Evolution of functional class, biochemical and echocardiographic parameters and clinical outcomes after sacubitril/valsartan initiation in daily practice

This was a retrospective and unicenter real-life study that included all patients with HF and left ventricular ejection fraction less than 40% taking sacubitril/valsartan (at least one dose) in the hospital of Denia, Alicante, Spain. Patients without complete data, inappropriate indication of prescription (i.e., preserved left ventricular ejection fraction), or withdrawal of the drug during the first 2 months without a clear reason were excluded. This study was approved by the Clinical Research Ethics Committee of the hospital.

Clinical data were collected from the medical history of the patients and during the physician interview at baseline and during follow-up visits. There was not a specific protocol, and patients were managed according to the clinical situation of the patients and physicians’ judgment.

At baseline, biodemographic data (age and sex), physical examination (blood pressure, heart rate and bodyweight), cardiovascular risk factors (hypertension, diabetes, dyslipidemia, smoking and alcoholism), concomitant conditions (history of myocardial infarction, stroke, atrial fibrillation and new-onset atrial fibrillation during the follow-up), the cause of HF, NYHA functional class and hospitalization or admission to the emergency department for HF in the previous year were recorded. Treatment for HF was also documented. Data from blood analysis (closest to the moment of starting treatment with sacubitril/valsartan, and during the follow-up) were recorded. Blood analysis included serum creatinine, urea, sodium, potassium, transaminases, glucose, HbA1c, hemoglobin, platelets count, NT-proBNP, CA 125, ferritin, transferrin saturation, serum iron and uric acid. Data from the echocardiogram (closest to the moment of starting treatment with sacubitril/valsartan and during the follow-up) that included left ventricular end-diastolic diameter, left ventricular ejection fraction, systolic pulmonary artery pressure and the presence of mitral insufficiency and its severity were also determined. All these data were analyzed in the overall study population, and according to the physician that attended the patient (HF unit vs other clinical settings) and the initial dose prescribed of sacubitril/valsartan.

The initial and the final doses of sacubitril/valsartan were recorded in the overall study population and according to the physician that started treatment with this drug. The reasons for prescribing sacubitril/valsartan 24/26 mg at baseline were specifically analyzed.

The evolution of functional class, physical examination, treatment with diuretics, biochemical and echocardiographic parameters during the study according to the physician that started treatment with sacubitril/valsartan and the final dose prescribed of the drug was analyzed.

Due to the retrospective nature of the study, measurements were taken in different time points. However, in order to standardize as much as possible data collection, three different time points were considered: baseline: last measurement of every single variable collected in the six previous months before sacubitril/valsartan initiation, regardless of whether it was carried out during the hospitalization or in an outpatient basis, and regardless the clinical setting (general cardiology, HF unit, internal medicine, hospitalization, etc.) where the drug was initially prescribed; first visit: first measurement of every single variable after sacubitril/valsartan initiation, regardless the clinical setting or the time after drug prescription; study end: last measurement of every single variable before the study end, regardless the clinical setting or the time after drug prescription. Baseline values were compared with those at study end.

Any side effect and specific adverse events during treatment with sacubitril/valsartan in the overall study population and according to the physician that started treatment with this drug and according to the final dose prescribed were determined.

Events during treatment with sacubitril/valsartan in the overall study population and according to the physician that started treatment with this drug and according to the final dose prescribed were analyzed. Rates of outcomes (any event, death from any cause, cardiovascular death, sudden cardiac death, any hospitalization and hospitalization for HF) were independently analyzed, but also the variable composed of any event, cardiovascular death or hospitalization for HF and the variable composed of any event, cardiovascular death, hospitalization for HF or decompensated HF.

For the descriptive analysis, quantitative variables were described with measures of central tendency (mean) and dispersion (standard deviation) and qualitative variables as absolute frequency (n) and proportions (%). When two means at baseline and during follow-up were compared, paired sample t-test was used. Categorical variables were compared with the χ² test or the Fisher exact test, according to the sample size. Outcomes during treatment with sacubitril/valsartan were calculated as events per 100 patient-years. Statistical significance was set at a p-value < 0.05. The statistical analysis was performed using SPSS 23.0 (SPSS, Inc., IL, USA).

From June 2016 to June 2018, 167 patients started treatment with sacubitril/valsartan. Of these, 18 patients were excluded (3 for incomplete data, 2 for inappropriate indication and 13 for premature discontinuation). Finally, a total of 149 patients were included in the study (mean duration of treatment of 316.1 ± 155.9 days); 42 patients (28.2%) were attended in the HF unit and the remaining 107 patients (71.8%) in other clinical settings (97 patients [65.1%] in general cardiology, 6 patients [4.0%] in internal medicine, 3 patients [2.0%] in primary care and 1 patient [0.7%] in nephrology).

Overall, at baseline, mean age was 70.7 ± 9.6 years, 69.8% of patients were men, 73.8% of patients had hypertension, 49.0% previous myocardial infarction and 53.7% had no history of atrial fibrillation. Of these patients, 8.9% developed new-onset atrial fibrillation during the follow-up. Ischemic heart disease was the most common cause of HF, followed by idiopathic cardiomyopathy. Totally, 26.8% patients had been hospitalized for HF in the previous year. With regard to the treatment for HF at baseline, 89.9% of patients were taking angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, 91.9% beta blockers, 79.9% aldosterone antagonists and 73.2% diuretics. Mean NT-proBNP was 3884 ± 4871 pg/ml and CA 125 25.6 ± 39.9 U/ml. Mean left ventricular ejection fraction was 31.2 ± 7.0% and 33.1% had moderate to severe mitral insufficiency (Table 1).

Baseline clinical characteristics of the study population were compared according to the physician that started treatment with sacubitril/valsartan (HF unit vs no HF unit). Those patients attended in the HF unit were younger (65.9 ± 10.1 vs 72.6 ± 8.8; p < 0.001), more commonly men (83.3% vs 64.5%; p = 0.01) and ischemic heart disease was the most common cause of HF (61.9% vs 55.1%; p = 0.01). Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, ivabradine, diuretics and implantable cardioverter–defibrillators were more commonly prescribed among patients attended in the HF unit than in other clinical settings, whereas NT-proBNP levels (2566.0 ± 3208.2 vs 4544.2 ± 5417.8 pg/ml; p = 0.02) were lower in those patients attended in the HF unit, CA 125 levels were higher (37.4 ± 54.3 vs 18.2 ± 25.6 pg/ml; p = 0.003). No other significant differences were found between both groups.

At baseline, 83.9% of patients were taking sacubitril/valsartan at a dose of 24/26 mg, 15.4% at a dose of 49/51 mg and 0.7% at a dose of 97/103 mg. At the end of the follow-up, these numbers moved to 38.9, 39.6 and 12.8%, respectively. Totally, 8.7% patients discontinued treatment with sacubitril/valsartan during the follow-up. Among the 125 patients that started treatment with sacubitril/valsartan at a dose of 24/26 mg, 90.4% met at least one criterion recommended for beginning with this dose. These data did not significantly differ between patients attended in the HF unit or other clinical settings.

Overall, mean duration of treatment with sacubitril/valsartan was 316.1 ± 155.9 days. With regard to the evolution of different clinical, biochemical and echocardiographic parameters during the follow-up (Table 2 & Figure 1), systolic and diastolic blood pressure significantly decreased (from 129.5 ± 16.3/78.9 ± 13.7 to 117.8 ± 15.5/73.0 ± 9.8 mmHg, respectively; p < 0.001), as well as the use of diuretics (from 73.2% to 57.4%; p = 0.008); NYHA functional class (from 2.3 ± 0.6 to 1.8 ± 0.5; p < 0.001) significantly improved; potassium levels significantly increased (from 4.47 ± 0.5 to 4.73 ± 0.4; p < 0.001), NT-proBNP (from 3884 ± 4871 to 1975.3 ± 3006.6 pg/ml; p = 0.0001), and CA 125 levels (from 25.6 ± 39.9 to 11.3 ± 21.5 U/ml; p = 0.0001); left ventricular end-diastolic diameter (from 59.3 ± 9.3 to 57.1 ± 7.6 mm; p = 0.02) and systolic pulmonary artery pressure (from 39.8 ± 12.4 to 35.3 ± 7.4 mmHg; p = 0.0002) significantly decreased and left ventricular ejection fraction significantly increased (from 31.2 ± 7.0 to 37.3 ± 10.5%; p < 0.001). In general, the positive evolution of the majority of these parameters occurred in both, patients attended in HF unit and other clinical settings and with all doses of sacubitril/valsartan. The increase of left ventricular ejection fraction and the decrease of NT-proBNP levels were slightly higher in patients with nonischemic cardiomyopathy than in patients with ischemic heart disease (from 28.8 ± 7.5% to 38.4 ± 12.9%; p < 0.001 and from 32.9% ± 6.0% to 36.3 ± 7.5%; p = 0.001, respectively and from 4279.4 ± 5130.3 to 1912.8 ± 3696.5 pg/ml; p = 0.003 and from 3527.8 ± 4638.1 to 2043.2 ± 2064.8 pg/ml; p = 0.007, respectively). In those patients who discontinued treatment with sacubitril/valsartan, NT-proBNP levels and left ventricular ejection fraction did not significantly change during the follow-up.

With regard to side effects associated with treatment with sacubitril/valsartan (Table 3), 21.5% of patients reported at least one adverse event, being hypotension (11.4%) and dizziness (4.0%) the most common ones. The proportion of patients with any side effect was independent of the clinical setting in which patients were attended. All patients that discontinued treatment with sacubitril/valsartan had at least one side effect, mainly hypotension, and renal insufficiency (Figure 2), and among patients who remained on sacubitril/valsartan, the frequency of any side effect was more common in those patients taking sacubitril/valsartan 24/26 mg (Table 3).

Events during treatment with sacubitril/valsartan were reported in Table 4. During the study period, the combined variable composed of any event, cardiovascular death, hospitalization for HF or decompensated HF occurred in 17 patients (11.4%; 13.2 events per 100 patient-years). No cardiovascular death, or sudden cardiac death occurred during the follow-up. There was a trend towards more events (the combined variable of any event, cardiovascular death, hospitalization for HF or decompensated HF) among patients taking sacubitril/valsartan 24/26 mg or those that discontinued treatment compared with patients taking sacubitril/valsartan 49/51 or 197/203 mg, that reached statistical significance when patients taking low doses of sacubitril/valsartan were combined with those patients that discontinued treatment compared with patients taking sacubitril/valsartan at moderate or high doses (16.9 vs 6.4%, p = 0.04). Rates of events were independent of the clinical setting in which patients were attended.

In the PARADIGM-HF trial [7], to be included, patients were required to take a stable dose of renin angiotensin system inhibitors (equivalent to at least 10 mg of enalapril daily). In addition, patients were excluded if they had a systolic blood pressure <100 mmHg at screening or 95 mmHg at randomization, an estimated glomerular filtration rate below 30 ml/min/1.73 m², a serum potassium level >5.2 mmol/l at screening (or >5.4 mmol/l at randomization) or unacceptable side effects during treatment with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. This means that in clinical practice, many patients with HF and reduced ejection fraction may not meet inclusion/exclusion criteria of the PARADIGM-HF trial. Thus, it has been reported that only 25% of real-world population with HF and reduced ejection fraction are eligible for sacubitril-valsartan according to the PARADIGM-HF criteria [13,23]. In addition, the adoption of sacubitril/valsartan in ‘real-life’ patients is slow [24], and this may limit the beneficial effects of sacubitril/valsartan in HF population in routine practice. This lack of data makes necessary the publication of observational studies that evaluate whether the results of the PARADIGM-HF trial may be extrapolated to the overall HF study population.

In our study, patients from different clinical settings (i.e., HF unit, general cardiology, internal medicine, primary care and nephrology) were included, and this makes our data more representative of the whole HF population. Our patients were old, 70% were men, the presence of cardiovascular risk factors and other comorbidities was common and ischemic heart disease was the main cause of HF. Compared with the PARADIGM-HF trial [7], our patients were older (71 vs 64 years), systolic blood pressure and NT-proBNP levels were higher (130 vs 122 mmHg and 3884 vs 1631 pg/ml, respectively). The majority of the studies that have analyzed the use of sacubitril/valsartan in clinical practice have included patients with a similar clinical profile than those included in the pivotal trial [17,20,22,25], and only a small number of studies have included patients with a worse clinical profile. Thus, in a prospective study in 10 hospitals with 427 patients, subjects had a higher risk profile than in the pivotal trial, poorer functional class, higher levels of natriuretic peptides and worse renal function [18]. In addition, in our study, the clinical profile of patients according to the clinical setting in which patients were attended was also analyzed. Patients attended in the HF unit were younger, more commonly men and the use of guideline-directed medical therapies was higher than in other clinical settings. As a result, our data may provide relevant information about the effectiveness and safety of sacubitril/valsartan in more complex patients attended by different specialties and clinical scenarios.

In our study, in the great majority of patients (84%), the initial dose of sacubitril/valsartan was 24/26 mg, and at the end of the follow-up, 39% remained on this dose. As our study showed, there were some relevant differences regarding the clinical profile of patients according to the initial dose prescribed of sacubitril/valsartan. Remarkably, this dose was not tested in the PARADIGM-HF trial [7]. Thus, in the PARADIGM-HF trial [7], during the run-in period, patients received enalapril (at a dose of 10 mg twice daily) for 2 weeks and if no unacceptable side effects occurred, this regimen was followed by treatment with sacubitril/valsartan for an additional 4–6 weeks (initially at a dose of 49/51 mg twice daily, which was increased to 97/103 mg twice daily). Since no specific Phase III clinical trial has been developed for the lowest dose of sacubitril/valsartan, the effectiveness and safety of this dose can only be analyzed in observational studies at this moment. According to the summary of product characteristics [26], those patients with HF and reduced ejection fraction that are naive for sacubitril/valsartan, receive low doses of angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, have an estimated glomerular filtration rate <60 ml/min/1.73 m², a systolic blood pressure 100–110 mmHg, or moderate hepatic insufficiency (or AST/ALT levels >2 upper limit of normality [ULN]) should start with sacubitril/valsartan at a dose of 24/26 mg. In our study, more than 90% of patients that started treatment with this dose met at least one of these criteria. In these patients, treatment should be titrated gradually to facilitate achieving the target dose of sacubitril/valsartan [27]. Some titration algorithms have been proposed for reaching the maximal dosage with a good tolerability profile [28].

In the PARADIGM-HF trial, compared with enalapril, sacubitril/valsartan improved functional class and health-related quality of life outcomes [29,30]. In our study, treatment with sacubitril/valsartan was associated with a significant symptom improvement (NYHA functional class improved from 2.3 to 1.8; p < 0.001), that was independent of the clinical setting in which patients were attended, and that occurred with all doses of sacubitril/valsartan. Similarly, other studies have shown that sacubitril/valsartan relieves symptoms, improves functional class and physical capability in routine practice [16,19,22,31,32]. Thus, a recent study has shown that after only 1 month of therapy with sacubitril/valsartan 97/103 mg, the mean distance walked on the 6-min walk test increased from 129 to 436 m [31].

Although some physicians may consider that many outpatients with HF are stable, the fact is that HF has a progressive nature [33]. Experimental data have shown that sacubitril/valsartan attenuates cardiac hypertrophy, fibrosis, vasculopathy, inflammation, oxidative stress and mitochondrial depletion/dysfunction [34,35]. This translates into a reduction of cardiac filling pressures and a beneficial reverse remodeling of both, systolic as diastolic function in patients with HF and reduced ejection fraction [36,37]. A small study showed that after only 3 months of treatment, sacubitril/valsartan was associated with a significant increase of left ventricular ejection fraction and a significant reduction of left ventricular diameters. In addition, the increase in ejection fraction tended to be marginally greater in the medium/high dose group as compared with the low dose group [38]. In our study, NT-proBNP levels significantly decreased, and echocardiographic parameters significantly improved. Left ventricular improvement was higher as the dose of sacubitril/valsartan increased but left ventricular ejection fraction remained unchanged among those patients that discontinued treatment with sacubitril/valsartan. These data emphasize the importance of remaining on treatment with sacubitril/valsartan when possible. Increase in CA 125 levels have been associated with adverse outcomes in patients with HF [39]. In our study, overall, CA 125 significantly decreased by 56%, and this decrease tended to be higher as the dose of sacubitril/valsartan increased.

In the PARADIGM-HF [7], during the run-in period, 12% of the patients withdrew treatment because of an adverse event (most frequently cough, hyperkalemia, renal dysfunction or hypotension). In our study that included patients without a run-in period, but the majority started with low doses, less than 9% of patients discontinued treatment with sacubitril/valsartan. Other studies have confirmed the safety of sacubitril/valsartan in routine practice [40–42].

Despite our cohort patients was old and had many comorbidities, the number of events that occurred after almost 1 year of treatment was low. Thus, during this period, the combined variable composed of any event, cardiovascular death, hospitalization for HF or decompensated HF occurred only in 17 patients (13.2 events per 100 patient-years), and no cardiovascular death occurred. The majority of events were related with hospitalizations for HF. Remarkably, in our study, although 26.8% of patients had been hospitalized for HF in the year before taking sacubitril/valsartan, only 5.4% of patients were hospitalized for HF during treatment with sacubitril/valsartan. During this period, the HF unit was established, and this could have had an impact on the reduction of outcomes. On the other hand, events were more common among patients taking sacubitril/valsartan 24/26 mg or discontinued treatment than patients taking moderate or high doses of sacubitril/valsartan. Recent studies have shown that sacubitril/valsartan is associated with a significant and early reduction in incident HF-hospitalizations in real-world patients with HF and reduced ejection fraction, but higher doses might be associated with more reduction in HF admissions [42,43]. In our study, no sudden cardiac deaths were reported. A reduction of sudden cardiac death was observed in the PARADIGM-HF [7] with sacubitril/valsartan compared with enalapril. Remarkably, it has recently been reported that in patients with HF and reduced left ventricular ejection fraction and an implantable cardioverter–defibrillator with remote monitoring, sacubitril/valsartan significantly decreases ventricular arrhythmias and appropriate implantable cardioverter–defibrillator shocks [44].

This study has some limitations. First, the study had a retrospective design without a comparator group. As a result, some bias could not be controlled. However, the aim of the study was not to ascertain causality between sacubitril/valsartan and changes of different variables over time, but to describe sacubitril/valsartan use in real-world practice. In this context, the retrospective design is the best to ascertain the clinical management of patients as physicians and patients are not influenced by participating in the study. Second, although the sample size was quite small, it was sufficient to respond the main objectives of the study. In addition, our results were in line with previous publications, which showed an improvement in clinical measurements and laboratory data after sacubitril/valsartan prescription in daily practice, as well as good drug tolerability. Third, the healthcare area of the hospital of Denia includes a significant proportion of foreign patients, not only from different countries but also from other parts of Spain. In some cases, sacubitril/valsartan was started but the patient moved to another city. Due to the retrospective design of our study, we were unable to collect reliable data of these patients, and we decided to exclude them from the analysis, as we preferred to analyze only those patients with complete data in order to assure a higher robustness of the results. Fourth, although the measures were taken at different relative times, the standardization of the protocol made that the variability of the data was small, but the robustness of the results high. Finally, this was a unicenter study. Although this may assure a higher homogeneity in the management of these patients, this could limit the generalization of the results.

In conclusion, in this cohort of ‘real-life’ patients with HF and reduced ejection fraction, improvements in NYHA functional class, surrogate variables such as NT-proBNP, CA 125 or left ventricular ejection fraction were observed after the initiation of sacubitril/valsartan. The safety profile of sacubitril/valsartan was remarkable. In addition, cardiovascular events were low, mainly associated with HF hospitalizations. Up to 39% of patients remained on the lowest dose of sacubitril/valsartan during the follow-up. More efforts should be performed to uptitrate the dose of sacubitril/valsartan as recommended by guidelines.

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing and editorial assistance was provided by Content Ed Net (Madrid, Spain), with funding from Sociedad Valenciana de Cardiología.

The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations.