Outcomes Associated With Early Initiation and Rapid Uptitration of Heart Failure Treatment in Patients With Ejection Fraction <50%: Real‐Life Data From the Swedish Heart Failure Registry 2003 to 2022

Abstract

Background

Heart failure (HF) is associated with high mortality and significant health care expenditures. Although guideline‐directed medical therapy (GDMT) improves outcomes in patients with reduced ejection fraction, many patients receive suboptimal doses. This study compared real‐world effectiveness of low versus medium/optimal GDMT doses in patients with newly diagnosed HF with ejection fraction <50%, focusing on all‐cause and cardiovascular mortality.

Methods

This retrospective study included 6847 patients from the Swedish Heart Failure Registry, integrated with national health registries. Participants were categorized into 2 treatment cohorts: low dose and medium/optimal treatment dose. All‐cause and cardiovascular mortality were analyzed at 3, 12, and 24 months. Cox proportional hazards models that incorporated inverse probability treatment weighting were used.

Results

At 24 months, medium/optimal GDMT was associated with a 23% reduction in all‐cause mortality (adjusted hazard ratio [aHR], 0.77 [95% CI, 0.67–0.90], P<0.001) and a 24% reduction in CV mortality (aHR, 0.76 [95% CI, 0.62–0.93], P=0.008) compared with low‐dose therapy. Mortality benefits were observed as early as 3 months, although this difference was not statistically significant after adjustment for confounding factors. These findings align with the STRONG‐HF (Safety, Tolerability and Efficacy of Rapid Optimization, Helped by NT‐proBNP [N‐Terminal‐Pro‐B‐Type Natriuretic Peptide] Testing, of Heart Failure Therapies) trial, thereby confirming the clinical efficacy of rapid GDMT uptitration in clinical practice.

Conclusions

In real‐world patients with heart failure with reduced ejection fraction, medium or optimal GDMT dosages significantly reduced all‐cause and cardiovascular mortality at 12 and 24 months. These results underscore the importance of early and aggressive GDMT titration in improving patient outcomes and bridging the gap between clinical guidelines and real‐world application.

Nonstandard Abbreviations and Acronyms

ARNI

angiotensin‐receptor neprilysin inhibitor

GDMT

guideline‐directed medical therapy

HFrEF

heart failure with reduced ejection fraction

IPTW

inverse probability treatment weighting

STRONG‐HF

Safety, Tolerability and Efficacy of Rapid Optimization, Helped by NT‐proBNP [N‐Terminal‐Pro‐B‐Type Natriuretic Peptide] Testing, of Heart Failure Therapies

SwedeHF

Swedish Heart Failure Registry

Clinical Perspective.

What Is New?

• We evaluated the comparative real‐world efficacy of low versus medium or optimal guideline‐directed medical therapy doses in patients with newly diagnosed heart failure with an ejection fraction <50%, focusing on all‐cause and cardiovascular mortality.

• We analyzed the 2 treatment cohorts: low‐dose and medium or optimal treatment dose for all‐cause and cardiovascular mortality at 3, 12, and 24 months. At 24 months, medium or optimal guideline‐directed medical therapy was associated with a reduction in all‐cause and cardiovascular mortality compared with low‐dose therapy. Early mortality benefits were observed as early as 3 months, although this difference was not statistically significant after adjustment for confounding factors.

What Are the Clinical Implications?

• In patients with heart failure with reduced ejection fraction rapid up‐titration of guideline‐directed medical therapy for heart failure are associated with significantly reduced all‐cause and cardiovascular mortality at 12 and 24 months. These results underscore the importance of early and aggressive GDMT titration in improving patient outcomes.

Heart failure (HF) constitutes a major clinical and public health concern, distinguished by high prevalence, unfavorable prognosis, and substantial societal costs. ¹ , ² One‐year mortality rates following incident HF are approximately 15% to 30%. ² , ³ Guideline‐directed medical therapy (GDMT) is the cornerstone for HF treatment. ⁴ , ⁵ Ouwerkerk et al showed that patients treated with <50% of the recommended treatment doses of beta blockers (BBs) and angiotensin‐converting‐enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs) were associated with an increased risk of mortality or HF hospitalization. ⁶ The latest HF guidelines recommend that patients with HF with reduced ejection fraction (HFrEF) should be treated to trial‐directed doses of ACEIs, BBs, mineralocorticoid receptor antagonists (MRAs), and SGLT2 (sodium‐glucose cotransporter 2) inhibitors. ⁴ , ⁵ The TITRATE‐HF (Guideline Implementation and Quality of Care in Patients With Heart Failure: the TITRATE‐HF Registry) study reported high use of GDMT, but there was still room for improvement regarding the quadruple medical therapy. ⁷ The STRONG‐HF (Safety, Tolerability, and Efficacy of Up‐Titration of GDMTs for Acute HF) trial reported that accelerated uptitration of HF medication reduced the risk of HF readmissions and all‐cause mortality compared with usual care. ⁸

Despite current evidence, a significant proportion of patients with HF receive suboptimal medication regimens, failing to meet guideline recommendations for the 4 recommended drug classes or receiving inadequate dosages. ⁹ The effect of HF medication acts rapidly to reduce mortality and morbidity. ¹⁰ Expeditious initiation and escalation of quadruple HF therapy are therefore paramount. The vulnerable period after a newly diagnosed HF is a time of increased risk for mortality and HF‐related hospitalizations, particularly before the establishment of optimal pharmacologic management. ¹¹ , ¹² Although robust evidence supports the life‐saving potential of these pharmacological treatments, real‐world studies consistently reveal suboptimal use and optimization in clinical practice. ⁹ , ¹³

Randomized controlled trials are widely regarded as the gold standard for evaluating treatment efficacy. Still, register‐derived real‐world data can play a vital role in supplementing our understanding. This study leveraged data from the SwedeHF (Swedish Heart Failure Registry) and clinical outcomes from 3 additional Swedish national administrative health registries. Given the granularity and broad coverage of these databases, registry‐based findings in Sweden provide one of the most reliable real‐world observations.

This study evaluates the real‐world efficacy of varying GDMT dosages at 3, 12, and 24 months in patients with newly diagnosed HF and an EF <50%, focusing on all‐cause and cardiovascular mortality.

METHODS

Study Material

Data for this retrospective observational study were extracted from the SwedeHF, previously detailed elsewhere. ¹⁴ The sole inclusion criterion was clinician‐determined HF up to 2017, which was defined since 2017 by the following International Classification of Diseases, Tenth Revision (ICD‐10) codes: I11.0, I13.0, I13.2, I25.5, I42.0, I42.6, I42.7, I50.0, I50.1, and I50.9. In 2020, the SwedeHF contained roughly 31% of Sweden’s population with prevalent HF. Patients have been included in the SwedeHF since 2003. Following hospital discharge or an outpatient clinic visit, approximately 80 variables are recorded and entered into a database maintained by the Uppsala Clinical Research Center, Uppsala, Sweden (Swedehf.se). To access diverse information (eg, comorbidities, cause of death, medication dispensation from pharmacies), SwedeHF was linked to the Swedish National Patient Register, including both in‐ and outpatient visits, the Cause of Death Register, and the National Prescribed Drug Register using the unique personal identification number assigned to all Swedish residents.

The extracted comorbidities and causes of death were identified using ICD‐10 codes and medications classified according to the Anatomical Therapeutic Chemical system (see Table S1). The authors declare that all supporting data are available within the article (and its online supplementary files).

The work complied with Strengthening the Reporting of Observational Studies in Epidemiology guidelines. ¹⁵ The study adhered to the principles outlined in the Declaration of Helsinki, and ethical approval was obtained from the Swedish Ethical Review Authority. Although individual patient consent was not mandated, all patients were informed about their inclusion in the SwedeHF and were afforded the option to withdraw.

Study Patients

Of the 120 655 patients registered in the SwedeHF registry between 2003 and 2022, a study cohort of 6847 patients was selected for analysis. (Figure 1).

Figure 1. Study flow chart.

HF indicates heart failure; LVEF, left ventricular ejection fraction; and Swede‐HF, Swedish Heart Failure Registry.

In accordance with the European Society of Cardiology (ESC) Heart Failure Guidelines of 2005 and 2016, ¹⁶ , ¹⁷ the then‐current therapeutic approach for patients with symptomatic HF was a stepwise algorithm. BBs and ACEIs/ARBs constituted first‐line therapy to reduce mortality and morbidity in symptomatic patients with HFrEF. In patients with HFrEF and left ventricular (LV) EF ≤35%, MRAs (spironolactone and eplerenone) were recommended as add‐on therapy to ACEIs and BBs in all symptomatic patients to mitigate mortality and HF hospitalization. Angiotensin receptor neprilysin inhibitor (ARNIs) were suggested solely as an add‐on treatment in patients with HFrEF who exhibited persistent symptoms (New York Heart Association class II) and an LVEF ≤35% despite optimal therapy with BBs, ACEIs/ARBs, and MRAs in the ESC HF Guidelines 2016. ¹⁶ SGLT2 inhibitors were introduced later in the ESC Heart Failure Guidelines of 2021. ⁴

Definitions of Treatment Strategy

This definition was based on previous ESC Heart Failure Guidelines of 2005 and 2016 ¹⁶ , ¹⁷ as the database was extracted from SwedeHF during period 2003 to 2022. Two treatment strategy groups were formed: (1) low‐dose group, and (2) medium/optimal dose group.

The low‐dose group was defined to represent suboptimal clinical implementation of GDMT in real‐world. In this low‐dose group, BB and at least 1 of the following medications: ACEI/ARB/ARNI with dose <50% of GDMT, MRA <25 mg, and no SGLT2 inhibitors.

Remaining patients belonged to the medium/optimal dose group including (1) medium dose: treatment with BB, MRA, and at least 1 of the following medications—ACEI/ARB/ARNI with dose ≥50% and <100% of the target dose; and (2) optimal dose: BB and at least 1 of the following medications—ACEI/ARB/ARNI with a dose of ≥100% of GDMT and MRA ≥25 mg.

The intention‐to‐treat approach was applied, assigning patients to the respective treatment groups based on the medication information in SwedeHF and medication dispensation within 3 months before and after the index date.

Outcomes

The studied outcomes were all‐cause and cardiovascular mortality (ICD‐10 code I), followed up from 3 months after the index date (baseline) until censoring or occurrence of an event. The early deaths during the first 3 months from the index are excluded from the analyses. The outcomes were also studied at 3, 12, and 24 months from the baseline date.

Statistical Analysis

Descriptively, continuous variables were presented by mean±SD or median and interquartile range as applicable, and categorical variables by frequency and percentage.

For test between low dose, medium dose and optimal treatment groups Fisher’s exact test was used for binary variables, Mantel–Haenszel chi‐square trend test was used for ordered categorical variables, and Mann–Whitney U test for continuous variables.

Event rates were calculated as number of events divided by the total number of follow‐up years, expressed as 100 person‐years. Lower and upper 95% confidence limits were estimated using exact Poisson limits.

Time‐to‐event analyses were performed using Cox proportional hazards models adjusted for age and sex. Hazard ratio (HR) with 95% CI were presented. The assumption of proportional hazards was checked by studying the interaction between the treatment group variable and the log(follow‐up time) in the Cox regression models. To account for the effect of confounding, inverse probability treatment weighting (IPTW) with the stabilized weights was applied using age, sex, body mass index category, New York Heart Association, LVEF, NT‐proBNP (N‐terminal pro‐B‐type natriuretic peptide) category, estimated glomerular filtration rate category, ischemic heart disease, any valve surgery or disease, hypertension, atrial fibrillation, diabetes, implantable cardioverter‐defibrillator, and cardiac resynchronization therapy. Unadjusted cumulative incidence curves were presented for the treatment groups. All tests were 2 tailed. P values <0.05 were considered significant. All analyses were performed using SAS Software v9.4 (SAS Institute Inc., Cary, NC, USA).

RESULTS

Patient Population and Baseline Characteristics

A total of 6847 patients were identified (Figure 1). Among these 6847 patients, 4638 received low‐dose and 2209 medium‐dose or optimal treatment. In the overall group, 5008 patients (73.1%) had an EF <40%, and 1839 (26.9%) had an EF of 40% to 49%. Patients in the low‐dose group, aside from having twice as many individuals having an EF of 40% to 49%, were generally older, had a higher proportion of women, had a lower body mass index, a lower estimated glomerular filtration rate, more noncardiovascular comorbidities, lower blood pressure, and lower heart rate and were less symptomatic (Table 1).

Table 1.

Baseline Demographics, Clinical Data, and Comorbidities Stratified by 2 Dose Groups (Low‐Dose and Medium/Optimal‐Dose) in Patients With HF <6 Months Duration and LVEF <50%

Variable	Low dose N=4638	Medium dose or optimal treatment N=2209	P value
Patient demographics
Sex, Male	2933 (63.2%)	1610 (72.9%)	<0.0001
Age, y	71.7±13.4 74 (18–98)	67.2±11.9 68 (20–96)	<0.0001
≥70 y	2813 (60.7%)	1005 (45.5%)	<0.0001
Clinical data at the index visit
Body mass index, kg/m2	26.0 (5.0) n=3241	28.1 (5.8) n=1686	<0.0001
Systolic blood pressure, mm Hg	125.0 (21.0)	127.6 (20.9)	<0.0001
Diastolic blood pressure, mm Hg	73.9 (12.5)	77.6 (13.3)	<0.0001
Heart rate, bpm	76.1 (18.3) n=4580	77.9 (19.5) n=2168	0.0008
New York Heart Association functional class			0.0045
I	460 (13.8%)	182 (9.8%)
II	1822 (54.5%)	1047 (56.6%)
III	1009 (30.2%)	597 (32.3%)
IV	52 (1.6%)	24 (1.3%)
LVEF (%)			<0.0001
40% to ≤50%	1502 (32.4%)	337 (15.3%)
30% to ≤ 40%	1745 (37.6%)	772 (34.9%)
<30%	1391 (30.0%)	1100 (49.8%)
N‐terminal‐pro‐B‐type natriuretic peptide, pg/mL	4777.3 (7001.8) 2430 (940; 5439)	4234.6 (5273.8) 2384 (1161; 5193)	0.29
Estimated glomerular filtration rate (Chronic Kidney Disease Epidemiology Collaboration) <60	1591 (34.7%)	560 (25.7%)	<0.0001
Medical history at the index visit
Ischemic heart disease	2074 (44.7%)	853 (38.6%)	<0.0001
Valve surgery or disease	613 (13.2%)	235 (10.6%)	0.0025
Hypertension	2228 (48.0%)	1388 (62.8%)	<0.0001
Atrial fibrillation	1815 (39.1%)	1187 (53.7%)	<0.0001
Chronic obstructive lung disease	675 (14.6%)	309 (14.0%)	0.56
Diabetes	774 (16.7%)	512 (23.2%)	<0.0001
Stroke/transient ischemic attack	525 (11.3%)	226 (10.2%)	0.19
Musculoskeletal diseases in the 3 y before admission	341 (7.4%)	108 (4.9%)	0.0001
Malignant cancer within the past 3 y before admission	507 (10.9%)	175 (7.9%)	<0.0001
Implantable cardioverter‐defibrillator	60 (1.3%)	92 (4.2%)	<0.0001
Cardiac resynchronization therapy	27 (0.6%)	47 (2.1%)	<0.0001

Data are presented as mean±SD, or number (percentage). For testing between 2 groups for binary variables, Fisher’s exact test was used; for ordered categorical variables, the Mantel–Haenszel chi‐square trend test was used; and for continuous variables, the Mann–Whitney U test was used. Prescriptions were determined from 3 months before the index date to 3 months after. The dose level was defined as the most recent dose administered during the 3 months preceding the index to 3 months after. For beta blockers, the dose level was referred to as metoprolol or equivalent; for angiotensin‐converting enzyme inhibitors, the dose level was referred to as ramipril or equivalent; For mineralocorticoid receptor antagonists, the dose level was referred to as either spironolactone or eplerenone; for angiotensin receptor blockers, the dose level was referred to as candesartan or equivalent. HF indicates heart failure; and LVEF, left ventricular ejection fraction.

In the low‐dose group, all patients received BB (100%) and at least 1 of the following medications: ACEI (91.2%), ARB (9.7%), and ARNI (1.9%). In contrast, in the medium‐dose or optimal treatment group, all patients received BBs (100%), and 84.2% received ACEIs; additionally, 11.0% received ARBs, and an increased prescription with ARNIs was noted in 14.4% of cases, 100% of participants received an MRA, and 12.8% received an SGLT2 inhibitor (Table 2).

Table 2.

Guideline‐Directed Medical Therapy Dosing Regimens Stratified by 2 Dose Groups

	Low dose N=4638	Medium dose/optimal treatment N=2209	P value
Prescriptions
BB	4638 (100.0%)	2209 (100.0%)
ACE	4231 (91.2%)	1861 (84.2%)	<0.0001
MRA	0 (0.0%)	2209 (100.0%)	<0.0001
ARB	451 (9.7%)	244 (11.0%)	0.10
ARNI	86 (1.9%)	319 (14.4%)	<0.0001
SGLT2i	0 (0.0%)	283 (12.8%)	<0.0001
Dose levels (mg) (most recent dose)
BB	27.7 (21.7)	55.8 (55.9)	<0.0001
ACE	2.7 (1.5)	6.9 (4.9)	<0.0001
MRA		25.5 (3.6)
ARB	4.3 (14.0)	10.4 (31.4)	0.0050
SGLT2i		1.4 (3.8)

Data are presented as mean±SD, or number (percentage). For tests comparing 2 groups on binary variables, Fisher’s exact test was employed, and the Mann–Whitney U test was used for continuous variables. Prescriptions were determined from 3 months before the index date to 3 months after. The dose level was defined as the most recent dose administered within a 3‐month period before the index date to 3 months after. For BB, the dose level was referred to as metoprolol or equivalent; for ACEI, the dose level was referred to as ramipril or equivalent; for MRA, the dose level was referred to as either spironolactone or eplerenone; for ARB, the dose level was referred to as candesartan or equivalent. ACEi indicates angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor neprilysin inhibitor; BB, beta blocker; MRA, mineralocorticoid receptor antagonist; and SGLT2i, sodium‐glucose cotransporter 2 inhibitors.

All‐Cause Mortality in 2 Dose Groups With Follow‐Up From 3 Up to 24 Months

All‐cause mortality occurred in 1930 patients (41.6%; event rate per 100 person‐years, 8.4 [95% CI, 8.0–8.8] at a median follow‐up time of 4.35 [95% CI, 2.06–7.30] years) in the low‐dose group and 636 patients (28.8%; event rate per 100 person‐years, 5.9 [95% CI, 5.4–6.4], the median follow‐up time 3.99 [95% CI, 1.81–7.32] years) in the medium‐dose or optimal treatment group. After IPTW adjustment for confounders, the adjusted HR (aHR) was 0.91 (95% CI, 0.83–0.99), P=0.029.

Crude all‐cause mortality during the first 3 months was 45% lower in the medium‐ and optimal‐dose group, with 28 deaths (1.3%), corresponding to an event rate of 5.2 (95% CI, 3.5–7.6) per 100 person‐years, compared with 107 deaths (2.3%) and an event rate of 9.5 (95% CI, 7.8–11.5) per 100 person‐years in the low‐dose group (unadjusted HR, 0.55, P=0.005). However, after adjustment for age, sex, and IPTW, the association failed to achieve statistical significance (aHR, 0.70, P=0.08).

At the 12‐month follow‐up, the medium‐dose or optimal treatment group showed a 51% decrease in crude all‐cause mortality, with 90 deaths (4.1%), in comparison with 388 deaths (8.4%) in the low‐dose group. By 24 months, the medium‐dose or optimal treatment group exhibited a 46% reduction in mortality, with 178 deaths (8.1%) versus 708 deaths (15.3%) in the low‐dose group. The aHR for the medium‐dose or optimal treatment group was thus reduced to 0.62 (95% CI, 0.50–0.77, P<0.001) at the 12‐month follow‐up and 0.77 (95% CI, 0.67–0.90, P<0.001) at the 24‐month follow‐up (Table 3, Figure 2).

Table 3.

Event Rates, Number of Events, Follow‐Up Time, Unadjusted, Age and Sex Adjusted, and IPTW‐Adjusted HRs (Cox Regression) Stratified by 2 Dose Groups

End point	Low‐dose group			Medium‐dose or optimal treatment group			Medium dose or optimal treatment vs low dose
	no./No. (%) events No. events/follow‐up time	Follow‐up time median (IQR)	Event rate (95% CI) per 100 person‐years	No. (%) events No. events/follow‐up time	Follow‐up time median (IQR)	Event rate (95% CI) per 100 person‐years	Hazard ratio* (95% CI) P value	Hazard ratio† (95% CI) P value	Hazard ratio‡ (95% CI) P value
All‐cause death	1930/4638 (41.6%)	4.35 (2.06–7.30)	8.4 (8.0–8.8)	636/2209 (28.8%)	3.99 (1.81–7.32)	5.9 (5.4–6.4)	0.70 (0.64–0.77) P<0.0001	0.95 (0.86–1.04) P=0.23	0.91 (0.83–0.99) P=0.029
All‐cause death within 3 mo	107/4638 (2.3%)	0.25 (0.25–0.25)	9.5 (7.8–11.5)	28/2209 (1.3%)	0.25 (0.25–0.25)	5.2 (3.5–7.6)	0.55 (0.36–0.84) P=0.0051	0.77 (0.50–1.18) P=0.23	0.70 (0.47–1.05) P=0.08
All‐cause death within 12 mo	388/4638 (8.4%)	1.00 (1.00–1.00)	8.9 (8.1–9.9)	90/2209 (4.1%)	1.00 (1.00–1.00)	4.4 (3.5–5.4)	0.49 (0.39–0.62) P<0.0001	0.66 (0.52–0.84) P=0.0006	0.62 (0.50–0.77) P<0.0001
All‐cause death within 24 mo	708/4638 (15.3%)	2.00 (2.00–2.00)	8.7 (8.1–9.4)	178/2209 (8.1%)	2.00 (1.81–2.00)	4.7 (4.0–5.5)	0.54 (0.46–0.64) P<0.0001	0.73 (0.61–0.86) P=0.0002	0.77 (0.67–0.90) P=0.0008
Cardiovascular death	1006/4638 (21.7%)	4.35 (2.06–7.30)	4.4 (4.1–4.6)	346/2209 (15.7%)	3.99 (1.81–7.32)	3.2 (2.9–3.6)	0.73 (0.65–0.83) P<0.0001	1.05 (0.93–1.19) P=0.43	0.98 (0.87–1.10) P=0.75
Cardiovascular death within 3 mo	67/4638 (1.4%)	0.25 (0.25–0.25)	6.0 (4.6–7.6)	16/2209 (0.7%)	0.25 (0.25–0.25)	3.0 (1.7–4.9)	0.50 (0.29–0.87) P=0.014	0.73 (0.42–1.27) P=0.27	0.68 (0.41–1.13) P=0.14
Cardiovascular death within 12 mo	214/4638 (4.6%)	1.00 (1.00–1.00)	4.9 (4.3–5.6)	53/2209 (2.4%)	1.00 (1.00–1.00)	2.6 (1.9–3.4)	0.53 (0.39–0.71) P<0.0001	0.77 (0.57–1.05) P=0.10	0.64 (0.47–0.85) P=0.0022
Cardiovascular death within 24 mo	383/4638 (8.3%)	2.00 (2.00–2.00)	4.7 (4.2–5.2)	94/2209 (4.3%)	2.00 (1.81–2.00)	2.5 (2.0–3.0)	0.53 (0.42–0.66) P<0.0001	0.77 (0.61–0.98) P=0.030	0.76 (0.62–0.93) P=0.0076

CIs for unadjusted event rates per 100 person‐years are obtained from exact Poisson confidence limits. Cox regression was used for time to any event presented by HR. For the construction of the above treatment groups, the most recent dose found per medication group, during the period from 3 months before to within 3 months after the index date, was used. Low dose: BB and at least 1 of the following medications—ACEI/ARB/ARNI with dose <50% of the target dose. No sodium glucose transporter 2 inhibitor and no MRA dose at ≥25 mg. Medium dose: BB, MRA and at least 1 of the following medications—ACEI/ARB/ARNI with dose ≥50% and <100% the target dose. Optimal treatment: BB and at least 1 of the following medications—ACEI/ARB/ARNI with dose on the target dose or more. MRA dose on at least 25 mg. ACEi indicates angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor neprilysin inhibitor; BB, beta blocker; HR, hazard ratio; IPTW, inverse probability treatment weighting; IQR, interquartile range; and MRA, mineralocorticoid receptor antagonist.

^*Unadjusted.

^† Adjusted for age and sex.

^‡ Adjusted by stabilized IPTW using the covariates age, sex, body mass index category, New York Heart Association, left ventricular ejection fraction, N‐terminal‐pro‐B‐type natriuretic peptide category, estimated glomerular filtration rate category, ischemic heart disease, any valve surgery or disease, hypertension, atrial fibrillation, diabetes, implantable cardioverter‐defibrillator, and cardiac resynchronization therapy.

Figure 2. Cumulative incidence curve for low‐dose, medium/optimal dose, and treatment groups.

Furthermore, as shown in Figure 3, the forest plots of IPTW‐adjusted analyses for time to all‐cause death within 2 years, comparing the low‐dose group with the medium/optimal‐dose group, revealed no significant interactions except in the subgroups of female patients and those with LVEF 40% to 50%. Although statistical significance was not achieved likely due to the limited number of events there was still a notable nominal reduction in all‐cause mortality: 5.6% among women and 4% among patients with LVEF 40% to 50% (Figure 3).

Figure 3. Forest plots for IPTW‐adjusted analysis of time to the event of all‐cause death within 2 years by subgroups for low‐dose group versus medium/optimal dose group.

CKD‐EPI indicates Chronic Kidney Disease Epidemiology Collaboration; HR, hazard ratio; IHD, ischemic heart disease; IPTW, inverse probability treatment weighting; and LVEF, left ventricular ejection fraction.

Cardiovascular Mortality in 2 Dose Groups With Follow‐Up From 3 Up to 24 Months

Cardiovascular mortality was noted in 1006 patients (21.7%; event rate per 100 person‐years, 4.4 [95% CI, 4.1–4.6], the median follow‐up time, 4.35 [95% CI, 2.06–7.30] years) in the low‐dose group and 346 patients (15.7%; event rate per 100 person‐years, 3.2 [95% CI, 2.9–3.6], the median follow‐up time, 3.99 [95% CI, 1.81–7.32] years) in the medium‐dose or optimal treatment group. In the adjusted analyses, no statistically significant differences were observed, with an aHR of 0.98 (95% CI, 0.87–1.10), P=0.75.

At the 3‐month follow‐up, the medium‐dose/optimal treatment group exhibited a 50% reduction in crude cardiovascular mortality (16 deaths, 0.7%; event rate per 100 person‐years, 3.0 [95% CI, 1.7–4.9]) compared with the low‐dose group (67 deaths, 1.4%; event rate, 6.0 [95% CI, 4.6–7.6]; P=0.014). After IPTW adjustment, this difference did not reach statistical significance (aHR, 0.68 [95% CI, 0.41–1.13], P=0.14).

At the 12‐month follow‐up, crude cardiovascular mortality was 47% lower in the medium‐dose or optimal treatment group, with 53 deaths (2.4%), compared with 214 deaths (4.6%) in the low‐dose group. By 24 months, crude cardiovascular mortality was 48% lower in the medium‐dose or optimal treatment group, with 94 deaths (4.3%) versus 383 deaths (8.3%) in the low‐dose group.

The aHR for cardiovascular mortality was statistically significant for both the 12‐month (0.64 [95% CI 0.47–0.85], P=0.002) and 24‐month (0.76 [95% CI 0.62–0.93], P=0.008) follow‐ups (Table 3, Figure 2).

DISCUSSION

Our study demonstrates that prompt initiation and rapid dose up‐titration of GDMT are associated with improved survival rates in a real‐world population with HF. More specifically, a statistically significant reduction in all‐cause mortality was observed at the 12‐ and 24‐month follow‐ups among patients receiving at least a medium dose or optimal treatment of GDMT compared with those receiving a low dose of GDMT. This risk reduction began to show a strong trend as early as 3 months, although it did not reach statistical significance after adjusting for age, sex, and comorbidities.

Our results corroborate align with the results of the STRONG‐HF study, which demonstrated that an intensive treatment strategy, involving rapid uptitration of GDMT and close follow‐up following an acute HF admission, led to a reduced risk of 180‐day all‐cause mortality or HF readmission compared with usual care. Although the reduction in all‐cause mortality in the STRONG‐HF study was as high as 18%, it did not achieve statistical significance. Conversely, our study shows a 30% nonsignificant risk reduction during the first 3 months and a statistically significant 23% reduction after 24 months. This finding underscores the potential for early benefits from initiating GDMT and uptitration to higher doses, even before adjustment for confounding variables such as age, sex, and comorbidities. The unadjusted results could be interpreted as indicative of a true therapeutic effect, though the influence of these confounding factors must still be considered.

Our study differs from the STRONG‐HF study in several important ways. (1) Study Population: Our cohort comprises real‐world patients with HF without any exclusions, exhibiting a higher mean age than the patients in the STRONG‐HF study. (2) Study Design and Treatment Methodology: In contrast to the prospective, open‐label, randomized STRONG‐HF trial, the present study employed a retrospective design.

Additionally, patients in our study were treated according to the therapeutic algorithm outlined in the 2016 ESC HF guidelines. At that time, HF was managed stepwise, with BBs and ACEIs/ARBs as first‐line therapy. By contrast, MRAs and ARNIs were considered add‐on treatments for patients with HFrEF who remained symptomatic and maintained an LVEF ≤35% despite initial therapy. Group stratification in our study was therefore based on the use of BBs and ACEIs/ARBs as the primary treatment. However, ARNI was also included, notwithstanding its classification as an adjunctive therapy, due to its 2016 introduction in Sweden. In select cases, it was used as an initial treatment instead of an ACEI or ARB. (3) Timing of Patient Enrollment: In the STRONG‐HF study, patients were enrolled after an acute HF admission. In contrast, our study, due to its retrospective nature, included patients who were index‐registered within 6 months of their HF diagnosis. Despite these differences, our findings align with those of the STRONG‐HF study. Our real‐world data complement the randomized study findings, providing additional insights into HF management in routine clinical practice.

Strengths and Limitations

In this real‐world cohort, the absence of exclusion criteria enhances the generalizability of findings to the broader population with HF, but it also introduces concomitant analytical challenges. Our cohort, composed of older patients with multiple comorbidities, may not be adequately represented in the more rigorously controlled STRONG‐HF study. Another strength of this study is the access to several large, high‐quality national registries that are linked to each other, enabling the identification of clinical data, drug dispensation data, and mortality information.

Although our study provides valuable insights into the benefits of early GDMT uptitration, it is essential to acknowledge certain limitations. First, the observational nature of our research means that causality cannot be definitively established. Additionally, the lack of randomization introduces potential biases, particularly regarding patient adherence to medication, health care access, and variations in treatment protocols across different centers.

One limitation is that patients receiving the medium dose or optimal treatment were healthier and had less comorbidity. However, we adjusted for confounding factors including comorbidity.

The absence of a control group restricts the ability for direct comparison, thus necessitating cautious interpretation of the findings. Despite extensive adjustments, residual and unmeasured confounding factors remain a possibility.

CONCLUSIONS

In summary, our results from a real‐world clinical study are largely congruent with existing randomized studies. These complementary findings underscore the importance of early initiation and rapid up‐titration of GDMT for HF in real‐world settings. Therefore, our results bolster the accumulating evidence supporting the implementation of earlier and more intensive HF management strategies.

Sources of Funding

Support was provided solely from institutional and departmental sources.

Disclosures

Dr Karlström has received lecture fees from AstraZeneca and Boehringer Ingelheim, and has served on advisory boards for Pharmacosmos, Novartis, and AstraZeneca. Dr Fu has received research grants from the Swedish Heart‐Lung Foundation and AstraZeneca, and has received fees for lectures and as an advisory board member from Boehringer Ingelheim, AstraZeneca, and Novo Nordisk. Dr Fu is the Swedish National Coordinator for the multicenter study EMPULSE, sponsored by Boehringer Ingelheim, and the Swedish National Coordinator for the multicenter study STEP‐HFPEF‐DM, sponsored by Novo Nordisk. Dr Pivodic, Backström, and Dr Louca declared no conflicts of interest regarding the content of this article.

Supporting information

Table S1