Changes in BNP levels from discharge to 6-month visit predict subsequent outcomes in patients with acute heart failure.

BACKGROUND: This study aimed to investigate the association between changes in brain natriuretic peptide (BNP) from discharge to 6-month visit and subsequent clinical outcomes in patients with acute heart failure (AHF).
METHODS: Among 1246 patients enrolled in the prospective longitudinal follow-up study nested from the Kyoto Congestive Heart Failure registry, this study population included 446 patients with available paired BNP data at discharge and 6-month index visit. This study population was classified into 3 groups by percent change in BNP from discharge to 6-month visit; the low tertile (≤-44%, N = 149), the middle tertile (>-44% and ≤22%, N = 149) and the high tertile (>22%, N = 148).
FINDINGS: The cumulative 180-day incidence after the index visit of the primary outcome measure (a composite endpoint of all-cause death or hospitalization for HF) was significantly higher in the high and middle tertiles than in the low tertile (26.8% and 14.4% versus 6.9%, log-rank P<0.0001). The adjusted excess risk of the high tertile relative to the low tertile remained significant for the primary outcome measure (hazard ratio: 3.43, 95% confidence interval: 1.51-8.46, P = 0.003).
CONCLUSIONS: Percent change in BNP was associated with a subsequent risk for a composite of all-cause death and hospitalization for HF after adjustment of the absolute BNP values, suggesting that observing the change in BNP levels, in addition to absolute BNP levels themselves, helps us to manage patient with HF.

Introduction

Natriuretic peptide is the powerful biomarker for diagnosis of acute and chronic heart failure (HF) [1, 2]. Brain natriuretic peptide (BNP) measurement is strongly recommended and widely applied in daily clinical HF management [3, 4]. A high value of BNP level is an important parameter for worsening HF [5]. However, absolute BNP values in a compensated condition varied in each patient due to the underlying cardiac disease, the extent of ventricular hypertrophy, pre-loads and after-loads as well as many non-cardiac factors such as age, renal failure and obesity [6, 7].

In addition to baseline BNP levels, changes in BNP are also important in the management for HF [4, 8]. Previous studies showed that a change in BNP from admission for acute HF (AHF) to follow-up after discharge was associated with clinical events in HF patients [8]. However, in clinical practice, HF management was supported by BNP levels at follow-up, compared with BNP levels at discharge in a steady condition and the prognostic value of the changes in BNP levels from discharge to follow-up remains to be elucidated. Therefore, in the present study, we investigated the association between the changes in BNP levels and subsequent clinical outcomes in patients with AHF.

Materials and methods

Patient population

In the Kyoto Congestive Heart Failure (KCHF) registry, we enrolled consecutive 4,056 patients who were hospitalized for AHF as index hospitalization between 1 October 2014 and 31 March 2016. Identifiable patient records were anonymized before analysis. The detailed description of rationale, design and enrollment of the KCHF registry have been previously showed [9, 10]. In the prospective longitudinal follow-up study parallel with the main KCHF study, we enrolled 1,246 patients who were to have a visit at 6 +/- one month after excluding 271 patients who died during index hospitalization and 2,539 patients corresponding to exclusion criteria [10]. The design and exclusion criteria for the prospective longitudinal follow-up study has been specifically described in our previous reports [10]. After excluding 99 patients who were lost to follow within 6 months after the index hospitalization or after a 6-month visit, 23 patients who died within 6 months after the index hospitalization, and 678 patients with missing BNP data at discharge and/or at 6-month visit (S1 Table), the present study population consisted of 446 patients with paired data available for serum BNP (Figs 1 and 2). This study population was classified into the 3 groups by tertiles of percent change in BNP levels from discharge to 6-month visit.

Fig 1

Study flowchart.

AHF, acute heart failure; KCHF, Kyoto Congestive Heart Failure; s-Cr, serum creatinine; BNP, brain natriuretic peptide.

Fig 2

Scheme of the present analysis.

BNP, brain natriuretic peptide.

Ethics

The investigation conformed with the principles outlined in the Declaration of Helsinki. The study protocol was approved by the ethical committees of the Kyoto University Hospital (local identifier: E2311) and each participating hospital. Written informed consent was obtained from patients enrolled in the longitudinal prospective cohort study.

Outcomes

The date of the 6-month visit was considered as time zero for evaluating the clinical events censored at 210 days after the 6-month visit in this study (Fig 2). The primary outcome measure in this study was defined as a composite of all-cause death or hospitalization for HF [10]. The secondary outcome measure was all-cause death and hospitalization for HF, respectively.

Definitions

AHF is defined as de novo HF or worsening signs and symptoms of HF [11]. A change in BNP levels was calculated as follows: (BNP level at 6-month visit)—(BNP level at discharge). Percent change in BNP level was calculated by dividing the change in BNP by the BNP level at discharge and multiplying the result by one hundred to make it a percentage. The detailed definitions of baseline patient characteristics were previously described [9, 10]. HF was divided into HF with reduced left ventricular ejection fraction (LVEF) (<40%) (HFrEF) and HF with non-reduced LVEF (≥40%) (non-HFrEF), based on LVEF at the 6-month visit. Atrial arrythmias including atrial fibrillation and flutter were counted base on medical history and their events during index hospitalization and electrocardiography at 6-month visit.

Statistical analysis

Continuous variables were expressed as mean and standard deviation or median with interquartile range (IQR) and categorical variables are expressed as counts and percentages. Differences among the 3 groups were evaluated by means of the one-way analysis of variance, the Kruskal-Wallis test or the chi-square test, as appropriate. A paired t test was used for continuous variables and Sign test was used for binary variables to compare those at discharge and those at 6-month visit. Cumulative incidences were calculated by means of the Kaplan–Meier analysis and the among-groups differences are tested by means of the log-rank test.

We set the low tertile of percent change in BNP as reference and evaluated the adjusted risks of high tertile versus low tertile, and middle tertile versus low tertile for the primary and secondary outcome measures. The Cox proportional hazards regression models were utilized to assess the association between percent change in BNP levels and the clinical events after adjusting for 10 clinically relevant risk variables: age ≥80 years, sex, LVEF <40% by echocardiography, BNP levels at discharge, eGFR <30ml/min/1.73m2, albumin <3.0 g/dL and medications at 6-month visit (diuretics, angiotensin converting-enzyme inhibitor [ACE-I] or angiotensin-receptor blocker [ARB], β-blocker, and mineralocorticoid receptor antagonist [MRA]). As a sensitivity analysis, we included age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable in the adjusted model in patients with available data. The results were expressed as the hazard ratios (HRs) and their 95% CIs. Post-hoc subgroup analyses were performed in the 5 clinically relevant subgroups including the tertiles of BNP levels at 6-month visit (≤120 ng/L, >120 ng/L and ≤295 ng/L, and >295 ng/L), atrial arrythmias, LVEF <40%, use of ACE-I or ARB, and use of β-blocker at 6-month visit. Effects of percent change in BNP levels-by-subgroup interactions were evaluated by means of the Cox proportional hazards regression model. In additional subpopulation analyses, this study population (446 patients) was classified into 3 subpopulations according to BNP level at discharge; low tertile (≤155 ng/L, N = 150), middle tertile (>155ng/L and ≤350 ng/L, N = 147), and high tertile (>350 ng/L, N = 149). We compared cumulative incidences of the tertiles of percent change in BNP by means of the log-rank test in each BNP level at discharge. Statistical analyses were performed using JMP pro software, version 14 (SAS Corp., Cary, NC, USA). A two-tailed P value <0.05 was considered as statistically significant in all analyses.

Results

Clinical characteristics, laboratory test results, and medications at 6-month visit

The study population was classified into the 3 groups by tertiles of percent change in BNP levels; the low tertile, the marked BNP improvement group (≤-44%, N = 149), the middle tertile, the no-marked BNP change group (>-44% and ≤22%, N = 149) and the high tertile, the BNP worsening group (>22%, N = 148) (S1 Fig). Baseline characteristics at 6-month visit were significantly different across the 3 groups (Table 1). Compared with the marked BNP improvement group, the BNP worsening and no-marked BNP change groups were older and had higher prevalence of woman, a history of atrial arrhythmia, and myocardial infarction (Table 1). Compared with the marked BNP improvement group, the BNP worsening and no-marked BNP change groups had lower serum albumin, hemoglobin and eGFR and had a lower prevalence of cardiomyopathy etiology and β-blocker use and a higher proportion of diuretics use (Table 1).

Table 1

Patient characteristics at 6-month visit.

Variable	Marked BNP improvement (low tertile) (N = 149)	No-marked BNP change (middle tertile) (N = 149)	BNP worsening (high tertile) (N = 148)	P value	N of patients analyzed
Clinical characteristics
Age (years)	69.5 ± 13.8	76.8 ±11.7	78.3 ± 9.4	<0.0001	446
Age ≥80 years a	39 (26%)	77 (52%)	79 (53%)	<0.0001	446
Women a	56 (38%)	69 (46%)	84 (57%)	0.004	446
BMI (kg/m2)	23.2 ± 5.3	22.7 ± 4.9	22.9 ± 4.7	0.48	353
BMI ≤22 kg/m2	49 (40%)	61 (52%)	56 (49%)	0.19	353
Etiology
Coronary artery disease	34 (23%)	36 (24%)	38 (26%)	0.85	446
Hypertensive heart disease	41 (28%)	44 (30%)	47 (32%)	0.73	446
Cardiomyopathy	50 (34%)	32 (21%)	24 (16%)	0.002	446
Valvular heart disease	17 (11%)	26 (17%)	25 (17%)	0.26	446
Arrythmia	4 (2.7%)	8 (5.4%)	10 (6.8%)	0.23	446
Other diseases	3 (2.0%)	3 (2.0%)	4 (2.7%)	0.90	446
Medical history
AF or AFL	69 (46%)	102 (68%)	98 (66%)	0.0001	446
Hypertension	103 (69%)	110 (74%)	114 (77%)	0.30	446
Diabetes	50 (34%)	54 (36%)	62 (42%)	0.32	446
Dyslipidemia	57 (38%)	62 (42%)	54 (36%)	0.65	446
Previous myocardial infarction	23 (15%)	34 (23%)	44 (30%)	0.01	446
Previous ischemic stroke or ICH	21 (14%)	21 (14%)	20 (14%)	0.99	446
Chronic lung disease	20 (13%)	14 (9.4%)	17 (11%)	0.55	446
Vital signs at 6-month visit after discharge
Systolic BP (mmHg)	124.7 ± 19.8	119.5 ± 21.0	122.7 ± 21.5	0.04	387
HR (bpm)	74.0 ± 12.7	73.5 ± 14.2	77.6 ± 16.2	0.16	384
BNP values at discharge and 6-month visit
BNP at discharge (ng/L) a	318 (157–537)	239 (150–447)	160 (87.3–301)	<0.0001	446
BNP at 6-month visit (ng/L)	73.1 (28.6–149)	218 (128–380)	370 (185–727)	<0.0001	446
Change in BNP (ng/L)	-212 (-396- -97.7)	-27.7 (-71.1–5)	153 (72.1–361)	<0.0001	446
% change in BNP (%)	-72.3 (-84.2- -59.3)	-12.7 (-31.1–2.2)	92.0 (44.6–197)	<0.0001	446
Tests at 6-month visit after discharge
eGFR (mL/min/1.73m2)	51.0 ± 22.6	41.2 ± 16.1	43.6 ± 20.3	0.0002	445
eGFR <30 mL/min/1.73m2 a	24 (16%)	40 (27%)	44 (30%)	0.01	445
Albumin (g/dL)	4.1 ± 0.50	3.9 ± 0.56	3.8 ± 0.53	<0.0001	427
Albumin <3 g/dL a	3 (2.1%)	5 (3.5%)	9 (6.3%)	0.19	427
Sodium (mEq/L)	139.1 ± 3.5	140.1 ± 3.1	140.2± 3.1	0.01	445
Hemoglobin (g/dL)	12.6 ± 2.0	11.9 ± 1.9	11.3 ± 2.2	<0.0001	440
Medications at 6-month visit after discharge
ACE-I or ARB a	80 (67%)	73 (61%)	73 (62%)	0.56	356
MRA a	58 (49%)	51 (42%)	53 (45%)	0.55	356
β-blocker a	102 (85%)	93 (78%)	84 (72%)	0.04	357
Diuretics a	94 (79%)	102 (84%)	107 (91%)	0.04	358

Categorical variables are presented as number (%), and continuous variables are presented as mean ± SD or median (interquartile range).

Diuretics included loop diuretic, thiazide and tolvaptan.

a Risk-adjusting variables selected for the Cox proportional hazards regression model: age ≥80 years, sex, BNP values at discharge as a continuous variable, eGFR <30 mL/min/1.73m2, albumin <3 g/dL, ACE-I or ARB, MRA, β-blockers and diuretics, in addition to LVEF <40% at 6-month visit echocardiography in Table 2.

BMI, body mass index; AF, atrial fibrillation; AFL, atrial flutter; ICH, intracranial hemorrhage; BP, blood pressure; HR, heart rate; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; MRA, mineralocorticoid receptor antagonist; LVEF, left ventricular ejection fraction; SD, standard deviation.

Table 2

Changes in echocardiographic parameters from discharge to 6-month visit.

	Marked BNP improvement (low tertile) (N = 149)				No-marked BNP change (middle tertile) (N = 149)				BNP worsening (high tertile) (N = 148)				Between-groups comparison
Variable	Discharge	6-month visit	Delta #	P value (paired)	Discharge	6-month visit	Delta #	P value (paired)	Discharge	6-month visit	Delta #	P value (paired)	P value (discharge)	P value (6-month visit)	P value (delta)
LVEDD (mm)	55.1 ± 8.5	49.5 ± 8.8	-5.6 ± 6.4	<0.0001	52.4 ± 9.9	51.0 ± 10.2	-1.7 ± 4.8	<0.0001	50.6 ± 9.1	49.7 ± 9.1	-0.9 ± 4.8	0.03	<0.0001	0.70	<0.0001
LVESD (mm)	44.0 ± 10.9	35.7 ± 10.1	-8.2 ± 8.3	<0.0001	40.0 ± 12.2	38.0 ± 12.7	-2.4 ± 5.9	<0.0001	37.4 ± 10.6	36.0 ± 11.2	-1.2 ± 6.3	0.04	<0.0001	0.66	<0.0001
IVST (mm)	9.6 ± 2.1	10.1 ± 2.2	-0.3 ± 1.6	0.04	9.5 ± 2.0	10.3 ± 2.4	-0.1 ± 1.4	0.42	10.1 ± 2.2	10.3 ± 2.3	-0.2 ± 1.5	0.16	0.16	0.10	0.81
LVMI (g/m2)	135.4 ± 40.0	108.6 ± 31.4	-29.3 ± 32.6	<0.0001	123.0 ± 35.5	116.2 ± 36.7	-11.6 ± 26.9	<0.0001	126.6 ± 40.6	124.4 ± 40.7	-7.9 ± 27.6	0.004	0.01	0.005	<0.0001
LVEF (%)	38.8 ± 16.7	51.8 ± 13.5	13.0 ± 14.3	<0.0001	46.4 ± 15.7	49.6 ± 16.4	3.8 ± 11.9	0.0003	48.7 ± 15.6	50.3 ± 16.2	2.0 ± 10.1	0.02	<0.0001	0.66	<0.0001
LVEF <40% a	82/137 (60%)	29/137 (21%)	-53 (-39%)	<0.0001	52/137 (38%)	41/137 (30%)	-11 (-8.0%)	0.03	37/139 (27%)	39/139 (28%)	2 (1.4%)	0.64	<0.0001	0.22	<0.0001
LAD (mm)	43.8 ± 7.2	39.3 ± 8.6	-4.2 ± 6.8	<0.0001	46.5 ± 9.3	45.2 ± 9.1	-1.7 ± 6.5	0.003	46.7 ± 9.6	47.1 ± 9.6	0.04 ± 6.6	0.94	0.02	<0.0001	<0.0001
Moderate/Severe MR	43/136 (32%)	18/136 (13%)	-25 (-18%)	<0.0001	56/127 (44%)	47/127 (37%)	-9 (-7.1%)	0.08	48/130 (37%)	48/130 (37%)	0 (0%)	1.0	0.09	<0.0001	0.01
Moderate/Severe TR	40/135 (30%)	19/135 (14%)	-21 (-16%)	0.0003	38/133 (29%)	38/133 (29%)	0 (0%)	1.0	43/135 (32%)	49/135 (36%)	6 (4.4%)	0.22	0.90	<0.0001	0.001
TRPG (mmHg)	33.0 ± 12.5	22.6 ± 10.3	-9 ± 12.6	<0.0001	32.3 ± 12.0	28.5 ± 12.0	-3.1 ± 13.1	0.02	33.2 ± 13.1	31.8 ± 13.6	-0.1 ± 13.9	0.94	0.90	<0.0001	<0.0001
IVC (mm)	17.0 ± 5.4	13.8 ± 4.1	-3.4 ± 5.4	<0.0001	16.5 ± 4.6	15.5 ± 4.4	-1.1 ± 5.3	0.02	16.0 ± 4.6	16.7 ± 5.1	0.8 ± 4.9	0.07	0.23	<0.0001	<0.0001

Categorical variables are presented as number (%), and continuous variables are presented as mean ± SD.

# Delta is calculated for continuous variables according to the following equation: (the value at 6-month visit)–(the value at discharge) and for binary variables according to the following equation: (the numbers at 6-month visit)–(the numbers at discharge).

a Risk-adjusting variables selected for the Cox proportional hazards regression model: LVEF <40% at 6-month visit echocardiography in addition to variables in Table 1.

BNP, brain natriuretic peptide; LVEDD, left ventricular end-diastolic dimension; LVESD, left ventricular end-systolic dimension; IVST, intraventricular septum thickness; LVMI, left ventricular mass index; LVEF, left ventricular ejection fraction; LAD, left atrial diameter; MR, mitral regurgitation; TR, tricuspid regurgitation; TRPG, tricuspid regurgitant pressure gradient; IVC, inferior vena cava; SD, standard deviation; n/a, not available.

Echocardiographic findings at discharge and at 6-month visit

The BNP worsening group had a larger left atrial diameter (LAD) and a smaller left ventricular end-diastolic dimension (LVEDD), lower left ventricular mass index (LVMI) and higher LVEF at discharge than the marked BNP improvement and no-marked BNP change groups (Table 2). At 6-month visit, the BNP worsening group had a higher LVMI and tricuspid regurgitation pressure gradient (TRPG), a greater LAD and diameter of inferior vena cava (IVC), and a higher prevalence of moderate/severe mitral regurgitation (MR) and tricuspid regurgitation (TR) than the marked BNP improvement and no-marked BNP change groups. On the other hand, there were no significant differences in LVEDD and LVEF at the 6-month visit among the 3 groups (Table 2). From discharge to 6-month visit, the BNP worsening group had a minimal increase in LVEF and a minimal decrease in LVEDD, LVMI and TRPG, and had a numerical increase in LAD, IVC diameter and the prevalence of moderate/severe TR (Table 2 and representative values in Fig 3).

Fig 3

Changes of echocardiographic parameters from discharge to 6-month visit.

Changes of each echocardiographic parameters are represented as mean values. LVEDD, left ventricular end-diastolic dimension; LAD, left atrial diameter; IVC, inferior vena cava; LVEF, left ventricular ejection fraction.

Clinical outcomes

The follow-up rate at 180-day after the 6-month visit was 97.3%. During the 180-day follow-up, 39 patients in the marked BNP improvement group, 21 patients in the no-marked BNP change group and 10 patients in the BNP worsening group encountered all-cause death or hospitalization for HF (Fig 4A and Table 3). The cumulative 180-day incidences of the primary outcome measure were significantly higher in the BNP worsening group and the no-marked BNP change group than in the marked BNP improvement group (26.8% in the BNP worsening group and 14.4% in the no-marked BNP change group versus 6.9% in the marked BNP improvement group, log-rank P <0.0001) (Fig 4A). With respect to the secondary outcome measures, the cumulative 180-day incidence of all-cause death was significantly higher in the BNP worsening group than in the no-marked BNP change group and the marked BNP improvement group (9.6%, 4.8%, and 4.1%, respectively, log-rank P = 0.04) (Fig 4B) and the cumulative 180-day incidence of hospitalization for HF was also significantly higher in BNP worsening group and the no-marked BNP change group than in the marked BNP improvement group (21.2%, 9.9%, and 2.8%, respectively, log-rank P<0.0001) (Fig 4C).

Fig 4

Kaplan Meier curves for (A) the primary outcome measure, (B) all-cause death, and (C) hospitalization for heart failure.

The primary outcome measure was defined as a composite of all-cause death or hospitalization for heart failure. BNP, brain natriuretic peptide.

Table 3

Clinical outcomes.

Clinical outcome measures	Categorized group	N of patients with event/N of patients at risk (Cumulative 180-day incidence)	Crude HR (95% CI)	P value	Adjusted HR (95% CI)	P value
Primary outcome measure (a composite of all-cause death or hospitalization for heart failure)
	BNP worsening	39/101 (26.8%)	4.55 (2.44–9.29)	<0.0001	3.43 (1.51–8.46)	0.003
	No-marked BNP change	21/119 (14.4%)	2.39 (1.20–5.05)	0.01	1.79 (0.76–4.45)	0.19
	Marked BNP improvement	10/129 (6.9%)	1 (Reference)		1 (Reference)
All-cause death
	BNP worsening	14/125 (9.6%)	2.94 (1.22–8.14)	0.02	1.81 (0.49–7.30)	0.38
	No-marked BNP change	7/133 (4.8%)	1.51 (0.54–4.51)	0.43	1.67 (0.46–6.51)	0.43
	Marked BNP improvement	6/133 (4.1%)	1 (Reference)		1 (Reference)
Hospitalization for heart failure
	BNP worsening	30/101 (21.2%)	7.45 (3.17–21.8)	<0.0001	5.35 (1.83–19.7)	0.001
	No-marked BNP change	14/119 (9.9%)	3.36 (1.32–10.3)	0.01	1.87 (0.58–7.22)	0.30
	Marked BNP improvement	4/129 (2.8%)	1 (Reference)		1 (Reference)

The Cox proportional hazards regression model was constructed adjusting for 10 clinically relevant risk-adjusting variables: age ≥80 years, sex, LVEF <40% by echocardiography, BNP at discharge, eGFR <30ml/min/1.73m2, albumin <3.0 g/dL, diuretics, ACE-I or ARB, β-blocker and MRA.

Diuretics included loop diuretic, thiazide and tolvaptan.

LVEF, left ventricular ejection fraction; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; MRA, mineralocorticoid receptor antagonist; HR, hazard ratio; CI, confidence interval.

After adjusting for confounding variables, the excess risk of the BNP worsening group relative to the marked BNP improvement group remained significant for the primary outcome measure (HR: 3.43, 95%CI: 1.51–8.46, P = 0.003) and for hospitalization for HF (HR: 5.35, 95%CI: 1.83–19.7, P = 0.0001), whereas the adjusted risk of the BNP worsening group relative to the marked BNP improvement group was no longer significant for all-cause death (HR: 1.81, 95%CI: 0.49–7.30 P = 0.38) (Table 3). We showed the figures of changes in BNP during discharge and 6-month visit in each group (S2 Fig).

Sensitivity analyses

When we evaluated age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable, the excess risk of the BNP worsening group relative to the marked BNP improvement group remained significant for the primary outcome measure (HR: 3.47, 95%CI: 1.46–8.21, P = 0.005) and for hospitalization for HF (HR: 5.19, 95%CI: 1.59–17.0, P = 0.007), whereas the adjusted risk of the BNP worsening group relative to the marked BNP improvement group was not significant for all-cause death (HR: 1.94, 95%CI: 0.52–7.27 P = 0.32) (S2 Table), which was consistent with the main analysis.

Post-hoc subgroup analyses

There were no significant interactions between the risk of the percent change in BNP for the primary outcome measure and all the subgroup factors except for the use of ACE-I or ARB (S3 Table); the magnitude of the effect of the BNP worsening group for the primary outcome measure was greater in patients with the use of ACE-I or ARB.

Clinical outcomes in the subpopulations according to tertiles of BNP level at discharge

In all the subpopulations, the cumulative 180-day incidences of the primary outcome measure were significantly or numerically higher in the BNP worsening group and the no-marked BNP change group than in the marked BNP improvement group (S3 Fig).

Discussion

The main findings of the present study are as follows; 1) Patients in the BNP worsening group had higher prevalence of non-HFrEF at discharge with minimal change in LAD and LVEDD from discharge to 6-month visit; 2) Percent change in BNP was associated with a subsequent risk for a composite of all-cause death or hospitalization for HF after adjustment of the absolute BNP values at discharge; 3) The direction of BNP changes from discharge to 6-month visit might be affected by regression to the mean.

There are large systemic differences among BNP levels provided by commercial immunoassay methods because of considerable chemical and structural heterogeneity of BNP circulating in human blood [12]. Franzini et al. reported that the IRMA method (by Shionogi’s Diagnostic Division, Japan), the ADVIA method for the Centaur platform (by Siemens Health Care Diagnostics) and the ST-AIA-PACK method for the AIA platform (by TOSOH Corporation, Tokyo, Japan) measured greatly lower (up to the half) BNP values in comparison with other immunoassays, such as the POCT Triage method (by Alere Diagnostics), the BNP Triage Biosite for Access and UniCell DxI platforms (by Beckman Coulter Diagnostics), the MEIA method for the AxSYM platform and the chemiluminescent microparticle immunoassay for ARCHITECT platform (both by Abbotts Diagnostics) [13]. Additionally, BNP levels were affected by sex, age, heart rate, renal function and body mass index [12]. In Japan, BNP level is measured by the former immunoassay methods. The reference interval of ST-AIA-PACK method in a healthy population is ≤18.4 ng/L. The sensitivity and specificity of BNP at a threshold of ≤100 ng/L were 0.95 [95% confidence interval (CI): 0.93–0.96] and 0.63 (95% CI: 0.52 to 0.73), respectively [1].

Many previous small or large-scale studies showed that change in natriuretic peptides from hospital admission to follow-up after discharge was associated with clinical outcomes in HF patients [8, 14, 15]. Kagiyama et al. evaluated change in BNP during hospitalization for AHF as a prognostic biomarker for all-cause death [16]. It is obvious that patients with higher BNP levels than those in acute phase of HF result in unfavorable outcomes. Bettencourt et al. conducted a single-center retrospective study and numerically observed that BNP levels at AHF admission were more than 2.5 times higher than those at a stable HF condition in the AHF hospitalization group [17]. To the best of our knowledge, no previous study statistically evaluated the association of the changes in BNP from discharge to follow-up with subsequent clinical outcomes in patients with AHF.

This study showed that percent change in BNP was independently associated with the primary composite outcome measure and HF hospitalization, even after adjusting for medications for HF and BNP levels at discharge. This finding may be supported by the observation that the risk for adverse clinical events in the BNP worsening group tended to be greater in patients using ACE-I or ARB (S3 Table). On the other hand, Zhang et al. pointed out that although serial measurement of NT-proBNP is useful, the most recent value of NT-proBNP has similar predictive power [18]. The event occurred in the BNP worsening group despite of the low BNP levels in the present study (S2 Fig). Volume expansion and pressure overload caused by worsening HF and leading to wall stress stimulate synthesis and secretion of BNP mainly from cardiac ventricular myocytes [19, 20]. Conversely, increase in BNP reflects volume expansion and pressure overload, which may attribute to clinical events. In asymptomatic HF patients, cardiac remodeling was an independent predictor of clinical events [21].

As shown in Table 1, although BNP level at discharge in the marked BNP improvement group was significantly higher, BNP level at 6-month visit was significantly lower than the other groups. This reverse association may be affected by regression to the mean and attributed to a higher prevalence of cardioprotective drugs use. Those with higher BNP level were more likely to be treated intensively; thus, if HF management was successful, they were more likely to be in the marked BNP improvement group with better final outcomes.

At 6-month visit, regardless of no difference in LVEDD and LVEF among 3 groups, the BNP worsening group had a higher LVMI indicating pressure overload and other echocardiographic findings of congestive status including a higher TRPG, greater LAD and IVC diameter and a higher prevalence of MR and TR, which indicate volume expansion. These features may be linked to the increased BNP value. With reference to echocardiographic changes from discharge to 6-month visit, the BNP worsening group relative to the other groups showed minimal improvement of echocardiographic parameters, indicating a lack of LV and LA reverse remodeling. There might be several reasons for this lack of LV and LA reverse remodeling. First, the BNP worsening group had a higher prevalence of previous myocardial infarction and atrial arrythmias at the 6-month visit. Ischemic cardiomyopathy is known to be associated with the absence of LV reverse remodeling [22]. Atrial fibrillation is associated with atrial enlargement [23]. Second, the BNP worsening group had a lower prevalence of β-blocker use at 6-month visit, which is one of the key drugs for cardiac reverse remodeling [24]. Third, the BNP worsening group had smaller LVEDD and higher LVEF at discharge, indicating that there was a possibility of little room of reverse remodeling. Further, the BNP worsening group had a numerical increase in LAD, which was considered to be the reflection of elevated end-diastolic pressure of LV.

Regardless of absolute BNP levels, the direction of BNP changes from a stable condition at discharge may indicate disease progression or successful management of HF. The more decrease in BNP levels means the more favorable outcomes in the present study. Thus, we can modify the intensity of management for congestion if we know the changes of BNP levels in each patient. Further studies are needed to research improvement of clinical outcomes in patients with HF by adjusting HF management based on change in BNP.

Limitations

The present study has several limitations that should be addressed. First, it is possible that absent data can alter the study results (i.e. selection bias); the present study population only comprised 446 patients of the 4056 patients enrolled in the KCHF registry or of the 1246 patients scheduled for a 6-month follow-up. Although 99 loss to follow-up and 23 death were excluded, there was no significant difference in BNP at discharge between 122 excluded patients and 446 analyzed patients (248 [IQR, 90.7–502] versus 234 [IQR, 127–443], P = 0.93). Data on changes in BNP were not available in a substantial proportion of the cohort scheduled for a 6-month follow-up. BNP were not measured in a substantial proportion of patients who were followed by NT-proBNP. The measurements of BNP or NT-proBNP were basically dependent on the availability in each participating hospital. Nevertheless, the patients without the data on the change in BNP levels were older and less likely to be women, and had a lower prevalence of atrial arrythmias (S1 Table). These very significant selection of patients remains a major limitation to this study. Second, data on medications at the 6-month visit were also not available in a substantial proportion of patients, although the characteristics of patients with available data on medications (N = 352) and without data on medications (N = 94) were not significantly different (S4 Table). A very advanced age of the study population might be a reason for us not to collect the detailed data in all patients, even if they were prospectively enrolled. Additionally, a proportion of those who used cardioprotective drugs was relatively low because the present population included many non-HFrEF patients. There is a possibility that missing detailed data might alter the study results and adjustment of medications might be inadequate. Third, the follow-up period was relatively short and the number of clinical events was relatively small in this study, which made it difficult to make extensive adjustment. There may be residual and unmeasured confounding factors related to outcomes. Forth, the BNP immunoassay methods were not collected and not designed to be uniformed among the 19 participating hospitals. Finally, high-sensitivity cardiac troponin was not included into the adjustment model because of many absent data on cardiac troponin, which might have a better cardiovascular risk stratification [25].

Conclusion

Percent change in BNP was associated with a risk for a composite of all-cause death or hospitalization for HF after adjustment of the absolute BNP values, suggesting that observing the change in BNP levels, in addition to absolute BNP levels themselves, helps us to manage patient with HF.

Patient characteristics compared between patients with available data on change in BNP and those without data.

Categorical variables are presented as number (%), and continuous variables are presented as mean ± SD. BNP, brain natriuretic peptide; BMI, body mass index; AF, atrial fibrillation; AFL, atrial flutter; eGFR, estimated glomerular filtration rate; ACE-I, angiotensin converting-enzyme inhibitor; ARB, angiotensin-receptor blocker; MRA, mineralocorticoid receptor antagonist; SD, standard deviation. Diuretics included loop diuretic, thiazide or tolvaptan.

(PDF)

Click here for additional data file.

Sensitivity analyses.

The Cox proportional hazards regression model was constructed adjusting for 10 clinically relevant risk-adjusting variables: age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable and sex, diuretics, ACE-I or ARB, β-blocker and MRA. Diuretics included loop diuretic, thiazide and tolvaptan. LVEF, left ventricular ejection fraction; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; MRA, mineralocorticoid receptor antagonist; HR, hazard ratio; CI, confidence interval.

(PDF)

Click here for additional data file.

Subgroup analysis for the primary outcome measure according to the tertiles of percent change in BNP.

Values are n/n (%). BNP, brain natriuretic peptide; LVEF, left ventricular ejection fraction; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; HR, hazard ratio; CI, confidence interval.

(PDF)

Click here for additional data file.

Patient characteristics compared between patients with available data on medications at 6-month visit and those without data.

Categorical variables are presented as number (%), and continuous variables are presented as mean ± SD or median (interquartile range). BMI, body mass index; AF, atrial fibrillation; AFL, atrial flutter; ICH, intracranial hemorrhage; BP, blood pressure; HR, heart rate; BNP, brain natriuretic peptide; eGFR, estimated glomerular filtration rate; SD, standard deviation.

(PDF)

Click here for additional data file.

Histogram of percent change in BNP from discharge to 6-month visit.

% change in BNP, percent change in brain natriuretic peptide.

(PDF)

Click here for additional data file.

Changes in BNP during discharge and 6-month visit in (A) the marked BNP improvement group, (B) the no-marked BNP change group, and (C) the BNP worsening group.

This study population was classified into the 3 groups by percent change in BNP during discharge and 6-month visit; the marked BNP improvement group (≤-44%, N = 149), the no-marked BNP change group (>-44% and ≤22%, N = 149) and the BNP worsening group (>22%, N = 148). Red lines indicate patients with events. Blue lines indicate patients without events. BNP, brain natriuretic peptide; HF, heart failure.

(PDF)

Click here for additional data file.

Kaplan Meier curves for the primary outcome measure in three subpopulations according to BNP level at discharge.

(PDF)

Click here for additional data file.

23 Nov 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Claudio Passino, MD

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Acknowledgments Section of your manuscript:

(This work was supported by the Japan Agency for Medical Research and Development [18059186] (Drs T. Kato, Kuwahara, and Ozasa))

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

(This work was supported by the Japan Agency for Medical Research and Development [18059186] (T.K, K.K, and N.O).

https://www.amed.go.jp/en/

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.)

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: To the Authors

General Considerations

The aim of this study was to investigate the association between changes in BNP from discharge to 6-month visit and subsequent clinical outcomes in patients with acute HF. Authors enrolled 446 patients with available paired BNP data at discharge and 6-month index visit, classified according to the tertiles of BNP concentrations: low (≤-44 %, N=149), middle tertile (>-44 % and ≤22 %, N=149) and high (>22 %, N=148). The main results of this study were that % changes in BNP among the 3 groups of HF patients were associated with a subsequent risk for a composite of all-cause death and hospitalization for HF. Authors concluded that changes in BNP levels may help to manage patient with HF.

There is an extensive literature supporting the clinical relevance of changes of BNP levels in monitoring patients with an acute episode of HF (249 articles in PubMed using the key words: acute heart failure, BNP changes; 101 articles using the key words: acute heart failure, BNP changes, outcome). Authors should better indicate in Introduction (or Discussion) sections of the revised manuscript the originality of this article compared to other previous studies or to other more recently published (for example: Bettencourt P et al. JACEP Open 2021;2:e12448).

I have some specific points to address to the Authors in order to further improve the scientific message of this article.

Specific Points

1. BNP assay. Authors compare 3 different groups of HF patients divided according to the levels of BNP. It is well known that there are several assay methods for BNP showing very large systematic differences (up to folds) in the concentrations measured in healthy subjects and HF patients (for reviews about this issue: Clerico A et al. Clin Chim Acta 2015;443:17-24; Clerico A. et al. Future Cardiol 2016:12:573-84)). Authors should add a specific paragraph in the revised manuscript reporting the analytical characteristics and performances and reference interval values measured in a healthy population (compared to age and sex to HF patients) with the BNP method used in this study. The international guidelines by IFCC (International Federation of Clinical Chemistry) recommend that the BNP concentration should be reported as ng/L (not pg/mL) (Apple FS et al. Circulation 2007;116;e95-e98).

2. Table 1, BNP at discharge. An interesting data reported in Table 1 is that patients included in the first tertile have a BNP discharge values significantly lower than the other two tertiles. These results should be discussed more in detail in the revised manuscript. Authors should better explain this reverse association between elevated BNP at discharge and better final outcome.

3. A limitation of this study is that the measurement of cardiac troponins using high-sensitivity assays (i.e., hs-cTnI and hs-cTnT methods) was not performed in this study. Several recent studies have demonstrated that hs-cTnI and hs-cTnT may have a better risk stratification both in apparently healthy subjects and patients with cardiac disease (including HF) than natriuretic peptides (BNP or NT-proBNP) (Farmakis D et al. Eur Heart J 2020;41:4050-6; Clerico A et al. Clin Chem Lab Med 2020;59:79-90; Harrison N et al. Curr Heart Fail Rep 2019;16:21-31; Rosa GM et al. Eur J Clin Invest 2019;49:e13044; Perna ER Minerva Carioangiol 2016;64:165-80; Aimo A et al. Circulation 2018;137:286.297; Aimo A et al. Int J Cardiol 2019;277:166-172). This important point should be discussed by the Authors.

Reviewer #2: In the present paper, Shiba and colleagues aim to investigate "the association between the changes in BNP levels and subsequent clinical outcomes in patients with AHF". The Authors collect data from the Kyoto Congestive Heart Failure (KCHF) registry (n=446) whe received BNP testing at discharge after an AHF episode and at 6 month. They report that the incidence of the primary end-point (a composite of all-cause death or hospitalization for HF) was significantly higher in patients with stable/increased BNP vs those with decreasing BNP levels.

The message of the paper is rather clear, and the conclusions are supported by the provided evidence. Still, some major points, as listed below, need to be addressed.

Major points

- How did the Authors select the variables to be included in the multivariable model? Why did they dichotomized variables such as LVEF or eGFR?

- Further to the pre-specified subgroups, a subset analysis should have been performed in patients with/without atrial fibrillation

- How/when was exactly defined atrial fibrillation? How did the Authors account for eventual changes in background rhythm

- The cause for AHF should be reported in the study population. Moreover, whether it was a de novo vs an acutely decompensated HF should be also mentioned.

Minor points

- In the methods section, end-stage renal disease was defined according to eGFR, while exsclusion criteria were based on serum creatinine. Please clarify.

- The number of events in the whole population and in subgroups according to BNP changes should be clearly reported in the Methods section.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

12 Dec 2021

Response

We appreciate the information about your journal format and the editor`s and reviewers` careful evaluation and suggestion. We have replayed to the reviewers’ comments and revised our manuscript.

We have modified font size of heading, figure captions, acknowledgements (Page 33, line 9-11) and supporting information captions (Page 40, line 1-Page 42, line 15). We have removed any funding-related text from the manuscript. We have added the section of Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files (Page 10, line 15-16).

Reviewers' comments:

Reviewer's Responses to Questions

Reviewer #1: To the Authors

General Considerations

The aim of this study was to investigate the association between changes in BNP from discharge to 6-month visit and subsequent clinical outcomes in patients with acute HF. Authors enrolled 446 patients with available paired BNP data at discharge and 6-month index visit, classified according to the tertiles of

BNP concentrations: low (≤-44 %, N=149), middle tertile (>-44 % and ≤22 %, N=149) and high (>22 %, N=148). The main results of this study were that % changes in BNP among the 3 groups of HF patients were associated with a subsequent risk for a composite of all-cause death and hospitalization for HF. Authors concluded that changes in BNP levels may help to manage patient with HF.

There is an extensive literature supporting the clinical relevance of changes of BNP levels in monitoring patients with an acute episode of HF (249 articles in PubMed using the key words: acute heart failure, BNP changes; 101 articles using the key words: acute heart failure, BNP changes, outcome). Authors

should better indicate in Introduction (or Discussion) sections of the revised manuscript the originality of this article compared to other previous studies or to other more recently published (for example: Bettencourt P et al. JACEP Open 2021;2:e12448).

I have some specific points to address to the Authors in order to further improve the scientific message of this article.

Response

We thank the reviewer for the careful assessment, the appropriate suggestion and showing many specific references. We have described the originality of this article compared to the previous study (Bettencourt P et al. JACEP Open.2021;2:e12448) in the section of Discussion. Bettencourt`s concept and purpose was the same as ours. Bettencourt et al. conducted a single-center retrospective study and numerically observed that BNP levels at AHF admission were more than 2.5 times higher than those at a stable HF condition in the AHF hospitalization group. However, he didn`t statistically evaluate the relationship between change in BNP from a stable condition and the risk for clinical outcomes. This is our advantage. (Page 28, line 11-15).

Specific Points

1. BNP assay. Authors compare 3 different groups of HF patients divided according to the levels of BNP. It is well known that there are several assay methods for BNP showing very large systematic differences (up to folds) in the concentrations measured in healthy subjects and HF patients (for reviews about this issue: Clerico A et al. Clin Chim Acta 2015;443:17-24; Clerico A. et al. Future Cardiol 2016:12:573-84)). Authors should add a specific paragraph in the revised manuscript reporting the analytical characteristics and performances and reference interval values measured in a healthy population (compared to age and sex to HF patients) with the BNP method used in this study. The international guidelines by IFCC (International Federation of Clinical Chemistry) recommend that the BNP concentration should be reported as ng/L (not pg/mL) (Apple FS et al. Circulation 2007;116;e95-e98).

Response

We appreciate your comment. We have change from pg/mL to ng/L along with your recommendation. We have added the paragraph about BNP Assay in the section of Discussion and reported the analytical characteristics and performances and reference interval values measured in a healthy population (Page 27, line 8-Page 28, line 5). The BNP immunoassay methods were not collected and not designed to be uniformed among the 19 participating hospitals. We have added this into the section of Limitation (Page 32, line 15-17).

2. Table 1, BNP at discharge. An interesting data reported in Table 1 is that patients included in the first tertile have a BNP discharge values significantly lower than the other two tertiles. These results should be discussed more in detail in the revised manuscript. Authors should better explain this reverse association between elevated BNP at discharge and better final outcome.

Response

We agree on your great insight. We think that this reverse association may be affected by regression to the mean and attributed to a higher prevalence of cardioprotective drugs use (Page 29, line 11-17). We had added the following discussions:

In the section of Discussion

As shown in Table 1, although BNP level at discharge in the marked BNP improvement group was significantly higher, BNP level at 6-month visit was significantly lower than the other groups. This reverse association may be affected by regression to the mean and attributed to a higher prevalence of cardioprotective drugs use. Those with higher BNP levels were more likely to be treated intensively; thus, if HF management was successful, they were more likely to be in the marked BNP improvement group with better final outcomes.

3. A limitation of this study is that the measurement of cardiac troponins using high-sensitivity assays (i.e., hs-cTnI and hs-cTnT methods) was not performed in this study. Several recent studies have demonstrated that hs-cTnI and hs-cTnT may have a better risk stratification both in apparently healthy subjects and patients with cardiac disease (including HF) than natriuretic peptides (BNP or NT-proBNP) (Farmakis D et al. Eur Heart J 2020;41:4050-6; Clerico A et al. Clin Chem Lab Med 2020;59:79-90; Harrison N et al. Curr Heart Fail Rep 2019;16:21-31; Rosa GM et al. Eur J Clin Invest 2019;49:e13044; Perna ER Minerva Carioangiol 2016;64:165-80; Aimo A et al. Circulation 2018;137:286.297; Aimo A et al. Int J Cardiol 2019;277:166-172). This important point should be discussed by the Authors.

Response

We thank the reviewer for the valuable suggestion. We tried to include cardiac troponin into the adjustment model. However, we could not include because of many absent data on cardiac troponin. We have discussed this in the section of Limitation (Page 32, line 17-Page 33, line 1).

In the section of Limitation

Finally, high-sensitivity cardiac troponin was not included into the adjustment model because of many absent data on cardiac troponin, which might have a better cardiovascular risk stratification (25).

Reviewer #2: In the present paper, Shiba and colleagues aim to investigate "the association between the changes in BNP levels and subsequent clinical outcomes in patients with AHF". The Authors collect data from the Kyoto Congestive Heart Failure (KCHF) registry (n=446) who received BNP testing at discharge after an AHF episode and at 6 month. They report that the incidence of the primary end-point (a composite of all-cause death or hospitalization for HF) was significantly higher in patients with stable/increased BNP vs those with decreasing BNP levels.

The message of the paper is rather clear, and the conclusions are supported by the provided evidence. Still, some major points, as listed below, need to be addressed.

Response

We appreciate the positive comments, the careful assessment, and the clear suggestions.

Major points

- How did the Authors select the variables to be included in the multivariable model? Why did they dichotomized variables such as LVEF or eGFR?

Response:

We thank the reviewer for your comments. From the adjusting variables which was preliminarily designed in our previous studies, we selected 10 adjusting factors more closely-related with heart failure outcomes because a few clinical events occurred in this study population.

The use of dichotomization of continuous variables was almost consistent across our previous studies. We have added a sensitivity analysis with age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable. The results were still consistent with the main analysis (S4 Table).

In the section of Materials and methods (Page 9, line 16-18)

As a sensitivity analysis, we included age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable in the adjusted model in patients with available data.

In the section of Results (Page 25, line 11-18)

When we evaluated age, LVEF, BNP at discharge, eGFR and albumin as a continuous variable, the excess risk of the BNP worsening group relative to the marked BNP improvement group remained significant for the primary outcome measure (HR: 3.47, 95%CI: 1.46-8.21, P=0.005) and for hospitalization for HF (HR: 5.19, 95%CI: 1.59-17.0, P=0.007), whereas the adjusted risk of the BNP worsening group relative to the marked BNP improvement group was not significant for all-cause death (HR: 1.94, 95%CI: 0.52-7.27 P=0.32) (S4 Table), which was consistent with the main analysis.

- Further to the pre-specified subgroups, a subset analysis should have been performed in patients with/without atrial fibrillation

- How/when was exactly defined atrial fibrillation? How did the Authors account for eventual changes in background rhythm.

Response

We thank the reviewer for your comments. We have added the definition of atrial arrythmias. Atrial arrythmias including atrial fibrillation and flutter were counted base on medical history and their events during index hospitalization and electrocardiography at 6-month visit (Page 8, line 13-15). We additionally have evaluated atrial arrythmias in the post-hoc subgroup analysis (Page 10, line 3). There was no significant interaction between the risk of the percent change in BNP for the primary outcome measure and atrial arrythmias (S5 Table).

- The cause for AHF should be reported in the study population. Moreover, whether it was a de novo vs an acutely decompensated HF should be also

mentioned.

Response

We thank the reviewer for your comments. We have added the definition and etiology of AHF. AHF is defined as de novo HF or worsening signs and symptoms of HF (11) (Page 8, line 6). Compared with the marked BNP improvement group, the BNP worsening and no-marked BNP change groups had a lower prevalence of cardiomyopathy etiology (Table 1) (Page 11, line 11-14).

Minor points

- In the methods section, end-stage renal disease was defined according to eGFR, while exsclusion criteria were based on serum creatinine. Please clarify.

Response

We appreciate specific points and apologize for a double-standard definition of end-stage renal disease. We have added the event number of the primary endpoint in the Results (Page 21, line 8-11). We defined s-Cr >3.0 mg/dL or on hemodialysis as end-stage renal disease. eGFR <30ml/min/1.73m2 was included into the adjusting model along with our previous studies.

- The number of events in the whole population and in subgroups according to BNP changes should be clearly reported in the Methods section.

Response

We appreciate your comments. We have added the number of events in the Result.

In the section of Results

During the 180-day follow-up, 39 patients in the marked BNP improvement group, 21 patients in the no-marked BNP change group and 10 patients in the BNP worsening group encountered all-cause death or hospitalization for HF (Fig 4A and Table 3) (Page 21, line 8-11).

Submitted filename: Response to review comments.docx

Click here for additional data file.

13 Jan 2022

Changes in BNP levels from discharge to 6-month visit predict subsequent outcomes in patients with acute heart failure

PONE-D-21-35110R1

Dear Dr. Kato,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Claudio Passino, MD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Authors revised the manuscript in accordance of suggestions made by the two Reviewers. The scientific message of the article is significantly improved now.

Reviewer #2: The Authors have properly addressed all the comments raised in the previous revision, and the manuscript has now significantly improved.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

20 Jan 2022

PONE-D-21-35110R1

Changes in BNP levels from discharge to 6-month visit predict subsequent outcomes in patients with acute heart failure

Dear Dr. Kato:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Claudio Passino

Academic Editor

PLOS ONE