Use of sacubitril/valsartan in patients with heart failure: evidence from the real world

ABSTRACT

Background

To characterize the use of sacubitril/valsartan in a group of patients with heart failure in Colombia.

Research design and methods

Follow-up study of patients with heart failure who started sacubitril/valsartan and were affiliated with the Colombian health system between 2019 and 2021. Sociodemographic, clinical, and pharmacological variables and adherence and persistence of use were identified.

Results

A total of 514 patients were identified, with a mean age of 65.7 years, 73.7% of whom started sacubitril/valsartan at low doses, and only 12.5% reached the maximum dose. Adherence was 78.2% and persistence was 56.8% at 1 year of follow-up. The increase in systolic blood pressure (odds ratio (OR): 1.01; 95% CI: 1.00–1.03) and the use of β-blockers (OR: 2.63; 95% CI: 1.42–4.85) were correlated with a greater persistence, while receiving furosemide (OR: 0.59; 95% CI: 0.39–0.89) and not having received renin – angiotensin – aldosterone system inhibitors in the 3 months before starting sacubitril/valsartan (OR: 0.48; 95% CI: 0.31–0.76) were associated with lower persistence.

Conclusions

The persistence of treatment 1 year after starting sacubitril/valsartan was not high, and a small proportion of patients reached the target dose of the drug. Nontitration of the drug dose was common.

KEYWORDS:

• Heart failure
• sacubitril valsartan drug combination
• medication adherence
• pharmacoepidemiology
• Colombia

1. Introduction

Heart failure is a clinical syndrome with symptoms and signs caused by a structural or functional cardiac abnormality. It is associated with elevated levels of natriuretic peptide and objective evidence of pulmonary or systemic congestion [1]. This pathology leads to high morbidity and mortality, poor functional capacity, low quality of life, and high costs of care [2]. The global number of heart failure cases is estimated to be 64.3 million, with a prevalence of 831.0 cases per 100,000 people [3]. Its prevalence has been increasing due to the aging of the population, improved treatment of and survival from ischemic heart disease, and the availability of effective, evidence-based therapies that prolong the life of patients [2]. Even so, mortality at 1 year can be 15–30% and at 5 years 50–75% [2].

The treatment of heart failure has undergone important changes [4]. Sacubitril/valsartan is the first approved drug that increases serum levels of natriuretic peptides by inhibiting neprilysin and regulating the renin – angiotensin–aldosterone system by blocking AT1 receptors [5]. The clinical information on this innovative drug is based primarily on the Prospective Comparison of ARNI with angiotensin-converting enzyme inhibitor to Determine Impact on Global Mortality and Morbidity in Heart Failure (PARADIGM-HF) clinical trial, which included patients with heart failure with reduced ejection fraction (HFrEF) and demonstrated a 20% reduction in the primary composite outcome of hospitalization for heart failure and cardiovascular mortality compared to enalapril [6]. Other clinical trials have also shown the efficacy and safety of sacubitril/valsartan in the management of patients with heart failure [7].

According to clinical practice guidelines, sacubitril/valsartan is indicated as first-line therapy in patients with HFrEF or New York Heart Association (NYHA) functional class II-III [4]. However, real-world evidence has shown that the characteristics of these patients differ from those found in clinical trials [8,9]. Studies with real-world evidence allow for longer follow-up and provide information on effectiveness in a less controlled setting, including the complexities of less strict management and lower adherence rates than those of a clinical trial [10]. This can lead to wider variations in clinical outcomes [11].

The Colombian health system offers universal coverage to the entire population through two affiliation regimes: the contributory regime that is paid for by workers and employers and the subsidized regime that is responsible for the insurance of all people who cannot pay, which also includes a benefit plan that involves a significant number of medications approved for the management of heart failure, including sacubitril/valsartan. The only study with real-world evidence on sacubitril/valsartan conducted in Colombia included only 56 patients [12]. The lack of information on this topic in the rest of Latin America is evident [8,9]. The present study aims to characterize the usage of sacubitril/valsartan in a group of heart failure patients affiliated with the Colombian health system.

2. Patients and methods

2.1. Study design and patients

A retrospective follow-up study was carried out on a cohort of patients with a diagnosis of heart failure who began pharmacological management with sacubitril/valsartan. The cases were identified from a population-based drug dispensing database that collects information from approximately 9.2 million people affiliated with the Colombian health system. Included were patients affiliated with an insurer, who numbered approximately 3.8 million people distributed in most regions of the country and affiliated with the contributory (85.0%) and subsidized (15.0%) regimes of the Colombian health system. Out of all the patients who were receiving sacubitril/valsartan in the dispensation database of Audifarma SA, those aged 18 or over, of any sex and city of residence, who started the drug between 1 July 2019 and 30 June 2020 were selected. The first dispensation was set as the index date for each patient, and from there, follow-up was carried out for 1 year (latest: 1 July 2021) or until the drug was definitively suspended.

Subjects without medical records and those who were prescribed sacubitril/valsartan during the year before the index date were excluded. For the selected patients, we proceeded to review the electronic medical records that were completed during the observation and follow-up period (1 year).

2.2. Variables

Based on the information obtained, a database was designed for gathering the following groups of variables:

1. Sociodemographic: sex, age, education, occupation, affiliation regime, and department of residence. Departments are the regions of Colombia designated by the National Administrative Department of Statistics: Bogotá-Cundinamarca, Caribbean, Central, Eastern, Pacific, and Amazon-Orinoquía.

2. Clinical:

   • Vital signs: systolic blood pressure, diastolic blood pressure, and heart rate at the index date and at 3 months.

   • Anthropometric: weight, height, and body mass index (BMI). According to BMI, the patients were classified as normal (18.5-24.9 kg/m²), overweight (25.0-29.9 kg/m²), or obese (≥30.0 kg/m²).

   • Comorbidities: cardiovascular, renal, endocrine, respiratory, rheumatological, and digestive pathologies were identified. The age-adjusted Charlson comorbidity index was calculated.

   • Classification of heart failure: Heart failure was classified according to etiology (ischemic, nonischemic), left ventricular ejection fraction (LVEF) (reduced: LVEF ≤ 40%; mildly reduced: LVEF 41-49%; preserved: LVEF ≥ 50%), NYHA functional class (I, II, III, IV), and American Heart Association (AHA) disease state (A, B, C, D) [4].

   • Exacerbations: number of exacerbations of heart failure during the year before the index date and during the follow-up period.

   • Use of implantable cardiac devices: cardioverter, resynchronizer, or pacemaker.

   • Cardiovascular risk: Cardiovascular risk was calculated according to the Framingham scale calibrated for Colombia, where the original result was multiplied by 0.75 [13]. Patients were also classified according to the 2019 ESC/EAS Guidelines for the Management of Dyslipidemias into very high-risk, high-risk, moderate-risk, and low-risk groups [14].

3. Paraclinical:

   • Echocardiogram: LVEF before the index date and at the end of the follow-up

   • Clinical laboratory: lipid profile (total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides), glycated hemoglobin (HbA1c), serum electrolytes (sodium, potassium), natriuretic peptides (brain natriuretic peptide -BNP-, N-terminal proBNP), and creatinine. The glomerular filtration rate (GFR) was calculated using the chronic kidney disease epidemiology collaboration (CKD-EPI) 2021 equation at baseline and at 3 months.

4. Pharmacological (sacubitril/valsartan):

   • Initial prescribing physician: general practitioner, family physician, or specialist (internal medicine, geriatrics, cardiology, others).

   • Presentation: coated tablet 50 mg, 100 mg, or 200 mg.

   • Dosage: Initial dose and maximum dose reached. The outcome was achieving a dose of 200 mg twice daily.

   • Adherence was determined with the formula of the medication possession rate (MDR) (days of supply of medication dispensed/days from the first dispensing to the end of the follow-up period × 100).

   • Persistence was defined as the continuous use of sacubitril/valsartan during the year of follow-up and with an interval between dispensations of no more than 90 days (time between the completion of the medication and the next dispensation).

   • Indications: use in patients with or without prescription of renin – angiotensin–aldosterone system inhibitors, according to the LVEF, the NYHA functional class, and AHA disease stage.

   • Adverse drug reactions: hypotension, hyperkalemia, increased creatinine, cough, angioedema, dizziness, itching, rash, and others.

   • Other pharmacological therapies for heart failure include β-blockers (metoprolol, carvedilol, bisoprolol, others), mineralocorticoid receptor antagonists (MRAs) (spironolactone, eplerenone), diuretics (furosemide, others), sodium/glucose cotransporter 2 inhibitors (empagliflozin, dapagliflozin), and others (ivabradine, digoxin).

   • Previous use of angiotensin-converting enzyme inhibitors (ACEIs) (enalapril, captopril, others) or angiotensin II receptor antagonists (ARA II) (losartan, valsartan, candesartan, others).

   • Comedications: grouped into the following categories: a) analgesic, b) lipid-lowering, c) antiplatelet, d), anticoagulant, e), antiulcer, and f) antidiabetic drugs, among others.

2.3. Ethics statement

The protocol was approved by the Bioethics Committee of the Universidad Tecnologica de Pereira in the category of ‘risk-free research’ (approval code: 27–050421) and by the ethics committee of the insurer. The principles of confidentiality of information established by the Declaration of Helsinki were respected.

2.4. Data analysis

The data were analyzed with the statistical software SPSS Statistics, version 26.0 for Windows (IBM, U.S.A.). Qualitative variables are written as frequencies and proportions, and quantitative variables are written as measures of central tendency and dispersion (mean and standard deviation). Categorical variables were compared by the χ² test or Fisher’s exact test. Quantitative variables were compared by Student’s t test. Multivariate binary logistic regression models were built that included the significant variables in the bivariate analyses, as well as those that could plausibly be associated with the outcomes of reaching the target dose of 200 mg twice daily (yes/no) and persistence at 1 year (yes/no). The Hosmer‒Lemeshow test was performed to describe the goodness of fit. The predictive capacity of the model was determined according to the area under the curve. Statistical significance was accepted at p < 0.05. A bivariate analysis was performed to compare our results with the PARADIGM-HF results [6].

3. Results

3.1. Sociodemographic

A total of 514 patients living in 21 different cities were identified in the Audifarma database. The majority were men older than 65 years (range: 22.0–98.0 years). A total of 11.3% (n = 58) were <50 years, 34.0% (n = 175) were 50–64 years, 40.5% (n = 208) were between 65 and 79 years, and 14.2% (n = 73) were 80 years or older. The distribution according to region, education, occupation, and affiliation regime is given in Table 1.

Table 1. Sociodemographic characteristics of a group of patients with heart failure treated with sacubitril/valsartan, Colombia.

Variables	n = 514	%
Men	346	67.3
Women	168	32.7
Age, mean (SD)	65.7 ± 13.0
≥65 years	281	54.7
Origin	-	-
Bogotá-Cundinamarca region	172	33.5
Caribbean region	109	21.2
Central region	107	20.8
Eastern-Amazon-Orinoquía region	74	14.4
Pacific region	52	10.1
Scholarship	-	-
Primary	104	20.2
Secondary	104	20.2
University	64	12.5
Unknown	242	47.1
Occupation	-	-
Household activities	91	17.7
Pensioner	64	12.5
Driver	28	5.4
Others (n = 114)	243	47.3
Unknown	88	17.1
Affiliation regime	-	-
Contributory	454	88.3
Subsidized	60	11.7

SD:Standard Deviation.

3.2. Clinical

Most of the patients had heart failure of ischemic etiology, with reduced LVEF, AHA stage C, and NYHA functional class III (Table 2). Some 24.1% (n = 124) had an implantable device, most often a pacemaker (n = 66; 12.8%), followed by a cardioverter-defibrillator (n = 47; 9.1%) and resynchronizer (n = 13; 2, 5%). A total of 13.0% (n = 67) of the patients had exacerbations of heart failure during the year before the index date, while during the year of follow-up exacerbations occurred in 5.3% (n = 27) of the patients (p < 0.001). The average Charlson comorbidity index was 4.9 ± 2.2 points. The most common pathological antecedents were arterial hypertension, myocardial infarction, and diabetes mellitus (Table 2).

Table 2. Clinical and paraclinical characteristics of a group of patients with heart failure treated with sacubitril/valsartan, Colombia.

Variables	n = 514	%
Vital signs	-	-
Systolic blood pressure (mmHg), mean (SD)	115.9 ± 13.2
Diastolic blood pressure (mmHg), mean (SD)	72.5 ± 9.7
Heart rate (bpm), mean (SD)	74.3 ± 8.0
Anthropometric measures	-	-
Weight (kg), mean (SD)	71.8 ± 15.4
Body mass index (Kg/m2), mean (SD)	26.9 ± 5.0
Comorbidities	-	-
Arterial hypertension	460	89.5
Myocardial infarction	276	53.7
Diabetes mellitus	198	39.5
Dyslipidemia	145	28.2
Hypothyroidism	123	23.9
Charlson Comorbidity Index	-	-
1 point	28	5.4
2 points	51	9.9
≥3 points	435	84.6
Etiology of heart failure	-	-
Ischemic	320	62.3
Non ischemic	165	32.1
Unknown	29	5.6
NYHA functional class	-	-
I	34	6.6
II	145	28.2
II	240	46.7
IV	59	11.5
Unknown	36	7.0
AHA	-	-
A	0	0.0
B	34	6.6
C	400	77.8
D	44	8.6
Unknown	36	7.0
Left ventricular ejection fraction	-	-
≤40%	396	77.0
41–49%	42	8.2
≥50%	40	7.8
Unknown	36	7.0
Clinical laboratory	-	-
Total cholesterol (mg/dL), mean (SD); n = 316	160.9 ± 43.8
LDL cholesterol (mg/dL), mean (SD); 315	91.0 ± 38.4
HDL cholesterol (mg/dL), mean (SD); n = 314	40.6 ± 11.3
Triglycerides (mg/dL), mean (SD); 312	153.6 ± 76.8
Glycosylated hemoglobin (%), mean (SD); n = 125	6.5 ± 1.2
Potassium (mEq/L), mean (SD); n = 189	4.5 ± 0.5
Sodium (mEq/L), mean (SD); n = 41	129.9 ± 32.8
Creatinine (mg/dL), mean (SD); n = 376	1.17 ± 0.5
GFR (mL/min/1.73 m2); mean (SD)	68.4 ± 21.2

SD:Standard Deviation; NYHA: New York Heart Association; AHA: American Heart Association; GFR: glomerular filtration rate.

The average cardiovascular risk was 15.1 ± 6.4%. The majority had a very high (n = 357; 69.5%) or high cardiovascular risk (n = 59; 11.5%). Among the patients who continued taking sacubitril/valsartan, the mean systolic blood pressure (116.3 vs. 116.4; p = 0.906), diastolic blood pressure (72.3 vs. 73.1; p = 0.175), and frequency of cardiac arrest (74.5 vs. 74.7; p = 0.728) were similar between baseline and 3 months.

3.3. Paraclinical

The mean baseline LVEF was 32.4 ± 12.2%, and according to the baseline GFR, the mean stage of chronic kidney disease was 2. Table 2 lists some clinical laboratory data, such as the results of the lipid profile, HbA1c, electrolytes, and creatinine. In the subgroup of 52 patients who continued to receive sacubitril/valsartan and who had controlled creatinine, we compared the baseline GFR with that at 3 months of follow-up, without finding significant differences between them (63.4 mL/min/1.73 m² vs. 61.8 mL/min/1.73 m²; p = 0.702).

3.4. Pharmacological

Sacubitril/valsartan was prescribed mainly by cardiologists, especially to patients with heart failure who had previously been managed with an ACEI or ARA II (Table 3). Most started with a dose of 50 mg twice a day (n = 379; 73.7%), and only 12.5% (n = 64) reached the maximum dose (Figure 1; modified from Park et al [15]). The mean drug possession rate was 78.2 ± 20.8%, and 56.8% (n = 292) persisted with treatment 1 year after initiation. The causes of nonpersistence were mostly unknown (n = 171/222; 77.0%), followed by administrative problems (n = 23; 10.4%), poor patient adherence (n = 17; 7.7%), and intolerance to therapy due to adverse drug reactions (n = 11; 4.9%). Adverse drug reactions were recorded in the clinical history of 8.4% (n = 43) of the patients (Table 3). In addition to sacubitril/valsartan, the patients used an average of 2.4 ± 1.1 medications for the management of heart failure, mainly β-blockers and MRAs (Table 3). The most commonly used management regimens were the combination of angiotensin receptor/neprilysin inhibitor (ARNI) + β-blocker + MRA (n = 128; 24.9%) and ARNI + β-blocker + MRA + furosemide (n = 122; 23.7%).

Figure 1. Dose titration patterns of sacubitril-valsartan in the 12-month follow-up. Figure modified from Park et al. [15].

Table 3. Pharmacological characteristics of a group of patients with heart failure treated with sacubitril/valsartan, Colombia.

Variables	n = 514	%
Prescribing physician	-	-
Cardiologist	179	34.8
Internist/Geriatrician	112	21.8
Familiar	24	4.7
Unknown	199	38.7
Indication	-	-
Heart failure with previous use of ACE inhibitors or ARBs	374	72.8
Heart failure without prior use of ACE inhibitors or ARBs	140	27.2
Starting dose	-	-
50 mg once a day	21	4.1
50 mg twice a day	379	73.7
100 mg twice a day	90	17.5
200 mg twice a day	24	4.7
Maximum dose	-	-
50 mg once a day	1	0.2
50 mg twice a day	267	51.9
100 mg twice a day	182	35.4
200 mg twice a day	64	12.5
Months of treatment	-	-
≤3 months	105	20.4
4–6 months	79	15.4
7–9 months	150	29.2
10–12 months	180	35.0
Adverse drug reactions	-	-
Hyperkalemia	18	3.5
Increased creatinine	13	2.5
Arterial hypotension	10	1.9
Skin rash	3	0.6
Others (angioedema, cough, dizziness)	3	0.6
Other heart failure therapies	-	-
β-blocker	452	87.9
Mineralocorticoid receptor antagonists	379	73.7
Diuretics	246	47.9
Sodium/glucose cotransporter 2 inhibitors	59	11.5
Digital	44	8.6
Ivabradine	29	5.6
Other comedications	-	-
Lipid-lowering	429	83.5
Antiulcer	336	65.4
Antiplatelet agents	334	65.0
Analgesics and anti-inflammatories	302	58.8
Anticoagulants	242	47.1

ACEI: angiotensin converting enzyme inhibitor; ARBs: Angiotensin II receptor blockers.

3.5. Multivariate analysis

Multivariate analyses were adjusted for sex, age, origin, heart failure characteristics, comorbidities, and pharmacological variables (Tables 4 and 5). The binary logistic regression showed that the increase in systolic blood pressure and the concomitant use of β-blockers were correlated with a greater persistence of use at 1 year, while the concomitant use of furosemide and not having taken an ACEI or ARA II in the 3 months before the start of sacubitril/valsartan reduced this probability (Hosmer‒Lemeshow test p = 0.940, area under the curve = 0.678) (Table 4). Patients who lived in the Bogotá-Cundinamarca region and those who persisted for 1 year with sacubitril/valsartan were more likely to reach the maximum dose of the drug, while initiating therapy at low doses reduced this probability (Hosmer‒Lemeshow test p = 0.968, area under the curve = 0.826) (Table 5).

Table 4. Binary logistic regression of variables related to persistence at one year, in a group of patients with heart failure treated with sacubitril/valsartan, Colombia.

			95%CI
Variables	Sig.	OR	Lower	Upper
Men (yes/no)	0.079	1.456	0.958	2.213
Age <50 years	0.166	Reference	Reference	Reference
Age 50–64 years	0.237	1.699	0.705	4.093
Age 65–79 years	0.065	2.462	0.946	6.408
Age ≥80 years	0.054	2.844	0.983	8.229
Bogotá-Cundinamarca region (yes/no)	0.287	0.800	0.531	1.206
Charlson comorbidity index ≥ 3 points (yes/no)	0.596	0.801	0.352	1.823
Arterial hypertension (yes/no)	0.053	0.509	0.257	1.009
Chronic kidney disease (yes/no)	0.369	0.811	0.513	1.282
Very high cardiovascular risk (yes/no)	0.912	1.026	0.645	1.634
Systolic blood pressure (continuous)	0.038	1.016	1.001	1.031
Body mass index (continuous)	0.178	1.029	0.987	1.072
Heart failure with reduced LVEF (yes/no)	0.082	0.614	0.355	1.064
NYHA functional class III or IV (yes/no)	0.083	1.508	0.948	2.397
Implantable cardiac devices (yes/no)	0.336	1.252	0.792	1.979
No previous ACEI or ARBs (yes/no)	0.002	0.487	0.311	0.763
Start sacubitril/valsartan 50 mg twice daily (yes/no)	0.474	1.173	0.758	1.815
Adverse drug reactions (yes/no)	0.748	0.894	0.452	1.767
Concomitant use of β-blockers (yes/no)	0.002	2.632	1.425	4.859
Concomitant use of minieralocorticoid receptor antagonists (yes/no)	0.229	1.332	0.835	2.127
Concomitant use of furosemide (yes/no)	0.013	0.596	0.396	0.899
Concomitant use of analgesics/anti-inflammatories (yes/no)	0.893	0.973	0.654	1.449

Sig: Statistical Significance; OR: Odds Ratio; CI: Confidence Interval; LVEF: Left Ventricular Ejection Fraction; NYHA: New York Heart Association; ACEI: Angiotensin Converting Enzyme Inhibitor; ARBs: Angiotensin II receptor blockers.

Table 5. Binary logistic regression of the variables related to reaching the maximum dose of sacubitril/valsartan, in a group of patients with heart failure, Colombia.

			95%CI
Variables	Sig.	OR	Lower	Upper
Men (yes/no)	0.960	1.018	0.502	2.067
Age <50 years	0.799	Reference	Reference	Reference
Age 50–64 years	0.514	0.691	0.227	2.099
Age 65–79 years	0.885	0.906	0.236	3.477
Age ≥80 years	0.642	0.633	0.092	4.353
Bogotá-Cundinamarca region (yes/no)	0.005	2.507	1.328	4.734
Charlson Comorbidity Index (continuous)	0.230	0.853	0.658	1.106
Arterial hypertension (yes/no)	0.832	0.907	0.367	2.240
Chronic kidney disease (yes/no)	0.429	1.403	0.606	3.246
Obesity (yes/no)	0.892	1.052	0.504	2.199
Very high cardiovascular risk (yes/no)	0.540	1.276	0.585	2.785
Systolic blood pressure (continuous) (yes/no)	0.858	1.002	0.978	1.026
Heart failure with reduced LVEF (yes/no)	0.383	0.697	0.310	1.569
NYHA functional class III or IV (yes/no)	0.275	0.656	0.308	1.398
No previous ACEI or ARBs (yes/no)	0.650	0.843	0.403	1.765
Start sacubitril/valsartan 50 mg twice daily (yes/no)	<0.001	0.143	0.075	0.272
Sacubitril/valsartan persistence at one year (yes/no)	0.001	3.378	1.641	6.952
Adverse drug reactions (yes/no)	0.220	0.374	0.077	1.802
Number of medications for heart failure (continuous)	0.896	1.040	0.577	1.873
Concomitant use of β-blockers (yes/no)	0.084	0.406	0.146	1.129
Concomitant use of minieralocorticoid receptor antagonists (yes/no)	0.065	2.640	0.942	7.398
Concomitant use of furosemide (yes/no)	0.099	0.436	0.163	1.168
Concomitant use of analgesics/anti-inflammatories (yes/no)	0.267	0.700	0.373	1.313

Sig: Statistical Significance; OR: Odds Ratio; CI: Confidence Interval; LVEF: Left Ventricular Ejection Fraction; NYHA: New York Heart Association; ACEI: Angiotensin Converting Enzyme Inhibitor; ARBs: Angiotensin II receptor blockers.

3.6. Comparison with the PARADIGM-HF study

This cohort of patients had a greater proportion of women than the PARADIGM-HF study. The average age, the proportion of patients with arterial hypertension, the proportion with myocardial infarction, the proportion with NYHA functional class III or IV, and the average LVEF were also higher in our study. On the other hand, the PARADIGM-HF study had a higher proportion of patients with atrial fibrillation and had a higher mean systolic blood pressure and mean body mass index. More of their patients used β-blockers, diuretics, and digitalis (Supplementary Table S1).

4. Discussion

This study determined the form of sacubitril/valsartan use through real-world evidence in patients with heart failure. We compared these patients’ characteristics with those found in the PARADIGMA-HF study, and we explored the variables that were associated with persistence 1 year after starting the drug, as well as the variables associated with reaching its target dose (400 mg per day) in a group of patients affiliated with the Colombian health system. Our findings may be useful for healthcare, academic, and scientific personnel in making decisions about drug treatment and the risks their patients face.

The sociodemographic, clinical, and pharmacological characteristics of the patients who received sacubitril/valsartan in Colombia differ from those treated in the PARADIGM-HF study [6], though they are consistent with what has been reported in other studies of real-world practice [12,15–26]. Specifically, in this analysis and in others of this type, a higher proportion of women taking sacubitril/valsartan was found than the PARADIGM-HF study reported [12,16,24–26]. They were older [12,15,17,18,22,23,26], had a worse functional class [12,16,17,19–23,25], had a lower baseline systolic blood pressure [15–18,20,21,23–25], and took β-blockers less often [15,16,21,25] but MRAs more often [12,16–24]. Furthermore, not all patients in this cohort were taking ACEIs or ARBs before starting sacubitril/valsartan, which contrasts with patients in the PARADIGM-HF trial, where all of them were previously treated with enalapril (run-in phase) [6]. Some of these factors can negatively influence the persistence and maximum tolerated dose of the drug [15,17–19,21,23,24,26–29].

Some 56.8% of our patients persisted with their treatment 1 year after starting, which is lower than rates reported in Spain (85.7–87.5%) [29,30], Italy (84.6%) [17], China (84.3%) [16], Sweden (82.0%) [22], the United States, the United Kingdom, another study from Sweden (73%) [26], and Germany (71%) [28]. The reason for the low persistence is likely that the drug was not initially covered by the country’s health benefits plan [31] and the difficulties caused by the coronavirus disease 2019 (COVID-19) pandemic [32]. Those who had higher systolic blood pressure and those who used β-blockers were more likely to be persistent at 1 year, in line with Wachter et al. (Germany) [28] and Vicent et al. (Spain) [29]. Furthermore, not having taken an ACEI or ARB was correlated with nonpersistence, in line with the same German study [28]. Those who were prescribed furosemide were also less persistent, while Wachter et al. found that the use of diuretics was a predictor of not discontinuing sacubitril/valsartan [28]. In China, Chen et al. documented that previous hospitalizations and a low LVEF impacted persistence [16], while in Spain, Esteban-Fernandez et al found that the appearance of adverse drug reactions was related to the discontinuation of the medication [33].

A total of 12.5% of the patients reached the maximum dose of sacubitril/valsartan, which is consistent with the rate reported in Taiwan (15.8%) [27] but lower than what was found in European (21.0–38.0%) [17,23,26,28,29,33], North American (27.2–30.0%) [18,26], and Asian studies (24.8–55.2%) [15,19,21,24]. This may reflect the difficulties that mainly low- and middle-income countries (like ours) have in implementing heart failure clinical practice guidelines [34,35]. Some healthcare providers have multidisciplinary teams that can improve treatment optimization, adherence, and patient prognosis [34,35]. When patients are closely monitored, a greater proportion of them can achieve the target dose of sacubitril/valsartan (64.0–75.0%) [36,37]. This is concerning because a meta-analysis of studies with real-world evidence determined that low doses of sacubitril/valsartan were associated with higher risks of hospitalization for heart failure and all-cause mortality than the maximum dose [11]. Even so, a lower frequency of heart failure exacerbations was evidenced during treatment with sacubitril/valsartan than the patients had the previous year, which is consistent with other studies [17,20,21,30,33].

Some variables have been correlated with reaching the target dose, such as younger age [17,23,25], higher blood pressure [17,23,25], higher BMI [21], better kidney function [17,30], and higher doses of ACEIs or ARBs [23,30]. Such relationships were not found in this study, but we did find that those patients who persisted with their treatment for 1 year were more likely to reach the target dose. Furthermore, patients who were given low doses of sacubitril/valsartan were the least likely to reach the target dose. In this report, the vast majority of patients had no dose modification (87.4%), which is consistent with what was found in China (91.7%) [16] but contrasts with the rates reported in Germany (62.0%) [28] and Korea (39.8%) [15]. This may be due to clinical inertia, a condition characterized by the failure of physicians to initiate or intensify a therapy that is indicated when the goals established for the treatment of a disease have not been met [38–40]. In Colombia, this problem has been documented in patients with high blood pressure [38] and diabetes mellitus [39]. In the international literature, there has been discussion of clinical inertia in the setting of heart failure [40]. On the other hand, it is important to highlight that the low proportion of patients who reached the target doses of sacubitril/valsartan may also be due to the lack of availability of the drug and its high cost [31,41].

Adverse drug reactions were documented in 8.4% of cases, which is in line with data from Italy (10.6%) [17], Korea (10.5%) [15], and China (10.0%) [25] but was lower than that described in Spain (25,7–25.9%) [29,33] and a meta-analysis of 20 randomized controlled trials involving 22,510 participants (32.6%) [42]. Similarly, the different types of adverse reactions presented were consistent with reports from clinical trials [42,43] and from different pharmacovigilance programs [43]. It is possible that adverse reactions are underreported, as some patients may not report them, so they are not recorded in the medical records [44].

Some limitations should be considered when interpreting our results. The data were obtained from a group of patients insured mainly through the contributory regime of the Colombian health system, so the findings might not be extrapolated to patients in different insurance settings. For some variables, information was not available for every patient because the records were obtained retrospectively. No laboratory tests, such as for natriuretic peptides or echocardiograms, were performed during follow-up. The study included relatively few representatives from most geographic regions of Colombia, and there was a range of LVEFs and functional classes.

5. Conclusions

Patients taking sacubitril/valsartan in Colombia for the treatment of heart failure are mainly men, older than 65 years, with AHA stage C disease, with NYHA functional class III, and with 4.9 comorbidities on average. According to the Charlson comorbidity index, they have a high or very high cardiovascular risk and a reduced LVEF. They take low doses of the drug, and few reach the maximum dose, but 56.8% persist with the treatment 1 year after initiation, and adverse reactions are rare. The sociodemographic, clinical, and pharmacological characteristics of this group of patients differ significantly from those included in the PARADIGMA-HF study. Research should continue on the use of these drugs given the increase in their use, the profile of patients taking them and the potential safety risks they pose.

6. Expert opinion

The characteristics of patients receiving sacubitril/valsartan in the real world differ from those of patients included in the PARADIGM-HF study. These variations can impact the degree of tolerability and adherence to the medication. A very low proportion of patients reached the maximum dose, probably due to adverse drug reactions, lack of dose titration due to clinical inertia or lack of close medical follow-up, or factors related to the accessibility of the medication. High doses of sacubitril/valsartan result in a greater reduction in the risk of hospitalization and death compared to lower doses.

Declaration of interest

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.

Author contributions

L F Valladales-Restrepo: wrote manuscript, designed research, performed research, analyzed data. N Sánchez-Ramírez: analyzed data, wrote manuscript. A F Usma-Valencia: analyzed data, wrote manuscript. S Velásquez-Quirama: analyzed data, wrote manuscript. M Henao-Martínez: analyzed data, wrote manuscript. J A Castro-Rodriguez: analyzed data. A Gaviria-Mendoza: analyzed data, wrote manuscript. M E Machado-Duque: designed research, analyzed data. J E Machado-Alba: wrote manuscript, designed research, performed research.

Availability of data and material

https://www.protocols.io/private/81C23F205D8411EE80E00A58A9FEAC02

Acknowledgments

To Soffy Claritza López for her work in obtaining the database. To Camilo Alexander Constain Mosquera for his support in the preparation of the research protocol.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14656566.2023.2287667

Additional information

Funding

This paper was not funded.