in the surgical intensive-care unit by strict glycemic control. Endocrine Practice 2004; 10(Suppl. 2): 4652. 31. Bishop JR, Moul JW, Sihelnik SA et al. Use of glycosylated hemoglobin to identify diabetics at high risk for penile periprosthetic infections. The Journal of Urology 1992; 147: 386388. 32. Wilson SK, Carson CC, Cleves MA & Delk 2nd JR. Quantifying risk of penile prosthesis infection with elevated glycosylated hemoglobin. The Journal of Urology 1998; 159: 15371539 [discussion: 1539-1540]. *33. Latham R, Lancaster AD, Covington JF et al. The association of diabetes and glucose control with surgical-site infections among cardiothoracic surgery patients. Infection Control and Hospital Epidemiology 2001; 22: 607612. 34. Dronge AS, Perkal MF, Kancir S et al. Long-term glycemic control and postoperative infectious complications. Archives of Surgery 2006; 141: 375380 [discussion: 380]. 35. Guvener M, Pasaoglu I, Demircin M & Oc M. Perioperative hyperglycemia is a strong correlate of postoperative infection in type II diabetic patients after coronary artery bypass grafting. Endocrine Journal 2002; 49: 531537. 36. Alberti KG & Thomas DJ. The management of diabetes during surgery. British Journal of Anaesthesia 1979; 51: 693710. nome ` nes de la vie communs aux animaux et aux ve ge taux. Paris, France, 37. Bernard C Lec ons sur les phe Bailliere 1878, pp. 564. 38. Thorell A, Nygren J & Ljungqvist O. Insulin resistance: a marker of surgical stress. Current Opinion in Clinical Nutrition and Metabolic Care 1999; 2: 6977. 39. McCowen KC, Malhotra A & Bistrian BR. Stress-induced hyperglycemia. Critical Care Clinics 2001; 17: 107124. 40. Schricker T, Lattermann R, Schreiber M et al. The hyperglycaemic response to surgery: pathophysiology, clinical implications and modication by the anaesthetic technique. Clinical Intensive Care 1998; 9: 118128. 41. Lattermann R, Carli F, Wykes L & Schricker T. Epidural blockade modies perioperative glucose production without affecting protein catabolism. Anesthesiology 2002; 97: 374381. 42. Carvalho G, Moore A, Qizilbash B et al. Maintenance of normoglycemia during cardiac surgery. Anesthesia and Analgesia 2004; 99: 319324. 43. Chaney MA, Nikolov MP, Blakeman BP & Bakhos M. Attempting to maintain normoglycemia during cardiopulmonary bypass with insulin may initiate postoperative hypoglycemia. Anesthesia and Analgesia 1999; 89: 10911095. 44. Collier B, Dossett LA, May AK & Diaz JJ. Glucose control and the inammatory response. Nutrition in Clinical Practice 2008; 23: 315. *45. Esposito K, Nappo F, Marfella R et al. Inammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 2002; 106: 20672072. 46. Monnier L, Mas E, Ginet C et al. Activation of oxidative stress by acute glucose uctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. JAMA : The Journal of the American Medical Association 2006; 295: 16811687. 47. McManus LM, Bloodworth RC, Prihoda TJ et al. Agonist-dependent failure of neutrophil function in diabetes correlates with extent of hyperglycemia. Journal of Leukocyte Biology 2001; 70: 395 404. 48. Mowat A & Baum J. Chemotaxis of polymorphonuclear leukocytes from patients with diabetes mellitus. The New England Journal of Medicine 1971; 284: 621627.

534 A. M. Shilling and J. Raphael 49. Bagdade JD, Root RK & Bulger RJ. Impaired leukocyte function in patients with poorly controlled diabetes. Diabetes 1974; 23: 915. 50. Delamaire M, Maugendre D, Moreno M et al. Impaired leucocyte functions in diabetic patients. Diabetic Medicine 1997; 14: 2934. 51. Marhoffer W, Stein M, Maeser E & Federlin K. Impairment of polymorphonuclear leukocyte function and metabolic control of diabetes. Diabetes Care 1992; 15: 256260. 52. Gin H, Brottier E & Aubertin J. Inuence of glycaemic normalization by an articial pancreas on phagocytic and bactericidal functions of granulocytes in insulin dependent diabetic patients. Journal of Clinical Pathology 1984; 37: 10291031. 53. Ellger B, Debaveye Y, Vanhorebeek I et al. Survival benets of intensive insulin therapy in critical illness: impact of maintaining normoglycemia versus glycemia-independent actions of insulin. Diabetes 2006; 55: 10961105. 54. Weekers F, Giulietti AP, Michalaki M et al. Metabolic, endocrine, and immune effects of stress hyperglycemia in a rabbit model of prolonged critical illness. Endocrinology 2003; 144: 53295338. 55. Rassias AJ, Marrin CA, Arruda J et al. Insulin infusion improves neutrophil function in diabetic cardiac surgery patients. Anesthesia and Analgesia 1999; 88: 10111016. 56. Bagdade JD & Walters E. Impaired granulocyte adherence in mildly diabetic patients: effects of tolazamide treatment. Diabetes 1980; 29: 309311. 57. Bagdade JD, Stewart M & Walters E. Impaired granulocyte adherence. A reversible defect in host defense in patients with poorly controlled diabetes. Diabetes 1978; 27: 677681. *58. Turina M, Miller FN, Tucker C & Polk HC. Effects of hyperglycemia, hyperinsulinemia, and hyperosmolarity on neutrophil apoptosis. Surgical Infections 2006; 7: 111121. *59. Turina M, Miller FN, Tucker CF & Polk HC. Short-term hyperglycemia in surgical patients and a study of related cellular mechanisms. Annals of Surgery 2006; 243: 845851 [discussion: 851853]. 60. Polk Jr. HC, George CD, Wellhausen SR et al. A systematic study of host defense processes in badly injured patients. Annals of Surgery 1986; 204: 282299. 61. Ditschkowski M, Kreuzfelder E, Rebmann V et al. HLA-DR expression and soluble HLA-DR levels in septic patients after trauma. Annals of Surgery 1999; 229: 246254. 62. Balasoiu D, van Kessel KC, van Kats-Renaud HJ et al. Granulocyte function in women with diabetes and asymptomatic bacteriuria. Diabetes Care 1997; 20: 392395. 63. Tan JS, Anderson JL, Watanakunakorn C & Phair JP. Neutrophil dysfunction in diabetes mellitus. The Journal of Laboratory and Clinical Medicine 1975; 85: 2633. 64. von Kanel R, Mills PJ & Dimsdale JE. Short-term hyperglycemia induces lymphopenia and lymphocyte subset redistribution. Life Sciences 2001; 69: 255262. 65. Morigi M, Angioletti S, Imberti B et al. Leukocyte-endothelial interaction is augmented by high glucose concentrations and hyperglycemia in a NF-kB-dependent fashion. The Journal of Clinical Investigation 1998; 101: 19051915. 66. Nielson CP & Hindson DA. Inhibition of polymorphonuclear leukocyte respiratory burst by elevated glucose concentrations in vitro. Diabetes 1989; 38: 10311035. 67. Zozulinska D, Majchrzak A, Sobieska M et al. Serum interleukin-8 level is increased in diabetic patients. Diabetologia 1999; 42: 117118. 68. Mooradian AD, Reed RL, Meredith KE & Scuderi P. Serum levels of tumor necrosis factor and IL-1 alpha and IL-1 beta in diabetic patients. Diabetes Care 1991; 14: 6365. 69. Mohanty P, Hamouda W, Garg R et al. Glucose challenge stimulates reactive oxygen species (ROS) generation by leucocytes. The Journal of Clinical Endocrinology and Metabolism 2000; 85: 29702973. 70. Hostetter MK. Handicaps to host defense. Effects of hyperglycemia on C3 and Candida albicans. Diabetes 1990; 39: 271275. 71. Verhofstad MH & Hendriks T. Complete prevention of impaired anastomotic healing in diabetic rats requires preoperative blood glucose control. The British Journal of Surgery 1996; 83: 17171721. 72. Tesfamariam B, Brown ML, Deykin D & Cohen RA. Elevated glucose promotes generation of endothelium-derived vasoconstrictor prostanoids in rabbit aorta. The Journal of Clinical Investigation 1990; 85: 929932. 73. Bohlen HG & Lash JM. Topical hyperglycemia rapidly suppresses EDRF-mediated vasodilation of normal rat arterioles. The American Journal of Physiology 1993; 265: H219H225.

Diabetes, Hyperglycemia, and Infections 535 74. Williams SB, Goldne AB, Timimi FK et al. Acute hyperglycemia attenuates endothelium-dependent vasodilation in humans in vivo. Circulation 1998; 97: 16951701. 75. Tesfamariam B. Free radicals in diabetic endothelial cell dysfunction. Free Radical Biology & Medicine 1994; 16: 383391. 76. Richardson RM, Ptasienski J & Hosey MM. Functional effects of protein kinase C-mediated phosphorylation of chick heart muscarinic cholinergic receptors. The Journal of Biological Chemistry 1992; 267: 1012710132. 77. Krinsley JS. Association between hyperglycemia and increased hospital mortality in a heterogeneous population of critically ill patients. Mayo Clinic Proceedings 2003; 78: 14711478. 78. Talbot TR. Diabetes mellitus and cardiothoracic surgical site infections. American Journal of Infection Control 2005; 33: 353359. 79. Geerlings SE, Brouwer EC, Gaastra W et al. Effect of glucose and pH on uropathogenic and nonuropathogenic Escherichia coli: studies with urine from diabetic and non-diabetic individuals. Journal of Medical Microbiology 1999; 48: 535539. *80. Blondet JJ & Beilman GJ. Glycemic control and prevention of perioperative infection. Current Opinion in Critical Care 2007; 13: 421427. 81. Dandona P, Thusu K, Hafeez R et al. Effect of hydrocortisone on oxygen free radical generation by mononuclear cells. Metabolism 1998; 47: 788791. 82. Krinsley J. Perioperative glucose control. Current Opinion in Anaesthesiology 2006; 19: 111116. 83. Svedjeholm R, Svensson S, Ekroth R et al. Trauma metabolism and the heart: studies of heart and leg amino acid ux after cardiac surgery. The Thoracic and Cardiovascular Surgeon 1990; 38: 15.