Pathways to Quality Inpatient Management of Hyperglycemia ... · Table 1dKey issues in inpatient glucose management, suggested solutions, and areas in which future research is needed

Pathways to Quality InpatientManagement of Hyperglycemia andDiabetes: A Call to ActionBORIS DRAZNIN, MD, PHD

1

JANICE GILDEN, MS, MD2

SHERITA H. GOLDEN, MD, MHS3

SILVIO E. INZUCCHI, MD4

FOR THE PRIDE INVESTIGATORS*

Currently patients with diabetes comprise up to 25–30% of the census of adult wards and criticalcare units in our hospitals. Although evidence suggests that avoidance of hyperglycemia (.180mg/dL) and hypoglycemia (,70 mg/dL) is beneficial for positive outcomes in the hospitalizedpatient, much of this evidence remains controversial and at times somewhat contradictory. Wehave recently formed a consortium for Planning Research in Inpatient Diabetes (PRIDE) with thegoal of promoting clinical research in the area of management of hyperglycemia and diabetes in thehospital. In this article, we outline eight aspects of inpatient glucose management in which ran-domized clinical trials are needed. We refer to four as system-based issues and four as patient-basedissues. We urge further progress in the science of inpatient diabetes management. We hope this callto action is supported by the American Diabetes Association, The Endocrine Society, the AmericanAssociation of Clinical Endocrinologists, the American Heart Association, the European Associationfor the Study of Diabetes, the International Diabetes Federation, and the Society of Hospital Med-icine. Appropriate federal research funding in this area will help ensure high-quality investigations,the results of which will advance the field. Future clinical trials will allow practitioners to developoptimal approaches for the management of hyperglycemia in the hospitalized patient and lessenthe economic and human burden of poor glycemic control and its associated complications andcomorbidities in the inpatient setting.

Diabetes Care 36:1807–1814, 2013

Over the past decade, there has beenincreasing interest in glycemic man-agement of hospitalized patients.

There is now broad consensus that bothhyperglycemia and hypoglycemia in hospi-talized patients are associated with adverseoutcomes, including mortality. There is lessagreement, however, as to whether theseassociations actually reflect the effects of thequality of glucose management or aremerely underlying paraphenomena of theseverity of acute illness. Even more contro-versial is the actual potential impact ofglycemic control during these hospitaliza-tions that are often relatively brief, thespecific glucose ranges that should be tar-geted, and the methods by which cliniciansmight achieve these.

In the 1960s, research on the benefits ofglucose-insulin-potassium infusion duringacute myocardial infarction began, but thisline of inquiry was not focused on glucosecontrol per se (1). Interest in the generalfield of glycemic management in the inpa-tient setting began in themid 1990s (2). Thenext 10 yearsweremarkedbybothprospec-tive observational trials and randomizedclinical trials (RCTs), the majority of whichseemed to indicate that “lower is better”:hospital complications, length of stay, cost,and even mortality could be dramaticallydecreased in a variety of critical care settingsif mean glucose concentrations were re-duced, usually with intravenous insulin, to-ward or within the euglycemic range (3,4).Some results, however, seemed too good to

be true, especially in the context of suchshort hospital stays. This skepticism led toconfirmatory trials, most conducted using amulticenter design. These could not con-firm the initial positive findings fromsingle-center investigations (5–7). Therewas resulting confusion as to how these re-sults might shape clinical practice. Severalconsensus documents have emerged, eachendorsing amoremoderate approach to themanagement of glycemia in the hospitalizedpatient (8–11). Notably, all have called formore research in this area so that we canbetter understand the impact of both hyper-glycemia and hypoglycemia on inpatientoutcomes and better delineate evidence-based standards for hospital practice.

To date, most investigations have beenfunded through local resources or industry,as agencies appear reluctant to commitfinancial support for research in inpatientglycemic management. However, greaterefforts devoted to the study of diabetes inthe hospital setting would have broadimplications for our health care system(12). In addition to funding, the nascentdiscipline of inpatient glucose manage-ment will benefit from standardized no-menclature, consistent and meaningfulmetrics, and transparent study designsand analytical methods allowing for com-parison of study outcomes.

In this article, we outline eight aspectsof inpatient glucose management in whichRCTs and/or rigorously designed observa-tional studies are needed. We refer to fouras system-based issues and four as patient-based issues. Our goal was to identifyexisting research gaps and clinical carechallenges in inpatient glucose manage-ment and to suggest future directions foreach. These are summarized in Table 1.

System-based issues

Obstacles for glycemic control inthe hospitalDespite growing evidence supporting theimportance of glycemic control in thehospital setting (13–15), numerous obsta-cles stand in the way of its achievement.Major factors include unanticipatedchanges in nutrition; medication changes

c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c c

From the 1Division of Endocrinology, Diabetes and Metabolism, University of Colorado School of Medicine,Aurora, Colorado; the 2Division of Diabetes and Endocrinology, Department of Medicine, ChicagoMedicalSchool, Rosalind Franklin University of Medicine and Science, North Chicago, Illinois; the 3Division ofEndocrinology and Metabolism, Johns Hopkins University School of Medicine, Baltimore, Maryland; andthe 4Section of Endocrinology, Yale University School of Medicine, New Haven, Connecticut.

Corresponding author: Boris Draznin, [email protected]: 10.2337/dc12-2508*A complete list of the members of the PRIDE Writing Group can be found in the APPENDIX.© 2013 by the American Diabetes Association. Readers may use this article as long as the work is properly

cited, the use is educational and not for profit, and thework is not altered. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

See accompanying articles, pp. 2107 and 2112.

care.diabetesjournals.org DIABETES CARE, VOLUME 36, JULY 2013 1807

P E R S P E C T I V E S I N C A R E

Table 1dKey issues in inpatient glucose management, suggested solutions, and areas in which future research is needed

Suggested solutions and future research needed

System-based issuesObstacles for glycemic controlin the hospital

Development and evaluation of provider inpatient glucose management education tools.c Address barriers to achieving glycemic controlc Improve provider knowledge of critical glucose management topics toenhance patient safety

Development and evaluation of point-of-care electronic clinical decision aids toguide prescribers in implementing evidence-based guidelines.

Develop inpatient glucose management sign-out tools to enhance provider communication.c Electronic insulin/glucose data displayc Daily diabetes rounding sheet

Implementation and impact ofglycemic management teams

Use QI statistical methodologies that allow systematic evaluation of the impact ofmulticomponent glucose management programs on process, intermediary, andclinical outcome measures.

Perform multicenter, prospective “trials” of different team-based glycemic controlstrategies within single healthcare organizations.

Standardization of glucometricsand benchmarking amonghospitals

Determine an agreed-upon definition of inpatient hypoglycemia and hyperglycemiausing thresholds that are associated with adverse clinical and economic outcomes.

Collaboration of professional organizations with The Joint Commission to assist withstandardization of glucometrics to which hospitals will be held accountable.

Economic impact of glycemiccontrol in the hospital

Perform RCTs and observational studies to examine different treatment approaches toachieving glycemic control in noncritically ill patients on hospital costs and length of stay.

Design retrospective and prospective studies to determine which elements of multicomponentglucose management programs are most cost-effective.

Patient-based issuesSliding scale vs. nutritional/correctionalscheduled insulin administration

Perform comparative effectiveness RCTs, similar to RABBIT, comparing scheduled insulin(basal/nutritional/correctional) to sliding scale insulin (basal 1 sliding scale withoutnutritional insulin).

Health services studies to evaluate the impact of systems interventions on prescribing behavior.Glycemic control in special

patient populationsPatients on enteral andparenteral nutrition

Comparative effectiveness RCTs comparing several approaches to insulin management.c Premixed insulin vs. scheduled basal and short-acting insulinc Addition of total insulin dose to TPN bag vs. combination of basal insulin with nutritionalinsulin in TPN bag

Comparative effectiveness RCTs comparing insulin-based to incretin-based therapies.Patients on glucocorticoids Comparative effectiveness RCTs of approaches to insulin management in patients receiving

various types of glucocorticoids.Comparative effectiveness RCTs comparing insulin-based to incretin-based therapies.

Patients undergoing organtransplantation

Intervention studies (RCTs) to determine whether tight glycemic control followingtransplantation improves graft survival and total and cardiovascular mortality.

Intervention studies (RCTs) on key risk factors for NODAT to determine if NODATcan be prevented.

Use of incretin-based therapyin the hospital

Comparative effectiveness studies (RCTs) of incretin-basedversus insulin-based therapies on glycemic and clinical outcomes in critical careand noncritical care patient populations.

Transition from inpatient tooutpatient management

Develop and test algorithms to determine the most appropriate hospital andpostdischarge glycemic control regimen based on patients’ prehospitalizationglycemic control.

Design observational studies to better characterize risk factors for readmission inpatients with diabetes.

Comparative effectiveness studies of optimal hospital discharge approachesfor patients with diabetes to prevent readmission.

QI, quality improvement.

1808 DIABETES CARE, VOLUME 36, JULY 2013 care.diabetesjournals.org

Inpatient management of hyperglycemia and diabetes

and the use of medications associated withincreased insulin resistance such as gluco-corticoids, often in variable and changingdoses; physiologic stress responses to ill-ness; comorbid events such as acute orworsening renal insufficiency which mayheighten the risk for hypoglycemia; andmultiple system/organizational barriers,including the lack of communicationand/or diabetes management knowledgedeficits among providers and care givers.

Many patients experience changes innutritional status during the course of ahospitalization, switching from oral in-take to temporary nil per os status, andoccasionally to the initiation of enteral orparenteral feeding. Each of these changesrequires distinct, individualized, and spe-cific insulin regimens (16). In addition,patients admitted to the hospital mayhave inadequate outpatient glycemic con-trol, or need to be transitioned from oralantihyperglycemic agents to insulin.Stress can come from many sources in-cluding procedures, surgeries, infection,and pain, adding to the numerous factorsthat can unexpectedly raise or lower glu-cose values in the hospital setting.

Furthermore, communication is alwaysa challenge in the inpatient setting. Goodcommunication and coordination, as wellas agreement between the primary teamand the consulting diabetes specialist, areessential for appropriate insulin adjust-ments and anticipation of major changesin a patient’s status. In addition, syn-chrony between physicians and otherproviders including nurses, nurse practi-tioners, dietitians, pharmacists, as well asstaffing issues required for efficient coor-dination of care (orders, timing, dietaryrequirements, and effective protocol im-plementation) must be emphasized in or-der to lessen the risk of errors. Clearcommunication with the patient, devotingparticular attention to those withhypoglycemic unawareness, can be espe-cially helpful in preventing untowardoutcomes, particularly in the transitionto discharge.

Finally, system issues often presentfurther obstacles to glycemic control inthe hospital setting. Coordinating insulindosing with meals is often problematic.Hospital staff may have varying degreesof knowledge regarding proper manage-ment of both hyper- and hypoglycemia, in-cluding a lack of awareness of consensus- orevidence-based practices. The behaviorsand beliefs of physicians and other provid-ers are often difficult to change because ofclinical inertia. This is best demonstrated

by the continued use of “sliding-scale insu-lin” as a substitute for scheduled insulinwith correction doses. Insufficient under-standing of the importance of inpatient gly-cemic control and a fear of the risk ofinducing hypoglycemia may influence var-ious care practices.Suggested future directions. To addressand overcome these obstacles we recom-mend development and evaluation of1) provider education tools to improveknowledge of and address barriers toachieving glycemic control, 2) clinical de-cision aids at the point of care to guideprescribers in implementing evidence-based guidelines, and 3) inpatient glucosemanagement sign-out tools to enhanceprovider communication. These inter-ventions should target physicians, midle-vel care providers, nurses, pharmacists,and dietitians. Studies evaluating the in-terventions should assess their impact onprocesses of care (insulin prescribingpractices), intermediary outcomes (hy-perglycemia and hypoglycemia), clinicaloutcomes (in-hospital mortality, nosoco-mial infections), and economic outcomes(length of stay, hospital admission costs,readmissions) (Fig. 1).

Implementation and impact ofglycemic management teamsIn 2004, the American College of Endo-crinology published a position statementoutlining the rationale for inpatient gly-cemic control (17). Following this earlyconsensus conference, numerous suc-cessful campaigns were launched to in-spire and produce champions in theburgeoning field of inpatient diabetesmanagement. In 2009, the American Col-lege of Endocrinology partnered with theAmerican Diabetes Association andreleased a call to action that outlined strat-egies for successful implementation of in-patient glucose management programs(8). Although clinical studies evaluatinginsulin use in the hospital setting havecontinued to emerge since then, rigorousscientific studies examining the risks andbenefits of institutional programs withmultilevel interventions have lagged be-hind. The optimal study should evaluatevalid process (e.g., protocol adherence andprescribing practices), intermediary (e.g.,glycemic control), clinical (e.g., nosocomialinfections), and economic (e.g., length ofstay) outcomes, as well as institutionalacceptance and cost-effectiveness. Addi-tionally, the expertise of glycemic manage-ment teams may be particularly importantwhen managing patients treated with

continuous subcutaneous insulin infusionor other special populations, such as thosereceiving enteral and parenteral nutrition,glucocorticoids, and transplant medica-tions (see below). In this era of account-able care, high-quality research toidentify the most effective glycemic man-agement program characteristics andcomponents, and the information sys-tems required to maintain them, is abso-lutely imperative, since 30–50% of adultinpatients have diabetes and/or hypergly-cemia during their hospital stay.Suggested future directions. Given eth-ical considerations in not exposing allinpatients to the same quality of care andlogistical issues, RCTs in this area arechallenging. However, several groupshave described quality improvement statis-tical methodologies that allow systematicevaluation of a program’s impact on out-comes using process control charts (18).These methodologies should be used to1) systematically evaluate the impact ofmulticomponent glucose managementprograms on process, intermediary, andclinical outcomes before and after the inter-ventions, and (2) evaluate institutional in-terventions through regularly scheduledcycles of performance improvement (e.g.,Plan-Do-Study-Act) allowing real-time al-ternations to make interventions most ef-fective. Finally, multicenter prospective“trials” are needed to examine the impactof different team-based glycemic controlstrategies at single healthcare organizations(e.g., physician-led team vs. midlevel careprovider-led team) on process, intermedi-ary, and clinical outcomes.

Standardization of glucometrics andbenchmarking among hospitalsOver the past decade, a growing body ofknowledge has emerged regarding theimportance of the management of hyper-glycemia in hospitalized patients, both inintensive care units (ICUs) and in non-critical care settings (19). As a result ofthese data, clinical guidelines have beenpublished, and The Joint Commission haseven established an “Advanced Certifica-tion in Inpatient Diabetes” program. Glu-cose data collection and analysis are thefoundation of all of these efforts. Unfor-tunately, there is a surprising lack of stan-dardization that hampers benchmarkingbetween various hospitals and directcomparison of different protocols.

Point-of-care glucose values fromnoncritical care units are used for mostanalyses, whereas for critically ill pa-tients, both hospital laboratory plasma

care.diabetesjournals.org DIABETES CARE, VOLUME 36, JULY 2013 1809

Draznin and Associates

glucose values and point-of-care testingare often used. Even when the types ofglucose data are homogeneous, adverseoutcomes do not have agreed-upon def-initions. For example, hypoglycemia canvary from “glucose level ,70 mg/dL,”which encompasses the physiologicallevels of 60–70 mg/dL, to glucose levels,60 mg/dL. “Severe” hypoglycemia isfrequently described as glucose values,40 mg/dL, without adequate data sup-porting these particular glucose thresholds.

Glucometrics efforts by Yale, the Uni-versity Hospitals Consortium, and theSociety of Hospital Medicine (20–23)are important steps in the direction ofstandardization, but the definition of op-timal glucose control still differs amonghospitals and does not always reflect theglycemic targets published by the Ameri-can Association of Clinical Endocrinolo-gists, the American Diabetes Association,and The Endocrine Society.Suggested future directions. Since RCTsare not ethical, rigorously designed retro-spective and prospective observationalstudies are needed to determine hypogly-cemia and hyperglycemia thresholds thatare associated with adverse clinical (e.g.,mortality, nosocomial infections) and eco-nomic outcomes (e.g., length of stay,

hospital charges). These data can assistin achieving consensus universal defini-tions of glucose control or malglycemia(24). It is also important to achieve con-sensus on the most appropriate glucomet-ric model for describing hyperglycemia(e.g., population, patient-day, or patient-stay model) based upon which model ismost strongly associated with clinicaland economic outcomes. Finally, profes-sional organizations that develop inpatientglycemic guidelines need to collaboratewithThe Joint Commission to assist in standard-izing glycemic definitions across hospitals.

Economic impact of glycemiccontrol in the hospitalThe potential economic impact of glyce-mic inpatient management initiativesneeds to be established further. Availableevidence shows that in the critically illpatient, intensive insulin therapy (IIT)results in a reduction in ICU length ofstay and hospital costs. Van den Bergheet al. (25) estimated a reduction in ICUlength of stay of 2 days with an associatedreduction in costs of V2,680 (;$3,410)per admission in patients who were treatedwith IIT. Krinsley et al. (26) also reporteda reduction in ICU length of stay, but amore modest 0.3 days, with a cost savings

of $1,580 per patient after implement-ing IIT. Finally, Sadhu et al. (27) showedan impressive reduction in ICU length ofstay of 1.8 days with a savings in total hos-pital costs of $7,580 per patient in patientsreceiving IIT. Despite these encouragingdata, further studies are needed in otherpatient populations. This is particularlypressing in the noncritically ill, which com-prise the majority of hospitalized patients.Suggested future directions. RCTs ex-amining different treatment approachesto achieving glycemic control in noncriti-cally ill patients should also assess theimpact of the interventions on hospitalcosts and length of stay. Multilevel mod-eling approaches applied to retrospec-tively and prospectively collected datashould be used to examine the impactof hospital-wide glucose managementprograms on economic outcomes and todetermine which elements of multicom-ponents programs are most cost-effective.

Patient-based issues

Sliding scale versus nutritional/correctional scheduled insulinadministrationTo the great chagrin of the majority ofdiabetologists (28,29), the use of sliding

Figure 1dDiagram of a conceptual model for pathways to quality inpatient management of hyperglycemia and diabetes, adapted fromMunoz et al.(18).

1810 DIABETES CARE, VOLUME 36, JULY 2013 care.diabetesjournals.org

Inpatient management of hyperglycemia and diabetes

scale insulin administration persists, evenat the most prestigious teaching hospitals(23,30). This reflexive behavior of order-ing sliding scale may be harmful, compar-ing unfavorably to the more physiologicalscheduled insulin therapy. Several obsta-cles appear to exist in this seemingly per-petual and frustrating fight. The first is alack of agreement on and understandingof what the term sliding scale actuallymeans. Depending on the amount offood (or carbohydrates) to be consumedand the level of premeal glycemia, bothcomponents of scheduled bolus insulin,nutritional and correctional, may differfrom one injection to another. This maycreate the deceptive appearance of a slid-ing scale regimen in the eyes of nonspe-cialists. It is our task to clearly articulatethe differences between sliding scale in-sulin and scheduled insulin doses that in-clude both nutritional and correctional(supplemental) insulin for our colleaguesand the next generation of clinicians.

A significant challenge preventing theelimination of the sliding scale is the lackof clinical evidence for the superiority ofbasal-prandial correction therapy to thesliding scale approach during shorter hos-pital stays. The paucity of trials examiningthe potential impact of glycemic controlduring shorter stays raises reasonable ques-tions about the appropriate approach inthis setting. However, it may not be ethicalto test sliding scale insulin therapy withoutscheduled insulin in type 1 diabetes.

An effective solution to the use ofsliding scale insulin is the developmentand implementation of policies, proto-cols, and order sets for scheduled insulinadministration in various clinical situa-tions, as was demonstrated by the RABBIT(RAndomized Study of Basal-Bolus Insu-lin Therapy) studies (31,32). Unfortu-nately, these are not available in manyhospital systems: institutional acceptanceof protocols and order sets is far from uni-versal. This leaves these key clinical deci-sions in the hands of practitioners, whomay be reluctant to seek advice from aconsultant but are unwilling or lack theexpertise to implement proactive insulinstrategies themselves.Suggested future directions. Additionalwell-designed comparative effectivenessRCTs, similar to the RABBIT studies,examining the effect of scheduled insulin(basal/nutritional/correctional) to slidingscale insulin (basal 1 correctional with-out nutritional insulin) on intermediaryglycemic, clinical, and economic out-comes are needed to provide the evidence

base for promoting changes in prescribingpractices. In addition, health servicesstudies are needed to evaluate the impactof systems interventions, such as imple-mentation of policies, protocols, and or-der sets promoting scheduled insulin andprovider education, on changing pre-scribing behavior.

Glycemic control in special clinicalpopulationsA) Patients receiving enteral and par-enteral nutrition. Enteral and parenteralnutrition pose major challenges for glu-cose management in the hospital. Thesenutritional approaches frequently resultin hyperglycemia, even in patientswithout a history of diabetes. The resul-tant hyperglycemia has been associatedwith poor outcomes (33,34). However,scheduled subcutaneous insulin may bedifficult to implement safely because offrequent planned and unplanned inter-ruptions and titration of the nutritionalsource. In the case of enteral nutrition, asmall randomized clinical study demon-strated that subcutaneous basal insulin ismore effective than correction dosingalone (35). Another study had suggestedthat the use of premixed 70/30 insulintwice or three times daily may be saferthan the use of long-acting insulin in pa-tients on continuous tube feeding (36). Itis unclear whether approaches such asscheduled short-acting insulin may beequally safe and efficacious because stud-ies of other potential approaches have notbeen published. In the case of total par-enteral nutrition (TPN), no randomizedcontrolled trials comparing various ap-proaches to insulin therapy are available,although retrospective data suggest thatthe addition of insulin in the TPN bagprovides good control with less hypogly-cemia than the use of intravenous insulinor subcutaneous insulin alone (37).Suggested future directions. Compara-tive effectiveness RCTs are needed tocompare several approaches to insulinmanagement on intermediary glycemic,clinical, and economic outcomes for pa-tients receiving continuous enteral nutri-tion and TPN. For either nutritionalapproach, a computerized intravenousinsulin algorithm or noninsulin agents,such as incretin-based therapies, may pro-vide safer ormore effective alternatives, butvery few data are available. Therefore,comparative effectiveness RCTs comparingthe effect of insulin-based to incretin-basedtherapies on glycemic and clinical out-comes are needed for this population.

B) Patients receiving glucocorticoids.Administration of glucocorticoids has adetrimental effect on glycemic control inpatients with diabetes (38,39), presentinga significant challenge for both outpatientand inpatient management. The dose andfrequency of steroid administration varieswidely, and steroids may be tapered orstopped abruptly. Acute or short-termadministration of methylprednisolonecauses predominantly postprandial hy-perglycemia that lasts 6 to 12 h. Predni-sone and dexamethasone have even longerdurations of action. Because NPH insulinhas a duration of action of;8–12 h (40), ithas been used at the time of methylpred-nisolone administration to counteract thehyperglycemic effect of this glucocorticoid(41). NPH insulin may be stopped as soonas methylprednisolone is discontinued,providing a safer option than having re-sidual insulin action from high doses oflong-acting insulin lasting for hours afterdiscontinuation of steroids. However, nosystematic research has been published inpatients receiving either multiple dailydoses of steroids or in those receivingdexamethasone.Suggested future directions. Compara-tive effectiveness RCTs of approaches toinsulin management on glycemic, clini-cal, and economic outcomes in patientswith steroid-induced hyperglycemia re-ceiving various types of glucocorticoidsare needed. Recently published commu-nications suggest that either GLP-1agonists or dipeptidyl peptidase-4 inhib-itors may be useful in treating steroid-induced hyperglycemia (42). Thus,comparative effectiveness RCTs comparingthe effect of insulin-based to incretin-basedtherapies on glycemic, clinical, and eco-nomic outcomes are needed in this popu-lation.C) Patients undergoing organ trans-plantation. Hyperglycemia is a commonproblem inpatients undergoing solid organor hematopoietic stem cell transplantation.There is a significant prevalence of preex-isting diabetes in patients undergoingtransplantation as well as patients whodevelopnew-onset diabetes after transplan-tation (NODAT) (43–45). In a large epide-miological study of kidney transplantation,the cumulative incidence of NODAT was9% at 3 months and increased linearly to24% at 36 months (46). Common risk fac-tors for NODAT include hepatitis C infec-tion and steroid and calcineurin/c inhibitorcombination therapy.

In most studies, preexisting diabeteshad no effect on renal graft survival but

care.diabetesjournals.org DIABETES CARE, VOLUME 36, JULY 2013 1811

Draznin and Associates

was associated with increased mortality,which was mostly cardiovascular (46,47).NODAT has been associated with renalgraft failure and increased cardiovascularmortality (45,48). Whether hyperglyce-mia is causal or just a marker of graftrejection or cardiovascular mortality re-mains unknown.Suggested future directions. RCTs areneeded to determine whether tight glyce-mic control following transplantation im-proves graft survival and total andcardiovascular mortality. Interventionstudies on key risk factors for NODATare also needed to determine whether therisk of NODAT can be reduced by 1) treat-ing hepatitis C infection, 2) using steroid-free immunosuppression, 3) avoidingcalcineurin/mammalian target of rapamy-cin inhibitor combination therapy, or 4)hyperglycemia management with insulin.A recent pilot study demonstrated that inkidney transplant patients without priordiabetes, short-term (3 weeks) tight gly-cemic control using insulin reduced therisk of development of NODAT at 12months by 73% when compared withcontrol subjects with less tight glucosecontrol (49). The authors speculate thatearly insulin therapy is b-cell protective.This surprising finding needs to be repli-cated, and a larger multicenter trial is cur-rently underway.

Use of incretin-based therapy inthe hospitalIncretin-based therapies would be ex-pected to control hyperglycemia in thehospital setting in patients with type 2diabetes without risking hypoglycemia,when used in the absence of otherhypoglycemic agents such as insulin.The rationale for predicting thatincretin-based therapy might be saferthan insulin therapy lies in its glucose-dependent insulin secretion, a markedbenefit in reducing glycemic elevationsfrom two stress hormones (glucagon andglucocorticoids), decreased glycemic var-iability, elimination of need for insulin,decreased requirement for bolus insulineven if basal insulin is needed, and po-tential beneficial effects on cardiovascularfunction (50). As alluded to in prior sec-tions, the preliminary and mainly uncon-trolled studies published thus far must bevalidated in well-designed RCTs (51).Suggested future directions. Compara-tive effectiveness RCTs are needed toexamine the effect of incretin-basedversus insulin-based therapies on glyce-mic, clinical, and economic outcomes in

critical care and noncritical care patientpopulations.

Transition from inpatient tooutpatient managementFor a patient with diabetes being dis-charged to home from the hospital, thereare both short- and long-term concerns.The immediate issue is whether or not thepatient can safely return home. The long-term concern is how and when changes topreadmission medication regimensshould be implemented at hospital dis-charge.

The distinct feature of diabetes tran-sitional care is that the inpatient diabetesmedical regimen is often completely dif-ferent from what is used in an outpatientsetting in the majority of patients (16).Current standards for inpatient diabetesmanagement are based primarily on insu-lin therapy, regardless of whether theywere treated with insulin or oral agentsprior to admission (19). Conditions inthe hospital may cause dramatic differen-ces in glucose handling that may not re-turn to baseline after discharge. This canbe particularly challenging in patients oncontinuous subcutaneous insulin infu-sion who may need the devices temporar-ily removed or adjusted during certainprocedures or due to mental statuschanges during a hospitalization.

The discrepancy in diabetes drugtherapy between inpatient and outpatientcare poses a significant threat to patientsafety after patients are discharged. Thedoses of insulin recommended on dis-charge can be significantly higher orlower than those actually needed athome.

Furthermore, diabetes may contrib-ute to the high readmission rates associatedwith certain conditions, such as cardio-vascular disease. In addition, many pa-tients with uncontrolled diabetes do nothave their diabetes regimen adjustedproperly both prior to or during thehospitalization, possibly due to apparentclinical inertia (52). The hospitalizationper se and the design of a diabetes medi-cation regimen for the patient uponhospital discharge may represent op-portunities to modify previous outpatientdiabetes care. How the level of prehospitalcontrol may guide therapy during hospi-talization and after hospital discharge toprevent readmissions should also be asubject of intense investigation (53,54).Suggested future directions. Studies areneeded to develop and test algorithms todetermine the most appropriate hospital

and postdischarge glycemic control regi-men based upon patients’ prehospitaliza-tion glycemic control. Rigorously designedobservational studies are needed to bettercharacterize risk factors for readmissionamong diabetic patients because risk fac-tors and causes specific to this populationare not well characterized (55). Finally,comparative effectiveness studies areneeded to determine optimal hospital dis-charge approaches for patients with diabe-tes, to maintain continuity of care andprevent readmissions.

ConclusiondThe PRIDE group,a consortium for Planning Research inInpatient Diabetes, has been formed topromote clinical research in the manage-ment of hyperglycemia and diabetes inthe hospital. We urge further progress inthe science of inpatient diabetes manage-ment. We hope this call to action issupported by the American Diabetes As-sociation, The Endocrine Society, theAmerican Association of Clinical Endo-crinologists, the American Heart Associa-tion, the European Association for theStudy of Diabetes, the International Di-abetes Federation, and the Society ofHospital Medicine. Appropriate federal re-search funding in this area will help ensurehigh-quality investigations, the results ofwhichwill advance thefield. Future clinicaltrials will allow practitioners to developoptimal approaches for the management ofhyperglycemia in the hospitalized patientand lessen the economic and human bur-den of poor glycemic control and itsassociated complications and comorbidi-ties in the inpatient setting.

AcknowledgmentsdB.D. has received re-search grants from Novo Nordisk and Sanofi.S.E.I. has consulted for Takeda, Merck,Boehringer Ingelheim, and Janssen.B.D. drafted themanuscript. All members of

the writing group revised and edited themanuscript. J.G., S.H.G., and S.E.I. made ex-tensive revisions. B.D. reviewed and finalizedthe manuscript.

APPENDIXdThe PRIDE WritingGroup in alphabetical order (all authorsparticipated actively in writing and editingthe manuscript): David Baldwin, RushUniversity Medical School (research grantfrom Novo Nordisk); Bruce W. Bode, At-lanta Diabetes Associates (stock owner-ship in Aseko; consulting for Medtronic,DexCom,NovoNordisk, Sanofi, Halozyme;

1812 DIABETES CARE, VOLUME 36, JULY 2013 care.diabetesjournals.org

Inpatient management of hyperglycemia and diabetes

speaker for Medtronic, DexCom, NovoNordisk, Eli Lilly, Sanofi, Bristol-MyersSquibb, Amylin, Merck, Insulet; researchgrants from Medtronic, DexCom, NovoNordisk, Eli Lilly, Sanofi, Bristol-MyersSquibb, MannKind, Biodel, Halozyme,Macrogenetics, Merck, Abbott); Jeffrey B.Boord, Vanderbilt Heart and Vascular In-stitute; Susan S. Braithwaite, University ofIllinois at Chicago; Enrico Cagliero, Mas-sachusetts General Hospital, HarvardMedical School; Boris Draznin, Univer-sity of Colorado School of Medicine (re-search grants fromNovoNordisk, Sanofi);KathleenM. Dungan, Ohio State UniversitySchool of Medicine (consulting for Eli Lilly,Pfizer, Diabetes Technology Management;speaker for Medikinetics; research grantfrom Novo Nordisk); Mercedes Falciglia,University of Cincinnati College of Medi-cine; M. Kathleen Figaro, Vanderbilt Uni-versity School of Medicine; Janice Gilden,Rosalind Franklin University of Medicineand Science/Chicago Medical School andCaptain James A. Lovell Federal HealthCare Center; Sherita H. Golden, JohnsHopkins University School of Medicine; IrlB. Hirsch, University ofWashington Schoolof Medicine (consulting for Johnson &Johnson, Roche, Abbott; research grantfrom Sanofi); Silvio E. Inzucchi, Yale Uni-versity School of Medicine, Yale-NewHaven Hospital (consulting for Takeda,Merck, Boehringer Ingelheim, Janssen);David Klonoff, Mills-Peninsula HealthServices, Diabetes Research Institute (con-sulting for Bayer, Insulet, Google, Roche,Sanofi; research grants from Biodel, EliLilly, MannKind, Medtronic, Novo Nor-disk); Mary T. Korytkowski, University ofPittsburgh School of Medicine (consultingfor Regeneron; research grant from Sanofi);Mikhail Kosiborod, St. Luke’s MidAmericaHeart Institute, University of MissouriKansas City (consulting for Medtronic,Glumetrics, Gilead, Genentech, Hoffmann-La Roche, Sanofi, Boehringer Ingelheim,CardioMEMS; research grants from Med-tronic, Glumetrics, Gilead, Genentech,Sanofi); Lillian F. Lien, Duke UniversityMedical Center (consulting for Sanofi,Merck, and Eli Lilly); Michelle F. Magee,MedStar Health, Georgetown UniversitySchool of Medicine; Umesh Masharani,University of California, San Francisco;Gregory Maynard, University of Cal-ifornia, San Diego; Marie E. McDonnell,Boston University School of Medicine; EtiS. Moghissi, University of California LosAngeles; Neda Rasouli, University of Col-orado School ofMedicine; Daniel J. Rubin,Temple University School of Medicine;

Robert J. Rushakoff, University of Cal-ifornia, San Francisco (speaker for NovoNordisk and Merck); Archana R. Sadhu,Weill Cornell Medical College; StanleySchwartz, Main Line Health System,Emeritus, University of Pennsylvania(consulting for Santarus, Merck, Takeda,Janssen, Sanofi, Amylin; speaker forSantarus, Merck, Takeda, Janssen, Sanofi,Eli Lilly, Amylin, Bristol-Myers Squibb,AstraZeneca, Boehringer Ingelheim, NovoNordisk); Jane Jeffrie Seley, New York-Presbyterian/Weill Cornell; GuillermoE.Umpierrez, GradyHealth System, EmoryUniversity School of Medicine (researchgrants from Merck and Sanofi); RobertA. Vigersky, Walter Reed National Mili-tary Medical Center (research grant fromDexCom Corporation); Cecilia C. LowWang, University of Colorado School ofMedicine; Deborah J.Wexler,MassachusettsGeneral Hospital, Harvard Medical School.

References1. Sodi-Pallares D, Testelli MR, Fishleder BL,

et al. Effects of an intravenous infusion of apotassium-glucose-insulin solution on theelectrocardiographic signs of myocardialinfarction. A preliminary clinical report.Am J Cardiol 1962;9:166–181

2. Malmberg K, Rydén L, Efendic S, et al.Randomized trial of insulin-glucose in-fusion followed by subcutaneous insulintreatment in diabetic patients with acutemyocardial infarction (DIGAMI study):effects on mortality at 1 year. J Am CollCardiol 1995;26:57–65

3. Furnary AP, Zerr KJ, Grunkemeier GL, StarrA. Continuous intravenous insulin infusionreduces the incidence of deep sternal woundinfection in diabetic patients after cardiacsurgical procedures. Ann Thorac Surg 1999;67:352–360; discussion 360–362

4. van den Berghe G, Wouters P, Weekers F,et al. Intensive insulin therapy in critically illpatients. N Engl JMed 2001;345:1359–1367

5. Brunkhorst FM, Engel C, Bloos F, et al.;German Competence Network Sepsis(SepNet). Intensive insulin therapy andpentastarch resuscitation in severe sepsis.N Engl J Med 2008;358:125–139

6. Preiser JC, Devos P, Ruiz-Santana S, et al.A prospective randomised multi-centrecontrolled trial on tight glucose control byintensive insulin therapy in adult in-tensive care units: the Glucontrol study.Intensive Care Med 2009;35:1738–1748

7. Finfer S, Chittock DR, Su SY, et al.; NICE-SUGAR Study Investigators. Intensiveversus conventional glucose control incritically ill patients. N Engl J Med 2009;360:1283–1297

8. Moghissi ES, Korytkowski MT, DiNardoM, et al.; American Association of

Clinical Endocrinologists; American Di-abetes Association. American Associationof Clinical Endocrinologists and AmericanDiabetes Association consensus statementon inpatient glycemic control. DiabetesCare 2009;32:1119–1131

9. QaseemA,Humphrey LL,ChouR, SnowV,Shekelle P; Clinical Guidelines Committeeof the American College of Physicians. Useof intensive insulin therapy for the man-agement of glycemic control in hospitalizedpatients: a clinical practice guideline fromthe American College of Physicians. AnnIntern Med 2011;154:260–267

10. Umpierrez GE, Hellman R, KorytkowskiMT, et al.; Endocrine Society. Managementof hyperglycemia in hospitalized patients innon-critical care setting: an endocrine so-ciety clinical practice guideline. J Clin En-docrinol Metab 2012;97:16–38

11. Jacobi J, Bircher N, Krinsley J, et al.Guidelines for the use of an insulin in-fusion for the management of hypergly-cemia in critically ill patients. Crit CareMed 2012;40:3251–3276

12. Simmons D, Wenzel H. Diabetes in-patients: a case of lose, lose, lose. Is it timeto use a ‘diabetes-attributable hospitali-zation cost’ to assess the impact of di-abetes? Diabet Med 2011;28:1123–1130

13. Clement S, Braithwaite SS, Magee MF,et al.; American Diabetes Association Di-abetes in Hospitals Writing Committee.Management of diabetes and hyperglyce-mia in hospitals. Diabetes Care 2004;27:553–591

14. Inzucchi SE, Bergenstal RM, Buse JB, et al.;American Diabetes Association (ADA);European Association for the Study ofDiabetes (EASD). Management of hyper-glycemia in type 2 diabetes: a patient-centered approach: position statement ofthe American Diabetes Association (ADA)and the European Association for theStudy of Diabetes (EASD). Diabetes Care2012;35:1364–1379

15. Braithwaite SS, Magee M, Sharretts JM,Schnipper JL, Amin A, Maynard G; Soci-ety of Hospital Medicine Glycemic Con-trol Task Force. The case for supportinginpatient glycemic control programs now:the evidence and beyond. J Hosp Med2008;3(Suppl):6–16

16. Barnard K, Batch B, Lien LF. Sub-cutaneous insulin: a guide for dosingregimens in the hospital. Glycemic Controlin the Hospitalized Patient. 1st ed. Lien LF,Cox ME, Feinglos MN, Corsino L, Eds.New York, Springer, 2011, p. 7–16

17. Garber AJ, Moghissi ES, Bransome ED Jr,et al.; American College of EndocrinologyTask Force on Inpatient Diabetes MetabolicControl. American College of Endocrinologyposition statement on inpatient diabetesandmetabolic control. Endocr Pract 2004;10(Suppl. 2):4–9

18. Munoz M, Pronovost P, Dintzis J, et al. Im-plementing and evaluating amulticomponent

care.diabetesjournals.org DIABETES CARE, VOLUME 36, JULY 2013 1813

Draznin and Associates

inpatient diabetes management pro-gram: putting research into practice.Jt Comm J Qual Patient Saf 2012;38:195–206

19. McDonnell ME, Umpierrez GE. Insulintherapy for the management of hyper-glycemia in hospitalized patients. Endo-crinol Metab Clin North Am 2012;41:175–201

20. Goldberg PA, Bozzo JE, Thomas PG, et al.“Glucometrics”dassessing the quality ofinpatient glucose management. DiabetesTechnol Ther 2006;8:560–569

21. Kosiborod M, Inzucchi SE, KrumholzHM, et al. Glucometrics in patients hos-pitalizedwith acutemyocardial infarction:defining the optimal outcomes-basedmeasure of risk. Circulation 2008;117:1018–1027

22. Schnipper JL, Magee M, Larsen K,Inzucchi SE, Maynard G; Society of Hos-pital Medicine Glycemic Control TaskForce. Society of Hospital Medicine Gly-cemic Control Task Force summary:practical recommendations for assessingthe impact of glycemic control efforts.J Hosp Med 2008;3(Suppl):66–75

23. Wexler DJ, Meigs JB, Cagliero E, NathanDM, Grant RW. Prevalence of hyper- andhypoglycemia among inpatients with di-abetes: a national survey of 44 U.S. hos-pitals. Diabetes Care 2007;30:367–369

24. Hammer MJ, Casper C, Gooley TA,O’Donnell PV, Boeckh M, Hirsch IB. Thecontribution of malglycemia to morta-lity among allogeneic hematopoietic celltransplant recipients. Biol Blood MarrowTransplant 2009;15:344–351

25. Van den Berghe G, Wouters PJ, KestelootK, Hilleman DE. Analysis of healthcareresource utilization with intensive insulintherapy in critically ill patients. Crit CareMed 2006;34:612–616

26. Krinsley JS, Jones RL. Cost analysis ofintensive glycemic control in critically illadult patients. Chest 2006;129:644–650

27. Sadhu AR, Ang AC, Ingram-Drake LA,Martinez DS, Hsueh WA, Ettner SL. Eco-nomic benefits of intensive insulin ther-apy in critically Ill patients: the targetedinsulin therapy to improve hospital out-comes (TRIUMPH) project. Diabetes Care2008;31:1556–1561

28. Hirsch IB. Sliding scale insulindtimeto stop sliding. JAMA 2009;301:213–214

29. Schnipper JL, Barsky EE, Shaykevich S,Fitzmaurice G, Pendergrass ML. Inpatientmanagement of diabetes and hyperglyce-mia among general medicine patients at alarge teaching hospital. J Hosp Med 2006;1:145–150

30. Boord JB, Greevy RA, Braithwaite SS, et al.Evaluation of hospital glycemic control atUS academic medical centers. J Hosp Med2009;4:35–44

31. Umpierrez GE, Smiley D, Zisman A, et al.Randomized study of basal-bolus insulintherapy in the inpatient management of

patients with type 2 diabetes (RABBIT 2trial). Diabetes Care 2007;30:2181–2186

32. Umpierrez GE, Smiley D, Jacobs S, et al.Randomized study of basal-bolus insulintherapy in the inpatient management ofpatients with type 2 diabetes undergoinggeneral surgery (RABBIT 2 surgery). Di-abetes Care 2011;34:256–261

33. Cheung NW, Napier B, Zaccaria C,Fletcher JP. Hyperglycemia is associatedwith adverse outcomes in patients re-ceiving total parenteral nutrition. DiabetesCare 2005;28:2367–2371

34. Pasquel FJ, Spiegelman R, McCauley M,et al. Hyperglycemia during total paren-teral nutrition: an important marker ofpoor outcome and mortality in hospital-ized patients. Diabetes Care 2010;33:739–741

35. Korytkowski MT, Salata RJ, Koerbel GL,et al. Insulin therapy and glycemic controlin hospitalized patients with diabetesduring enteral nutrition therapy: a ran-domized controlled clinical trial. DiabetesCare 2009;32:594–596

36. Hsia E, Seggelke SA, Gibbs J, Rasouli N,Draznin B. Comparison of 70/30 biphasicinsulin with glargine/lispro regimen innon-critically ill diabetic patients on con-tinuous enteral nutrition therapy. NutrClin Pract 2011;26:714–717

37. Baldwin D, Kinnare K, Draznin B, et al.Insulin treatment of hyperglycemia inhospitalized patients receiving total par-enteral nutrition (TPN) (Abstract). Di-abetes 2012;61(Suppl. 1):A1070

38. Bevier WC, Zisser HC, Jovanovic L, et al.Use of continuous glucose monitoring toestimate insulin requirements in patientswith type 1 diabetes mellitus during ashort course of prednisone. J Diabetes SciTech 2008;2:578–583

39. Oyer DS, Shah A, Bettenhausen S. How tomanage steroid diabetes in the patientwith cancer. J Support Oncol 2006;4:479–483

40. Lepore M, Pampanelli S, Fanelli C, et al.Pharmacokinetics and pharmacody-namics of subcutaneous injection of long-acting human insulin analog glargine,NPH insulin, and ultralente human insulinand continuous subcutaneous infusion ofinsulin lispro. Diabetes 2000;49:2142–2148

41. Seggelke SA, Gibbs J, Draznin B. Pilotstudy of using neutral protamine Hage-dorn insulin to counteract the effect ofmethylprednisolone in hospitalized pa-tients with diabetes. J Hosp Med 2011;6:175–176

42. van Raalte DH, van Genugten RE, LinssenMM, Ouwens DM, Diamant M. Glucagon-like peptide-1 receptor agonist treatmentprevents glucocorticoid-induced glucoseintolerance and islet-cell dysfunction inhumans. Diabetes Care 2011;34:412–417

43. CosioFG, Pesavento TE,OseiK,HenryML,Ferguson RM. Post-transplant diabetes

mellitus: increasing incidence in renal al-lograft recipients transplanted in recentyears. Kidney Int 2001;59:732–737

44. Foo SM, Wong HS, Morad Z. Risk factorsand incidence of posttransplant diabetesmellitus in renal transplant recipients.Transplant Proc 2004;36:2139–2140

45. Schiel R, Heinrich S, Steiner T, Ott U,Stein G. Post-transplant diabetes mellitus:risk factors, frequency of transplant re-jections, and long-term prognosis. ClinExp Nephrol 2005;9:164–169

46. Kasiske BL, Snyder JJ, Gilbertson D, MatasAJ. Diabetes mellitus after kidney trans-plantation in the United States. AmJ Transplant 2003;3:178–185

47. Lufft V, Dannenberg B, Schlitt HJ,Pichlmayr R, Brunkhorst R. Cardiovas-cular morbidity and mortality in patientswith diabetes mellitus type I after kidneytransplantation: a case-control study. ClinNephrol 2004;61:238–245

48. Cole EH, Johnston O, Rose CL, Gill JS.Impact of acute rejection and new-onsetdiabetes on long-term transplant graft andpatient survival. Clin J Am Soc Nephrol2008;3:814–821

49. Kuo HT, Sampaio MS, Vincenti F,Bunnapradist S. Associations of pretrans-plant diabetes mellitus, new-onset di-abetes after transplant, and acute rejectionwith transplant outcomes: an analysis ofthe Organ Procurement and TransplantNetwork/United Network for Organ Shar-ing (OPTN/UNOS) database. Am J KidneyDis 2010;56:1127–1139

50. Schwartz S, Kohl BA. Type 2 diabetesmellitus and the cardiometabolic syn-drome: impact of incretin-based thera-pies. Diabetes Metab Syndr Obes 2010;3:227–242

51. Schwartz S, DeFronzo RA. Is incretin-based therapy ready for the care of hos-pitalized patients with type 2 diabetes?The time has come for GLP-1 receptoragonists! Diabetes Care 2013;36:2107–2111

52. Griffith ML, Boord JB, Eden SK, MathenyME. Clinical inertia of discharge planningamong patients with poorly controlleddiabetesmellitus. J Clin EndocrinolMetab2012;97:2019–2026

53. Stolker JM, Spertus JA, McGuire DK, et al.Relationship between glycosylated he-moglobin assessment and glucose therapyintensification in patients with diabe-tes hospitalized for acute myocardial in-farction. Diabetes Care 2012;35:991–993

54. Wu EQ, Zhou S, Yu A, et al. Outcomesassociated with insulin therapy disruptionafter hospital discharge among patientswith type 2 diabetes mellitus who hadused insulin before and during hospitali-zation. Endocr Pract 2012;18:651–659

55. Robbins JM, Webb DA. Diagnosing di-abetes and preventing rehospitalizations:the urban diabetes study. Med Care 2006;44:292–296

1814 DIABETES CARE, VOLUME 36, JULY 2013 care.diabetesjournals.org

Inpatient management of hyperglycemia and diabetes