Discussion
Currently, reports of trials or registries concerning treatments of patients with HF, as well as in other conditions, quantitatively define the use of guideline-recommended 'optimal therapy', usually as the proportion of patients who received these treatments, but, specifically for drugs, regardless of dose. Current observations in patients with HF, for which therapy is particularly complex, show that a large proportion of the affected population fail to receive the recommended drugs and substantially less than half receive the 'target doses'. However, reasons underlying this discrepancy between recommendations and actual clinical practice are not known. This ESC Long-Term Registry reflects 'real-world' management from a variety of hospitals, of all levels of complexity, and from all regions of Europe and Mediterranean countries. Detailed information has been obtained not only on patient characteristics and the use of procedures and treatments, but also on the reasons why recommended pharmacological or non-pharmacological treatments are not appropriately utilized in patients with HF. This information has generally not been available from previous administrative data sets, trials, or registries.
Characteristics of Patients
Patients included in the ESC Long-Term Registry present with baseline characteristics, clinical history, and co-morbidities which largely overlap the observations of other European or US registries. Patients hospitalized for acute HF show a more severe clinical profile, as well as a higher rate of co-morbidities than patients with chronic HF. The substantial similarity of this population of patients with respect to previous reports allows the findings of this registry to be considered largely applicable also to other clinical contexts.
Treatments of Hospitalized Patients With Acute Heart Failure
The treatment of patients hospitalized for acute HF remains largely opinion based, with little good evidence to guide therapy. Even in this context, the treatment algorithm suggested by the recent ESC guidelines recommends the administration of diuretics to all patients with congestion and, thereafter, different treatments according to the levels of SBP: specifically inotropes if SBP is < 85 mmHg; vasodilators (i.e. i.v. nitrates) if SBP is > 110 mmHg; and only diuretics with careful observation of the patients if SBP is between 85 and 110 mmHg.
Data from our registry show that > 80% of patients received a diuretic treatment, and that, despite the use of inotropes being more frequent, as expected, in patients with low SBP, a non-negligible proportion of patients received this treatment even with higher levels of SBP, and 6.8% of them even if SBP was > 110 mmHg. Similarly for i.v. nitrates, their prescription according to the levels of SBP does not appear adherent to the guideline suggestions.
Even considering the cut-off values of SBP suggested by previous guidelines, the general framework describing the use of inotropes and nitrates remains the same. Therefore, besides the fact that guidelines in this specific clinical context are not developed from evidence-based observations, the application in practice of these common-sense-based recommendations is not appropriate. This finding is similar to the observations of other registries.
With respect to oral treatments, the rate of use of RAS blockers, beta-blockers, and MRAs significantly increases at discharge after the index admission. This fact documents furthermore that the admission for HF is, for a relevant number of patients, a concrete opportunity to optimize their background therapy. This consideration is not valid for device implantation, which is not significantly modified during or after hospital admission.
Treatments of Ambulatory Patients With Chronic Heart Failure
The ESC-HF Long-Term Registry provides a clear picture of the rate of use of guideline- recommended, evidence-based treatments; and detailed data on the proportion of patients in whom the target suggested dose was reached as well as the adherence to recommendations on device implantation are available. However, unlike other registries or trials, it also exposes the reasons for non-adherence to the guidelines recommendations.
As was also demonstrated in the ESC-HF Pilot Survey, the rate of use of renin–angiotensin–aldosterone system blockers (ACE inhibitors, ARBs, and MRAs) and beta-adrenergic blockers is satisfactory. This registry can expand this observation showing that, among the relatively small proportion of patients not receiving ACE inhibitors, ARBs, or beta adrenergic blockers, the rate of real undertreatment is even smaller, a documented contraindication or intolerance being clearly reported in the majority of untreated patients. A different consideration must be made for MRAs, for which the recommendation for treatment, during the period of data collection, was still limited by previous guidelines to the most severe patients (NYHA class III–IV). Considering this context, the percentage of patients for whom there is no documented explanation for non-prescription was 5.4%.
Although the rate of use of treatments with a recommendation of type 1A was confirmed to be satisfactory, the number of patients treated with appropriate doses was very far from optimal, being no more than one-third. This is a common finding of most registries or trials on HF, confirming the gap between the dosages used in randomized controlled trials, performed in selected populations carefully monitored over time, and those provided by observational research reflecting routine clinical practice. This huge 'underdosing' might be related to physician inexperience, undesired effects, co-morbidities determining polypharmacy, and then limiting the up-titration generally performed in the original trials from which the targets were derived. If the recommended target doses in the current guidelines are appropriate, we might infer that most patients are currently not treated adequately. If we take the opposite point of view, the data collected from patients enrolled in the ESC-HF Long-Term Registry might suggest that the range of doses currently used in clinical practice appropriately represents patient needs and drug tolerability in current clinical practice. The evaluation of the reasons for non-adherence to recommended dosages shows that, in the majority of cases, in whom target doses were not achieved, reasonable clinical motivations related mostly to drug up-titration still ongoing or intolerance were reported, greatly reducing the number of patients for whom no documented explanations for low dosages have been reported.
Concerning the rate of implantation of devices (a CRT and/or ICD), those patients with a clinical profile theoretically corresponding to the indication for device implantation were identified. Of the 450 patients who qualified for CRT implantation, in 272 the procedure was planned, while in 178 (40%) patients doctor uncertainties or patient refusal were reported. Similar figures have been described for ICD implantation: of 731 patients indicated for ICD, in 408 the implantation was planned, while for 323 (44%) the procedure was not planned. As expected, logistical or cost issues were also reported as the cause of non-implantation of CRT or ICD.
This is certainly a gap between recommendations and actual clinical practice that should be considered, but underuse of device implantation seems to be less than described in previous observations.
Limitations
Some important limitations of our registry must be acknowledged.
First, criteria for HF diagnosis were discussed during the investigator meetings, and the Guidelines of reference were commented on and disseminated to all investigators. However, the diagnoses were made by the investigators according to their clinical judgement and not validated centrally. Secondly, although we tried to balance the methodological need for consecutiveness of enrolment with the practical feasibility, thereby decreasing the workload for centres by limiting recruitment to 1 day per week for 12 months, we cannot prove the consecutiveness of patient enrolment. Thirdly, representativeness is often recognized as a limitation in observational studies. To lessen this problem, the centres were selected in proportion to the size of the population of the participating countries, taking into account the different technological levels of the cardiology centres invited to participate. Fourthly, the patients were all enrolled in cardiology wards and clinics, and they generally did not include those presenting at the Emergency Department, and/or admitted to other hospital facilities. Accordingly, the population reported herein does not represent the whole gamut of patients with HF.