Intense optimization of oral therapy rapidly restores respiratory function in worsening heart failure patients

Abstract

Aims

Breathlessness is the primary symptom of decompensated heart failure (HF), but the prevalence and evolution of persistent respiratory distress post‐hospitalization remain unclear.

Methods and results

The Longitudinal Observational Study for mIddle Term Follow‐up Patients Admitted for Acute Dyspnea in TunIsia (SIDI) study is a prospective and multicentre study on HF patients with preserved ejection fraction admitted for acute dyspnoea. Patients were followed for 6 months. The primary endpoint was to assess the incidence of persistent respiratory distress and factors linked to respiratory recovery at Day 90. The secondary endpoint was rehospitalization and/or death at Day 90 and 180. Among 231 patients, 140 had SpO₂ and respiratory rate data at Day 90. Persistent respiratory distress was observed in 97 (69%). Multivariable analysis found no association between respiratory recovery and sex, diabetes, hypertension, kidney disease or NT‐proBNP levels. However, intensive follow‐up (n = 54) with oral neuroendocrine inhibition significantly improved respiratory parameters at Day 90 (adjusted HR: 5.70 [95% CI 2.13–18.28], P = 0.0001) compared with standard follow‐up (n = 86) and reduced rehospitalization and/or death at Day 90 and 180.

Conclusions

Persistent respiratory distress is frequent in the months following HF hospitalization. Optimized guideline‐directed therapy and close follow‐up improve respiratory recovery. SpO₂ and respiratory rate monitoring should be integrated into HF management.

The Longitudinal Observational Study for mIddle Term Follow‐up Patients Admitted for Acute Dyspnea in TunIsia (SIDI) study is a prospective and multicentre study on HF patients with preserved ejection fraction admitted for acute dyspnoea. The primary endpoint was to assess the incidence of persistent respiratory distress and factors linked to respiratory recovery at Day 90.

Introduction

Worsening heart failure (HF) patients are admitted in emergent conditions with the chief complaint of sudden aggravation of dyspnoea associated with signs of respiratory distress including low oxygen saturation and high respiratory rate. ¹ , ² Though recent progress has been recently made in the management of worsening HF by early optimisation of guideline‐directed medical therapy (GDMT), ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ time‐course and the prevalence of persistent respiratory distress in the months following discharge remains unexplored.

Worsening HF similarly affects patients with reduced or preserved left ventricular ejection fraction (LVEF). ² Though the pre‐ and post‐discharge management of HF with reduced EF (HFrEF) is well described and similar in both Europe and North America, ⁹ , ¹⁰ post‐discharge management of worsening HF with mildly reduced or preserved EF (HFmrEF and HFpEF) remains elusive. Indeed, concerning chronic HFmrEF and HFpEF, except for the use of sodium‐glucose transporter 2 inhibitors (SGLT2i) and management of comorbidities, there are no universal recommendations for long‐term therapies. Recently a novel MRA, finerenone, showed benefits in HFmrEF and HFpEF patients. ¹¹ Safety, tolerability and efficacy of up‐titration of guideline‐directed medical therapies for acute heart failure (STRONG‐HF) indicated that rapid and intensive implementation of GDMT for HF led to a similar improvement in outcome in all LVEF categories. ³ , ⁷ , ¹² Whether GDMT for HF improves outcome and especially the prevalence of persistent respiratory distress in worsening HFmrEF and HFpEF needs to be confirmed. ¹²

We designed the Longitudinal Observational Study for mIddle Term Follow‐up Patients Admitted for Acute Dyspnea in TunIsia (SIDI) study, to assess the rate of persistent respiratory distress within 3 months after discharge from a worsening HF episode in patients with HFmrEF and HFpEF and whether oral cardiovascular neuro‐endocrine inhibition associated with home visits may influence persistent respiratory distress, readmission and/or death.

Methods

The present study is a prospective, observational and multicentre study of worsening HF patients with HFmrEF and HFpEF admitted for acute dyspnea and followed 6 months after hospital discharge. The study was approved by local ethics committees and was registered in ClinicalTrials.gov (NCT04947358). Written information was given and written consent was obtained before inclusion.

Inclusion and exclusion criteria

Patients were included with the following criteria: (1) emergency department (ED) admission for unexpected respiratory difficulty with suffocation sensation and/or oxygen saturation (SpO₂) lower than 95% and/or respiration rate greater than 20/min and (2) elevated plasma natriuretic peptides (brain natriuretic peptide (BNP) > 200 pg/mL or N‐terminal pro brain natriuretic peptide (NT‐proBNP) > 800 pg/mL). Exclusion criteria were: age under 18 or greater than 85 years old, pregnancy, history of asthma, diagnostic retained of pulmonary embolism, LVEF below 40% known or assessed at admission at the ED.

ED and post‐ED management

At ED admission, all patients included in the study were treated in the first hours according to the latest guidelines ¹³ , ¹⁴ , ¹⁵ ; data on initial management was not recorded. When patients were leaving the ED to be discharged at home or transferred to the ward, after at least 24 h of stay in the ED, post‐discharge management was based on the physician's practice following two protocols: either (1) usual follow‐up (named ‘Usual FU’) group, meaning a report was sent to the general practitioner or cardiologist for follow‐up and one follow‐up visit was scheduled at Day 90 with the investigator team or (2) intensive FU (named ‘Intensive FU’) group for whom investigators felt a rapid inhibition of neuro‐hormonal pathways was needed that included combining, at the same time, half of doses of renin‐angiotensin‐aldosterone system inhibitors (RAASi), beta‐blockers (BB) and mineralocorticoid antagonists (MRA) prescribed before ED discharge and an increase to full recommended dose of the three classes of cardiovascular drugs, few (ideally 2) weeks after ED discharge, as previously described. ³ , ¹⁶ In the Intensive FU group, to ensure implementation and safety of rapid inhibition of neuro‐hormonal pathways, outpatient visits by the senior physician of the ED were planned at Days 15, 30, 60 and 90 days after discharge combined with scheduled home visits by a nurse at Days 7, 21, 45 and 75 after discharge. To note, since sodium‐glucose cotransporter 2 inhibitors (SGLT2i) were only recommended recently for the treatment of heart failure at discharge, patients of the Intensive FU received SGLT2i less frequently than the three other classes of HF medications.

Data collection at admission

For each patient, the following data have been collected at admission: medical history, clinical parameters, medications, including doses of RAASi, BB, MRA and SGLT2i, and routine biological assessment. Post‐discharge follow‐up visits recorded clinical parameters including those of persistent respiratory distress. Hence, for all visits—whether at home or in the hospital—the respiratory rate was manually measured by a nurse, counting chest rises over a full 60‐s period. Peripheral oxygen saturation (SpO₂, Wilbee) was assessed using the same pulse oximeter device for all patients as well as NT‐pro‐BNP levels, current treatment and doses of RAASi, BB, MRA and SGLT2i, vital status and recent hospital readmission. In the Intensive FU group, medication adherence was verified by the nurse during scheduled home visits through direct confirmation with the patient and their family, who were asked to demonstrate and confirm correct drug intake. Patients were included in three Tunisian emergency departments.

Endpoints

The primary endpoint was the incidence of persistent respiratory distress after hospital discharge for worsening heart failure and factors associated with restoration of respiratory parameters at Day 90. Secondary endpoint was the occurrence of re‐hospitalization and/or death at Day 90 and Day 180 after hospital discharge.

Statistical analysis

Values are expressed as medians and interquartile ranges (IQR) or counts and percentages, as appropriate. Group comparisons of continuous variables were performed using the Kruskal‐Wallis. Categorical data were compared using Pearson's Chi‐squared Test for Count Data. The correlation was calculated as Spearman rank correlation (r). All statistical tests were two‐tailed and a two‐sided P‐value of 0.05 was considered for significance. Variables associated with the occurrence of persistent respiratory distress defined as SpO₂ < 95% and/or a respiratory rate >15/min at Day 90 in univariate analysis were entered in a multivariate logistic regression model to identify factors independently associated with the outcome. Only the most clinically relevant variables were included in the multivariate model. The statistical analysis was performed using R software. ¹⁷

Results

Between February 2019 and December 2021, 231 patients were included in the study. Among studied patients, the primary endpoint was analysed in 140 patients for whom SpO₂ and respiratory rate data were available at Day 90 (Figure  1 ); of note their baseline characteristics were similar to those for whom SpO₂ and/or respiratory rate data were available at Day 90 (n = 91; Table S1). Due to follow‐up challenges during the pandemic, these data were not collected for all patients. However, information on re‐hospitalization and/or death on Day 90 and Day 180 after hospital discharge was available for 209 patients.

Figure 1.

Flow chart of the study.

Table 1 summarizes the clinical characteristics of the SIDI cohort (n = 231). Briefly, the cohort was equally composed of females (53%), with a median age of 70 [64–77] years. One hundred fifty‐four patients (70%) had a history of hypertension, 131 (59.5%) of chronic HF, 116 (52.7%) of diabetes mellitus, and 64 (29.1%) of coronary artery disease. Table 1 shows that acute dyspnoea, at ED admission, was associated with severe signs of respiratory distress: a median respiration rate of 26 [23–30] breaths/min (normal below 15 breaths/min) and a median of SpO₂ 89 [81–92] % at room air (normally greater than 95%) and almost no patient with RR < 15 breath/min and/or SpO₂ > 95%. Systolic blood pressure was 150 (130–170) mmHg, diastolic blood pressure 80 (70–97) mmHg, heart rate 100 (86–113) b.p.m. and body weight 80 (70–89) kg. On admission, median plasma natriuretic peptides was elevated (NT‐proBNP 3368 [1808, 6807] pg/mL and left ventricular ejection fraction (LVEF) was 55 [48, 62]%. Worsening heart failure was presumed to be related to cardiac in 142 (61%) patients or non‐cardiac origin in 89 (39%) patients (see baseline characteristics in Table S2).

Table 1.

Patients characteristics on admission in the entire cohort

	Overall
n	231
Female sex, n (%)	122 (53)
Age	70 (64–77)
Hypertension, n (%)	154 (70.0)
Chronic heart failure, n (%)	131 (59.5)
Diabetes mellitus, n (%)	116 (52.7)
Coronary artery disease, n (%)	64 (29.1)
Stroke, n (%)	15 (6.8)
Chronic kidney disease, n (%)	15 (6.8)
Atrial fibrillation, n (%)	48 (21.8)
Chronic obstructive pulmonary disease, n (%)	69 (31.4)
Tobacco consumption, n (%)	59 (26.8)
Weight (kg)	80 [70, 89]
Heart rate (b.p.m.)	100 [86, 113]
Diastolic blood pressure (mmHg)	80 [70, 97]
Systolic blood pressure (mmHg)	150 [130, 170]
Mean blood pressure (mmHg)	106 [93, 118]
Respiratory rate (breath/min)	26 (23–30)
SpO2 (%)	89 [81, 92]
NYHA (%)
I	54 (24.5)
II	114 (51.8)
III	50 (22.7)
IV	1 (0.5)
NTpro‐BNP (pg/mL)	3368 [1808, 6807]
D‐dimers (μg/mL)	350 [254, 1540]
Troponin I (ng/mL)	15 [0, 31]
Haemoglobin (g/dL)	11.9 [9.9, 13.1]
White blood count (G/L)	9.4 [6.9, 12.8]
Glycaemia (mmol/L)	7 [4.9, 11.3]
Natremia (mmol/L)	134 [128, 137]
Kaliemia (mmol/L)	4.1 [3.7, 4.7]
Creatininemia (μmol/L)	71 [39, 89]
Urea (mmol/L)	8 [6, 11]
Alanin aminotransferase (U/L)	20 [13, 25]
Aspartate aminotransferase (U/L)	24 [17, 35]
C reactive protein (mg/L)	22 [9, 56]
pH	7.38 [7.33, 7.43]
PaO2 (mmHg)	64 [53, 85]
PaCO2 (mmHg)	37 [32, 47]
Left ventricular ejection fraction (%)	55 [48, 62]

Of note, few parameters including ED management, such as diuretic therapies or signs of congestion, were not recorded.

The median length of stay in the emergency department was 40 h. At the time of ED discharge, 24% of patients were admitted to a cardiac ward (median length of stay: 5 days), while 76% were discharged home. Of note, two‐thirds of those discharged home were subsequently referred to Heart Failure clinics. Post‐discharge follow‐up was based on physician practice and consisted of an Usual FU for 165 (74%) patients, and an Intensive FU that included oral medical therapies to inhibit cardiovascular neuro‐endocrine pathways (Table S3) and several hospital and home visits for 66 (26%) patients; no differences were seen at baseline between Usual and Intensive FU (Table S4). At Day 90, a comparison of the proportion of patients receiving oral cardiovascular therapies showed a much greater implementation of GDMT for HF in Intensive FU patients: 93 versus 55% for BB, 95 versus 70% for ACE/ARNI/ARB, 67% versus 38% for MRA in Intensive versus Usual FU groups respectively (all P < 0.05; Figure S1). Of note, six (9%) patients among those with Intensive FU received empagliflozin (10 mg once daily), an SGLT2 inhibitor. Further, when considering optimal doses by classes of drugs, the Intensive FU group had more than 75% of patients getting full doses of beta‐blockers or ACE/ARNI/ARB contrasting with less than 10% for patients of the Usual FU group, at both Day 90 and Day 180 (Figure S1). Concerning MRA, more than half of Intensive FU and less than a quarter of Usual FU patients had 50% or more of the optimal dose.

Persistent respiratory distress at Day 90

Concerning the primary endpoint, among the 140 patients discharged from an episode of worsening heart failure and severe respiratory distress and for whom both SpO₂ and a respiratory rate were available at Day 90 (Figure  1 ), a large proportion (n = 97; 69%) had persistent respiratory distress, defined as SpO₂ < 95% and/or a respiratory rate > 15/min, at Day 90. Comparisons at baseline and at D90 between patients with persistent respiratory distress (n = 97) or those who restored respiratory parameters (n = 43) at Day 90 are shown in Table 2 . Univariate analysis showed that, though clinical characteristics were similar at baseline, patients who restored respiratory parameters had markedly better respiration rate, SpO₂ and heart rate at Day 90 (all P < 0.001) but similar NT‐proBNP (P = 0.8) and body weight (P = 0.8) compared with heart failure patients with prolonged respiratory distress on Day 90 (Table  2 ).

Table 2.

Characteristics at baseline and Day 90 of patients meeting or not the primary endpoint (persistence of respiratory distress at Day 90)

Parameters	Persistence of respiratory distress at Day 90 n = 97	Restoration of respiratory parameters at Day 90 n = 43	P‐value
Baseline
Body weight (kg)	79 [72, 87]	81 [71, 90]	0.5
Heart rate (b.p.m.)	98 [87, 117]	98 [84, 117]	0.7
Diastolic blood pressure (mmHg)	90 [73, 100]	80 [70, 100]	0.5
Systolic blood pressure (mmHg)	150 [131, 170]	150 [134, 190]	0.5
Mean blood pressure (mmHg)	107 [97, 116]	107 [92, 127]	0.8
SpO2 (%)	86 [79, 92]	90 [87, 92]	0.022
Respiratory rate	28 [22, 30]	26 [22, 28]	0.4
NT‐proBNP (pg/mL)	3990 [1897, 5686]	3328 [1391, 7594]	0.9
Plasmatic creatinine (μmol/L)	71 [41, 87]	79 [20, 86]	0.8
Day 90
Body weight (kg)	82 [70, 90]	79 [70, 90]	0.8
Heart rate (b.p.m.)	85 [71, 96]	69 [65, 80]	<0.001
Diastolic blood pressure (mmHg)	80 [73, 89]	69 [59, 82]	<0.001
Systolic blood pressure (mmHg)	139 [128, 158]	133 [122, 149]	0.3
Mean blood pressure (mmHg)	100 [90, 110]	93 [81, 102]	0.002
SpO2 (%)	93 [91, 94]	98 [96, 99]	<0.001
Respiratory rate	21 [18, 23]	14 [13, 20]	<0.001
NT‐proBNP (pg/mL)	1898 [584, 4415]	1560 [612, 4132]	0.8

Multivariable analysis showed that sex, history of diabetes, hypertension or kidney disease or NT‐proBNP at ED admission were not associated with the restoration of respiratory parameters at Day 90, while patients that received oral inhibition of cardiovascular neuro‐endocrine pathways under close follow‐up (ie Intensive FU group, n = 54/140, 39%) was associated with the restoration of respiratory parameters at Day 90 with an adjusted HR of 5.70 [IC95 2.13–18.28] with P = 0.0001 (Figure  2 ). Figure 3 confirmed that, at Day 90, respiratory rate was markedly lower and SpO₂ greater in the Intensive FU patients compared with Usual FU patients. In addition, systolic blood pressure and heart rate were lower in the Intensive FU group at Day 90 (Figure  3 ).

Figure 2.

Forest plot representing effect of sex, hypertension, diabetes mellitus, chronic kidney disease, heart failure and FU group (usual or intensive) on the association with restoration of respiratory parameters at Day 90 (respiratory rate > 15/min and/or SpO₂ < 95%) in patients initially admitted for AHF.

Figure 3.

Boxplots showing the evolution of the main respiratory and hemodynamic parameters according to the FU group (usual vs. intensive) in the 90 days following discharge, in the 140 patients analysed for the primary endpoint.

Hospital readmission or death at Day 90 and Day 180

Among the 209 patients with available data, 38 (16%) patients had hospital readmission or death at Day 90 and 39 (17%) at Day 180. Figure S2 shows a better restoration in respiratory parameters in the Intensive than Usual FU patients and Table 3 indicates that hospital readmission or death was markedly lower in Intensive than Usual follow‐up groups on both Day 90 (4.5% and 21% respectively, P = 0.002) and Day 180 (6.1% and 21% respectively, P = 0.005). No difference in outcome was seen in worsening HF patients from presumed cardiac or non‐cardiac origin (data not shown). When analysing the 140 patients of the primary endpoint, hospital readmission or death was lower in those who restored respiratory parameters and in those with the Intensive follow‐up at Day 90 and Day 180 (Table 3 and Table S5).

Table 3.

Readmission or death at Day 90 and Day 180

	Total events (n = 209)	Follow‐up groups		P‐value	Patients with primary endpoint available (n = 140)	Restoration of respiratory parameters at Day 90		P‐value
		Usual (n = 155)	Intensive (n = 54)			No (n = 97)	Yes (n = 43)
Readmission or death at Day 90	38 (16%)	35 (21%)	3 (4.5%)	0.002	7 (5.0%)	5 (12%)	2 (2.1%)	0.028

Data are presented for all patients with data on readmission or death available at Day 90 and Day 180 (n = 209) and by subgroups of follow‐up. Further, data are presented for the 140 patients of the primary endpoint analyses in total and divided by restoration or not of the respiratory parameters at Day 90.

Discussion

The multicentre, observational and prospective SIDI study, describes, that a large proportion of patients admitted to the ED with acute dyspnea related to worsening HF had persistent signs of respiratory distress, 90 days after hospital discharge. SIDI further indicates beneficial association between the oral inhibition of cardiovascular neuro‐endocrine pathways and rapid restoration of respiratory parameters at 90 days. This observation is novel and reinforces the recommendations of a rapid and intensive implementation of oral GDMT for HF before and after hospital discharge for a worsening heart failure.

Unscheduled admission for acute dyspnoea was repeatedly described to be associated with bad in‐hospital outcomes. ¹⁸ , ¹⁹ Furthermore, studies reported a high rate of hospital readmission and subsequent death in the months following discharge from acute dyspnoea whether the latter was primarily related to the decompensation of cardiac or pulmonary functions or both. ²⁰ Patients included in SIDI were admitted in the ED for acute respiratory distress, elevated natriuretic peptides and LVEF greater or equal than 40%. Those parameters indicate that most patients studied in SIDI were heart failure patients with mildly reduced or preserved EF as previously defined. ¹⁴ Using readily available respiratory parameters, SIDI indicated that a large proportion of patients discharged from an episode of worsening HF had signs of persistent respiratory distress 3 months later. This high prevalence of persistent respiratory distress was unexpected and calls to monitor respiratory distress namely respiratory rate and SpO₂ in addition to oedema, blood pressure and heart rate, already recommended by European guidelines, ¹⁴ during post‐discharge visits in HF patients.

Using an observational and pragmatic design, SIDI also indicates beneficial association between implementation and optimization of GDMT for HF and rapid and full restoration of respiratory distress, namely respiration rate and SpO₂ in most treated HF patients, 3 months after discharge. SIDI also indicates beneficial association between rapid up‐titration of GDMT for HF and cardiovascular efficacy outcomes, namely, improved long‐term post‐discharge rate of readmission and/or survival as recently described in STRONG‐HF trial. ³ SIDI also confirms safety of GDMT for HF when applied in the pre‐ and post‐discharge management of patients with HRmrEF and HFpEF. Accordingly, SIDI study suggests that the restoration of both respiration rate and SpO₂ associated with intensive cardiovascular neuroendocrine inhibition might participate to the improved cardiovascular outcome of rapid implementation of GDMT for HF in the present study and possibly in the high‐intensive care arm of STRONG‐HF trial. ³

SIDI showed that the marked beneficial association between inhibition of neuro‐endocrine pathways and respiratory function appears within 1 week of the beginning of oral therapies, in the Intensive FU group. Mechanisms of GDMT‐related improvement in respiratory distress remain elusive. It should be noted that, in the SIDI study, during the first days after discharge, medications such as BB, ACEi/ARNI/ARB and MRA were given mostly at half dose in the Intensive FU group and no visit was performed. In addition to the rapid improvement in respiratory distress, heart rate decreased toward normal values at Day 7 while no change was seen in blood pressure, body weight nor plasma NT‐proBNP. A decrease in heart rate may indicate that beta‐blockers were already active, as previously described. ²¹ , ²² Though one would have expected that the introduction of beta‐blockers might have led to some risk of new episodes of lung congestion after discharge, as previously described in STRONG‐HF, ²³ SIDI indicates that inhibition of beta‐blockers ‐ when associated with ACE/ARNI/ARA2 and MRA—did not lead to increased NT‐proBNP but rather to marked improvement in respiratory distress, as early as 7 days after the start of the combined cardiovascular medications, in SIDI study. Altogether, our data might indicate direct beneficial effects of the combination of HF therapies on the restoration of respiratory parameters early after discharge from worsening heart failure in patients with LVEF greater or equal to 40%. It might be possible that, in SIDI, the rapid restoration of oxygen saturation and of respiratory rate, both major determinants of body and myocardial oxygen supply/demand ratio have played an important role in preventing cardiovascular events such as readmission and/or death in patients of the Intensive FU group.

Results of SIDI are clinically relevant for multiple reasons. First, it shows that the intensive FU arm of SIDI achieved similar levels of rapid up‐titration of HF medications than in STRONG‐HF though half of the visits were performed by a nurse at home that transmitted clinical parameters to the ED physician in charge and the other half of the visits were performed by the ED physician in the outpatient centre. Second, SIDI shows that two readily available clinical parameters, RR and SpO₂, gave solid indications of patient improvement and should be included in any home or outpatient visit and maybe even self‐measured by the patients when discharged home from a worsening heart failure or any acute dyspnea episode. Third, in both STRONG‐HF and SIDI studies, medications, namely, BB, ACE/ARNI/ARA2, MRA and/or SGLT2 inhibitors, were taken orally at the same time suggesting that a fixed dose combination in a single capsule may facilitate drug intake and patient adherence, a major problem in heart failure therapy, and further improve outcome; this needs to be confirmed.

The present study had several limitations. Firstly, SIDI was an open‐label observational study where cardiovascular medications and doses were known to the treating physician. However, high medication observance, no crossover and no subjective endpoints would not have enhanced this study. In addition, treatment allocation was based on physicians' clinical judgement rather than randomization, which, while reflective of real‐world practice, remains a methodological limitation. Secondly, the sample size of 231 patients was relatively low with only 66 patients benefiting from Intensive FU. Moreover, only roughly two third of studied patients had complete data for the primary endpoint, which may reduce statistical power and introduce selection bias. However, despite the moderate number of patients, the power was sufficient to identify the association between oral neuro‐endocrine inhibition and rapid restoration of the main respiratory parameters. Thirdly, though the presumed cardiac or non‐cardiac origin of worsening heart failure was recorded, other parameters indicating the exact pathophysiological case of acute dyspnoea, such as pulmonary infection, or COPD exacerbation for non‐cardiac origin, were not recorded. Fourthly, it is difficult to distinguish whether the benefits of the intensive FU were related to full cardiovascular medications or close follow‐up combining home and outpatient visits. However, STRONG‐HF has already shown no benefits of repeated visits if no up‐titration of medication is planned. ³ Fifthly, echocardiographic data were not recorded as echocardiography at screening was only used to exclude AHF patients with LVEF below 40%. Recent publications shows in patients with reduced LVEF that GDMT was associated with improvement in LVEF, left atrial and ventricular volume and mitral regurgitation. ³ Whether similar reverse remodelling would be seen in HFmrEF and/or HFpEF need to be assessed.

Conclusions

In the multicenter, prospective and observational SIDI study of patients with worsening HF, we observed that persistent respiratory distress was common several months after discharge and that prompt optimization of GDMT for HF and close follow‐up was associated with a rapid and sustained improvement in respiratory parameters. This result needs to be confirmed in future studies but might already call to add respiration rate and SpO₂ for every post‐discharge visit of HF patients and also to add improved respiratory distress as treatment objective of future worsening HF trials.

Conflict of interest

AM received research contracts from 4TEEN4, Roche, Sphingotec, Abbott Diagnostics, Windtree; consultation fee from Roche, Corteria, Adrenomed, Fire, Abiomed; honorarium for lecture from Merck, Novartis, Roche, Bayer; is co‐inventor of patent on combined therapies to treat dyspnea, owned by S‐Form Pharma; member of Committee of trials for Secret‐HF, sponsored by the French Government, for S‐Form Pharma, for 4TEEN4 and Implicity. The rest of the authors declare no conflict of interest.

Funding

This study was sponsored by S‐Form Pharma, and data collection was organized by Eshmoun.

Supporting information

Table S1. Comparison of baseline characteristics between patients analysed for the primary endpoint (n = 140) and those for whom SpO2 and/or respiration rate at Day 90 were not available (n = 91).

Table S2. Comparison of baseline characteristics of patients admitted for Worsening of HF from either cardiac (n = 142) or non‐cardiac (n = 89) origin. Pathophysiology of cardiac and non‐cardiac causes were not recorded.

Table S3. half and full optimal doses used in Intensive follow‐up patients. ACEi, angiotensin converting enzyme inhibitor; MRA, mineralocorticoides antagonists; ARB: angiotensin receptor blockers.

Table S4. Comparison of baseline characteristics between patients of Usual or Intensive follow up (FU) groups.

Figure S1. Oral guideline direct medical therapies prescribed for AHF patients at discharge (Day 0) and Day 90 and Day 180, in Usual vs Intensive FU group.

Figure S2. Boxplots showing the evolution of the main parameters according to the FU group (Usual vs Intensive) in the 90 days following discharge, in the 209 patients analysed for the secondary endpoint.

Yaakoubi, H. , Boukadida, L. , Toumia, M. , Caillard, A. , Soltane, H. B. , Jaballah, R. , Trabelsi, I. , Dhaoui, R. , Zorgati, A. , Boubaker, H. , Khrouf, M. , Mezgar, Z. , Salah, H. B. , Mebazaa, A. , Deniau, B. , Nouira, S. , and Boukef, R. (2025) Intense optimization of oral therapy rapidly restores respiratory function in worsening heart failure patients. ESC Heart Failure, 12: 3569–3578. 10.1002/ehf2.15373.