Luigi TAVAZZI, MD, PhD
GVM Care & Research
Fondazione Ettore Sansavini per la Ricerca Scientifica
Health Science Foundation Onlus
Observational research in heart failure
Observational research today, particularly in the form of surveys and registries, which are widely used by the clinical/scientific community, has several applications.Well-definedmethodologies, which vary according to the specific aim(s) of the research, must be applied. This paper reports some observational findings on heart failure (HF) that were collected in a recent survey conducted by the European Society of Cardiology (ESC). This survey, with a multinational European network structure, was based on the principles of observational research. Its aim was to get a picture of real clinical profiles of both acute and chronic HF patients and to compare current therapeutic regimens in HF with ESC guideline recommendations. The main observations are as follows: (i) the clinical profiles of acute HF suggested by the ESC broadly identify small subsets of subjects with different outcomes, but this still leaves large clinical areas that are not fully characterized; (ii) acute HF therapy has remained practically unchanged in the last few decades, and we require proper randomized controlled trials for a better understanding of pathophysiological profiles; and (iii) although chronic HF therapy is evidence-based, drug doses currently used in clinical practice are far from recommended doses. In spite of these shortcomings, recently tested new drugs have been found to be both effective and safe, and with rapid incorporation in guidelines these should improve patient outcomes.
Over a decade has passed since Sakett advised, “If you find that a study was not randomized, we’d suggest that you stop reading it and go on to the next article.” What was the case in 1997 is not the case today. Today, observational research, particularly in the form of surveys and registries, which are widely used by the clinical/scientific community, has several applications, even though randomized controlled trials (RCTs), which protect against otherwise unavoidable bias, remain essential. RCTs are still the key to reliably evaluating the efficacy and safety of new drugs and diagnostic/therapeutic procedures and interventions.
It is important that this point is clear because many of the reservations concerning observational research derive from its inappropriate use. Observational research has its own objectives and well-defined methodologies that vary with the specific aim(s) of the research. This paper reports a few observational findings on heart failure (HF) collected in a recent survey conducted in several European countries.1 Firstly, a few criteria that should be taken into consideration when designing an observational clinical study to obtain reliable and informative data are discussed. These criteria have been applied in the above-mentioned survey.
Table I. Preliminary questions that should be answered before starting an observational study.
Table II. In-hospital mortality registries.
General outline of a registry
Some preliminary questions that should be answered before starting an observational study are listed and briefly commented on below (Table I).
Outline of heart failure in Europe today
Several registries and surveys have been conducted in patients with either chronic or acute HF, but they all had one or some of the limitations highlighted above. In response to these shortcomings, the European Society of Cardiology (ESC) planned a new registry that attempted to overcome these limitations.1 This study was conducted in 136 centers in 12 European countries. The countries, selected on the basis of their geographical distribution, provide a reasonably accurate representation of Europe overall. The centers, selected in each country according to a defined hospital/inhabitant ratio to provide an appropriate balance, were hospitals with dif- ferent types of cardiology facilities (with or without cardiac surgery and interventional facilities). The aim of this selection was to provide an up-to-date picture of HF in Europe by establishing a network with a broad spectrum of cardiology units capable of consecutively enrolling and following outpatients with HF and admitting patients with acute/worsening HF. From October 2009 to May 2010, 5118 patients were included in a pilot phase of this registry, and follow-up is ongoing. A few of the findings recorded at the enrollment of these patients, which are worthy of further consideration, are discussed below.1
Some 1892 patients were admitted to hospital for acute/ worsening HF. Mean age was 70 years, and about a third of the patients enrolled were female. More than half of the patients (64%) had an ischemic etiology confirmed by coronary angiography. At hospital entry, either pulmonary or peripheral congestion was detected in 82% of cases, and clinical signs of peripheral hypoperfusion were reported in 9% of the patients.
Atrial fibrillation was detected in 35% of the patients as well as a large QRS (≥120 ms). An echocardiographic examination was performed in 75% of the patients. The median ejection fraction (EF) was 38% (interquartile range [IQR] 27%- 52%); 39% of the patients had preserved EF, defined as EF >40%.Moderate-to-severemitral regurgitation was diagnosed in 43% of the patients. Echocardiograms were performed during hospitalization, but not necessarily on admission. As a result, we do not know what the EF value was when the clinical status of the patient was at its worst. However, the proportion of patients with preserved or mildly compromised ventricular systolic function observed in this cohort of acute or worsening HF patients (approximately 40%) is similar to that found in other comparable surveys,2-5 confirming that a drop in ventricular systolic contractile performance is not necessarily the precipitating cause of decompensation.
Comorbidities were frequent. Anemia, defined as a hemoglobin level <12 g/dL, was detected in 31% of patients; an estimated glomerular filtration rate (eGFR) <50 mL/min/1.73 m2 and <30 mL/min/1.73 m2 was reported in 33% and 10% of patients, respectively. Over a third (35%) of patients had a history of diabetes, while 54% had hyperglycemia on admission.
N-terminal prohormone of brain natriuretic peptide (NT-pro- BNP) and brain natriuretic peptide (BNP) were measured at entry in 489 and 204 patients only. The median values were 4007 pg/mL (IQR 2043-9487 pg/mL) and 870 pg/mL (IQR 423-1950 pg/mL), indicating the severity of patients’ clinical condition at hospital admission. Troponin (I or T) was measured in 987 patients with a median value of 0.04 ng/mL (IQR 0.01-0.29 ng/mL). These important markers of ventricular stress and compromise have not so far been fully incorporated in clinical practice.
The definition of acute HF reported in international guidelines is quite vague,6,7 which has led to the inclusion of heterogeneous populations under the umbrella of “acute heart failure.” This is one of most important reasons for the failure to have developed active drugs in the setting of acute HF, as most RCTs adopted an “all comers” approach. The current European guidelines for the diagnosis and treatment of HF propose a stratification of patients admitted for acute HF.2 Figure 1A (page 380) shows the stratification of patients enrolled in the ESC-HF (European Society of Cardiology Heart Failure) survey, according to the clinical profiles in the ESC guidelines.1 Decompensated HF was most frequent clinical profile (75% of the cases), while pulmonary edema and cardiogenic shock were reported in 13% and 2% of patients, respectively. Figure 1B shows the overall rate of in-hospital mortality stratified by clinical profile. As expected, patients with cardiogenic shock have the worst short-term prognosis. For this reason, patients presenting with this clinical profile should be managed with specific, intensive approaches. Patients with hypertensive HF are the other extreme, showing the most favorable in-hospital survival.
It is unlikely that the inclusion of both these patient categories in the same trial testing the same drug could lead to meaningful and applicable results. Whether the category of “decompensated heart failure” (75% of patients) represents a homogeneous group of patients is difficult to envisage, and to believe. There is the need, in future, to better clarify the relationship between clinical pictures and the definitions of profiles suggested by the guidelines to obtain a patients’ categorization that allows both individual decision-making and the identification of patient populations appropriate for testing specific new drugs.
Overall, 71 patients died during their hospital stay. When age and two major markers of cardiovascular function (systolic blood pressure and reduced renal function) were considered, 93% of deaths could be explained by the presence of at least one of these factors. The median length of hospitalization was 8 days (IQR 5-11 days), and 48% of patients were managed in the intensive care unit for amedian of 4 days (IQR 2-7 days). The median body weight reduction during the hospital stay was 2 kg. At discharge, pulmonary congestion, peripheral congestion, or both were still present in 10%, 18%, and 24% of cases, respectively.
Figures 1A and 1B. Clinical profiles and in-hospital mortality rates of patients with acute/worsening heart failure.
Modified from reference 1: Maggioni et al. Eur J Heart Fail. 2010;12:1076-1084. © 2010, European Society of Cardiology.
The mean duration of hospitalization, compared to that recorded in a previous ESC survey on acute HF,4 was shortened by a day (from 9 to 8 days). The length of hospitalization in Europe remains approximately twice the length of hospitalization reported in US surveys.8,9 However, the number of patients showing clinical signs of congestion at discharge is still elevated and may account, at least in part, for the high rates of hard events and rehospitalizations observed in the few months after discharge in previous studies.10,11
Intravenous diuretics were used in 85% of cases. The median dose of furosemide used during the hospital stay was 60 mg per day (IQR 40-100 mg per day). Nitrates and inotropes were administered in 18% and 10% of patients. Of the inotropes, the most used was dobutamine (in 4.6% of patients) followed by levosimendan (in 2.4% of patients). In fact, in spite of many trials testing a number of new drugs in acute HF patients in the last decade, these patients are still treated in the same way today as the same patients were 20 years ago.
Chronic heart failure in outpatients
In this population (3226 patients enrolled), the rates of moderate (New York Hospital Association [NYHA] class I-II) and severe HF (NYHA class III-IV) were 72% and 28%, respectively. Ischemic etiology accounted for just 40% of cases, with angiographic confirmation in 85% of cases. EF was available for 2857 patients (89%): its median value was 36%(IQR 30%- 46%) and preserved EF (>40%) was reported in 36% of cases, a figure not far from that observed in patients with acute/ worsening HF. A hemoglobin level <12 g/dL was reported in 19% of cases, and an eGFR <60 and <30 mL/min/1.73 m2 was reported in 41% and 5% of patients. NT-proBNP and BNP were measured in a minority of cases (747 and 285 patients). Median values were 1387 pg/mL (IQR 485-3381 pg/ mL) and 390 pg/mL (IQR 133-870 pg/mL).
The use of pharmacological treatments is reported in Table III. A blocker of the renin-angiotensin system, β-adrenergic blockers, and aldosterone blockers were prescribed in 88%, 87%, and 44% of cases, respectively. The combination of renin-angiotensin system blockers, β-adrenergic blockers, and aldosterone blockers was prescribed in 35% of patients, where- as the combination of β-adrenergic blockers, angiotensinconverting enzyme (ACE) inhibitors, and angiotensin receptor blockers was reported in just 3% of patients. Table IV reports the doses of evidence-based treatments prescribed.
Table III. Prescribed pharmacological treatment for chronic heart failure (n=3226 patients).
Table IV. Doses of evidence-based treatments used in the ESC-HF Pilot Survey with respect to the recommended target doses.
Prescribed doses of recommended treatments represent a major issue in chronic HF therapy. Ramipril (never tested in HF patients in RCTs) and enalapril were the most prescribed ACE inhibitors; the target dose of these drugs (assuming the target dose advised for ischemic heart disease for ramipril) was achieved in 38% and 46% of cases. For angiotensin receptor blockers, the target dose of candesartan, losartan, and valsartan was reached in 28%, 19%, and 17% of cases.
The target dose of carvedilol, bisoprolol, and metoprolol was reached in 37%, 21%, and 21% of patients, whereas the target dose of spironolactone, canrenone, and eplerenone was prescribed in 22%, 61%, and 33% of patients, respectively. Overall, beside a much improved rate of prescription of recommended treatments compared to previous European surveys, the drugs’ doses did not change at all. They actually correspond to about a third of the suggested optimal dose, according to current guidelines.
This behavior of physicians, which is common to all European countries, should be seriously considered. The first obvious question to ask is whether the expected drug effects, from the results of RCTs in which most patients were treated at target doses, are actually achieved. We have no answer to this key question. A second question is why do physicians not uptitrate drug doses as they are consistently recommended to? Are they not properly informed? If this was the case, unlikely though this possibility is, the matter could be dealt with via education and our goal would be to find more effective approaches in continuous medical education. Alternatively, perhaps physicians are satisfied with the clinical results achieved with the doses prescribed? If this is the case, we should consider and test whether lower doses than those recommended may be enough in the polypharmaceutical approach that we nowadays implement in HF therapy. Finally, could it be that full recommended doses are not tolerated by most HF patients in the real world?
We know that patients enrolled in clinical trials according to a number of inclusion/exclusion criteria represent a small, selected portion of the universe of HF patients. Moreover, the RCTs that tested recommended drugs against placebo were performed many years ago, with a background therapy that was markedly different to that which is currently used. The target doses recommended today were defined in those trials, which were not validated afterwards. Maybe, contrary to our expectations, these doses are not optimal for many patients.
This may particularly be the case for β-blockers. The history of β-blocker implementation is represented in Figure 2 (page 381),12 showing data recorded in a long-term Italian registry since 1995. Gradually, the prescription of the β-blockers increased until now about 80% of patients are taking these drugs. However, the doses have remained unchanged, at about one third of those recommended. A recent instructive experience occurred during SHIFT (Systolic Heart failure treatment with the If inhibitor ivabradine Trial), which tested ivabradine, a pure bradycardic agent, versus placebo on top of recommended treatments, including β-blockers.
Because both ivabradine and β-blockers reduce heart rate, an important goal of the trial (both for safety and efficacy) was to investigate the tested drug—ivabradine—when added to the highest tolerated dose of β-blockers. All the investigators were kept aware of this important condition and they were encouraged to modulate the β-blocker dose according to guideline recommendations, and required to report the reasons for not achieving target dose if this was not reached. In spite of this cogent approach, only 56% of patients were treated with at least half the target dose of β-blocker13—44% of patients were given less than half target dose. The main reasons for the failure to reach the goal were collateral effects, such as hypotension, asthenia, or comorbidities. The association of ivabradine with the best tolerated dose of β-blocker improved outcome without increasing undesired effects.
Figure 2. Yearly prescription rate of β-blockers in patients with chronic heart failure recorded in the Italian IN-CHF registry over time (1995- 2010). Based on data from reference 12.
Interestingly, a similar dissociation between recommendations and clinical practice was noted in a European survey of electrical device implantation. According to current guidelines,5 37% of patients had clinical characteristics that suggested an implantable cardioverter-defibrillator (ICD) was potentially suitable. Of these patients with a theoretical indication for the implant, only one third (33%) actually received an ICD implant. Similarly, of the 6% of patients with a clinical profile suitable for a CRT (cardiac resynchronization therapy) device, only 2.2% actually received an implant.
The ESC-HF survey commented on above provides a clear picture of the clinical profiles of both acute and chronic HF patients in Europe, the rate of use of guideline-recommended evidence-based treatments, and detailed data on the proportion of patients in whom target dose was reached, as well as the adherence to recommendations on device implantation. New therapeutic options have been shown to be as effective and safe in two recent large trials. The EMPHASIS-HF (Eplerenone in Mild Patients Hospitalization And SurvIval Study in Heart Failure) trial14 showed that eplerenone, a selective aldosterone receptor blocker, proved to be as effective on hard end points in patients with postacute myocardial infarction and left ventricular dysfunction and is also beneficial in patients with mild-to-moderate HF of any etiology.
The SHIFT trial showed that ivabradine, an If current blocker that selectively decreases heart rate, significantly reduces the incidence of both HF mortality and hospitalization in a broad population of patients with HF of any etiology and any severity.13 The important incremental knowledge provided by this trial is that the modulation of heart rate per se, independent of any other effect, can delay the progression of HF and prevent HF-related adverse events.15 Future guidelines on HF management will probably include this new information in terms of recommendations, and future observational studies will show the impact of their incorporation in clinical practice. _
1. Maggioni AP, Dahlström U, Filippatos G, et al. EURObservational Research Programme: The Heart Failure Pilot Survey (ESC-HF Pilot). Eur J Heart Fail. 2010;12:1076-1084.
2. Nieminen MS, Böhm M, Cowie MR, et al; Committee for Practice Guideline (CPG). Executive summary of the guidelines on the diagnosis and treatment of acute heart failure: the Task Force on Acute Heart Failure of the European Society of Cardiology. Eur Heart J. 2005;26:384-416.
3. Tavazzi L, Maggioni AP, Lucci D, et al; Italian Survey on Acute Heart Failure. Nationwide survey on acute heart failure in cardiology ward services in Italy. Eur Heart J. 2006;27:1207-1215.
4. Nieminen MS, Brutsaert D, Dickstein K, et al. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population. Eur Heart J. 2006;27:2725-2736.
5. Yancy CW, Lopatin M, Stevenson LW, De Marco T, Fonarow GC. Clinical presentation, management, and in-hospital outcomes of patients admitted with acute decompensated heart failure with preserved systolic function: a report from the Acute Decompensated Heart Failure National Registry (ADHERE) Database. J Am Coll Cardiol. 2006;47:76-84.
6. Dickstein K, Cohen-Solal A, Filippatos G, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2008. The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2008 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association of the ESC (HFA) and endorsed by the European Society of Intensive Care Medicine (ESICM). Eur J Heart Fail. 2008;10:933-989.
7. Hunt SA, Abraham WT, Chin MH, et al. 2009 focused update incorporated into the ACC/AHA 2005 guidelines for the diagnosis and management of heart failure in the adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2009;53:e1–e90.
8. Adams KF Jr, Fonarow GC, Emerman CL, et al; ADHERE Scientific Advisory Committee and Investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United Stated: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Survey (ADHERE). Am Heart J. 2005;149:209-216.
9. Gheorghiade M, AbrahamWT, Albert NM, et al; OPTIMIZE-HF Investigators and Coordinators. Systolic blood pressure at admission, clinical characteristics, and outcomes in patients hospitalized with acute heart failure. JAMA. 2006;296: 2217-2226.
10. Zannad F, Mebazaa A, Juilliere Y, et al. Clinical profile, contemporary management and one-year mortality in patients with severe acute heart failure syndromes: the EFICA study. Eur J Heart Fail. 2006;8:697-705.
11. Siirila-Waris K, Lassus J, Melin J, Peuhkurinen K, Nieminen MS, Harjola VP. Characteristics, outcomes, and predictors of 1-year mortality in patients hospitalized for acute heart failure. Eur Heart J. 2006;27:3011-3017.
12. Fabbri G, Gorini M, Maggioni AP, Cacciatore G, Di Lenarda A, Tavazzi L. Unpublished.
13. Swedberg K, Komajda M, Böhm M, et al; SHIFT Investigators. Ivabradine and outcomes in chronic heart failure (SHIFT): a randomised placebo-controlled study. Lancet. 2010;376:875-885.
14. Zannad F, McMurray J, Krum H, et al; EMPHASIS-HF Study Group. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med. 2011; 364:11-21.
15. Böhm M, Swedberg K, Komajda M, et al; SHIFT Investigators. Heart rate as a risk factor in chronic heart failure (SHIFT): the association between heart rate and outcomes in a randomised placebo-controlled trial. Lancet. 2010;376: 886-894.
Keywords: observational research; heart failure