J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5

ª 2 0 1 5 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 2 2 1 3 - 1 7 7 9 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j c h f . 2 0 1 4 . 0 8 . 0 0 4

STATE-OF-THE-ART PAPER

Impact of Diabetes on Epidemiology,Treatment, and Outcomes of PatientsWith Heart Failure

Alessandra Dei Cas, MD, PHD,* Sadiya S. Khan, MD,y Javed Butler, MD, MPH,z Robert J. Mentz, MD,xRobert O. Bonow, MD, MS,k Angelo Avogaro, MD, PHD,{ Diethelm Tschoepe, MD,# Wolfram Doehner, MD, PHD,**Stephen J. Greene, MD,k Michele Senni, MD,yy Mihai Gheorghiade, MD,k Gregg C. Fonarow, MDzz

ABSTRACT

Fro

of

Ch

Me

Un

an

Re

Be

An

Tre

co

No

for

rel

Ma

The prevalence of patients with concomitant heart failure (HF) and diabetes mellitus (DM) continues to increase with the

general aging of the population. In patients with chronic HF, prevalence of DM is 24% compared with 40% in those

hospitalized with worsening HF. Patients with concomitant HF and DM have diverse pathophysiologic, metabolic, and

neurohormonal abnormalities that potentially contribute to worse outcomes than those without comorbid DM. In

addition, although stable HF outpatients with DM show responses that are similar to those of patients without DM

undergoing evidence-based therapies, it is unclear whether hospitalized HF patients with DM will respond similarly to

novel investigational therapies. These data support the need to re-evaluate the epidemiology, pathophysiology, and

therapy of HF patients with concomitant DM. This paper discusses the role of DM in HF patients and underscores the

potential need for the development of targeted therapies. (J Am Coll Cardiol HF 2015;3:136–45) © 2015 by the American

College of Cardiology Foundation.

H eart failure (HF) is a clinical manifestationof diverse cardiac and noncardiac abnor-malities and represents a heterogeneous

group of patients ranging from stable outpatientswith chronic HF to those with worsening symptomsrequiring hospitalization for HF. In hospitalizedpatients with HF, post-discharge readmission andmortality rates approach 15% and 30%, respectively,

m the *Unit of Diabetes and Prevention of Associated Diseases, Departme

Parma, Parma, Italy; yDivision of Cardiology, Department of Medicine, No

icago, Illinois; zCardiology Division, Stony Brook University, Stony Brook

dicine, Duke University Medical Center, Durham, North Carolina; kCeiversity Feinberg School of Medicine, Chicago, Illinois; {Department of M

d Diabetes Center North Rhine Westfalia, University Clinic of the Ruhr

search, Charite Universitätsmedizin, Berlin, Germany; yyCardiovascular Dergamo, Italy; and the zzAhmanson-UCLA Cardiomyopathy Center, Ronald

geles, Los Angeles, California. Dr. Butler is a consultant for Amgen, Baye

vena. Dr. Mentz has consulting relationships with Bristol-Myers Squibb,

nsultant for Bayer, Novartis, Takeda, Cardiocell, and Johnson & Johns

vartis, Medtronic, Amgen, Bayer, Boston Scientific, Johnson & Johnson, an

Healthcare Research and Quality and National Institutes of Health. A

ationships relevant to the contents of this paper to disclose.

nuscript received July 25, 2014; revised manuscript received August 18, 2

at 60 to 90 days, but no specific therapy has beenshown definitively to improve post-discharge read-mission and mortality rates. Furthermore, HF withconcomitant diabetes mellitus (DM) may have furtherincrease risk through different pathophysiologic,hemodynamic, and neurohormonal abnormalities.This is especially critical as approximately 24% ofHF patients overall and 40% of hospitalized HF

nt of Clinical and Experimental Medicine, University

rthwestern University Feinberg School of Medicine,

, New York; xDivision of Cardiology, Department of

nter for Cardiovascular Innovation, Northwestern

edicine, University of Padova, Padova, Italy; #Heart

University, Bochum, Germany; **Center for Stroke

partment, Azienda Ospedaliera Papa Giovanni XXIII,

Reagan-University of California Medical Center Los

r, Celladon, GE Healthcare, Medtronic, Takeda, and

GlaxoSmithKline, and Novartis. Dr. Gheorghiade is a

on. Dr. Fonarow has consulting relationships with

d The Medicines Co.; and is a consultant for Agency

ll other authors have reported that they have no

014, accepted August 22, 2014.

AB BR E V I A T I O N S

AND ACRONYM S

ACEI = angiotensin-converting

enzyme inhibitor

ARB = angiotensin II receptor

blocker

ATP = adenosine triphosphate

CV = cardiovascular

DM = diabetes mellitus

EF = ejection fraction

FFA = free fatty acid

HF = heart failure

LV = left ventricular

RAAS = renin-angiotensin-

aldosterone system

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5 Dei Cas et al.F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5 Heart Failure and Diabetes

137

patients have DM, and these figures are expected togrow exponentially in the next decades with thegrowth of an aging population. HF patients with DMmay represent a different pathophysiologic popula-tion than those without DM, which affects intracel-lular calcium release, myocardial lipid metabolism,and impaired endothelial cell function. HospitalizedHF patients with DM show an even worse prognosiswith increased rates of cardiovascular (CV) mortalityand HF hospitalization post-discharge (1).

Moreover, data for possible differential effectsof drugs in HF patients with or without DM areemerging from the aliskiren drug study in the sub-group analysis of the ASTRONAUT (Aliskiren Trial onAcute Heart Failure Outcomes) trial in patientsadmitted to the hospital with reduced ejection frac-tion (EF) (2) and from possible increased risk of HF forantidiabetic drugs, as in the SAVOR-TIMI (Saxagliptinand Cardiovascular Outcomes in Patients withType 2 Diabetes Mellitus-Thrombolysis In Myocar-dial Infarction) trial for the dipeptidyl peptidase(DPP)-4 saxagliptin (3). Although these resultsshould be interpreted with caution and viewed inthe context of a subgroup analysis of a secondaryendpoint with corresponding statistical limitations,the epidemiology, pathophysiology, prognosis, andmanagement of HF patients with DM should becarefully evaluated in an effort to improve theirprognosis and outcomes. This paper discusses therole of DM in HF patients and underscores the needfor development of targeted therapies.

EPIDEMIOLOGY

HF is the primary cause of more than 1 million hos-pitalizations in the United States annually (4). Hos-pitalization for HF is associated with unacceptablyhigh post-discharge mortality and re-hospitalizationrates (5), with figures that have remained largelyunchanged over the last 2 decades. CV mortality andreadmission rates for clinically stable patientsrecently discharged following a HF hospitalizationare approximately 25% at 6 months, and all-causemortality exceeds 30% at 1 year. Hospitalized HFpatients show heterogeneous clinical profiles in termsof HF cause and pathophysiology and both CV andnon-CV comorbidities (6).

Approximately 40% of hospitalized HF patientswith low EF have DM (1). Notably, DM in hospi-talized patients is associated with worse prognosis(1), increased risk for combined CV mortality andHF-related hospitalization (7), and longer hospitalstay (8), despite receiving care that is similar tothat for patients without DM (1,9) although this

paradigm is not a consistent finding (10),particularly in women (11).

In a pre-specified subgroup analysis of theASTRONAUT study, patients hospitalizedfor HF with DM were older and had highersystolic blood pressure, more frequentischemic HF etiology, renal impairment, andgreater likelihood of receiving treatmentwith an angiotensin II receptor blocker (ARB)drug (2).

In patients with DM, the prevalence of HFis greater than that of the general population(12), and a 1% increase in glycosylatedhemoglobin in DM patients is associated withan 8% increased risk of HF (13). Patients withHF show marked insulin resistance (14),

which increases their risk of developing type 2 DMcompared to that of normal individuals or patientswith coronary artery disease. Similarly, insulin resis-tance in HF patients, even in the absence of overt DM,is an independent predictor of worse prognosis, sug-gesting a pathophysiologic involvement of insulinresistance in HF progression (14). Thus, the relation-ship between DM and HF is bidirectional, with eachdisease independently increasing the risk for theother.

PATHOPHYSIOLOGY OF HF IN DIABETES

Insulin resistance and hyperglycemia are the centralmetabolic perturbations that accompany type 2 DM,driving several adaptive and maladaptive cellularresponses that lead to specific changes in myocardialstructure and function (diabetic cardiomyopathy)(Figure 1). Longstanding metabolic and functionalalterations ultimately lead to irreversible structuralchanges. At this later stage, comorbidities seencommonly in patients with DM, such as hypertension,dyslipidemia, microvascular dysfunction, autonomicdysfunction, and renal impairment may accele-rate the progression of cardiac dysfunction towardadvanced disease (6).

METABOLIC ALTERATIONS. One consequence ofinsulin resistance is enhanced release of free fattyacid (FFA) and reduction in myocardial glucosetransporter expression and glucose uptake. Thischange in substrate availability leads to an imbalanceof glucose and FFA uptake and use in the heart. Asthese metabolic alterations become longstanding,high FFA levels activate myocyte expression ofperoxisome proliferator-activated receptor-a thatstimulates transcription of multiple genes respon-sible for an increase in mitochondrial FFA transportand oxidation (15). FFA b-oxidation is less favorable

FIGURE 1 Pathophysiologic Abnormalities in Diabetes and Heart Failure

Multiple pathophysiologic mechanisms are activated and modulated by the presence of diabetes mellitus, resulting in abnormalities in the myocardial substrate unique to

patients with diabetes mellitus and heart failure. Adapted with permission from Falcao-Pires I, Leite-Moreira AF. Heart Fail Rev 2012;17:325–44. AGE ¼ advanced

glycation end product; FFA ¼ free fatty acid; GLUT ¼ glucose transporter facilitators; PDH ¼ pyruvate dehydrogenase; PKC ¼ protein kinase C; PPAR ¼ peroxisome

proliferator-activated receptor; RAAS ¼ renin-angiotensin-aldosterone system; REDOX ¼ oxidation and reduction; ROS ¼ reactive oxygen species; TG ¼ triglyceride;

VGEF ¼ vascular endothelial growth factor.

Dei Cas et al. J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5

Heart Failure and Diabetes F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5

138

under stress conditions due to significantly higheroxygen consumption and reduction in cardiac effi-ciency. FFA myocardial uptake may exceed FFAb-oxidation capacity, leading to triglyceride accumu-lation within myocytes and epicardial tissue andproduction of toxic lipid intermediates such as diac-ylglycerol and ceramides, both of which promoteoxidative stress and cardiomyocyte apoptosis.Epicardial fat tissue is therefore expanded in thecontext of insulin resistance and DM, resulting in anincreased paracrine secretion of proinflammatorycytokines and macrophage infiltration, with detri-mental consequences for myocardial and vasculaturefunctions (16).

Notably, the failing myocardium without insulinresistance or DM activates an adaptive cardio-protective metabolic response by enhancing glucoseuptake and reducing FFA oxidation. This reduction inFFA and glucose oxidation balance improvesmyocardial contractile efficiency by increasing theATP production-to-oxygen consumption ratio. Thesechanges are not due to a lack of substrate availability

but rather to a maladaptive change in substrate fluxthrough enhanced glucose uptake and activation ofenzymes involved in the glycolytic pathway (17). AsHF progresses to its advanced stages, compensatoryadrenergic drive leads to increased lipolysis and FFAconcentrations with a consequent shift in cardiacmetabolic profile toward that resembling those withinsulin resistance/DM (18).

The exaggerated FFA uptake and use in HF withDM ultimately leads to dysfunction of myocardialmitochondria. Increased FFA oxidation in the insulin-resistant myocardium increases generation of reactiveoxygen species, thereby promoting mitochondrialuncoupling and leading to increased oxygen con-sumption and reduced myocardial efficiency.FUNCTIONAL ALTERATIONS. DM is associated withabnormalities in contractile and regulatory proteinexpression and in cardiomyocyte Ca2þ sensitivity.Major changes in DM include a shift in myosinisoenzyme composition (from V1 to V3 isoforms) andthe predominance of the fetal b myosin heavy chainexpression with respect to a-myosin heavy chain (19),

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5 Dei Cas et al.F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5 Heart Failure and Diabetes

139

leading to depressed ATPase activity of myofibrilsand reduced contractile force. In addition, sarco-plasmic (endoplasmic) reticulum Ca2þ ATPase(SERCA) 2 activity has been shown to be inactivatedby oxidative stress in cardiomyocytes incubated inhigh-glucose medium (20) with subsequent ineffi-cient sequestration of Ca2þ in the sarcoplasmaticreticulum, resulting in Ca2þ overload in the cytosoland impaired relaxation. These perturbations in Ca2þ

homeostasis are associated with myofibrillar remod-eling and have been correlated with the clinicalfinding of diastolic dysfunction (Figure 2).

HYPERGLYCEMIA-INDUCED STRUCTURAL ALTER-

ATIONS. Hyperglycemia is the main pathogenicmechanism in HF and DM that leads to structuralalterations. Four main hypotheses and pathologicmechanisms have been proposed to explain howhyperglycemia can cause all of the diabetes-relatedcomplications, which include the formation ofadvanced glycation endproducts resulting from non-enzymatic glycation and oxidation of proteins andlipids; activation of the protein kinase C/diacylglycerol

FIGURE 2 Echocardiographic Abnormalities in Diabetes and Heart F

An example from a diabetic patient with a high global contrast-enhanced

a corresponding global longitudinal strain of �19.0% (left) with norma

permission from Ng AC, et al. Circ Cardiovasc Imaging 2012;5:51–9.

signaling pathway; increased levels of poly(ADP-ribose) polymerase enzymes that are involved incellular processes, including DNA repair and pro-grammed cell death; and the oxidative stressburden from both mitochondrial and nonmito-chondrial sources. Structural modifications resultingfrom glycation are likely to be the consequencesof the altered cardiac metabolism found in type 2DM, with reactive metabolites like methylglyoxalbeing the most prominent actor in advancedglycation (21).

MICROVASCULAR DYSFUNCTION. Hyperglycemia isknown to induce diabetes-specific changes in micro-vascular architecture, mainly through advancedglycation end-product formation, and is character-ized by microvascular remodeling with thickening ofthe capillary basement membrane and formation ofmicroaneurysms. These structural alterations causefunctional modification such as impaired nitric oxideproduction with consequent endothelial dysfunction,decreased vessel density, and increased permeability.The consequent deficiency in coronary blood flow

ailure

myocardial T1 time (508 ms), suggesting less interstitial fibrosis and

l left ventricular ejection fraction of 60% for both. Adapted with

Dei Cas et al. J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5

Heart Failure and Diabetes F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5

140

reserve and myocardial ischemia contribute to loss ofcontractile proteins and myocyte necrosis with reac-tive focal perivascular and interstitial fibrosis,collagen deposition, and hypertrophy of myocardialcells (Figure 3).

CARDIAC AUTONOMIC NEUROPATHY. Insulin-resis-tance is associated with activation of the sympatheticnervous system. An early manifestation of cardiacautonomic neuropathy is parasympathetic denerva-tion with an imbalance toward higher relative sym-pathetic drive, which increases lipolysis, FFAoverflow that influences use of myocardial substrate,mitochondrial uncoupling, and oxidative stress. Pa-tients with severe cardiac autonomic neuropathy mayhave distal sympathetic denervation associated withproximal ventricular islands of hyperinnervation thatmay produce myocardial regions with electricalinstability. The estimated 8-year survival rate inpatients with cardiac autonomic neuropathy is 77%compared with 97% in those with normal autonomicfunction, in which most deaths are sudden andunexpected (22). HF is also characterized by over-activity of the sympathetic nervous system, whichleads to inappropriate sodium retention, renin stim-ulation, and diminished renal function through renalartery constriction.

NEUROHORMONAL ABNORMALITIES. The renin-angiotensin-aldosterone system (RAAS) is activatedearly in DM with overproduction of angiotensin II,favoring cardiac fibrosis by increases in synthesis ofcollagen, myocyte apoptosis, vascular inflammation,and oxidative damage. Activation of the sympatheticnervous system and RAAS increases as HF progresses,exacerbated by eventual aldosterone breakthroughthat results from angiotensin-converting enzymeinhibitor (ACE-I) or ARB therapy and high-dose loopdiuretics, with worsening CV and renal function.

FIGURE 3 Immunohistochemical Abnormalities in Diabetes and Hea

Immunohistochemical study of the biopsied myocardium in a patient wi

mulation of collagen with perimysial fibrosis (A), endomysial fibrosis (B

method. Adapted with permission from Shimizu M, et al. J Clin Pathol 1

In patients with congestion associated with wors-ening HF, hemodynamic changes (increased renalvein pressure and/or reduced renal perfusion); sym-pathetic and RAAS hyperactivity and consequentstimulation of the inflammatory pathways; andfibrosis, oxidative stress, and endothelial dysfunctionare major determinants of cardiorenal syndrome (23).Many drugs used for acute HF treatment, particularlydiuretics, may affect renal function and induceincreased sympathetic and RAAS activation, which inturn may decrease glomerular function. Deteriora-tion in renal function may limit administration ofthe evidence-based guideline therapies known toimprove survival, such as ACE-I and mineralocorti-coid receptor antagonists, further exacerbatingneurohormonal activation and HF progression.

RENAL INSUFFICIENCY. Renal function on admissionis independently associated with adverse outcomesin hospitalized HF patients. DM is strongly associatedwith the risk of worsening renal function in thesetting of decompensated HF (24). The underlyingrenal substrate, diabetic nephropathy seems toworsen kidney dysfunction in decompensated HF.Two aspects of diabetic nephropathy, albuminuriaand reduced glomerular filtration rate, reflect differ-ences in renal pathology that define the 2 chronickidney disease phenotypes (albuminuric and non-albuminuric). This heterogeneity was specificallyinvestigated in a study conducted in a large cohortof patients with type 2 DM which suggested thatmicroalbuminuria is an expression of glomerulardamage and a reliable biomarker of renal microvas-cular disease, whereas glomerular filtration rate is afunctional parameter inversely related with indicesof intrarenal as well as systemic atherosclerosis(25). Thus, nonalbuminuric renal impairment, as amanifestation of prevailing renal macrovascular

rt Failure

th diabetes demonstrates interstitial fibrosis with substantial accu-

), and perivascular fibrosis (C) stained with Mallory’s AZAN trichrome

993;46:32–6.

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5 Dei Cas et al.F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5 Heart Failure and Diabetes

141

involvement, was frequently associated with coro-nary atherosclerosis compared with the albuminuricforms, which is an indicator of microangiopathy.

Renal involvement in type 1 DM is different fromthat in type 2. In type 1 DM, functional and structuralchanges appear to be induced and sustained primarilyby direct and indirect effects of hyperglycemia. In type2 DM, functional and structural changes are moreheterogeneous because of different interactionsamong aging, insulin resistance, hypertension and,potentially, obesity and dyslipidemia (“fatty kidney”).A recent study demonstrated that nearly 70% of pa-tients presenting with acute HF had abnormal albuminexcretion rate on admission, which decreased to 40%after 7 days of treatment, similar to that observed inchronic HF patients (26). Whether the pathogenesis ofincreased albumin excretion in HF is the same as DM isunclear. The same pathophysiologic mechanismsinvolved in DM, such as vascular dysfunction, sys-temic inflammation, and neurohumoral activation, arehypothesized to also be operative in HF.

RAAS inhibition exerts specific effects in thevasculature and kidney, specifically, decreasedintraglomerular pressure and improved glomerularbarrier size selectivity, which eventually result inreduction of proteinuria beyond reduction in bloodpressure. Drugs that inhibit the RAAS and reducealbuminuria may prevent the onset and worsening ofdiabetic nephropathy, HF, and CV disease.

HEART FAILURE TREATMENT IN

PATIENTS WITH CONCOMITANT DM

Outpatients with stable HF have responses similar tothose from patients receiving evidence-based therapyfor HF with or without concomitant DM (Table 1).Conversely, patients who are hospitalized for decom-pensated HF remain at significant risk for adverseoutcomes despite receiving guideline-recommendedtreatments, highlighting the need for novel therapies(27). Results from subgroup analyses suggest thathospitalized HF patients with DM may be prone todevelop side effects or have responses to novel ther-apies that diverge from those without DM. These datamay be explained by the underlying pathophysiologicdifferences in HF patients with concomitant DM interms of severity of microvascular and end-organcomplications, comorbidities, and degree of neuro-hormonal activation. Also, drug interactions betweentherapies for DM and HF may blunt clinical efficacy,increase the risk of adverse effects, or promote wors-ening HF.

BETA-ADRENERGIC BLOCKING AGENTS. Adminis-tration of b-blockers to patients with DM has been

questioned given concerns that b-blockers mayworsen insulin sensitivity and blunt symptoms ofhypoglycemia. As a result, prescription rates ofb-blockers was historically very low (approximately25%) in patients with DM, despite evidence ofthe overall benefits of b-blockers in HF for morbidityand mortality. In 2003, Haas et al. (28) reportedresults of a meta-analysis of large-scale b-blockertrials, providing clear evidence of beneficial responseof b-blockers in patients with DM and HF, althoughthe magnitude of that benefit was lower in subjectswith DM.

In contrast, there is now a large body of data sup-porting blockade of the sympathetic nervous systemsin HF patients with and without DM (29–32). COMET(Carvedilol or Metoprolol European Trial) suggestedthat the combined a- and b-blocker with carvedilolwas more beneficial than the selective b-1 antagonistmetoprolol in reducing mortality in HF patients withconcomitant DM (33). Importantly, administration ofthe b-selective agents metoprolol and bisoprolol hasnot been associated with any increase in the inci-dence of side effects in patients with DM compared topatients without (30,32). Therapy with b-blockersshould therefore be prescribed to HF patients withDM (34), unless contraindicated by no clear evidenceof efficacy of 1 evidence-based b-blocker over anotherin this population (34).RAAS INHIBITION. ACE-I shows effects in HF pa-tients with DM that are similar to those in patientswithout DM (35–39). Similar effects are achieved withARB therapy (40), although in some studies, ARBswere not as effective in the DM subgroup (41,42).Although subgroup analyses from the SOLVD (Studiesof Left Ventricular Dysfunction) and SAVE (Survivaland Ventricular Enlargement) trials did not showsignificant effects with ACE-I in patients with DM,this was likely related to the small sample size thatwas not powered to detect this effect (35,36). Guide-lines recommend that treatment with ACE-I/ARB inDM patients should be initiated at low doses, withgradual titration to doses used in clinical trials or tomaximally tolerated doses with frequent monitoringof renal function and electrolytes (34). Patients withcomorbid DM and HF may also benefit from theaddition of mineralocorticoid receptor antagonisttherapy to ACE-I/ARBs (43,44), although their use incombination with ACE-I and/or ARBs has been limitedby concerns of worsening renal function and hyper-kalemia (45).

Aliskiren is a first-in-class direct renin inhibitorthat blocks the RAAS at the most proximal step,offering the theoretical advantage of preventing thecompensatory rise in plasma renin activity that can

TABLE 1 Major Clinical Trials in Patients With Heart Failure and Subgroup Analyses With Concomitant Diabetes Mellitus

Trial Primary OutcomeSubgroup of Patients With

or Without DMAverage Follow-Up

(Months) TreatmentResults

RR or HR (95% CI) (Ref. #)

SOLVD Mortality andhospitalization forworsening HF

HF patients withEF <35%

41.4 Enalapril vs. placebo RR of 0.84 (0.74–0.95) in noDM vs. 1.01 (0.85–1.21)in DM

(35)

SAVE CV mortality Recent survivors of MIwith EF #40%

42 Captopril vs. placebo RR of 0.82 (0.68–0.99) in noDM vs. 0.89 (0.68–1.16)in DM

(36)

TRACE All-cause mortality Patients with LVEF <35%after acute MI

26 Trandalapril vs. placebo RR of 0.82 (0.69–0.97) in noDM vs. 0.64 (0.45–0.91)in DM (interaction analysisp ¼ 0.3)

(37)

SMILE Progression to HF Patients with anterioracute MI not eligiblefor thrombolytictreatment

12 Zofenopril vs. placebo RR of 0.79 (0.54–1.15) in noDM vs. 0.44 (0.22–0.87)in DM

(38)

ATLAS All-cause mortality High-risk HF patients 46 Lisinopril high dose(32.5–35 mg day�1)vs. low-dose (2.5–5mg day�1)

Y6% RR in no DM patients vs.Y14% RR in DM high-dosevs. low dose (interactionanalysis p ¼ 0.3)

(39)

CHARM CV death orhospitalization for HF

NYHA functional classII–IV patients(LVEF <40%)

40 Candesartan vs. placebo RR not significant in no DM vs.that in DM (interactionanalysis p ¼ 0.12)

(41)

Val-HeFT Combined endpoint of CVmortality andmorbidity

NYHA functional classII–IV patients

23 Valsartan vs. placebo YRR in no DM vs. that in DM(although not significant)

(42)

EPHESUS Mortality and CVmorbidity

Postacute MI patientswith LVEF #40% andclinical HF

16 Eplerenone vs. placebo YRR in no DM vs. that in DM(although not significant;p ¼ 0.59)

(43)

Subgroupanalysis

All-cause mortality NYHA functional classIII–IV patients withLVEF <35%

24 Spironolactone vs.placebo

RR of 0.70 (0.60–0.82) innon-DM vs. HR of0.70 (0.52–0.94) in DM

(44)

BEST All-cause mortality NYHA functional classIII–IV patients withEF #35%

24 Bucindolol vs. placebo RR of 0.91 (0.74–1.13) in noDM vs. HR of 0.87(0.73–1.03) in DM

(29)

MERIT-HF Risk of hospitalizationfor HF

NYHA functional classIII–IV patients withEF <40%

12 Metoprolol vs. placebo Y37% RR in DM vs. Y35% inno DM; test of diabetes bytreatment interaction wasnonsignificant

(30)

CIBIS II All-cause mortality NYHA functional classIII–IV patients withEF #35%

16 Bisoprolol vs. placebo RR 0.66 (95% CI 0.54–0.81)in no DM vs. RR 0.81 (95%CI 0.51–1.28) in DM(heterogeneity test forinteraction p ¼ 0.48)

(32)

COPERNICUS Combined endpointAll-cause mortality orhospitalization for HF

NYHA functional class IVpatients withEF <25%

10.4 Carvedilol vs. placebo RR of 0.67 (0.52–0.85) in noDM vs. 0.68 (0.47–1.00)in DM

(31)

ASTRONAUT CV death or HFrehospitalization

Hemodynamically stablepatients hospitalizedfor HF

11.3 Aliskiren vs. placebo HR of 0.80 (0.64–0.99) in noDM vs. 1.16 (0.91–1.47) inDM; test for interactionp ¼ 0.03

(2)

CI ¼ confidence interval; CV ¼ cardiovascular; DM ¼ diabetes mellitus; EF ¼ ejection fraction; HF ¼ heart failure; HR ¼ hazard ratio; LVEF ¼ left ventricular ejection fraction; MI ¼ myocardial infarction;NYHA ¼ New York Heart Association functional classification; RR ¼ relative risk.

Dei Cas et al. J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5

Heart Failure and Diabetes F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5

142

mediate aldosterone escape in the setting of otherRAAS inhibitors. The effects of aliskiren in type 2 DMwith proteinuria or CV disease were investigated inthe ALTITUDE (Aliskiren Trial in Type 2 DiabetesUsing Cardiorenal Endpoints) study (46), which wasstopped early because of an excess risk of renaldysfunction, hyperkalemia (8.8% vs. 5.6%), hypo-tension (12.1% vs. 8.0%), and stroke. The subsequentASTRONAUT trial with aliskiren included patientshospitalized for HF and 41% with comorbid DM (2).

Pre-specified subgroup analysis by DM status sug-gested an increased risk of all-cause mortality within12 months in DM patients receiving aliskiren incontrast with data for the non-DM cohort. Oneexplanation for the possible negative effects of alis-kiren in DM patients includes increased adverseevents including severe hyperkalemia (serum potas-sium of $6.0 mmol/l; 9.7% vs. 4.7% in DM with alis-kiren vs. placebo, respectively) with concomitantACE-I/ARB (85%) and mineralocorticoid receptor

FIGURE 4 Heart Failure Rates in Diabetes Clinical Trials

Differences in overall heart failure (HF) event rates are shown in contemporary

clinical diabetes trials.

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5 Dei Cas et al.F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5 Heart Failure and Diabetes

143

antagonist (55%) therapy similar to the excess riskseen in the ALTITUDE trial. A second possible mech-anism is the differential effect on neurohormonalprofiles in patients with diabetes. Although N-termi-nal pro–B-type natriuretic peptide and plasma aldo-sterone levels were reduced in non-DM patientsreceiving aliskiren at 6 months, there were no dif-ferences in these biomarker levels in patients withDM, suggesting differential effects in the RAAScascade (46). These results should be interpreted withcaution and viewed in the context of a subgroupanalysis of a secondary endpoint, with the corre-sponding statistical limitations including a smallernumber of patients per treatment group and issueswith multiplicity. Further dedicated trials in patientswith DM and HF are needed.

DIURESIS. The unfavorable association of diuretictherapy with impaired glucose metabolism is wellestablished with both thiazide and loop diuretics.There is a dose-dependent relationship with use ofthiazide diuretics and a less prominent effect withloop diuretics. Diuretics are, however, unavoidable inthe treatment of HF patients hospitalized with vol-ume overload for symptomatic relief of congestion, aswell as, in many patients with chronic HF, for main-taining a stable compensated euvolemic state. Nodata are available to indicate a possible differentialefficacy in patients with or without DM.

HEART FAILURE EVENT RATES IN

LANDMARK TRIALS OF DIABETES THERAPY

Improvement in glycemic control enhances myocar-dial contractility with improved metabolic control,likely due to more efficient use of myocardial energysubstrate and improved microvascular perfusion (47).In the ORIGIN (Outcome Reduction with InitialGlargine Intervention) trial, the basal insulin glarginewas associated with a nonsignificant trend toward areduction in hospitalizations for HF compared to thatwith standard care (48). However, results of previoustrials challenge this hypothesis (49,50), as strict gly-cemic control was not associated with change in therate of onset of HF. Conversely, recent trials haveshown increased HF event rates with thiazolidine-diones. The use of rosiglitazone (51) and pioglitazone(52) has been associated with a more than 2-fold risk ofHF, including HF hospitalizations due to fluid reten-tion and peripheral edema, even in patients withoutpre-existing LV systolic dysfunction and the riskincreased further in patients with a history of HF.

Recently, the DM agent saxagliptin, a DPP-4inhibitor, was also associated with an increased riskof HF, despite significantly improved glycemic

control and reduction in development and progres-sion of microalbuminuria (3). Although this findingshould be considered within the context of theheterogeneity of a large clinical trial with HF as 1 ofseveral secondary endpoints, it merits further inves-tigation regarding the risk of HF with DM therapiesand highlights the need to include HF and HF hospi-talizations as endpoints in DM trials (Figure 4). On thecontrary, the DPP-4 alogliptin showed no significantexcess of HF (53). A recent meta-analysis showed thatthe overall risk of acute HF was higher in patientstreated with DPP-4 inhibitors than in those treatedwith placebo or active comparators (Mantel-Haenszelodds ratio [MH–OR] 1.19 [95% confidence interval:1.03 to 1.37]; p < 0.05) without any clear evidence ofdifferences among drugs of the same class (54). Thepossible mechanisms are unclear, but it is importantto note that BNP is a substrate for DPP-4 inhibitors.Ongoing trials, including EXSCEL (Exenatide Study ofCardiovascular Event Lowering) and TECOS (Sita-gliptin Cardiovascular Outcome Study), whichinclude pre-specified endpoints of HF hospitaliza-tion, will add knowledge to these previous findings.In contrast, studies are underway to investigate GLP-1agonists to improve HF (e.g., FIGHT [FunctionalImpact of GLP-1 for Heart Failure Treatment]).

CONCLUSIONS

HF in the setting of DM may have a myocardial sub-strate that is characterized by specific myocardial

Dei Cas et al. J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5

Heart Failure and Diabetes F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5

144

metabolic, functional, neurohormonal, and hemody-namic perturbations. The prevalence of comorbid DMexceeds 40% in patients with HF hospitalizations,highlighting the need for targeted therapies to addressHF and DM in this high-risk subset. Post-dischargeevent rates after a hospitalization for HF remain highand have not changed in recent decades, and no spe-cific therapies definitively proven to improve out-comes remain. The broad phenotypic heterogeneity ofpatients hospitalized for HF necessitates that futuredrug development programs aim to match novel ther-apies with the specific patient subgroup most likely

to benefit. Patients with comorbid HF and DM repre-sent a growing patient population, and future clinicaltrials in both HF and DM are encouraged to refocusand differentially investigate the efficacy, safety, andoutcomes in HF patients with concomitant DM.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Gregg C. Fonarow, Ahmanson-UCLA CardiomyopathyCenter, Ronald Reagan UCLA Medical Center, 10833LeConte Avenue, Room 47-123 CHS, Los Angeles,California 90095-1679. E-mail: gfonarow@mednet.ucla.edu.

RE F E RENCE S

1. Sarma S, Mentz RJ, Kwasny MJ, et al. Associa-tion between diabetes mellitus and post-dischargeoutcomes in patients hospitalized with heart fail-ure: findings from the EVEREST trial. Eur J HeartFail 2013;15:194–202.

2. Gheorghiade M, BöhmM, Greene SJ, et al. Effectof aliskiren on postdischarge mortality and heartfailure readmissions among patients hospitalizedfor heart failure: the ASTRONAUT randomized trial.JAMA 2013;309:1125–35.

3. Scirica BM, Bhatt DL, Braunwald E, et al.,SAVOR-TIMI 53 Steering Committee andInvestigators. Saxagliptin and cardiovascular out-comes in patients with type 2 diabetes mellitus.N Engl J Med 2013;369:1317–26.

4. Blecker S, Paul M, Taksler G, Ogedegbe G,Katz S. Heart failure associated hospitalizations inthe United States. J Am Coll Cardiol 2013;61:1259–67.

5. Gheorghiade M, Vaduganathan M, Fonarow GC,Bonow RO. Rehospitalization for heart failure:problems and perspectives. J Am Coll Cardiol2013;61:391–403.

6. Metra M, Zacà V, Parati G, et al. Cardiovascularand noncardiovascular comorbidities in patientswith chronic heart failure. Heart Failure StudyGroup of the Italian Society of Cardiology.J Cardiovasc Med (Hagerstown) 2011;12:76–84.

7. Nieminen MS, Brutsaert D, Dickstein K, et al.EuroHeart Failure Survey II (EHFS II): a survey onhospitalized acute heart failure patients: descrip-tion of population. Eur Heart J 2006;27:2725–36.

8. Greenberg BH, Abraham WT, Albert NM, et al.Influence of diabetes on characteristics and out-comes in patients hospitalized with heart failure: areport from the Organized Program to InitiateLifesaving Treatment in Hospitalized Patients withHeart Failure (OPTIMIZE-HF). Am Heart J 2007;154:e271–8.

9. Kapoor JR, Fonarow GC, Zhao X, Kapoor R,Hernandez AF, Heidenreich PA. Diabetes, qualityof care, and in-hospital outcomes in patientshospitalized with heart failure. Am Heart J 2011;162:480–486.e3.

10. MacDonald MR, Petrie MC, Varyani F, et al.Impact of diabetes on outcomes in patients withlow and preserved ejection fraction heart failure:

an analysis of the Candesartan in Heart failure:Assessment of Reduction in Mortality andmorbidity (CHARM) programme. Eur Heart J2008;29:1377–85.

11. Gustafsson I, Brendorp B, Seibaek M, et al.Influence of diabetes and diabetes-gender inter-action on the risk of death in patients hospitalizedwith congestive heart failure. J Am Coll Cardiol2004;43:771–7.

12. Nichols GA, Gullion CM, Koro CE, Ephross SA,Brown JB. The incidence of congestive heartfailure in type 2 diabetes: an update. Diabetes Care2004;27:1879–84.

13. Iribarren C, Karter AJ, Go AS, et al. Glycemiccontrol and heart failure among adult patientswith diabetes. Circulation 2001;103:2668–73.

14. Doehner W, Rauchhaus M, Ponikowski P, et al.Impaired insulin sensitivity as an independent riskfactor for mortality in patient with stable chronicheart failure. J Am Coll Cardiol 2005;46:1019–26.

15. Lopaschuk GD, Ussher JR, Folmes CD,Jaswal JS, Stanley WC. Myocardial fatty acidmetabolism in health and disease. Physiol Rev2010;90:207–58.

16. Cherian S, Lopaschuk GD, Carvalho E. Cellularcross-talk between epicardial adipose tissue andmyocardium in relation to the pathogenesis ofcardiovascular disease. Am J Physiol EndocrinolMetab 2012;303:E937–49.

17. Rajabi M, Kassiotis C, Razeghi P, Taegtmeyer H.Return to the fetal gene program protects thestressed heart: a strong hypothesis. Heart Fail Rev2007;12:331–43.

18. Metzler B, Schocke MF, Steinboeck P, et al.Decreased high-energy phosphate ratios in themyocardium of men with diabetes mellitus type I.J Cardiovasc Magn Reson 2002;4:493–502.

19. Malhotra A, Sanghi V. Regulation of contractileproteins in diabetic heart. Cardiovasc Res 1997;34:34–40.

20. Teshima Y, Takahashi N, Saikawa T, et al.Diminished expression of sarcoplasmic reticulumCa(2þ)-ATPase and ryanodine sensitive Ca(2þ)channel mRNA in streptozotocin-induceddiabetic rat heart. J Mol Cell Cardiol 2000;32:655–64.

21. Gawlowski T, Stratmann B, Stork I, et al. Heatshock protein 27 modification is increased in thehuman diabetic failing heart. Horm Metab Res2009;41:594–9.

22. Givertz MM, Sawyer DB, Colucci WS. Antioxi-dants and myocardial contractility: illuminatingthe “Dark Side” of beta-adrenergic receptoractivation? Circulation 2001;103:782–3.

23. Metra M, Cotter G, Gheorghiade M, Dei Cas L,Voors AA. The role of the kidney in heart failure.Eur Heart J 2012;33:2135–42.

24. Forman DE, Butler J, Wang Y, et al. Incidence,predictors at admission, and impact of worseningrenal function among patients hospitalised withheart failure. J Am Coll Cardiol 2004;43:61–7.

25. Solini A, Penno G, Bonora E, et al. RenalInsufficiency And Cardiovascular Events (RIACE)Study Group. Diverging association of reducedglomerular filtration rate and albuminuria withcoronary and noncoronary events in patients withtype 2 diabetes: the Renal Insufficiency andCardiovascular Events (RIACE) Italian multicenterstudy. Diabetes Care 2012;35:143–9.

26. Jackson CE, Solomon SD, Gerstein HC, et al.,CHARM Investigators and Committees. Albumin-uria in chronic heart failure: prevalence andprognostic importance. Lancet 2009;374:543–50.

27. Butler J, Fonarow GC, Gheorghiade M. Needfor increased awareness and evidence-basedtherapies for patients hospitalized for heartfailure. JAMA 2013;310:2035–6.

28. Haas SJ, Vos T, Gilbert RE, et al. Are beta-blockers as efficacious in patients with diabetesmellitus as in patients without diabetes mellituswho have chronic heart failure? A meta-analysis oflarge-scale clinical trials. Am Heart J 2003;146:848–53.

29. Domanski M, Krause-Steinrauf H,Deedwania P, et al. The effect of diabetes onoutcomes of patients with advanced heart failurein the BEST trial. J Am Coll Cardiol 2003;42:914–22.

30. Deedwania PC, Giles TD, Klibaner M, et al.Efficacy, safety and tolerability of metoprololCR/XL in patients with diabetes and chronic heartfailure: experiences from MERIT-HF. Am Heart J2005;149:159–67.

J A C C : H E A R T F A I L U R E V O L . 3 , N O . 2 , 2 0 1 5 Dei Cas et al.F E B R U A R Y 2 0 1 5 : 1 3 6 – 4 5 Heart Failure and Diabetes

145

31. Packer M, Fowler MB, Roecker EB, et al. Effectof carvedilol on the morbidity of patients withsevere chronic heart failure: results of the carve-dilol prospective randomized cumulative sur-vival (COPERNICUS) study. Circulation 2002;106:2194–9.

32. Erdmann E, Lechat P, Verkenne P, Wiemann H.Results from post-hoc analyses of the CIBIS II trial:effect of bisoprolol in high-risk patient groupswith chronic heart failure. Eur J Heart Fail 2001;3:469–79.

33. Torp-Pedersen C, Metra M, Charlesworth A,et al. Effects of metoprolol and carvedilol on pre-existing and new onset diabetes in patients withchronic heart failure: data from the Carvedilol OrMetoprolol European Trial (COMET). Heart 2007;93:968–73.

34. Yancy CW, Jessup M, Bozkurt B, et al. 2013ACCF/AHA guideline for the management of heartfailure: executive summary: a report of theAmerican College of Cardiology Foundation/American Heart Association Task Force on practiceguidelines. J Am Coll Cardiol 2013;62:1495–539.

35. Effect of enalapril on survival in patients withreduced left ventricular ejection fractions andcongestive heart failure. The SOLVD Investigators.N Engl J Med 1991;325:293–302.

36. Moyé LA, Pfeffer MA, Wun CC. Uniformityof captopril benefit in the SAVE study: subgroupanalysis. Survival and Ventricular EnlargementStudy. Eur Heart J 1994;15 Suppl B:2–8;26–30.

37. Gustafsson I, Torp-Pedersen C, Køber L,Gustafsson F, Hildebrandt P. Effect of theangiotensin-converting enzyme inhibitor trando-lapril on mortality and morbidity in diabetic pa-tients with left ventricular dysfunction after acutemyocardial infarction. Trace Study Group. J AmColl Cardiol 1999;34:83–9.

38. Borghi C, Ambrosioni E, Magnani B. Effects ofthe early administration of zofenopril on onset andprogression of congestive heart failure in patientswith anterior wall acute myocardial infarction. The

SMILE Study Investigators. Survival of MyocardialInfarction Long-term Evaluation. Am J Cardiol1996;78:317–22.

39. Ryden L, Armstrong PW, Cleland JG, et al.Efficacy and safety of high-dose lisinopril inchronic heart failure patients at high cardiovas-cular risk, including those with diabetes mellitus.Results from the ATLAS trial. Eur Heart J 2000;21:1967–78.

40. Carr AA, Kowey PR, Devereux RB, et al.Hospitalizations for new heart failure amongsubjects with diabetes mellitus in the RENAAL andLIFE studies. Am J Cardiol 2005;96:1530–6.

41. Young JB, Dunlap ME, Pfeffer MA, et al.Mortality and morbidity reduction with cande-sartan in patients with chronic heart failure andleft ventricular systolic dysfunction: results of theCHARM low-left ventricular ejection fraction trials.Circulation 2004;110:2618–26.

42. Cohn JN, Tognoni G. A randomized trial ofthe angiotensin-receptor blocker valsartan inchronic heart failure. N Engl J Med 2001;345:1667–75.

43. Pitt B, Remme W, Zannad F, et al. Eplerenone,a selective aldosterone blocker, in patients withleft ventricular dysfunction after myocardialinfarction. N Engl J Med 2003;348:1309–21.

44. Pitt B, Zannad F, Remme WJ, et al. The effectof spironolactone on morbidity and mortality inpatients with severe heart failure. RandomizedAldactone Evaluation study investigators. N Engl JMed 1999;341:709–17.

45. Parving HH, Brenner BM, McMurray JJ, et al.Cardiorenal end points in a trial of aliskirenfor type 2 diabetes. N Engl J Med 2012;367:2204–13.

46. Maggioni AP, Greene SJ, Fonarow GC, et al.Effect of aliskiren on post-discharge outcomesamong diabetic and non-diabetic patients hospi-talized for heart failure: insights from theASTRONAUT trial. Eur Heart J 2013;34:3117–27.

47. von Bibra H, Hansen A, Dounis V, Bystedt T,Malmberg K, Rydén L. Augmented metaboliccontrol improves myocardial diastolic function andperfusion in patients with non-insulin dependentdiabetes. Heart 2004;90:1483–4.

48. The ORIGIN Trial Investigators. Basal insulinand cardiovascular and other outcomes in dys-glycemia. N Engl J Med 2012;367:319–28.

49. Patel A, MacMahon S, Chalmers J, et al.Intensive blood glucose control and vascularoutcomes in patients with type 2 diabetes. N EnglJ Med 2008;358:2560–72.

50. Gerstein HC, Miller ME, Byington RP, et al., Ac-tion toControl Cardiovascular Risk inDiabetes StudyGroup. Effects of intensive glucose lowering in type2 diabetes. N Engl J Med 2008;358:2545–59.

51. Home PD, Pocock SJ, Beck-Nielsen H, et al.,RECORD Study Team. Rosiglitazone evaluated forcardiovascular outcomes in oral agent combina-tion therapy for type 2 diabetes (RECORD): amulticentre, randomised, open-label trial. Lancet2009;373:2125–35.

52. Dormandy JA, Charbonnel B, Eckland DJ,et al., PROactive investigators. Secondary pre-vention of macrovascular events in patientswith type 2 diabetes in the PROactive study(PROspective pioglitAzone Clinical Trial In macro-Vascular Events): a randomised controlled trial.Lancet 2005;366:1279–89.

53. White WB, Cannon CP, Heller SR, Nissen SE,Bergenstal RM, Bakris GL, et al., EXAMINEInvestigators. Alogliptin after acute coronarysyndrome in patients with type 2 diabetes. N EnglJ Med 2013;369:1327–35.

54. MonamiM,Dicembrini I,Mannucci E. Dipeptidylpeptidase-4 inhibitors and heart failure: a meta-analysis of randomized clinical trials. Nutr MetabCardiovasc Dis 2014;24:689–97.

KEY WORDS diabetic cardiomyopathy,diabetes mellitus, heart failure