Perioperative clinicians commonly manage patients with dysglycemia, defined as hyperglycemia, hypoglycemia, and glucose variability during preoperative, hospital, intraoperative, postoperative, and intensive care unit (ICU) encounters. With the expanding incidence and prevalence of diabetes mellitus (DM), patients at risk for the development or presence of unrecognized hyperglycemia should be anticipated and identified as part of routine preoperative assessment. This review will provide background on DM and discuss the complications and anesthetic implications of the disease. It will then review glycemic management strategies and address specific challenges in managing perioperative patients with dysglycemia.
Types of diabetes and screening recommendations
DM is classified as type 1, type 2, or variations of one of these types. Type 1 DM (DM1) most often develops in children and adolescents, but is increasingly being diagnosed in adults. Most commonly, DM1 is due to autoimmune destruction of pancreatic islet cells. Patients subsequently have an obligate need for insulin to prevent cellular starvation, severe hyperglycemia, and life-threatening metabolic ketoacidosis. The incidence of DM1 has not changed over the past century, currently representing ∼3% to 5% of individuals with DM.1
Type 2 DM (DM2), however, has markedly increased in incidence and prevalence. This has occurred in part due to increased inactivity, obesity, aging, and comorbidities of the population. Presently, almost 10% of Americans have DM2. Many are unaware of its presence, including up to 10% of patients undergoing noncardiac surgery.2
Patients with DM2 may be able to control their DM through weight loss, diet, and/or medications, sometimes including insulin. Preoperative assessment provides clinicians with a ready opportunity to identify at-risk patients who may have unrecognized DM or prediabetes via blood glucose sampling and/or glycated hemoglobin (HbA1c) measurement (Table 1).3
Complications of diabetes and chronic hyperglycemia
Chronic elevations in glucose injure numerous organ systems and pose unique perioperative risks (Table 2). Most well known is an increased risk of surgical-site infections (SSIs),4 but understanding other organ system involvement is paramount to mitigating risk.
Chronic hyperglycemia induces tissue glycation, inflammatory pathway activation through protein kinase C, and overall oxidative stress. Nonenzymatic tissue glycation occurs when excess glucose is chemically attached to proteins. This progressive, concentration- and time-dependent process occurs in a myriad of cells that make up organs, nerves, vessels, and soft tissue. Glycation results in arterial and capillary narrowing, stiff joints and soft tissue, and altered airway anatomy that increases the risk of difficult intubation even in patients without diabetes who have equal degrees of obesity.5 Nerve injury results in autonomic neuropathy, which can delay gastric emptying and increase the risk of aspiration.
Microvascular disease in the kidneys, retina, and nerves, and macrovascular disease in peripheral vessels and most notably the cerebrovascular and coronary arteries, increase perioperative morbidity and risk for cardiovascular events in patients with DM.6 Diabetic autonomic neuropathy can result in heart rate and blood pressure alterations that yield impaired feedback mechanisms including orthostatic hypotension and resting tachycardia. These effects, particularly when combined with arterial disease, can result in myocardial ischemia and increased risk of cardiopulmonary arrest.7 Finally, microvascular disease in the kidneys increases the rate of perioperative acute kidney injury (AKI) independent of pre-existing renal dysfunction.8
Patients with DM show impaired wound healing, decreased soft tissue tensile strength (despite stiffness), and compromised tissue perfusion due to vascular disease. Immune function is depressed, with decrements in neutrophil function, deficiencies in antioxidant systems, and alterations in humoral immunity.9 These factors result in increased risk for SSIs.4
Acute dysglycemia in the perioperative window
Dysglycemia of all types including swings in glucose concentrations (termed “glycemic variability”) can impact patient outcomes.10 The perioperative state can result in aberrant glucose homeostasis through a myriad of mechanisms, including excessive glucose administration, altered insulin release and resistance, stress and other hormonal changes, medication induced (eg, steroids), and as part of the surgical or infectious-based stress response.11
Optimal perioperative management of hyperglycemia is not yet clearly elucidated. Initial consideration for the increased risks inherent to dysglycemia emphasized normalizing glucose in critical illness. In the single-center Leuven trial, surgical ICU patients—two thirds of whom were cardiac surgical patients—were managed prospectively with intensive insulin therapy toward a strict glycemic goal of 80 to 110 mg/dL or to a liberal goal of 180 to 200 mg/dL. Intensive insulin therapy resulted in reduced mortality and decreased morbidity including infections, transfusions, AKI, polyneuropathy, and ventilator time, all despite a 6-fold increased risk of hypoglycemia.12 Unfortunately, these results were not confirmed in repeated trials, including the large follow-up multicenter, multinational trial (NICE-Sugar) comparing intermediate glycemic goals (<180 mg/dL) to strict control, which had a much higher rate of hypoglycemia (13-fold) and an unexpected 2.6% absolute increased risk for mortality in the more strictly controlled group, driven by nonspecific cardiovascular events.13 The reasoning behind these discrepant data is likely multifactorial. Contributors could include hypoglycemic injury, hypoglycemia, followed by glucose reperfusion and resultant oxidative stress, or even tissue hypoglycemia despite systemic euglycemia. Further, hypokalemic cardiovascular morbidity may have resulted from insulin-induced hypokalemia. Even the glucose-measuring devices utilized may have played a role. For example, blood gas meters used to measure glucose in the Van den Berghe et al trial12 co-test for potassium, whereas point-of-care glucose meters, utilized in NICE-Sugar,13 do not and are potentially less accurate. In addition, the overall strategy for nutrition and dextrose also differed considerably, with Leuven’s study utilizing early aggressive glucose intake and parenteral nutrition.12
Nonetheless, in the perioperative environment, hyperglycemia has been found to increase the risk of infectious complications and overall morbidity and mortality in many noncardiac surgery groups.14,15 In cardiac and vascular surgery, hyperglycemia is associated with infectious complications including deep sternal wound infections,16 and importantly, stroke and even death.17 Despite these potential risks, as glucose goals are lowered, the risk of iatrogenic injury and errors, monitoring inaccuracies that result in hypoglycemia, and potentially devastating cardiac or neural injury all increase. Further, even when “euglycemic”, patients may show relative tissue hypoglycemia in certain disease states, such as traumatic brain injury.18
In cardiac surgery, intensive insulin strategies have been trialed prospectively and were without benefit.19 However, with glucose co-infusion (hyperinsulinemic euglycemia), a later study showed improved composite outcomes with tighter control.20 Meta-analyses evaluating tight glucose control in the perioperative window suggest potentially improved rates of infection.21
Finally, study interpretation to guide perioperative glycemic goals is nuanced, as the specific glucose goals are not the only contributor to patient outcomes. Evolving literature reports that preillness glycemic status impacts the relationship of glycemia with outcomes, particularly in the critically ill. Specifically, the benefits of tight control are manifest mainly among patients without diabetes.22 Similarly, hyperglycemia in patients without diabetes is associated with worse outcomes, whereas the same relationship is not seen in patients with diabetes.23 In fact, the latter may fare better, with less hypoglycemia, using liberalized glycemic strategies.24 Beyond patient factors, center expertise and resources are important, as is the glucose monitoring approach, nutritional, and dextrose infusion strategies, and the method used to achieve euglycemia.
Hypoglycemia avoidance and glucose variability
Because intensive insulin strategies often increase the risk of hypoglycemia, alternative approaches have been developed to gain the benefits of tight glycemic control without the risks of hypoglycemia. As above, a glucose co-infusion can be used along with insulin and titrated to glycemic goals. This is known as hyperinsulinemic euglycemia.20,25 This strategy can decrease hypoglycemia while also limiting glycemic variability, a problem inherent to tight control and associated with increased ICU mortality26 and perioperative morbidity including AKI.27
The second strategy to mitigate hypoglycemia is to utilize continuous glucose monitors (CGMs). These devices “spot check” glucose with enough frequency to serve as trend monitors, giving hints as to the direction and speed of patient glucose concentration changes. A myriad of CGMs sample different sources including venous and arterial blood or even interstitial fluid. Blood sampling has several weaknesses, including catheter-related problems such as infection, clotting, or heparin exposure. Subcutaneous devices sensing interstitial fluid sample downstream from plasma glucose and are subject to delays, which, along with inaccurate calibration, can result in unanticipated hypoglycemia. These sensors can cause acute inflammation with placement, thus limiting immediate use, and then develop inflammatory surface biofilm over time, limiting duration of use, requiring calibration for most devices, and potentially causing a drift in glucose results. All current CGMs sacrifice accuracy for frequency of sampling in hopes that the gained safety from trend-monitoring will overcome potential risks of inaccuracy. CGM technology and accuracy continue to improve and their use will likely become more frequent in the future.28
The degree to which dysglycemia is a marker of a stress or inflammatory state, thus correlating to poor outcomes, or a direct instigator of injury causing poor outcomes, is unclear. Determining an appropriate glucose target must balance the risk of hypoglycemia, glycemic variability, and associated untoward cardiovascular or neurologic events versus the benefit of improved glycemia. This comparison must consider the center’s resources and protocols, glucose monitors, and experience in the management of dysglycemia; the type of surgery and expected postoperative course and discharge; and the patients themselves, including their preoperative glycemic state, type of DM if present, complications from DM, and other comorbidities.
Hypoglycemia: iatrogenic injury and adaptive treatment
Hypoglycemia can occur with abnormal glucose homeostasis as a result of, or despite, therapies. Hypoglycemia—even in the moderate range and even with single episodes—is associated with ICU mortality29 and perioperative surgical complications.30 It is imperative to understand that tight glycemic control could impose an increased risk of hypoglycemia and glycemic variability. To date, no strategy has been proven to totally mitigate these risks.
Management of hypoglycemia should be quick and effective, yet cautious, so as not to overshoot and induce glycemic swings and variability. Guidelines vary in their thresholds for defining hypoglycemia warranting treatment or even for delineating severe hypoglycemia. Nonetheless, progressive treatment protocols suggest first stopping the insulin infusion and then administering 10 to 20 g aliquots of dextrose titrated to effect, with follow-up blood glucose measurements taken every 5 to 15 minutes as needed.31 Certain populations, however, may warrant individualized consideration. Patients with traumatic brain injury, for example, should be treated at a higher hypoglycemic threshold (eg, <100 mg/dL) to prevent local tissue hypoglycemia.18
Anesthesiologists should identify patients with known DM and screen for the disease in patients at risk.3 This can improve both perioperative morbidity and long-term health. In patients with a history of DM, a systematic approach to the prepreprocedural evaluation is warranted, beginning with the type of DM, current medication management including insulin, and baseline glycemic control. It is important to note that poorer chronic control of DM as assessed by HbA1c measurement has been associated previously with worsened outcomes.32 Next, specific end-organ effects of the disease should be elucidated including screening for autonomic neuropathy,7 kidney dysfunction,8 cardiac, cerebrovascular and peripheral vascular disease, and wound and infection history. A complete physical examination should be performed, but with a focus on the airway examination, joint mobility, peripheral or autonomic neuropathy, and extremity perfusion.33
Hyperglycemia encountered in the preoperative area in the patient without known DM should trigger a more complete evaluation for NPO-status, DM/prediabetes status, medication use, and meter accuracy. Etiologies for stress-induced or steroid-induced hyperglycemia must be assessed. In the patient with DM, hyperglycemia in the preoperative area may reflect periprocedural medication changes, poor baseline control, or stress/medication-induced elevations. Hyperglycemia in patients with and without DM is associated with poorer surgical outcomes including infection, reoperations, and mortality.14 The specific etiology of this increased risk is not clear and may be related to decrease in neutrophil function, oxidative stress, or inflammation.34 Limited data suggest that glycemic control can mitigate these risks.14,35
In patients with known DM, early scheduling of procedures is optimal to decrease fasting time and the inherent risk of hypoglycemia. Second, a clear and concise medication recommendation for home antihyperglycemics is necessary; many guidelines exist.34,36,37 Hypoglycemia in the preoperative area should be managed quickly either enterally or intravenously, and followed up in a timely way. However, treatment of hyperglycemia in the preoperative area has potential challenges as follow-up glucose checks may be delayed due to preprocedure checklists and anesthetic induction. Nonetheless, insulin is likely indicated to improve outcomes.14,35 It is widely accepted that hospitals should develop and maintain strategies and algorithms for management of hyperglycemia in various windows of time, including the preprocedure area. In general, it is recommended to treat glucose levels above >180 mg/dL per the American Diabetes Association (ADA) guidelines. There are limited data for deferring surgery to optimize management.37 On a case-by-case basis, surgery can still proceed even with marked hyperglycemia (>250 mg/dL) in the absence of stress or infection-induced hyperglycemia, ketosis, hyperosmolar state, or significant dehydration. The patient’s primary physician and/or a hospital-based DM management team should be involved early for close postoperative and outpatient follow-up. Treatment for hyperglycemia should be instituted with caution, given the lack of knowledge of the patient’s insulin sensitivity and the aforementioned timeliness of follow-up checks. A conservative insulin regimen should be used and an insulin infusion protocol considered on the basis of the type and duration of surgery and frequency of planned monitoring.
Insulin pumps and artificial pancreas devices (APDs)
There is increasing use of insulin pumps for patients with DM1 and even, more recently, DM2. These deliver continuous and, frequently, bolus subcutaneous insulin (generally ultra-short duration insulin variants) through small programmable devices. These have been safely utilized intraoperatively and many guidelines exist to offer direction for the selection of appropriate surgeries for use, noting contraindications and offering guidance toward perioperative device management (Fig. 1).37 In short, a “basal test” is performed when fasting in the days before surgery to ensure adequate glycemia during prolonged fasting. Next, the insulin infusion cannula portion of the device is changed, preferably to a plastic-only set, the day before surgery, far from the site of surgery and yet within reach for intraoperative checks. This change should occur with time enough to have multiple glucose checks after replacement.
The day of surgery, after satisfactory preoperative glucose control is established, the insulin infusion is continued at the basal rate for short, morning surgeries, or decreased for longer surgeries (eg, to 75% to 90% the lowest fasting basal rate). Most experts recommend disconnecting the pump for surgeries lasting over 3 hours.37 Treatment should be initiated for recurrent hypoglycemia along with cessation of pump therapy with consideration for use of an intravenous insulin infusion. Glucose should be checked frequently through the procedure, with slowly decreasing frequency over 1 to 2 days postoperatively.
APDs are programmable devices that combine a continuously sensing, typically subcutaneously placed, glucose meter with a hormonal pump. Most APDs are small devices placed onto the skin and contain insulin only, although devices do exist that use insulin plus glucagon or other hormones, and bedside devices even exist that sense venous blood glucose and infuse glucose parenterally. Complex algorithmic strategies are used to titrate infusions to maintain glycemic goals. APDs can be closed-loop (ie, not requiring user input), or hybrid closed-loop, where user input can augment insulin delivery.38 Although not approved for use perioperatively, there are reports of safe use and limited studies have shown potential benefits toward both glycemic and patient outcomes.28
Glycemic goals and management strategies in the operating room must consider the type and duration of surgery, relevant complications, and the perioperative implications of dysglycemia and its treatment. One important caveat to intensive intraoperative insulin therapy in this population is the anesthetized patient’s inability to declare symptoms of hypoglycemia, and their blunted reaction to it. Further, the typical use of intravenous regular insulin yields effects in 10 to 15 minutes, peaks 10 to 15 minutes later, and results in pharmacologic swings that are likely shorter in duration than sampling frequency.
An insulin-glucose co-infusion strategy utilizing tight glycemic control for a brief outpatient surgery will escalate workload and potentially expose patients to more risk of harm than benefit. However, patients undergoing cardiac surgery have well-established risks of hyperglycemia,19 and close monitoring and intervention to treat dysglycemia is warranted.
Choosing a glycemic goal
Randomized trials of intensive insulin infusions compared with conservative management aiming for intermediate glucose concentrations are largely confined to the cardiac surgical population. As mentioned previously, intensive insulin therapy may increase the risk of hypoglycemia and glycemic variability,19 although infectious complications may decrease.21,35 Hyperinsulinemic euglycemic strategies exist that may improve this risk to benefit ratio,20 but at this time, evidence is not adequate to suggest non–research-based intensive insulin therapies in the operating room. A more conservative glucose goal of 130 to 180 mg/dL is recommended and notably, hospital protocols and systematic algorithms exist that assist in glycemic management. Perioperatively, frequent glucose measurement augments safety and improves glycemic control, particularly in light of the pharmacokinetics of insulin. In the operating room, point-of-care glucose meters are utilized frequently and although technology continues to improve, these devices have some potential noteworthy limitations and interferences.39
Management of dysglycemia in the postoperative period must include consideration of the course of the patient and timeline of transfer to the hospital floor, ICU, or discharge to home or another facility. Aggressively controlling glycemia may pose a risk of hypoglycemia that is dangerous to the sedated patient. In patients with poor DM control, it may be appropriate to allow patients to be hyperglycemic into their chronic range.22,40
Routine protocols should be followed for recovery from anesthesia, including perioperative nausea and vomiting prophylaxis and treatment and non–narcotic-based analgesics, to improve early resumption of enteral intake. Patients with perioperative dysglycemia should be monitored with, at minimum, hourly glucose checks in an ambulatory facility until they are able to adequately self-manage hyperglycemia, tolerate enteral intake to provide a glucose source if needed, recognize hypoglycemia, and have adequate postdischarge follow-up in place. Observation is warranted until risks of hypoglycemia related to exogenous insulin delivery resolve. This may extend to 4 hours after the last dose of subcutaneous regular insulin.
For patients with expected transitions to the medical or surgical hospital floor, or the ICU, hospital-based policies developed in conjunction with endocrinologists, DM care teams, and the accepting physician teams should be followed. These policies should prepare the patient to transition from the postanesthesia care unit to the subsequent area, establish reasonable standards for glucose goals, frequency, and method of monitoring, and standardize interventions with insulin dosing. Protocols should establish the timing and transitioning off of insulin infusions. Utilizing quality improvement-based interventions can markedly improve glycemic outcomes in the hospital and decrease the frequency of hypoglycemic events.41
DM is the most commonly encountered endocrinopathy, increasing in prevalence worldwide and ubiquitous in our hospitals. Patients are often undiagnosed, many remain poorly controlled, and perioperative dysglycemia is increasingly common. Not only can the anesthesiologist aid by screening for the disease, but knowledge of this metabolic process and its medical and pharmacologic management is paramount to providing optimal perioperative care.
An organized, organ-specific preoperative approach aids assessment of patients with DM and helps recognize and mitigate perioperative risks. Patients with DM are at increased risk for difficult intubation, aspiration, stroke, cardiac instability and arrest, postoperative impaired wound healing, AKI, and SSIs.4,5,7–9 Providers also need to be increasingly aware of the perioperative use of insulin pumps, CGMs, and APDs.28
Numerous perioperative glycemic management recommendations exist.34 Current guidelines suggest largely nonindividualized intermediate glycemic goals, balancing the risk of SSIs and other morbidity due to hyperglycemia to the divergent risk of hypoglycemia and glycemic variability associated with intensive insulin strategies. Hospital-based policies and multidisciplinary protocols help standardize management locally.
It has been recognized that hyperglycemic critically ill patients with DM benefit less from tighter glycemic control, with tighter glucose goals being associated with mortality in patients with higher HbA1c results.42,43 A revisiting of tight glycemic control, however, has begun, particularly in trauma and cardiac surgery patients, and critically ill patients without diabetes, but with stress-induced hyperglycemia.44 As glycemic target goals are lowered, however, the risk of iatrogenic harm increases. Thus, the use of hypoglycemia and glycemic variability mitigation strategies such as hyperinsulinemic euglycemia, CGMs, and APDs are increasingly being considered.
The optimal management of perioperative dysglycemia is continuing to evolve with enhanced insulin-delivery approaches and glycemic monitoring enabling individualized glycemic management with a range of glycemic goals while also limiting the risk of hypoglycemia and glycemic variability. Recognition of perioperative patients at risk of dysglycemia and appropriate treatment when it occurs will hopefully result in limitations in deleterious side effects and improve patient outcomes.
Conflict of interest disclosure
The authors declare that they have nothing to disclose.
1. American Diabetes Association. Statistics about diabetes: overall numbers, diabetes and prediabetes. American Diabetes Association. 2018. Available at: www.diabetes.org/diabetes-basics/statistics/
. Accessed March 24, 2019.
2. Abdelmalak B, Abdelmalak JB, Knittel J, et al. The prevalence of undiagnosed diabetes in non-cardiac surgery patients, an observational study. Can J Anaesth. 2010;57:1058–1064.
3. Joffe AM, Coursin DB, Long MTMurray MJ, Harrison BA, Mueller JT, Rose SH, Wass CT, Wedel DJ. Anesthesia for patients with diabetes. Faust’s Anesthesiology Review, 5th ed. Philadelphia, PA: W.B. Saunders (Elsevier Imprint); 2018:550–552.
4. Martin ET, Kaye KS, Knott C, et al. Diabetes and risk of surgical site infection: a systematic review and meta-analysis. Infect Control Hosp Epidemiol. 2016;37:88–99.
5. Reissell E, Orko R, Maunuksela EL, et al. Predictability of difficult laryngoscopy in patients with long-term diabetes mellitus. Anaesthesia. 1990;45:1024–1027.
6. Newman JD, Wilcox T, Smilowitz NR, et al. Influence of diabetes on trends in perioperative cardiovascular events. Diabetes Care. 2018;41:1268–1274.
7. Burgos LG, Ebert TJ, Asiddao C, et al. Increased intraoperative cardiovascular morbidity in diabetics with autonomic neuropathy. Anesthesiology. 1989;70:591–597.
8. Biteker M, Dayan A, Tekkesin AI, et al. Incidence, risk factors, and outcomes of perioperative acute kidney injury in noncardiac and nonvascular surgery. Am J Surg. 2014;207:53–59.
9. Casqueiro J, Casqueiro J, Alves C. Infections in patients with diabetes mellitus: a review of pathogenesis. Indian J Endocrinol Metab. 2012;16 (suppl 1):S27–S36.
10. Nair BG, Neradilek MB, Newman SF, et al. Association between acute phase perioperative glucose parameters and postoperative outcomes in diabetic and non-diabetic patients undergoing non-cardiac surgery. Am J Surg. 2019;218:302–310.
11. Akhtar S, Barash PG, Inzucchi SE. Scientific principles and clinical implications of perioperative glucose regulation and control. Anesth Analg. 2010;110:478–497.
12. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001;345:1359–1367.
13. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283–1297.
14. Kwon S, Thompson R, Dellinger P, 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.
15. Long CA, Fang ZB, Hu FY, et al. Poor glycemic control is a strong predictor of postoperative morbidity and mortality in patients undergoing vascular surgery. J Vasc Surg. 2019;69:1219–1226.
16. Perrault LP, Kirkwood KA, Chang HL, et al. A prospective multi-institutional cohort study of mediastinal infections after cardiac operations. Ann Thorac Surg. 2018;105:461–468.
17. McGirt MJ, Woodworth GF, Brooke BS, et al. Hyperglycemia independently increases the risk of perioperative stroke, myocardial infarction, and death after carotid endarterectomy. Neurosurgery. 2006;58:1066–1073; discussion 1066–1073.
18. Vespa P, McArthur DL, Stein N, et al. Tight glycemic control increases metabolic distress in traumatic brain injury: a randomized controlled within-subjects trial. Crit Care Med. 2012;40:1923–1929.
19. Gandhi GY, Nuttall GA, Abel MD, et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clin Proc. 2005;80:862–866.
20. Duncan AE, Sessler DI, Sato H, et al. Hyperinsulinemic normoglycemia during cardiac surgery reduces a composite of 30-day mortality and serious in-hospital complications: a randomized clinical trial. Anesthesiology. 2018;128:1125–1139.
21. Wang YY, Hu SF, Ying HM, et al. Postoperative tight glycemic control significantly reduces postoperative infection rates in patients undergoing surgery: a meta-analysis. BMC Endocr Disord. 2018;18:42.
22. Krinsley JS, Meyfroidt G, Van den Berghe G, et al. The impact of premorbid diabetic status on the relationship between the three domains of glycemic control and mortality in critically ill patients. Curr Opin Clin Nutr Metab Care. 2012;15:151–160.
23. Krinsley JS, Egi M, Kiss A, et al. Diabetic status and the relation of the three domains of glycemic control to mortality in critically ill patients: an international multicenter cohort study. Crit Care. 2013;17:R37.
24. Luethi N, Cioccari L, Biesenbach P, et al. Liberal glucose control in ICU patients with diabetes: a before-and-after study. Crit Care Med. 2018;46:935–942.
25. Slob EMA, Shulman R, Singer M. Experience using high-dose glucose-insulin-potassium (GIK) in critically ill patients. J Crit Care. 2017;41:72–77.
26. Hermanides J, Vriesendorp TM, Bosman RJ, et al. Glucose variability is associated with intensive care unit mortality. Crit Care Med. 2010;38:838–842.
27. Nam K, Jeon Y, Kim WH, et al. Intraoperative glucose variability, but not average glucose concentration, may be a risk factor for acute kidney injury after cardiac surgery: a retrospective study. Can J Anaesth. 2019;66:921–933.
28. Long MT, Coursin DB, Rice MJ. Perioperative considerations for evolving artificial pancreas devices. Anesth Analg. 2019;128:902–906.
29. Finfer S, Liu B, Chittock DR, et al. Hypoglycemia and risk of death in critically ill patients. N Engl J Med. 2012;367:1108–1118.
30. Goh SN, Yeoh E, Tan KY. Impact of perioperative hypoglycaemia in subjects with diabetes undergoing colorectal surgery. Int J Colorectal Dis. 2017;32:209–214.
31. American Diabetes Association. 15. Diabetes Care in the Hospital: standards of Medical Care in Diabetes-2019. Diabetes Care. 2019;42(suppl 1):S173–S181.
32. Jehan F, Khan M, Sakran JV, et al. Perioperative glycemic control and postoperative complications in patients undergoing emergency general surgery: What is the role of Plasma Hemoglobin A1c? J Trauma Acute Care Surg. 2018;84:112–117.
33. Stram ML, Long MT, Coursin DBSweitzer B. Diabetes mellitus. Preoperative Assessment and Management, 3rd ed. Philadelphia, PA: Wolters Kluwer; 2019:309–320.
34. Duggan EW, Carlson K, Umpierrez GE. Perioperative hyperglycemia management: an update. Anesthesiology. 2017;126:547–560.
35. Buchleitner AM, Martinez-Alonso M, Hernandez M, et al. Perioperative glycaemic control for diabetic patients undergoing surgery. Cochrane Database Syst Rev. 2012:Cd007315.
36. Simha V, Shah P. Perioperative glucose control in patients with diabetes undergoing elective surgery. JAMA. 2019;321:399–400.
37. Sreedharan R, Abdelmalak B. Diabetes mellitus: preoperative concerns and evaluation. Anesthesiol Clin. 2018;36:581–597.
38. Long MT, Coursin DB, Rice MJ. Perioperative considerations for evolving artificial pancreas devices. Anesth Analg. 2019;128:902–906.
39. Rice MJ, Coursin DB. Glucose meters: here today, gone tomorrow? Crit Care Med. 2016;44:e97–e100.
40. Krinsley JS, Maurer P, Holewinski S, et al. Glucose control, diabetes status, and mortality in critically ill patients: the Continuum From Intensive Care Unit Admission to Hospital Discharge. Mayo Clin Proc. 2017;92:1019–1029.
41. Singh A, Adams A, Dudley B, et al. Making surgical wards safer for patients with diabetes: reducing hypoglycaemia and insulin errors. BMJ Open Qual. 2018;7:e000312.
42. Egi M, Bellomo R, Stachowski E, et al. The interaction of chronic and acute glycemia with mortality in critically ill patients with diabetes. Crit Care Med. 2011;39:105–111.
43. Egi M, Krinsley JS, Maurer P, et al. Pre-morbid glycemic control modifies the interaction between acute hypoglycemia and mortality. Intensive Care Med. 2016;42:562–571.
44. Krinsley JS, Chase JG, Gunst J, et al. Continuous glucose monitoring in the ICU: clinical considerations and consensus. Crit Care. 2017;21:197.