Short- and Long-term Prognosis ofPatients With Acute Heart FailureWith and Without Diabetes:Changes Over the Last ThreeDecadesDiabetes Care 2018;41:143–149 | https://doi.org/10.2337/dc17-0544

OBJECTIVE

We studied differences in long-term (i.e., 10 year) prognosis among patients withacute heart failure (HF) with and without diabetes over the last three decades. Inaddition, we investigated whether the degree of prognostic improvement in thatperiod was comparable between patients with and without diabetes.

RESEARCH DESIGN AND METHODS

This prospective registry included all consecutive patients aged 18 years and olderadmitted to the Intensive Coronary Care Unit with acute HF in the period of 1985–2008. A total of 1,810 patients were included; 384 patients (21%) had diabetes. Theoutcome measure was the composite of all-cause mortality, heart transplantation,and left ventricular assist device implantation after 10-year follow-up.

RESULTS

The 10-year outcome in patients with diabetes was significantly worse than in thosewithout diabetes (87% vs. 76%; adjusted hazard ratio [HR] 1.17 [95% CI 1.02–1.33]).Patients admitted in the last decade had a significantly lower 10-year event rate thanpatients admitted in the first two decades, both among patients without diabetes(adjusted HR 0.86 [95% CI 0.75–0.99]) and patients with diabetes (adjusted HR 0.80[95% CI 0.63–1.00]).

CONCLUSIONS

The long-term outcome of patients with diabetes is worse than that of patientswithout diabetes. However, the long-term prognosis improved over time in bothgroups. Importantly, this improvement in long-term prognosis was comparable inpatients with and without diabetes. Despite these promising results, more aware-ness for diabetes in patients with acute HF is necessary and there is still need foroptimal treatment of diabetes in acute HF.

The prevalence of both heart failure (HF) and diabetes has increased over the lastdecades and is expected to do so in the upcoming decades (1,2). Therefore, the pres-ence of diabetes in patients with HF is also likely to increase and this is anticipated tobecome a major health concern. The actual prevalence of diabetes in patients withacute HF in different registries has varied but may be as high as 45% (3). Because the

Department of Cardiology, Thoraxcenter, Eras-mus Medical Center, Rotterdam, the Netherlands

Corresponding author: Jan C. van den Berge,j.vandenberge@erasmusmc.nl.

Received 16 March 2017 and accepted 9 Sep-tember 2017.

This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dc17-0544/-/DC1.

© 2017 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and thework is notaltered.More infor-mation is available at http://www.diabetesjournals.org/content/license.

See accompanying articles, pp. 11, 14,136, 150, and 156.

Jan C. van den Berge,

Alina A. Constantinescu, Hendrik J. Boiten,

Ron T. van Domburg, Jaap W. Deckers,

and K. Martijn Akkerhuis

Diabetes Care Volume 41, January 2018 143

CARDIOVASCULA

RANDMETA

BOLIC

RISK

structure and function of the heart is di-rectly influenced by the presence of di-abetes, diabetes is to be considered torepresentmore than just a comorbid con-dition in HF (4).Diabetes has been shown to be an in-

dependent risk factor for the developmentof HF (5,6). Moreover, this risk has beenshown to be age and sex dependent.Compared with patients without diabe-tes, the presence of diabetes doublesthe risk of HF in men, and the risk of de-veloping HF in women may be as muchas four times higher (5). These associa-tions may even be stronger in youngerpatients (5). Furthermore, the presenceof diabetes has been associated with alonger duration of hospitalization andhigher rates of rehospitalization amongpatients with acute HF (7,8). Importantly,in patients with HF, it has been estab-lished that the presence of diabetes isnot only associated with an increased car-diovascular morbidity but also with an in-creased mortality (9–13). However, theprognostic value of diabetes on in-hospitaland long-term mortality among patientswithacuteHF is still controversial (7,14–19).Since the 2000s, several new treatment

modalities have been added to the ther-apeutic regimen of chronic HF, resultingin an improved prognosis of these pa-tients (20–23). However, it has not beenestablished whether the improvement ofprognosis in patients with acute HF wasinfluenced by the presence of diabetes.For these reasons, we studied differencesin long-term prognosis among patientswith acute HF with and without diabetesover the last three decades. In addition,we investigated whether the degree ofprognostic improvement in that periodwas comparable between patients withand without diabetes.

RESEARCH DESIGN AND METHODS

InclusionThe study population and the design ofthe study have been described previously(23). In brief, all patients aged 18 yearsand older admitted to the Intensive Cor-onary Care Unit (ICCU) with acute HFwere included in this prospective registry.The inclusion period was from 1985 until2008. All patientswere recruited from theErasmus Medical Center.Patients were included when the ad-

mitting physician established a diagnosisof acute HF or cardiogenic shock. We in-cluded patients with acute, new-onset HF

as well as patients with decompensatedchronic HF. Patients admitted for acuteHF caused by an acute coronary syndromewithout evidence of sustained systolic ordiastolic dysfunction were excluded. If apatient was admitted more than oncewith acute HF, only the first admissionwas taken into account.

Ethics StatementThis was a prospective cohort registry.During the enrollment of the patients, ap-proval from the local research ethics com-mittee to conduct this study was notrequired. The study was conducted ac-cording to the Declaration ofHelsinki (24).

Baseline VariablesPatient records and discharge letterswere used for the collection of the base-line variables. Age and sex were collectedas demographic variables. The followingclinical variables were collected: previousmedical history, etiology of HF, left ven-tricular ejection fraction (LVEF), and BMI.Also, the type of treatment at the ICCUwas registered.

Diabetes was considered to be presentwhen patients received oral antidiabetictherapy and/or subcutaneous insulinprior to admission. The LVEF was classifiedinto the following qualitative categories:good, moderate, and poor. Quantitativemeasures for LVEF were categorized asfollows: .45%, 30–44%, and ,30% forpreserved,moderately depressed, and se-verely depressed LVEF, respectively. Theetiology of HF was categorized into ische-mic HF and nonischemic HF.

End PointThe outcome measure of this study wasthe composite of all-cause mortality,heart transplantation, and implantationof a left ventricular assist device (LVAD)10 years after the initial hospitalization.Survival status was assessed by usingthe Municipal Civil Registries in January2017 and was available for 98% of theincluded patients.

Statistical AnalysisThe study population was categorizedinto three groups: patients admitted in1985–1989, 1990–1999, and 2000–2008.We chose these time frames according tothe development of HF therapy (in partic-ular, ACE inhibitors and b-blockers). Inthe 1980s, there was no evidence-basedtherapy for HF. Newdrug therapy like ACEinhibitors and b-blockers was developedand subsequently implemented in daily

practice in the 1990s, and it had becomecommonpractice toprescribe them in the2000s. Therefore, we hypothesized thatthe prognosis of patients admitted inthe first decade was worse and that theprognosis would improve in the seconddecade and continued to improve in thelast decade. Moreover, we have also per-formed additional statistical analyseswith the study population divided intothree equal groups of periods of 8 years(1985–1992, 1993–2000, and 2001–2008) in order to make sure the resultswere not depending on the chosen timeperiod.

Data were summarized as mean withSD for continuous variables and as fre-quencies with percentage for categorizedvariables. The Student t test or ANOVAwas used for comparing continuous vari-ables and the x2 test for comparing cate-gorized variables.

Since the LVEFwas not reported in 28%of the patients and the etiology was notreported in 12% of the patients, we ap-plied multiple imputation. Baseline char-acteristicswere used as predictors. Pooledmeans were given for LVEF and etiology.

The Kaplan-Meier method was used topresent the cumulative event curves. Thelog-rank test was applied for comparingthe Kaplan-Meier curves. Landmark anal-yses for the 30-day event-free survivors(defined as patients who did not reachthe composite end point) were performedas secondary analyses.We used logistic re-gression for the 30-day outcome and theCox proportional hazards model for the1- and 10-year outcomes in order to eval-uate the independent association betweendiabetes and the composite end point ofall-cause mortality, heart transplantation,and LVAD implantation. In multivariableanalysis for 30-day and 1-year outcome,adjustments were made for age, sex,BMI, atrial fibrillation at admission, etiol-ogyofHF, LVEF, andahistoryofHF, rhythmor conduction disorder, and hypertension.In the analysis of the 10-year outcome,corrections were made for age, sex, BMI,etiology of HF, LVEF, period of admission,and a history of myocardial infarction, HF,and rhythmor conduction disorder. All var-iables were categorical, except for age,which was retained as a continuous vari-able. Results of logistic regression and theCox proportional hazards model were re-ported as odds ratios (ORs) and hazardratios (HRs), respectively, with their corre-sponding 95% CI.

144 Diabetes and Prognostic Trends in Acute HF Diabetes Care Volume 41, January 2018

All tests were two tailed, and P val-ues ,0.05 were considered statisticallysignificant. All data were analyzed usingSPSS software (SPSS 21.0, IBM Corp.,Armonk, NY).

RESULTS

Baseline CharacteristicsWe identified 1,810 patients admittedwith acute HF to our ICCU in the periodof 1985 until 2008. Of these, 384 pa-tients (21%) had diabetes. The preva-lence of diabetes increased; in the1980s, 14% of the patients had diabetes,comparedwith 21% in the 1990s and 27%in the most recent study period (P fortrend ,0.001). Baseline characteristicsof patients with and without diabeteswere different (Table 1). On average, pa-tients with diabetes were 5 years older,weremore often female, andhada higherBMI as compared with patients withoutdiabetes. Furthermore, patients with di-abetes more frequently had a historyof hypertension, myocardial infarction,and coronary revascularization. Pres-ence of diabetes was more commonlyassociated with ischemic HF, whereaspatients without diabetes more oftensustained HF of a nonischemic origin. Thedistribution of the left ventricular func-tion was not influenced by the presenceof diabetes.

Diabetes and MortalityCompared with patients without diabe-tes, patients with diabetes less frequentlyreached the composite end point of all-cause mortality, heart transplantation,and LVAD implantation at 30 days (9%vs. 16%; unadjusted OR 0.51 [95% CI0.35–0.75]) (Fig. 1). After multivariableadjustment, the difference in the 30-dayevent rate was somewhat attenuated butremained lower in patients with diabetes(adjusted OR 0.61 [95% CI 0.41–0.92]).The cumulative 1-year event rate wascomparable between patients with andwithout diabetes (P = 0.13) (Fig. 1A).When the analysis was restricted to the30-day event-free survivors only, the num-ber of patients who reached the com-posite end point was almost identical inpatients with diabetes and in those with-out diabetes (26% and 25%, respectively;P = 0.63) (Fig. 1B).

The number of patients who reachedthe composite end point of all-causemor-tality, heart transplantation, and LVADimplantation after 10 years of follow-upwas higher in patients with diabetes thanin those without diabetes (87% vs. 76%;unadjusted HR 1.19 [95% CI 1.06–1.36])(Fig. 1A), and this remained the case aftermultivariable adjustment (adjusted HR1.17 [95% CI 1.02–1.33]). A more pro-nounced difference in the 10-year event

rate between patients with and withoutdiabetes became apparent when theanalysis was restricted to the 30-dayevent-free survivors (adjusted HR 1.33[95% CI 1.15–1.53]) (Fig. 1B).

Patients with diabetesmore frequentlyhad prior myocardial infarction, and an is-chemic cause of HF was more commonamong these patients. However, since wedid not find a significant interaction in themultivariable Cox proportional hazardsmodel, neither between diabetes and pre-vious myocardial infarction (P = 0.95) norbetween diabetes and etiology of HF (P =0.95), therewas no difference in the impacton long-term outcome of these factors be-tween patients with and without diabetes.

Long-term Prognosis Over TimeThe baseline characteristics of patientswith and without diabetes changed dur-ing the three decades of observation(Supplementary Table 1). With time, thepresence of coronary revascularization,rhythmor conduction disorder, andhyper-tension became more frequent amongboth subgroups. In addition, the patientswithout diabetes were more commonlywomen and had less prior myocardialinfarction over time. The distribution ofthe etiology of HF and LVEF remainedstable over time in both patients withand without diabetes.

Table 1—Baseline characteristics of patients with and without diabetes

Total population Patients with diabetes Patients without diabetes P value*

No. of patients 1,810 384 1,426

Age (years) 63.5 6 14.8 67.1 6 11.1 62.3 6 15.5 ,0.001

Male sex 1,153 (64) 223 (58) 930 (65) 0.01

BMI 25.1 6 4.9 27.1 6 6.3 24.6 6 4.5 ,0.001

Medical historyMyocardial infarction 714 (39) 188 (49) 526 (37) ,0.001Coronary revascularization† 390 (22) 108 (28) 282 (20) ,0.001Heart surgery (not CABG) 237 (13) 38 (10) 199 (14) 0.04Heart transplantation 9 (0.5) 1 (0.3) 8 (0.6) 0.69Waiting for heart transplantation 35 (2) 4 (1) 31 (2) 0.15HF 888 (49) 184 (48) 704 (49) 0.61Rhythm or conduction disorder 445 (25) 83 (22) 362 (25) 0.13Hypertension 590 (33) 184 (48) 406 (29) ,0.001

HFEtiology of HF ,0.001Ischemic origin 845 (47) 239 (62) 606 (42)Nonischemic origin 965 (53) 145 (38) 820 (58)

Atrial fibrillation at admission 391 (22) 75 (20) 316 (22) 0.27LVEF .0.05Preserved 522 (29) 122 (32) 400 (28)Moderately depressed 427 (24) 91 (24) 336 (24)Severely depressed 861 (48) 171 (44) 690 (48)

Data are n (%) or mean6 SD unless otherwise indicated. CABG, coronary artery bypass graft. *Comparison between patients with and without diabetes.†Percutaneous coronary intervention and/or CABG.

care.diabetesjournals.org van den Berge and Associates 145

The short- and long-term event rate ofpatients admitted in the second decadewas comparable with the outcome in thefirst decade. This holds for both patientswith and without diabetes. For the pur-pose of comparison with the outcome ofpatients studied in the most recent timeperiod, patients admitted from 1985until 1999 were pooled into one group.This comparison demonstrated that the1-year outcome did not significantly im-prove over timedneither in patientswithdiabetes nor in those without diabetes(adjusted HR 0.93 [95% CI 0.64–1.36]and adjusted HR 0.92 [95% CI 0.76–1.12],respectively). In contrast, the long-termevent rate showed improvement, both

in acute HF patients with and without di-abetes (Fig. 2). Patients without diabetesadmitted in the last decade less fre-quently reached the composite end point10 years after initial hospitalization thanpatients admitted in the first two decades(adjusted HR 0.86 [95% CI 0.75–0.99]). Asimilar improvement in long-termoutcomewas found among the patients with diabe-tes (adjusted HR 0.80 [95% CI 0.63–1.00]).This improvement in long-term outcomeover time was more pronounced in bothpatients without and with diabetes whenthe analysis was restricted to the 30-dayevent-free survivors (adjusted HR 0.81[95% CI 0.69–0.95] and adjusted HR 0.74[95% CI 0.58–0.95], respectively).

When we analyzed the temporal trendin the long-term prognosis of the patientsdivided into three equal groups of periodsof 8 years, we found that the prognosisof patients admitted in the first period(1985–1992) was comparable with theoutcome in the patients admitted in theperiod 1993–2000. Therefore, we pooledthese patients into one group and com-pared their long-term outcome with theoutcome in patients admitted in the pe-riod 2001–2008 (Fig. 3). The outcome ofthese analyses was comparable to theresults of the analyses in patients dividedaccording to the decades. Hence, we canconclude that the resultswere not depen-dent on the division of the data set in thechosen time periods.

CONCLUSIONS

In this cohort study of patients with acuteHF, studied over a period of 24 years, theprevalence of diabetes increased overtime, with almost 30% of the patientsfound to have diabetes in the last decade.This study shows that among patients withacute HF, the presence of diabetes is asso-ciatedwith a clear prognostic disadvantagelong-term (i.e., at 10 years) when com-pared with those without diabetes. Moreimportant, we clearly demonstrated thatthe temporal reduction in long-term out-come (i.e., the composite of all-cause mor-tality, heart transplantation, and LVADimplantation) in patients admitted withacute HF, achieved in the last decade,was at least as high in patients with diabe-tes and in those without diabetes.

Short- and Long-term Outcome inPatients With and Without DiabetesThis study added results to the controver-sial evidence available in literature aboutthe prognostic impact of diabetes on the1-year prognosis among patients admit-ted with acute HF. We found that pa-tients with and without diabetes equallyreached the composite end point of all-cause mortality, heart transplantation,and LVAD implantation after 1-yearfollow-up. This endorsed the findings byothers who reported a comparable prog-nosis in patients with and without dia-betes (19). However, our findings werealso in contrast with previous studies.DIAMOND-CHF (Danish Investigations ofArrhythmia and Mortality on Dofetilidein Congestive Heart Failure) was a largeDanish trial with.5,000 patients report-ing higher 1-year mortality rates among

Figure 1—Kaplan-Meier curve of the cumulative incidence of reaching the composite end point ofall-cause mortality, heart transplantation, and LVAD implantation in the total population (A) and alandmark analysis in the 30-day event-free survivors (B): patients with diabetes vs. those withoutdiabetes.

146 Diabetes and Prognostic Trends in Acute HF Diabetes Care Volume 41, January 2018

patients with diabetes (14). Two otherEuropean registries also found a prognos-tic disadvantage of diabetes on the 1-yearprognosis (15,18). Last, a large retro-spective Scottish population study alsofound a prognostic disadvantage after1-year follow-upof patientswith diabetes(16). A potential reason for these discrep-ancies is the difference in the study pop-ulation. Generally, patients in our studywere younger, more commonly had amyocardial infarction or HF in their his-tory, and were less frequently found tohave a history of hypertension.Importantly, after a longer follow-up

duration, acute HF patients with diabeteshad a prognostic disadvantage comparedwith those without diabetes. This resultedin higher 10-year event rates among pa-tientswithdiabetes. These results confirmedthe results reported by others (14,25,26).The poorer long-term prognosis in pa-

tients with diabetes is an important

finding and has implications for the fu-ture. Since the incidence of diabetes inHF patients is likely to further increasein the future, this will become a majorhealth care problem with high morbidityandmortality, as well as high costs for so-ciety (27). Therefore, it is important torecognize diabetes in patients with HFand to start an adequate therapy for thediabetes. However, there is little evidencefor the best therapy of glycemic controlin HF patients in practice (28). For thatreason,more future clinical research is re-quired for the medical treatment of dia-betes in patients with HF.Webelieve thatbetter glycemic control in that specificsubset of patients may contribute to afurther improvement in prognosis.

Temporal Trends in Long-termPrognosisSeveral previous studies have reportedtrends in long-term outcome (20–23),

but temporal trends among acute HF pa-tients stratified by the presence of diabe-tes have not been described previouslydneither short-term trends nor long-termtrends. Novel treatment modalities forHF, like ACE inhibitors, b-blockers, miner-alocorticoid receptor antagonists, im-plantable cardioverter defibrillators, andcardiac resynchronization therapy, haveall been implemented in clinical practicein the last decades. This change in thetherapeutic regimen was associated witha lower long-term mortality in the totalpopulation with acute HF (20–23), and itcaused the improved long-termprognosisof the patients without diabetes.

However, our study cannot definitelyelucidate the mechanisms that resultedin the improved long-term prognosisamong the subgroup with diabetes. Sincethe novel treatment modalities havebeen found to have comparablemortalitybenefit in patients both with and withoutdiabetes (27), it is possible that this is the(only) reason for the improved long-termprognosis among the acute HF patientswith diabetes. On the other hand, previ-ous studies among patients with diabetes(not in an acute HF population) showedan improved survival over the last de-cades (29,30), which was attributed to agrowing awareness of diabetes, more fo-cus on cardiovascular preventionby treat-ing comorbidities, and an improvement inthe treatment of acute myocardial infarc-tion (29–31). We hypothesized that theimproved prognosis among acute HF pa-tients with diabetes may be a result of thecombination of both above-mentioned po-tential mechanisms. Therefore, despite theimpaired prognosis associated with diabe-tes in acute HF, patients admitted withacute HF both with diabetes and withoutdiabetes showed a comparable improve-ment over time in long-term prognosis.

Thirty-Day Outcome of Patients Withand Without DiabetesContrary to expectations, we found thatpatients with acute HF with diabetes lessfrequently reached the composite endpoint at 30 days than those without diabe-tes. This is not a uniquefinding, but data onthis topic vary. Somestudies reported com-parable in-hospital mortality in patientswith and without diabetes (7), but otheranalyses described worse in-hospital out-come in patients with diabetes (17,18). Alarge Scottish database reported that di-abetes was associated with lower 30-day

Figure 2—Multivariable-adjusted trends in 10-year prognosis among patients with acute HF: pa-tients with diabetes vs. those without diabetes. Analyses were separately done in the total pop-ulation (A) and in the 30-dayevent-free survivors (B). The data setwasdividedaccording to the threedecades.

care.diabetesjournals.org van den Berge and Associates 147

mortality (16). The hypothesis put forwardby these investigators was that patientswith diabeteswouldmost likely have a bet-ter ejection fraction than subjects withoutdiabetes. Since the authors were unable toadjust for LVEF, they could not establishthis hypothesis. When we adjusted forLVEF in our analyses, we found that, de-spite this, patients with diabetes continuedto have a better 30-day outcome.We think there might be two potential

reasons for this prognostic disadvan-tage of the patients without diabetes inour study. First, the patients without di-abetes who reached the composite endpoint within 30 days were most likelypredominantly patients with end-stageHF or, given the fact that patientswithoutdiabetes were more often treated withmechanical circulatory support, patientswith cardiogenic shock. It is well knownthat cardiogenic shock is associated with

elevated in-hospital mortality (32,33).The second reason why patients withoutdiabetes were found to have a poorer30-day prognosis may be due to the factthat the patients in our survey were ad-mitted to an intensive care unit. A lowerin-hospital mortality among patients withdiabetes admitted at a general intensivecare unit has also been described byGraham et al. (34) and Martin et al. (35).These two studies constitute the largestreports that investigated the in-hospitalprognosis among patients with diabetesadmitted to the general intensive careunit. However, the mechanism of thelower in-hospital mortality among thesepatients with diabetes admitted to an in-tensive care unit was not stipulated.

Strengths and LimitationsThis study has several strengths. First, westudied long-termoutcome (i.e., 10 years)

in a large population of patients withacute HF over a study period of 24 years.Furthermore, this is the first study re-porting temporal trends in prognosisamong acute HF patients with and with-out diabetes.

Despite these unique strengths, somelimitations should be acknowledged.First, this study was performed in a singlecenter. Therefore, this could result in alower external validity. Furthermore, nodistinctionwasmade between type 1 andtype 2 diabetes. Also, any information re-garding the development of diabetes dur-ing follow-up among the patients withoutdiabetes at baseline was not available.The development of diabetes duringfollow-up may have also influenced theprognosis. Finally, information aboutdrug therapy of diabetes was not reportedin our database. This may be of addedvalue because the type of diabetes treat-ment could influence HF symptoms, hos-pitalization, and mortality (36–38).

ConclusionIn conclusion, we showed that patientsadmitted with acute HF had a poor prog-nosis. Moreover, the long-term outcome(i.e., the composite of all-cause mortality,heart transplantation, and LVAD implan-tation) of patients with diabetes is worseas compared with those without diabe-tes. However, the long-term mortalityprognosis improved over time as a resultof an improved treatment of HF. Impor-tantly, this improvement in long-termprognosis was at least as high in patientswithout diabetes as in those with diabe-tes. This study not only emphasizes theneed to improve the treatment of HFbut also emphasizes the need for optimaltreatment of diabetes in acute HF as wellas the need to create more awareness fordiabetes in patients with acute HF.

Duality of Interest. No potential conflicts of in-terest relevant to this article were reported.AuthorContributions. J.C.v.d.B. acquired data,performed the statistical analysis, interpretedthe data, and wrote the manuscript. A.A.C. andH.J.B. reviewed the manuscript. R.T.v.D. de-signed the study, contributed to analysis, inter-preted the data, and reviewed the manuscript.J.W.D. designed the study, interpreted thedata, and reviewed andedited themanuscript.K.M.A. interpreted the data and reviewed andedited the manuscript. J.W.D. is the guarantorof this work and, as such, had full access to all thedata in the study and takes responsibility for theintegrity of the data and the accuracy of the dataanalysis.

Figure 3—Multivariable-adjusted trends in 10-year prognosis among patients with acute HF: pa-tients with diabetes vs. those without diabetes. Analyses were separately done in the total pop-ulation (A) and in the 30-dayevent-free survivors (B). The data setwasdividedaccording to the threegroups of equal length in years.

148 Diabetes and Prognostic Trends in Acute HF Diabetes Care Volume 41, January 2018

References1. Bui AL, Horwich TB, FonarowGC. Epidemiologyand risk profile of heart failure. Nat Rev Cardiol2011;8:30–412. Shaw JE, Sicree RA, Zimmet PZ. Global esti-mates of the prevalence of diabetes for 2010and 2030. Diabetes Res Clin Pract 2010;87:4–143. Ambrosy AP, Fonarow GC, Butler J, et al. Theglobal health and economic burden of hospitali-zations for heart failure: lessons learned fromhos-pitalized heart failure registries. J Am Coll Cardiol2014;63:1123–11334. Cook SA, Varela-Carver A, Mongillo M, et al.Abnormal myocardial insulin signalling in type 2diabetes and left-ventricular dysfunction. EurHeart J 2010;31:100–1115. Kannel WB, McGee DL. Diabetes and cardio-vascular disease. The Framingham study. JAMA1979;241:2035–20386. Gottdiener JS, Arnold AM, Aurigemma GP,et al. Predictors of congestive heart failure inthe elderly: the Cardiovascular Health Study.J Am Coll Cardiol 2000;35:1628–16377. Greenberg BH, AbrahamWT, Albert NM, et al.Influence of diabetes on characteristics and out-comes inpatientshospitalizedwithheart failure: areport from the Organized Program to InitiateLifesaving Treatment in Hospitalized Patientswith Heart Failure (OPTIMIZE-HF). Am Heart J2007;154:277.e1–277.e88. Sarma S, Mentz RJ, KwasnyMJ, et al.; EVERESTinvestigators. Association between diabetes mel-litus andpost-dischargeoutcomes in patients hos-pitalized with heart failure: findings from theEVEREST trial. Eur J Heart Fail 2013;15:194–2029. Bertoni AG, HundleyWG,MassingMW, BondsDE, Burke GL, Goff DC Jr. Heart failure prevalence,incidence, andmortality in the elderly with diabe-tes. Diabetes Care 2004;27:699–70310. Ho KK, Anderson KM, Kannel WB, GrossmanW, Levy D. Survival after the onset of congestiveheart failure in FraminghamHeart Study subjects.Circulation 1993;88:107–11511. From AM, Leibson CL, Bursi F, et al. Diabetesin heart failure: prevalence and impact on out-come in the population. Am J Med 2006;119:591–59912. MacDonald MR, Petrie MC, Varyani F, et al.;CHARM Investigators. Impact of diabetes on out-comes in patientswith lowandpreservedejectionfraction heart failure: an analysis of the Candesar-tan in Heart failure: Assessment of Reduction inMortality and morbidity (CHARM) programme.Eur Heart J 2008;29:1377–138513. De Groote P, Lamblin N, Mouquet F, et al.Impact of diabetes mellitus on long-term survivalin patientswith congestiveheart failure. EurHeartJ 2004;25:656–66214. Gustafsson I, Brendorp B, Seibaek M, et al.;Danish Investigatord of Arrhythmia and Mortalityon Dofetilde Study Group. Influence of diabetesand diabetes-gender interaction on the risk of

death in patients hospitalized with congestiveheart failure. J Am Coll Cardiol 2004;43:771–77715. Harjola VP, Follath F, Nieminen MS, et al.Characteristics, outcomes, and predictors of mor-tality at 3 months and 1 year in patients hospital-ized for acute heart failure. Eur J Heart Fail 2010;12:239–24816. MacDonald MR, Jhund PS, Petrie MC, et al.Discordant short- and long-term outcomes asso-ciated with diabetes in patientswith heart failure:importance of age and sex: a population study of5.1 million people in Scotland. Circ Heart Fail2008;1:234–24117. Parissis JT, Rafouli-Stergiou P, Mebazaa A,et al. Acute heart failure in patients with diabetesmellitus: clinical characteristics and predictors ofin-hospital mortality. Int J Cardiol 2012;157:108–11318. Targher G, Dauriz M, Laroche C, et al. In-hospital and 1-year mortality associated with di-abetes in patients with acute heart failure: resultsfrom the ESC-HFA Heart Failure Long-Term Regis-try. Eur J Heart Fail 2017;19:54–6519. Targher G, Dauriz M, Tavazzi L, et al.; IN-HFOutcome Investigators. Prognostic impact ofin-hospital hyperglycemia in hospitalized patientswith acute heart failure: results of the IN-HF (Ital-ian Network on Heart Failure) outcome registry.Int J Cardiol 2016;203:587–59320. Jhund PS, Macintyre K, Simpson CR, et al.Long-term trends in first hospitalization for heartfailure and subsequent survival between1986 and 2003: a population study of 5.1 millionpeople. Circulation 2009;119:515–52321. Shafazand M, Schaufelberger M, Lappas G,Swedberg K, Rosengren A. Survival trends inmen and women with heart failure of ischaemicand non-ischaemic origin: data for the period1987-2003 from the Swedish Hospital DischargeRegistry. Eur Heart J 2009;30:671–67822. Teng TH, Hung J, Knuiman M, et al. Trends inlong-term cardiovascular mortality and morbidityin men and women with heart failure of ischemicversus non-ischemic aetiology in Western Aus-tralia between 1990 and 2005. Int J Cardiol2012;158:405–41023. van den Berge JC, Akkerhuis MK, Con-stantinescu AA, Kors JA, van Domburg RT,Deckers JW. Temporal trends in long-term mor-tality of patients with acute heart failure: datafrom 1985-2008. Int J Cardiol 2016;224:456–46024. GoodyearMD, Krleza-Jeric K, Lemmens T. TheDeclaration of Helsinki. BMJ 2007;335:624–62525. Cubbon RM, Adams B, Rajwani A, et al. Di-abetes mellitus is associated with adverse prog-nosis in chronic heart failure of ischaemic andnon-ischaemic aetiology. Diab Vasc Dis Res2013;10:330–33626. Varela-Roman A, Grigorian Shamagian L,Barge Caballero E, Mazon Ramos P, RigueiroVeloso P, Gonzalez-Juanatey JR. Influence of di-abetes on the survival of patients hospitalized

with heart failure: a 12-year study. Eur J HeartFail 2005;7:859–86427. Baliga V, Sapsford R. Review article: diabetesmellitus and heart failuredan overview of epide-miology and management. Diab Vasc Dis Res2009;6:164–17128. Ponikowski P, Voors AA, Anker SD, et al.;Authors/Task Force Members; Document Re-viewers. 2016 ESC Guidelines for the diagnosisand treatment of acute and chronic heart failure:the Task Force for the diagnosis and treatment ofacute and chronic heart failure of the EuropeanSocietyofCardiology (ESC).Developedwith the spe-cial contribution of the Heart Failure Association(HFA) of the ESC. Eur J Heart Fail 2016;18:891–97529. Eliasson M, Talback M, Rosen M. Improvedsurvival in both men and women with diabetesbetween 1980 and 2004–a cohort study in Swe-den. Cardiovasc Diabetol 2008;7:3230. Lipscombe LL, Hux JE. Trends in diabetesprevalence, incidence, and mortality in Ontario,Canada 1995-2005: a population-based study.Lancet 2007;369:750–75631. Nauta ST, Deckers JW, Akkerhuis KM, vanDomburg RT. Short- and long-termmortality aftermyocardial infarction in patientswith andwithoutdiabetes: changes from 1985 to 2008. DiabetesCare 2012;35:2043–204732. Goldberg RJ, Spencer FA, Gore JM, Lessard D,Yarzebski J. Thirty-year trends (1975 to 2005) inthe magnitude of, management of, and hospitaldeath rates associated with cardiogenic shockin patients with acute myocardial infarction: apopulation-based perspective. Circulation 2009;119:1211–121933. Zannad F, Mebazaa A, Juilliere Y, et al.; EFICAInvestigators. Clinical profile, contemporary man-agement and one-year mortality in patients withsevere acute heart failure syndromes: the EFICAstudy. Eur J Heart Fail 2006;8:697–70534. GrahamBB, Keniston A, Gajic O, Trillo AlvarezCA, Medvedev S, Douglas IS. Diabetes mellitusdoes not adversely affect outcomes from a criticalillness. Crit Care Med 2010;38:16–2435. Martin GS, Mannino DM, MossM. The effectof age on the development and outcome of adultsepsis. Crit Care Med 2006;34:15–2136. Eurich DT, Majumdar SR, McAlister FA,Tsuyuki RT, Johnson JA. Improved clinical out-comes associated with metformin in patientswith diabetes and heart failure. Diabetes Care2005;28:2345–235137. Giles TD, Miller AB, Elkayam U, BhattacharyaM, Perez A. Pioglitazone and heart failure: resultsfrom a controlled study in patients with type 2diabetes mellitus and systolic dysfunction. J CardFail 2008;14:445–45238. Masoudi FA, Inzucchi SE, Wang Y, HavranekEP, Foody JM, Krumholz HM. Thiazolidinediones,metformin, and outcomes in older patients withdiabetes andheart failure: anobservational study.Circulation 2005;111:583–590

care.diabetesjournals.org van den Berge and Associates 149