Pharmacotherapy for Hyperglycemia in Noncritically Ill Hospitalized Patients

The positive effects of insulin administration include correction of hyperglycemia, as well as anti-inflammatory, vasodilatory, and antioxidant effects and inhibition of lipolysis and platelet aggregation.^16–18,21  Hyperglycemia is associated with impaired leukocyte function, including decreased phagocytosis, impaired bacterial killing, and chemotaxis, which can increase the number of hospital infections.²²  Hyperglycemia has also been shown to impair collagen synthesis and to impair wound-healing in patients with poorly controlled diabetes.²³  In addition, acute hyperglycemia results in NF-κB activation and production of inflammatory cytokines such as TNF-α; interleukin-6 (IL-6); plasminogen activator inhibitior-1, which causes increased vascular permeability; and leukocyte and platelet activation resulting in an inflammatory and pro-thrombotic state.²⁴  Several studies have reported rapid improvement in high levels of inflammatory and oxidative stress markers after insulin administration and correction of hyperglycemia, although these levels remained higher than those of control subjects without diabetes.²⁵ 

No single insulin regimen meets the needs of all patients with hyperglycemia. Scheduled subcutaneous insulin therapy with basal or intermediateacting insulin given once or twice per day in combination with short- or rapid-acting insulin administered before meals is the preferred strategy for glycemic management in non-critically ill patients.^5,26  Subcutaneous insulin regimens should address the three components of a patient's total insulin requirement: basal (what is required in the fasting state), nutritional (what is required to dispose of glucose contained in nutrition), and correctional or supplemental (what is required to correct for glucose elevations caused by changes in insulin requirements).²⁶ 

The practice of discontinuing oral diabetes medications and/or existing insulin therapy and starting sliding-scale insulin (SSI) results in undesirable levels of hypoglycemia and hyperglycemia.^27,28  SSI insulin regimens involve the administration of regular or rapid-acting insulin before meals or every 4–6 hours if patients are on NPO (nothing by mouth) status to correct hyperglycemia. Although straightforward and easy to use, SSI regimens are fraught with challenges, including inadequate coverage of glycemic excursions and insulin stacking (accumulation of insulin in subcutaneous tissues).²⁹  We previously reported the results of a prospective, randomized, multicenter trial comparing the efficacy and safety of a basal-bolus insulin regimen to that of SSI in inpatients with type 2 diabetes admitted to general medicine and surgery wards.³⁰  We found that, among 130 insulinnaive patients with an admission blood glucose level between 140 and 400 mg/dl, the use of a basal-bolus insulin regimen led to greater improvement in blood glucose control than SSI alone. A blood glucose target of < 140 mg/dl was achieved in 66% of patients in the glargine-plus-glulisine basal-bolus group and 38% of the SSI group. One-fifth of patients treated with an SSI protocol without a basal component had persistently elevated blood glucose levels > 240 mg/dl during their hospital stay. The incidence of hypoglycemia, defined in this study as a blood glucose level < 60 mg/dl, was low (3%) and was not different between groups.

In general surgery patients, the RABBIT 2 Surgery (Randomized Study of Basal Bolus Insulin Therapy in the Inpatient Management of Patients with Type 2 Diabetes Undergoing General Surgery) trial¹²  compared the efficacy and safety of a basal-bolus regimen to that of SSI in 211 patients with type 2 diabetes. Study outcomes included differences in daily glucose levels and a composite of postoperative complications, including wound infection, pneumonia, respiratory failure, acute renal failure, and bacteremia. Patients were randomized to receive a basal-bolus regimen with glargine once daily and glulisine before meals at a starting dose of 0.5 units/kg or to receive SSI four times daily for glucose > 140 mg/dl. The basal-bolus regimen resulted in significant improvement in glucose control and a significant reduction in the frequency of the composite of hospital complications. The results of these trials indicate that a basal-bolus regimen is preferred to SSI and results in improved glycemic control and lower rates of hospital complications in general medical and surgical inpatients with type 2 diabetes.

The DEAN (Insulin Detemir Versus NPH Insulin in Hospitalized Patients With Diabetes) trial, an open-label, controlled, multicenter trial, randomly assigned 130 medical patients with type 2 diabetes to receive either detemir once daily and aspart before meals or NPH and regular insulin twice daily.³¹  Both regimens resulted in significant improvements in inpatient glycemic control, with a glucose target of < 140 mg/dl before meals achieved in 45% in the detemir/aspart group and in 48% of NPH/regular insulin group. Hypoglycemia (< 60 mg/dl) was observed in approximately one-fourth of patients treated with detemir/aspart and NPH/regular insulin during the hospital stay. There was no difference in length of hospital stay or mortality between groups. Thus, a similar improvement in glycemic control can be achieved with basal-bolus therapy with detemir/aspart or with NPH/regular insulin in general medicine patients with type 2 diabetes.

In insulin-naive patients, initial total daily insulin doses have varied widely in different protocols from 0.3 to 1.5 units/kg/day.^30,32–34  However, only doses of 0.4 and 0.5 units/kg/day have been studied prospectively.^12,30  A case-control analysis of 1,990 patients with diabetes reported that insulin doses > 0.6 units/kg/day were associated with higher odds of hypoglycemia than doses < 0.2 units/kg/day, independent of the types of insulin used.³⁵  In addition, higher rates of hypoglycemia have been seen with increasing age and in patients with impaired renal function. A recent, randomized trial compared the efficacy of basal-bolus regimens at 0.5 and 0.25 units/kg/day in patients with diabetes with moderate renal failure and a glomerular filtration rate (GFR) < 45 ml/min. There were no differences in mean daily glucose concentration, but those in the higher-dose insulin group had twice the rate of hypoglycemia (30 vs. 15.8%) during their hospital stay.³⁶  Thus, lower initial daily doses (0.25–0.3 units/kg/day) should be used in patients with moderate renal failure who are at risk of hypoglycemia.

The recently reported Basal Plus trial³⁷  recruited 375 patients with type 2 diabetes treated with diet, oral antidiabetic agents, or low-dose insulin (≤ 0.4 unit/kg/day) to receive a basal-bolus regimen with glargine once daily and glulisine before meals, a “basal-plus” regimen with glargine once daily and supplemental doses of glulisine for correction of hyperglycemia (> 140 mg/dl) per a sliding scale, and SSI. This trial reported that the basal-plus regimen resulted in improvements in glycemic control and frequency of hypoglycemia that were similar to those of the standard basal-bolus regimen. In addition, treatment with basal-bolus and basal-plus regimens resulted in fewer treatment failures than treatment with SSI. Thus, in insulin-naive patients or in those receiving low-dose insulin on admission (< 0.4 units/kg/day), as well as in patients with reduced oral intake, a basal-plus regimen is an effective alternative to basal-bolus insulin. Patients with persistent hyperglycemia or with regular caloric intake could be moved up from basal-plus to a basal-bolus regimen if needed.

Hyperglycemia in the hospital setting affects 38–46% of noncritically ill hospitalized patients.^38,39  Evidence from observational studies, including our own, indicates that hospital hyperglycemia in patients with and without diabetes is associated with increased risks of complications and mortality, longer hospital stays, higher admission rates to the intensive care unit (ICU), and a greater need for transitional or nursing home care after hospital discharge^38,40,41  The mechanisms implicated in the detrimental effects of hyperglycemia during acute illness are not completely understood. Current evidence indicates that hyperglycemia results in impaired neutrophil granulocyte function, overproduction of ROS, circulating free fatty acids, inflammatory mediators that can result in direct cellular damage, and vascular and immune dysfunction.³² 

Several intervention studies and meta-analyses have reported that intensive insulin therapy, defined as the use of intravenous (IV) insulin infusion protocols to achieve strict glycemic control, improves clinical outcomes in surgical ICU patients but not in medical or mixed ICU settings.²  A recent meta-analysis of seven randomized, controlled studies found that intensive insulin therapy during or after cardiac surgery reduced mortality in the ICU, postsurgical atrial fibrillation, use of epicardial pacing, duration of mechanical ventilation, and length of ICU stay.⁴²  Recent studies of mediators of inflammation have also shown that insulin exerts anti-inflammatory, antithrombotic, and anti-atherogenic effects.⁴³  Insulin counteracts many of the detrimental effects of hyperglycemia, including free radical formation, oxidative stress, apoptotic cell death, and increased levels of proinflammatory cytokines.⁴⁴  Additionally, insulin has been shown to increase cardiac contractility and coronary vasodilation and to preserve endothelial function.⁴⁵  The antioxidant and anti-inflammatory effects of insulin are mediated through a number of pathways. Insulin inhibits NF-κB, which is at the center of many proinflammatory pathways, thereby affecting a range of downstream events. Inhibition of mitogen-activated protein kinase suppresses IL-6 signaling, whereas inhibition of c-Jun N-terminal kinase mediates inflammatory processes downstream of TNF-α. TNF-α itself is suppressed via phosphatidyl-inositol 3-kinase-AKT–mediated activation of endothelial nitric oxide synthase.⁴³  Taken together, these findings suggest a cardiac-specific mechanism that may contribute to improved clinical outcomes in insulin-treated patients with hyperglycemia.

Although insulin therapy is the standard of care in hospitals, it is a source of medication errors and increased risk of hypoglycemia. An analysis of medication errors between 2006 and 2008 revealed that insulin was the drug with the greatest number of medication errors in hospitals.⁴⁶  Hypoglycemia in the hospital has been associated with adverse cardiovascular outcomes such as prolonged QT intervals, ischemic electrocardiogram changes/angina, arrhythmias, sudden death, and increased inflammation.^47,48  In addition, insulin-induced hypoglycemia is associated with increases in C-reactive protein and proinflammatory cytokines (TNF-α, interleukin-1β, IL-6, and interleukin-8), markers of lipid peroxidation, ROS, and leukocytosis.^49,50 

In non-ICU settings, hypoglycemia has been associated with increased lengths of hospital stay, greater costs of hospitalization, and higher mortality both during hospitalization and after discharge.^51,52  Observational studies have demonstrated a J-shaped relationship between inpatient glycemic control and mortality, with increased risk of death at both the high and low extremes of dysglycemia.⁵³  Most of these studies reported that inpatient mortality risk is greater in patients with spontaneous hypoglycemia than in those with insulin-associated hypoglycemia.^53–55  This suggests that spontaneous hypoglycemia may be a marker of disease burden instead of a direct cause of mortality.

The use of oral antidiabetic agents is not recommended in hospitals because few data are available regarding their safety and efficacy in the inpatient setting. Major limitations to the use of oral agents in the hospital include their side effect profiles and slow onset of action, which does not allow for rapid attainment of glycemic control or dose adjustments to meet the changing needs of acutely ill patients.^5,6  Table 1 depicts the advantages and disadvantages of available antidiabetic drugs compared to insulin therapy for the inpatient management of hyperglycemia in noncritical care settings.

Metformin is the most commonly prescribed glucose-lowering agent for the outpatient treatment of type 2 diabetes.⁵⁶  Despite a lack of randomized studies on its safety and efficacy in hospitals, metformin is also commonly used in the inpatient setting,^57,58  and it is estimated that up to one-fourth of hospitalized patients with type 2 diabetes in the United States are treated with metformin, even in the presence of contraindications.⁵⁹  Metformin exerts its antidiabetic effect by suppressing excessive hepatic glucose production through a reduction in gluconeogenesis.⁶⁰  Metformin might also increase glucose utilization in peripheral tissues and possibly reduces food intake and intestinal glucose absorption, resulting in weight loss.⁶¹  Because metformin does not stimulate endogenous insulin secretion, it does not cause hypoglycemia when used as monotherapy.⁶²  Gastrointestinal discomfort in the form of abdominal pain, flatulence, and diarrhea occurs in ~ 25% of patients taking metformin.⁶³ 

Metformin has been reported to increase the risk of lactic acidosis. The estimated rate of lactic acidosis in patients receiving metformin is of 2–5 cases per 100,000 patient-years, which is similar to that reported for patients with diabetes who do not take metformin.^64,65  A 2010 Cochrane systematic review reported no increase in the number of cases of lactic acidosis in ambulatory patients across 347 clinical trials with 70,490 patient-years of metformin use.⁵⁸  Patients with diabetes, however, are at increased risk of developing lactic acidosis if safety guidelines are ignored and if the drug is continued in the presence of contraindications.⁶⁶  Impaired kidney function, decompensated heart failure, hypoxemia, alcoholism, cirrhosis, contrast dye exposure, sepsis, and shock are associated with an increased risk of lactic acidosis and are absolute contraindications to the use of metformin therapy.^58,67  The importance of adhering to safety recommendations for metformin was illustrated by Pasquel et al.,⁶⁶  who recently reported a series of cases of metformin-induced lactic acidosis when the drug was prescribed in the hospital despite the presence of contraindications.

Sulfonylureas are the second most commonly prescribed antidiabetic agents in patients with type 2 diabetes. Insulin secretagogues exert their antidiabetic effect by stimulating endogenous insulin production.⁶⁷  The insulinotropic effect of these medications is exerted through binding to a receptor-like structure in the β-cell surface and closing of the ATP-dependent potassium channels in the β-cell membrane.^68,69  This results in depolarization of the cell, influx of calcium, and subsequent stimulation of insulin secretion.⁷⁰  In the outpatient setting, sulfonylureas have proven to be effective in improving glucose control, with an A1C reduction of 1–2 percentage points, and in reducing microvascular complications.^71,72  However, sulfonylureas increase the risk of hypoglycemia, especially in elderly patients and in those with impaired renal function or poor oral intake.⁷⁰  In addition, there is concern that sulfonylureas may worsen cardiac and cerebral ischemia^73,74  by inhibiting ATP-sensitive potassium channels, resulting in cell membrane depolarization and increased intracellular calcium concentration.⁷⁵ 

Use of sulfonylureas in the inpatient setting is discouraged because of the potential risk of hypoglycemia secondary to long duration of action and variable meal availability in the inpatient setting.^5,6  Results of a nested case-control study composed of hospitalized patients who received a sulfonylurea as part of their inpatient regimen reported a prevalence of hypoglycemia in 130 of 692 (19%) of patients.⁷⁶  Hypoglycemia was highest with the use of glyburide (22%), followed by glimepiride (19%), and glipizide (16%), especially in patients > 65 years of age and with a GFR < 30 ml/min.⁷⁶ 

Significant drug-to-drug interactions have also been described with the use of sulfonylureas in hospitalized patients. The use of sulfonylureas in patients receiving fluoroquinolone antibiotics increases the risk of hypoglycemia.⁷⁷  Additionally, drugs that inhibit hepatic cytochrome CYP2C9, including metronidazole, fluconazole, amiodarone, miconazole, trimethoprim-sulphamethoxazole, valproate, and gemfibrozil, can also exaggerate the effect of sulfonylureas, thereby increasing the risk of hypoglycemia.⁷⁸ 

Glinides available in the United States include repaglinide and nateglinide. Repaglinide and nateglinide enhance β-cell insulin production but differ from sulfonylureas in receptor affinity, binding sites, and absorption and elimination rates.⁷⁹  These differences result in faster onset and shorter duration of action, offering the advantage of better postprandial insulin response.⁸⁰  A retrospective review of hospitalized patients treated with repaglinide or nateglinide indicated a rate of hypoglycemia of 7%, similar to that reported with insulin therapy in the inpatient setting.⁸¹  These agents are not recommended in the inpatient setting because of the potential risk of hypoglycemia and lack of randomized trials evaluating their safety and efficacy in the inpatient management of hyperglycemia.

Pioglitazone is the only thiazolidinedione (TZD) widely available for clinical use in the United States. Pioglitazone is an insulin sensitizer and exerts its glucose-lowering effect through direct activation of the peroxisome proliferator–activated receptor γ.⁸²  TZDs are highly effective in improving glycemic control, with an average A1C reduction of 1–2 percentage points.⁸³  However, the time needed to reach its maximal antihyperglycemic effect can be up to 12 weeks, which makes it a less attractive agent for inpatient management of diabetes and hyperglycemia.⁵  In addition, TZD therapy is associated with fluid retention leading to edema, especially when used in combination with insulin, and worsening heart failure in some individuals.^32,84  TZDs are contraindicated in patients with congestive heart failure, hemodynamic instability, or evidence of hepatic dysfunction.⁶⁷ 

Sodium glucose co-transporter 2 (SGLT-2) inhibitors are a new class of oral antidiabetic medications that increase urinary glucose excretion by reducing renal glucose reabsorption in the proximal convoluted tubules.⁸⁵  Canaglifozin and dapaglifozin are the two available drugs approved by the U.S. Food and Drug Administration for management of type 2 diabetes.^86,87  Both agents are effective in reducing A1C by ~ 0.6–0.8%, with a low risk of hypoglycemia.⁸⁸  The use of SGLT-2 inhibitors, however, has been associated with an increased risk of urinary and genital tract infections compared to placebo (odds ratio 1.34–3.50, respectively).⁸⁸  In addition, because of their glycosuric effects, dehydration and hypotension have been reported.⁸⁹  The potential side effects of SGLT-2 inhibitors make their use less attractive for inpatient management.

Acarbose and miglitol, both α-glucosidase inhibitors (AGIs), decrease blood glucose by delaying the breakdown of carbohydrates in the gut and thus slowing the absorption of sugars. Although AGIs can significantly reduce postprandial glucose excursions with a low risk of hypoglycemia, the increased delivery of carbohydrates to the colon commonly results in increased gas production and gastrointestinal symptoms.⁹⁰  These agents are not recommended for use in the inpatient setting because of their mild antihyperglycemic effect, with most reports showing an A1C reduction of ~ 0.5% and minimal reduction in fasting plasma glucose.⁸³  AGIs are contraindicated in patients with gastrointestinal diseases such as inflammatory bowel disease, partial bowel obstruction, or severe renal or hepatic disease. In addition, intolerance to these agents is common and affects 25–45% of patients.⁸³ 

Two distinct classes of incretin-based therapies are available in the United States: glucagon-like peptide-1 (GLP-1) receptor agonists (i.e., exenatide and liraglutide) and the dipeptidyl peptidase-4 (DPP-4) inhibitors (i.e., sitagliptin, saxagliptin, alogliptin, and linagliptin).⁶⁷  The benefits and potential risks of incretin-based therapies in hospitalized patients have been reviewed elsewhere.^7,8,91,92  Although agents in both classes are well tolerated, the oral administration of DPP-4 inhibitors (as opposed to injectable GLP-1 receptor agonists) might be preferred for patients who are able to take oral medications. The efficacy and safety of incretin-based therapies in hospitalized patients has not been fully determined. However, the potential metabolic and cardiovascular benefits and the low risk of hypoglycemia of these agents make them an attractive possibility for inpatient management of diabetes.

In the hospital setting, the use of native GLP-1 infusion has been shown effective in improving the glycemic response after meals or enteral nutrition similar to insulin administration and also reduces exogenous insulin requirements.^93–97  The perioperative administration of exenatide has been shown to improve glucose levels in patients with diabetes undergoing general surgery.⁹⁸  In addition, exenatide effectively lowered glucose levels when given as an IV infusion for 48 hours to patients with diabetes who were admitted to a cardiac ICU.⁹⁹  Moreover, the infusion of native GLP-1 has been reported to improve left ventricular function and cardiac functional status in patients with severe heart failure.¹⁰⁰  The fact that these beneficial effects occurred in patients with or without diabetes supports a GLP-1 effect on the heart independent of glycemic control. GLP-1 receptor agonist therapy, however, is frequently associated with gastrointestinal side effects and decreased food intake, which could be undesirable in most hospitalized patients. In addition, there is an increased risk of hypoglycemia when these agents are used with insulin.

A recently published, randomized pilot study assessed the safety and efficacy of the DPP-4 inhibitor sitagliptin for the inpatient management of type 2 diabetes.¹⁰¹  In this trial, patients treated with diet, oral antidiabetic agents, or a low daily insulin dose (≤ 0.4 units/kg/day) were randomized to sitagliptin alone or in combination with low-dose insulin glargine or to a basal-bolus insulin regimen plus supplemental doses of insulin lispro. Glycemic control improved similarly in all treatment groups. It should be noted that most patients with an admission glucose < 180 mg/dl who were treated with sitagliptin plus correction doses of rapid-acting insulin responded as well as those treated with a basal-bolus insulin regimen. However, patients with an admission glucose > 180 mg/dl treated with sitagliptin alone had a higher mean daily blood glucose (182.7 ± 30 mg/dl) compared with patients treated with basal-bolus (168.1 ± 31 mg/dl) and sitagliptin plus glargine (161.8 ± 31 mg/dl) (P = 0.08).

Improving glycemic control while decreasing glycemic variability may be another advantage of using incretin-based therapies in the hospital. There is substantial evidence linking high glycemic variability with adverse outcomes in critically ill patients^102–104  and with increased lengths of stay and mortality rates in noncritically ill hospitalized patients.¹⁰⁵  Incretin therapies are known to reduce glucose fluctuations and glycemic variability measures,^99,106  but the potential benefits of these agents on glycemic variability has yet to be determined in the hospital setting.

Increased risk of acute pancreatitis, including fatal and nonfatal hemorrhagic or necrotizing pancreatitis has been reported with incretin therapies.^107–111  Although post-marketing reports of these occurrences are exceedingly rare, caution should be used in patients with a history of pancreatitis, abdominal pain, or postsurgical ileus. Additionally, increases of 2–5 bpm in heart rate have been observed in hospitalized patients with cardiovascular disease receiving GLP-1 receptor agonist therapy.^100,112  Although clinical trials have found no associations between adverse cardiovascular events and the use of GLP-1 receptor agonists¹¹³  or DPP-4 inhibitors,¹¹⁴  long-term studies are needed. Two prospective outpatient cardiovascular outcome trials recently reported that patients treatment with alogliptin or saxagliptin had no increased risk of ischemic events,^115,116  although rates of hospitalization for heart failure were increased with saxagliptin therapy.^115,116 

To avoid medication errors and hospital readmissions, preparation for transition to the outpatient setting should begin at the time of hospital admission.¹¹⁷  Based on a recent A1C, for patients previously well controlled (A1C < 7%) on oral medications, it is reasonable to resume their previous regimen at the time of discharge. However, patients with higher A1C values will often require intensification of their outpatient regimen or continuation of the inpatient insulin regimen.⁵ 

Patients newly started on insulin therapy during hospitalization will require substantial diabetes education before discharge. Although American Diabetes Association guidelines recommend that patients receive “survival skills” education to ensure safe care upon returning home,¹¹⁸  it is unknown whether this actually occurs and whether patients who do receive this training are able to apply it after discharge. There is limited information about follow-up of patients with diabetes after discharge, and there have been virtually no studies on methods to improve the transition of care from the inpatient to the outpatient setting. Nevertheless, current recommendations encourage that patients and their family or caregivers receive both written and oral instructions regarding their diabetes management regimen at the time of hospital discharge.⁵  In addition, clear communication with outpatient providers is crucial for ensuring a safe and successful transition to outpatient glycemic management.⁶ 

Inpatient hyperglycemia is common and associated with increased risks of complications in patients with or without diabetes. Correction of hyperglycemia with insulin administration has been shown to improve clinical outcomes,^5,6  and existing clinical evidence favors insulin over other antidiabetic agents in achieving and maintaining glycemic control in hospitalized patients. The basal-bolus approach has been shown to improve glycemic control and reduce perioperative complications in general medicine and surgery patients with diabetes.^12,30  Hypoglycemia is the most frequent side effect and the limiting factor for achieving glycemic control in the inpatient setting. Although, recent studies have reported an association between hypoglycemia and increased hospital mortality, it remains unclear whether hypoglycemia is a direct mediator of adverse outcomes and mortality or merely a biomarker of disease burden and severity.

Increasing preliminary data indicate that incretin-based therapies have the potential to improve glycemic control with a low risk of hypoglycemia. However, more prospective studies are needed to determine the safety and efficacy of these agents in the hospital setting. The remaining available antidiabetic drugs have a limited role in the hospital because of their potential for serious adverse side effects, including the risk of lactic acidosis with metformin, hypoglycemia with sulfonylureas and glinides, edema and heart failure with pioglitazone, and dehydration and urinary and genital infections with the SGLT-2 inhibitors.