Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2020 Dec 28;15(12):e0243818. doi: 10.1371/journal.pone.0243818

Association of an increase in serum albumin levels with positive 1-year outcomes in acute decompensated heart failure: A cohort study

Takao Kato 1,*, Hidenori Yaku 1, Takeshi Morimoto 2, Yasutaka Inuzuka 3, Yodo Tamaki 4, Neiko Ozasa 1, Erika Yamamoto 1, Yusuke Yoshikawa 1, Takeshi Kitai 5, Ryoji Taniguchi 6, Moritake Iguchi 7, Masashi Kato 8, Mamoru Takahashi 9, Toshikazu Jinnai 10, Tomoyuki Ikeda 11, Kazuya Nagao 12, Takafumi Kawai 13, Akihiro Komasa 14, Ryusuke Nishikawa 15, Yuichi Kawase 16, Takashi Morinaga 17, Mitsunori Kawato 18, Yuta Seko 19, Masayuki Shiba 1, Mamoru Toyofuku 20, Yutaka Furukawa 5, Kenji Ando 17, Kazushige Kadota 16, Yukihito Sato 6, Koichiro Kuwahara 21, Takeshi Kimura 1
Editor: Vincenzo Lionetti22
PMCID: PMC7769473  PMID: 33370299

Abstract

Background

Despite the prognostic importance of hypoalbuminemia, the prognostic implication of a change in albumin levels has not been fully investigated during hospitalization in patients with acute decompensated heart failure (ADHF).

Methods

Using the data from the Kyoto Congestive Heart Failure registry on 3160 patients who were discharged alive for acute heart failure hospitalization and in whom the change in albumin levels was calculated at discharge, we evaluated the association with an increase in serum albumin levels from admission to discharge and clinical outcomes by a multivariable Cox hazard model. The primary outcome measure was a composite of all-cause death or hospitalization for heart failure.

Findings

Patients with increased albumin levels (N = 1083, 34.3%) were younger and less often had smaller body mass index and renal dysfunction than those with no increase in albumin levels (N = 2077, 65.7%). Median follow-up was 475 days with a 96% 1-year follow-up rate. Relative to the group with no increase in albumin levels, the lower risk of the increased albumin group remained significant for the primary outcome measure (hazard ratio: 0.78, 95% confidence interval: 0.69–0.90: P = 0.0004) after adjusting for confounders including baseline albumin levels. When stratified by the quartiles of baseline albumin levels, the favorable effect of increased albumin was more pronounced in the lower quartiles of albumin levels, but without a significant interaction effect (interaction P = 0.49).

Conclusions

Independent of baseline albumin levels, an increase in albumin during index hospitalization was associated with a lower 1-year risk for a composite of all-cause death and hospitalization in patients with acute heart failure.

Introduction

Because of the influences of aging, heart failure (HF)-specific disability, and the progression of the disease, improving the long-term mortality in patients with HF hospitalization remains a challenging medical need in developed countries [1, 2]. Hypoalbuminemia is a well-known prognostic marker in patients hospitalized for acute medical illness, including acute decompensated heart failure (ADHF) [3–5]. This prognostic value probably refers primarily to the syndrome of malnutrition-inflammation and the severity of comorbidities [5]. Albumin decreased by the malnutrition, hemodilution, renal loss, and shortened half-life due to severe illness such as inflammation [6]. Serum albumin has many physiological properties, including in particular antioxidant, anti-inflammatory, anticoagulant and anti-platelet aggregation activity. It also plays an essential role in the exchange of fluids across the capillary membrane [6]. In addition to that, hypoalbuminemia may be a potentially modifiable risk factor through the disruption of the protective roles and the exacerbation of peripheral congestion and pulmonary edema [6].

Despite the prognostic importance of hypoalbuminemia, the prognostic implication of a change in albumin levels has not been fully investigated. We aimed to test the hypothesis that an increase as compared with no increase in albumin levels from at admission to at discharge is associated with better 1-year clinical outcome using a large contemporary all-comer registry of patients hospitalized due to ADHF in Japan.

Materials and methods

Study population

The Kyoto Congestive Heart Failure registry is a physician-initiated, prospective, observational, multicenter cohort study that enrolled consecutive patients hospitalized for ADHF for the first time between 1 October 2014 and 31 March 2016 without any exclusion criteria [2, 7, 8]. These patients were admitted into 19 secondary and tertiary hospitals, including rural and urban, large and small institutions, throughout Japan. The overall design of the Kyoto Congestive Heart Failure study and patient enrolment has been previously described in detail [2, 8, 9]. We enrolled consecutive patients admitted to the participating centers who had ADHF as defined by the modified Framingham criteria and underwent heart failure-specific treatment involving intravenous drugs within 24 hours of hospital presentation. Among 4056 patients enrolled in the registry, we excluded 271 patients who died during index hospitalization, 57 patients who were lost to follow-up, and 568 patients whose albumin levels at admission or at discharge were not available. Therefore, the current study population consisted of 3160 patients who were discharged alive and in whom the change in albumin levels was calculated at discharge (Fig 1). We classified the patients into 2 groups: those with increased albumin levels during hospitalization (increase group; N = 1083, 34.3%) and those with no increase in albumin levels (no-increase group; N = 2077, 65.7%) (Fig 1). We compared albumin levels at admission/hospital presentation (anytime of the day) and at the nearest time to discharge in the morning. Any increase in albumin level corresponded to albumin increase. This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Patient characteristics with missing values for albumin levels were provided in S1 Table. Analysis began in October 2019. Detailed definitions of baseline patient characteristics are as follows: 1) Anemia was defined using the World Health Organization criteria (hemoglobin <12.0 g/dL in women and <13.0 g/dL in men). 2) HF was classified according to left ventricular ejection fraction (LVEF) into HF with preserved LVEF (LVEF ≥ 50%), HF with mid-range LVEF (40%≤ LVEF < 50%), and HF with reduced LVEF (LVEF < 40%) [10].

Fig 1. Study flowchart.

Fig 1

Open in a new tab

KCHF, Kyoto Congestive Heart Failure.

Ethics

The investigation conformed with the principles outlined in the Declaration of Helsinki. The study protocol was approved by the ethical committees in Kyoto University Hospital Kyoto University Graduate School of Medicine (approval number: E2311) and participating hospitals (details in S1 File). Patient records were anonymized prior to analysis. As the study met the conditions of the Japanese Ethical Guidelines for Medical and Health Research Involving Human Subjects [11], a waiver of written informed consent from each patient was granted by the institutional review boards of Kyoto University and each participating center. We disclosed the details of the present study to the public as an opt-out method, and the notice clearly informed patients of their right to refuse enrollment. Details were described previously [11].

Outcomes

Clinical follow-up data were collected in October 2017. The attending physicians or research assistants at each participating hospital collected clinical events after index hospitalization from hospital charts or by contacting patients, their relatives, or their referring physicians with consent.

The primary outcome measure for the present analysis was a composite of all-cause death or hospitalization for HF after discharge following index hospitalization. Other outcome measures included all-cause death, cardiovascular death, sudden death, and HF hospitalization. Death was regarded as cardiovascular in origin unless obvious non-cardiovascular causes could be identified. Cardiovascular death included death related to HF, sudden death, death related to stroke, and death from other cardiovascular causes. Sudden death was an unexplained death in a previously stable patient. Stroke was defined as either ischemic or hemorrhagic stroke that required acute or prolonged hospitalization and had symptoms that lasted for more than 24 hours [8]. HF hospitalization was defined as hospitalization due to worsening of HF, requiring intravenous drug therapy [8]. The causes of death were adjudicated by a clinical event committee [12].

Statistical analyses

We compared the baseline characteristics between the increase and no-increase groups, and we compared 1-year clinical outcomes between the 2 groups and performed subgroup analysis stratified by the albumin levels at admission.

Categorical variables were presented as numbers and percentages, and were compared by the chi square test. Continuous variables are expressed as mean and standard deviation or median with interquartile interval. Based on their distributions, continuous variables were compared using Student’s t-test or the Wilcoxon rank-sum test between the 2 groups. The cumulative incidences of clinical events during 1-year after discharge were estimated using the Kaplan-Meier method and the intergroup differences assessed by the log-rank test. Cumulative incidence rates of HF hospitalization or cardiovascular death/ non-cardiovascular death were estimated by using the Gray method [13], accounting for the competing risk of all-cause death or non-cardiovascular death/ cardiovascular death, respectively. We regarded the date of discharge as time zero for clinical follow-up. To estimate the risk for the primary and secondary outcome measures in the increase group relative to the no-increase group, a multivariable Cox proportional-hazards model was developed adjusting for confounders. Consistent with our previous reports [7, 9], we included the following 23 clinically relevant risk-adjusting variables and two additional variables into the model: age ≥80 years, sex, body mass index<22kg/m2, variables related to medical history (previous HF hospitalization, etiology of HF hospitalization associated with acute coronary syndrome, atrial fibrillation or flutter, hypertension, diabetes mellitus, previous myocardial infarction, previous stroke, current smoking, and chronic lung disease), variables related to comorbidities (living alone, ambulatory, systolic blood pressure <90 mmHg, heart rate <60 bpm, LVEF<40%, estimated glomerular filtration rate [eGFR] <30 ml/min/1.73m2, sodium <135 mEq/L, anemia, and albumin levels at admission), and medications at discharge (angiotensin converting enzyme inhibitors or angiotensin II receptor blockers, and β-blockers), along with histories of liver cirrhosis, malignancy, and duration of hospital stay. Multivariable Cox proportional hazards models described by Fine and Gray sub-distribution hazard model [14] were developed for HF hospitalization or cardiovascular death/ non-cardiovascular death accounting for the competing risk of all-cause death or non-cardiovascular death/ cardiovascular death, respectively. Continuous variables were dichotomized by clinically meaningful reference values or median values. Albumin levels at admission were classified into quartiles: ≤ 3.2 g/dl (Q1), > 3.2 g/dl and ≤ 3.5 g/dl (Q2), ≥3.5 g/dl and ≤ 3.8 g/dl (Q3), and > 3.8 g/dl (Q4). The results were expressed as a hazard ratio (HR) and 95% confidence intervals (CIs).

In the sensitivity analysis, we stratified patients into 4 groups based on the quartiles of the percent change of albumin levels as follows: (albumin at discharge) -(albumin at presentation)/(albumin at presentation)*100 (%). Comparisons among 4 groups were performed using the chi-square test for categorical variables and 1-way ANOVA or Kruskal-Wallis test for continuous variables in addition to the Cochran-Armitage trend test in order to assess the trend across the 4 groups. A multivariable Cox proportional-hazards model was developed using the same variables mentioned above to estimate the risk for the primary outcome measures relative to the lowest quartile. In the subgroup analysis, we evaluated the interaction between the albumin levels at admission or the presence of acute coronary syndrome and the risk of the change of the albumin levels for the primary outcome measure. To analyze the baseline factors associated with high LVMI, we used a multivariable logistic regression model involving the following potential independent clinically relevant variable. The adjusted odds ratios and 95% CIs were calculated. All statistical analyses were conducted by a physician (T.K.) and a statistician (T.M.) using JMP 14 and EZR [15]. All the reported P values were two-tailed, and the level of statistical significance was set at P <0.05.

Results

Comparison of baseline characteristics between the increase and no-increase groups

Patients with increased albumin levels were younger and more likely to be men and current smokers, less often had a body mass index below 22 kg/m2, eGFR <30 ml/min/1.73m2 and hypertension, and had higher C-reactive protein levels and lower blood urea nitrogen levels and albumin levels at presentation than those with no increase in albumin levels (Table 1). At discharge, those with increased albumin levels had a higher prevalence of diuretics and mineralocorticoid receptor antagonist and had a higher BNP levels than those with no increase in albumin levels.

Table 1. Patient characteristics.

Increase in albumin (N = 1083, 34.3%) No-increase in albumin (N = 2077, 65.7%) P value N of patients analyzed
Clinical characteristics
    Age, years 78 [68–84] 81 [74–87] <0.0001 3,160
    Age >80 years* 475 (43.9) 1174 (56.5) <0.0001 3,160
    Men* 623 (57.5) 1125 (54.2) 0.08 3,160
    Body mass index || <22 kg/m2* 448 (42.8) 950 (48.4) 0.003 3,009
Prior hospitalization for heart failure* 373 (34.8) 747 (36.4) 0.39 3,124
Etiology
    Dilated cardiomyopathy 162 (15.0) 178 (8.6) <0.0001 3,160
    Acute coronary syndrome* 53 (4.9) 120 (5.8) 0.32 3,160
    Aortic stenosis 63 (5.8) 160 (7.7) 0.057 3,160
    Hypertensive 255 (23.6) 538 (25.9) 0.15 3,160
    Ischemic (not acute) 278 (25.7) 580 (27.9) 0.18 3,160
    Others 272 (25.1) 501 (24.1) 0.54 3,160
Medical history
    Atrial fibrillation or flutter* 444 (41.0) 874 (42.1) 0.57 3,160
    Hypertension* 747 (69.0) 1564 (75.3) 0.0002 3,160
    Diabetes mellitus* 408 (37.7) 788 (37.9) 0.91 3,160
    Dyslipidemia 445 (41.1) 809 (39.0) 0.25 3,160
    Prior myocardial infarction* 244 (22.5) 482 (23.2) 0.69 3,160
    Prior stroke* 180 (16.6) 332 (16.0) 0.65 3,160
    Current smoking* 165 (15.4) 215 (10.6) 0.0001 3,104
    Ventricular tachycardia/fibrillation 54 (5.0) 71 (3.4) 0.03 3,160
    Chronic lung disease* 136 (12.6) 291 (14.0) 0.27 3,160
    Liver cirrhosis* 12 (1.1) 28 (1.4) 0.62 3,160
    Malignancy* 133 (12.3) 328 (15.8) 0.008 3,160
Social backgrounds
    Poor medical adherence 208 (19.2) 339 (16.3) 0.048 3,160
    Living alone* 244 (22.5) 422 (20.3) 0.15 3,160
Daily life activities
    Ambulatory* 861 (80.1) 1627 (79.1) 0.55 3,132
    Use of wheelchair [outdoor only] 83 (7.7) 159 (7.7) 1.00 3,132
    Use of wheelchair [outdoor and indoor] 97 (9.0) 193 (9.4) 0.80 3,132
    Bedridden 34 (3.2) 78 (3.8) 0.42 3,132
Vital signs at presentation
    Systolic blood pressure, mmHg 145 ± 33 150 ± 36 0.0001 3,149
    Systolic blood pressure <90 mmHg* 33 (3.1) 45 (2.2) 0.15 3,153
    Heart rate, bpm 97 ± 29 96 ± 27 0.45 3,142
    Heart rate <60 bpm* 78 (7.2) 137 (6.6) 0.55 3,142
    Body temperature>37.5 degree Celsius 55 (5.3) 141 (7.1) 0.06 3,033
    NYHA Class III or IV 917 (85.0) 1838 (88.9) 0.0021 3,147
Tests at presentation
    LVEF 46.0 ± 16.4 46.6 ± 16.1 0.35 3,086
    HFrEF (LVEF <40%)* 397 (36.7) 761 (36.8) 1.00 3,153
    HFmrEF (LVEF 40–49%) 240 (22.2) 372 (18.0) 0.005 3,153
    HFpEF (LVEF ≥50%) 445 (41.1) 938 (45.3) 0.026 3,153
    BNP, ng/L 691 [378–1245] 724 [395–1268] 0.21 2,811
    Serum creatinine, mg/dL 1.05 [0.80–1.50] 1.12 [0.83–1.67] 0.0006 3,160
    eGFR <30 mL/min/1.73m2* 241 (22.3) 604 (29.1) <0.0001 3,160
    Blood urea nitrogen, mg/dL 22 [16–32] 25 [18–37] <0.0001 3,157
    Sodium <135 mEq/L* 126 (11.7) 255 (12.3) 0.61 3,155
    Anemia*§ 740 (68.4) 1374 (66.2) 0.23 3,157
    C reactive protein, mg/dL 0.69 [0.22–2.57] 0.60 [0.20–1.85] 0.0001 3,103
    Increase in albumin levels, g/dl☨
Albumin levels at presentation, g/dl *# 3.25 ± 0.48 3.58 ± 0.45 <0.0001 3,160
Q1: ≤ 3.2 g/dl 524 (48.4) 463 (22.3) <0.0001 987
Q2: > 3.2 g/dl and ≤ 3.5 g/dl 262 (24.2) 451 (21.7) 0.17 713
    Q3: ≥3.5 g/dl and ≤ 3.8 g/dl 193 (17.8) 555 (26.7) <0.0001 748
    Q4: > 3.8 g/dl 104 (9.6) 608 (29.3) <0.0001 712
Medications at discharge
    ACE-I or ARB* 634 (58.5) 1177 (56.7) 0.31 3,160
    Beta blocker* 740 (68.3) 1358 (65.4) 0.10 3,160
    MRA 520 (48.0) 898 (43.2) 0.010 3,160
    Diuretics (non-MRA) 945 (87.2) 1694 (81.6) <0.0001 3,160
BNP at discharge, ng/L 231 [118–475] 299 [147–543] <0.0001 2,148
Hospital stay >16 days* (median) 557 (51.4%) 979 (47.1%) 0.02 3,160
Open in a new tab

*26 risk-adjusting variables selected for COX hazard model including albumin quartiles.

|| Body mass index was calculated as weight in kilograms divided by height in meters squared.

§ Anemia was defined by the World Health Organization criteria (hemoglobin <12.0 g/dL in women and <13.0 g/dL in men).

☨ Any increase in albumin level corresponded to albumin increase.

#Albumin levels at admission were classified into quartiles: ≤ 3.2 g/dl (Q1), > 3.2 g/dl and ≤ 3.5 g/dl (Q2), ≥3.5 g/dl and ≤ 3.8 g/dl (Q3), and > 3.8 g/dl (Q4).

BP = blood pressure; bpm = beat per minute; NYHA = New York Heart Association, LVEF = left ventricular ejection fraction; HFrEF = heart failure with reduced ejection fraction; HFmrEF = heart failure with mid-range ejection fraction; HFpEF = heart failure with preserved ejection fraction; BNP = brain-type natriuretic peptide; eGFR = estimated glomerular filtration rate, ACE-I = angiotensin converting enzyme inhibitors, ARB = angiotensin II receptor blockers, MRA = mineralocorticoid receptor antagonist.

Clinical outcomes

The median follow-up duration after discharge was 15.8 months (interquartile interval: 12.1–20.9), with a 96.0% follow-up rate at 1 year. The cumulative 1-year incidence of the primary outcome measure (a composite endpoint of all-cause death or hospitalization for HF) was significantly lower in the increase group than that in the no-increase group (39.6% versus 36.1%, P <0.0001) (Fig 2A). After adjustment for confounders, the lower risk of the increase group relative to the no-increase group remained significant for the primary outcome measure (HR: 0.78, 95% CI: 0.69–0.90, P = 0.0004). For the secondary outcome measures, the cumulative 1-year incidences of all-cause death and hospitalization for HF were significantly lower in the increase group than in the no-increase group (14.4% versus 18.8%, P<0.0001 and 21.2% versus 23.4%, P = 0.0073, respectively) (Fig 2B and 2C). After adjustment for confounders, the lower risk of the increase group relative to the no-increase group remained significant for all-cause death (HR: 0.70, 95% CI: 0.58–0.83, P <0.0001), whereas the lower risk of the increase group relative to the no-increase group was no longer significant for hospitalization for HF (HR: 0.92, 95% CI: 0.72–1.18, P = 0.53) (Table 2 and S1 Fig). The trends for cardiovascular and non-cardiovascular death were mostly consistent with that for all-cause death (Table 2).

Fig 2.

Fig 2

Open in a new tab

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.

Table 2. Primary and secondary outcomes.

Increase in albumin No-increase in albumin Unadjusted Hazard Ratio (95%CI) P value Adjusted Hazard Ratio (95%CI) P value
N of patients with event/N of patients at risk (Cumulative 1-year incidence) N of patients with event/N of patients at risk (Cumulative 1-year incidence)
A composite of all-cause death or HF hospitalization 382/1083 (29.6%) 887/2077 (36.1%) 0.77 (0.68–0.87) <0.0001 0.78 (0.68–0.89) 0.0004
All-cause death 204/1083 (14.4%) 520/2077 (18.8%) 0.71 (0.60–0.83) <0.0001 0.69 (0.58–0.83) <0.0001
HF Hospitalization 247/1083 (21.2%) 545/2077 (23.4%) 0.85 (0.73–0.98) 0.03 0.92 (0.72–1.18) 0.53
Cardiovascular death 125/1083 (8.8%) 307/2077 (13.0%) 0.76 (0.61–0.93) 0.008 0.81 (0.64–1.04) 0.09
Non-cardiovascular death 79/1083 (5.5%) 213/2077 (7.6%) 0.69 (0.53–0.89) 0.0048 0.62 (0.46–0.82) 0.001
Open in a new tab

HF = heart failure, CI = confidence interval, HR = hazard ratio.

Sensitivity analysis based on the percent change of albumin levels

We stratified patients into 4 groups based on the quartiles of the percent change of albumin levels: the lowest quartile (≤ -11.1%), the lower quartile (> -11.1% and ≤ -3.0%), the higher quartile (> -3.0% and ≤ 5.3%), and the highest quartile (> 5.3%) (S2 Table). The cumulative 1-year incidence of the primary outcome measure (a composite endpoint of all-cause death or hospitalization for HF) was significantly lower in the higher and the highest quartiles than that in the lowest quartile (S2 Fig). The lower risk of the higher and the highest percent change groups relative to the lowest percent change group was significant after adjustment (HR: 0.70, 95% CI: 0.58–0.83, P <0.0001) (S3 Table).

Subgroup analysis based on the baseline albumin levels

When stratified by the quartiles of baseline albumin levels, the lower risk of the increase group relative to the no-increase group for the primary outcomes and all-cause death trended to be more prominent in the lower quartiles of albumin levels, but without significant interaction across the quartiles (interaction P = 0.49 for the primary outcomes, and interaction P = 0.89 for all-cause death) (Table 3).

Table 3. Subgroup analysis based on the baseline albumin levels.

Increase in albumin No-increase in albumin Unadjusted Adjusted
N of patients with events/N of patients at risk (Cumulative 1-year incidence) N of patients with events/N of patients at risk (Cumulative 1-year incidence) HR (95% CI) P value HR (95% CI) P value P value for interaction
A composite of death or HF hospitalization Q1 213/524 (33.9%) 251/463 (47.7%) 0.63 (0.52–0.76) <0.0001 0.75 (0.62–0.92) 0.006 0.49
Q2 94/262 (29.7%) 224/451 (41.7%) 0.65 (0.51–0.82) 0.0003 0.79 (0.61–1.03) 0.08
Q3 43/193 (18.4%) 250/555 (32.0%) 0.54 (0.39–0.75) 0.0001 0.79 (0.55–1.13 0.19
Q4 32/104 (28.5%) 207/608 (26.9%) 0.94 (0.63–1.34) 0.73 1.01 (0.64–1.52) 0.98
All-cause death Q1 128/524 (18.1%) 176/463 (30.3%) 0.55 (0.44–0.70) <0.0001 0.72 (0.55–0.92) 0.01 0.89
Q2 48/262 (14.5%) 143/451 (23.7%) 0.53 (0.38–0.72) <0.0001 0.65 (0.44–0.93) 0.02
Q3 16/193 (6.3%) 115/555 (14.9%) 0.36 (0.21–0.59) <0.0001 0.64 (0.37–1.12) 0.12
Q4 12/104 (10.0%) 86/608 (10.1%) 0.85 (0.44–1.48) 0.58 0.93 (0.44–1.83) 0.85
Open in a new tab

Albumin levels at admission were classified into quartiles: ≤ 3.2 g/dl (Q1), > 3.2 g/dl and ≤ 3.5 g/dl (Q2), ≥3.5 g/dl and ≤ 3.8 g/dl (Q3), and > 3.8 g/dl (Q4). CI = confidence interval, HR = hazard ratio.

Subgroup analysis based on the acute coronary syndrome

When stratified by the acute coronary syndrome, the lower risk of the increase group relative to the no-increase group for the primary outcomes and all-cause death was consistently seen in patients with or without acute coronary syndrome. The effect was directionally significant in patients with acute coronary syndrome (S4 Table).

Baseline factors associated with the increase in albumin levels

According to the multivariable logistic regression analysis, the albumin levels at admission < 3 g/dL and anemia were independent positive factors associated with the increase in albumin levels, while age ≥80 years, eGFR<30 ml/min/1.73m2, body mass index<22kg/m2, NYHA III/IV at presentation, and the history of diabetes mellitus were independent negative factors associated with the increase in albumin levels (Table 4).

Table 4. Factors associated with the increase in albumin levels.

Adjusted OR Lower 95% CI Upper 95% CI P value
Albumin <3 mg/dl 3.94 3.14 4.95 <0.0001
Age ≥80 years 0.54 0.46 0.65 <0.0001
eGFR<30 mL/min/1.73m2 0.65 0.53 0.80 <0.0001
BMI<22 kg/m2 0.80 0.67 0.94 0.0093
Anemia 1.26 1.05 1.51 0.01
NYHA III/IV 0.76 0.60 0.96 0.02
DM 0.83 0.70 0.997 0.046
COLD 0.84 0.66 1.07 0.16
SBP<90 mmHg 1.42 0.85 2.36 0.17
History of stroke 1.12 0.90 1.39 0.30
HFrEF 0.92 0.76 1.10 0.36
AF 1.04 0.88 1.23 0.61
History of MI 1.02 0.84 1.24 0.81
BNP or NT-pro BNP above median* 1.01 0.86 1.20 0.82
History of HF hospitalization 1.01 0.85 1.21 0.84
Female 1.01 0.85 1.20 0.85
Open in a new tab

*BNP value above 715.9 ng/L or NT-pro-BNP> 5744.1 pg/L.

OR = odds ratio, CI = confidence interval, eGFR = estimated glomerular filtration rate; DM = diabetes mellitus; BMI = body mass index; HFrEF = HF with reduced EF (<40%); COLD = chronic obstructive lung disease, SBP = systolic blood pressure; AF = atrial fibrillation/flutter; ACS = acute coronary syndrome; NYHA = New York Heart Association.

Discussions

The principal findings of the present study were as follows: 1) The increase as compared with no increase in albumin level during the index hospitalization for ADHF was associated with lower adjusted risk for the primary outcome measure (a composite of all-cause death or hospitalization for HF) as well as all-cause death, cardiovascular death, and non-cardiovascular death. 2) There was no interaction between the albumin levels at admission and the effects of the increase relative to no increase in albumin for the primary outcome measure and all-cause death.

In patients with ADHF, albumin levels are influenced by the production from the liver, hemodilution, loss from the kidney, digestive tract, and vascular bed, and shortened half-life due to severe illness or inflammation [3, 4, 16–18]. Comparisons of albumin levels between the decompensated state at admission and the compensated condition at discharge may enable assessment of the recovery of multi-organ damage due to the worsening of HF. In this study, about two-thirds of patients did not show increased albumin levels at discharge. They may be considered in the process of recovering from ADHF when their albumin levels were low. Nakayama et al reported that an increase in serum albumin during hospitalization had favorable long-term prognostic impact on the composite endpoint of all-cause death or HF hospitalization in 115 patients from a single center [19]. In the present study, we also showed a lower risk for all-cause death in the group with increased albumin levels during hospitalization after adjustment of baseline albumin levels. This impact seemed to be greater in patients with low serum albumin levels at admission, although the interaction was not significant. In addition, CRP levels at admission was significantly higher in the group with increased albumin levels at discharge. This finding implied that, patients with the increase in albumin levels showed the decreased albumin levels at admission due to the serious conditions; however, once recovered, they showed the less risk for mortality than those without. Factors positively associated with the increase in albumin levels were low baseline albumin levels and anemia in the present study. Due to the nature of the observational studies, a causal relationship could not be determined in this study. However, an attempt to increase the albumin levels through the improvement in nutritional status in patients with HF is under investigation [20, 21]. Whether nutritional support promotes the recovery of multi-organ damage due to worsening of HF needs to be determined by clinical trials, and dietary recommendations should be provided at discharge to patients hospitalized for ADHF [22]. In addition, patients with no increase in albumin levels had a higher BNP levels at discharge, although the difference was not large. From a clinical point of view, the assessment of intravascular volume status is crucial when we interpret the change of albumin levels.

We could not assess the intra-individual variability. In order not only to evaluate the prognostic impact of any increase in albumin but also to evaluate the increase that is above the intra-individual variability, we added the sensitivity analysis on the percent change of albumin levels. From a prognostic point of view, the cut-off point might exist around -3.0%; however, this value may be validated by further studies. In addition, whether the increase of albumin can be associated with a lower risk for HF hospitalization should be determined by further studies.

Limitations

Several limitations of the present study should be noted. First, we did not analyze nor did we have the data for the serial measurements of albumin, and thus only compared the values at admission and at discharge. Second, the importance of the increase in albumin levels in patients with normal albumin levels needs further investigation, although 47.9% of patients with ADHF showed decreased albumin levels [5]. We included the baseline albumin values in the risk-adjusting variables and also performed subgroup analysis based on the baseline albumin values, showing a lower risk in the group with increased albumin levels during hospitalization regardless of the baseline albumin levels. Third, there may be unadjusted confounding factors present in the current study. Finally, those excluded for missing data included patients without hypertension and anemia.

Conclusion

Independent of baseline albumin levels, an increase of albumin was associated with a lower 1-year risk for all-cause death in patients hospitalized for ADHF.

Supporting information

S1 Checklist. STROBE statement—checklist of items that should be included in reports of cohort studies.

(DOCX)

Click here for additional data file. (32.7KB, docx)
S1 File. Ethical approval of other participating centers.

(DOCX)

Click here for additional data file. (18.1KB, docx)
S1 Fig. Adjusted survival curves.

(A) A composite of all cause death and hospitalization for heart failure. (B) All-cause death. (C) Heart failure hospitalization. We adjusted outcomes for the same variables in COX hazard model.

(DOCX)

Click here for additional data file. (66.7KB, docx)
S2 Fig. Kaplan Meier curves for the primary outcome measure stratified by quartiles of the percent change of albumin levels.

(DOCX)

Click here for additional data file. (140.8KB, docx)
S1 Table. Baseline characteristics of the patients with versus without albumin data.

(DOCX)

Click here for additional data file. (22KB, docx)
S2 Table. Patient characteristics based on the quartiles of the percent change of albumin levels.

(DOCX)

Click here for additional data file. (26.8KB, docx)
S3 Table. Outcomes based on the quartiles of the percent change of albumin levels.

(DOCX)

Click here for additional data file. (23.8KB, docx)
S4 Table. Subgroup analysis with or without acute coronary syndrome.

(DOCX)

Click here for additional data file. (25.2KB, docx)

Data Availability

The minimal data set is ethically restricted by the Institutional Review Board of Kyoto University Hospital. This is because the secondary use of the data was to be reviewed by the Ethics Commission at the time of the initial application. Data are available from the Ethics Committee (contact via TK or directly to ethcom@kuhp.kyoto-u.ac.jp) for researchers who meet the criteria for access to confidential data.

Funding Statement

This work was supported by the Japan Agency for Medical Research and Development [18059186] (Drs T. Kato, T. Kuwahara, and N. Ozasa). The founder had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication.

References

  • 1.Ziaeian B, Fonarow GC. Epidemiology and aetiology of heart failure. Nature reviews Cardiology. 2016;13(6):368–78. 10.1038/nrcardio.2016.25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Yaku H, Ozasa N, Morimoto T, Inuzuka Y, Tamaki Y, Yamamoto E, et al. Demographics, Management, and In-Hospital Outcome of Hospitalized Acute Heart Failure Syndrome Patients in Contemporary Real Clinical Practice in Japan—Observations From the Prospective, Multicenter Kyoto Congestive Heart Failure (KCHF) Registry. Circulation journal: official journal of the Japanese Circulation Society. 2018;82(11):2811–9. [DOI] [PubMed] [Google Scholar]
  • 3.Arques S, Roux E, Sbragia P, Gelisse R, Pieri B, Ambrosi P. Usefulness of serum albumin concentration for in-hospital risk stratification in frail, elderly patients with acute heart failure. Insights from a prospective, monocenter study. International journal of cardiology. 2008;125(2):265–7. 10.1016/j.ijcard.2007.07.094 [DOI] [PubMed] [Google Scholar]
  • 4.Bonilla-Palomas JL, Gamez-Lopez AL, Moreno-Conde M, Lopez-Ibanez MC, Anguita-Sanchez M, Gallego de la Sacristana A, et al. Hypoalbuminemia in acute heart failure patients: causes and its impact on hospital and long-term mortality. Journal of cardiac failure. 2014;20(5):350–8. 10.1016/j.cardfail.2014.01.016 [DOI] [PubMed] [Google Scholar]
  • 5.Kato T, Yaku H, Morimoto T, Inuzuka Y, Tamaki Y, Yamamoto E, et al. Association with Controlling Nutritional Status (CONUT) Score and In-hospital Mortality and Infection in Acute Heart Failure. Scientific reports. 2020;10(1):3320 10.1038/s41598-020-60404-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Arques S, Ambrosi P. Human serum albumin in the clinical syndrome of heart failure. Journal of cardiac failure. 2011;17(6):451–8. 10.1016/j.cardfail.2011.02.010 [DOI] [PubMed] [Google Scholar]
  • 7.Su K, Kato T, Toyofuku M, Morimoto T, Yaku H, Inuzuka Y, et al. Association of Previous Hospitalization for Heart Failure With Increased Mortality in Patients Hospitalized for Acute Decompensated Heart Failure. Circulation Reports. 2019;1(11):517–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Yamamoto E, Kato T, Ozasa N, Yaku H, Inuzuka Y, Tamaki Y, et al. Kyoto Congestive Heart Failure (KCHF) study: rationale and design. ESC heart failure. 2017;4(3):216–23. 10.1002/ehf2.12138 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Yaku H, Kato T, Morimoto T, Inuzuka Y, Tamaki Y, Ozasa N, et al. Association of Mineralocorticoid Receptor Antagonist Use With All-Cause Mortality and Hospital Readmission in Older Adults With Acute Decompensated Heart Failure. JAMA network open. 2019;2(6):e195892 10.1001/jamanetworkopen.2019.5892 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. European journal of heart failure. 2016;18(8):891–975. 10.1002/ejhf.592 [DOI] [PubMed] [Google Scholar]
  • 11.Japan’s ethical guidelines for epidemiologic research. 2008;2002. Available at: http://www.lifescience.mext.go.jp/files/pdf/n796_01.pdf.
  • 12.Kitai T, Miyakoshi C, Morimoto T, Yaku H, Murai R, Kaji S, et al. Mode of Death Among Japanese Adults With Heart Failure With Preserved, Midrange, and Reduced Ejection Fraction. JAMA network open. 2020;3(5):e204296 10.1001/jamanetworkopen.2020.4296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.RJ G. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Statist. 1988;16:1141–54. [Google Scholar]
  • 14.Fine JP GR. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496–509. [Google Scholar]
  • 15.Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone marrow transplantation. 2013;48(3):452–8. 10.1038/bmt.2012.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Allen LA, Felker GM, Pocock S, McMurray JJ, Pfeffer MA, Swedberg K, et al. Liver function abnormalities and outcome in patients with chronic heart failure: data from the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program. European journal of heart failure. 2009;11(2):170–7. 10.1093/eurjhf/hfn031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kawamoto A, Kato T, Minamino-Muta E, Okano Y, Shioi T, Kimura T. Relationships between nutritional status and markers of congestion in patients with pulmonary arterial hypertension. International journal of cardiology. 2015;187:27–8. 10.1016/j.ijcard.2015.03.354 [DOI] [PubMed] [Google Scholar]
  • 18.Koyama S, Sato Y, Tanada Y, Fujiwara H, Takatsu Y. Early evolution and correlates of urine albumin excretion in patients presenting with acutely decompensated heart failure. Circ Heart Fail. 2013;6(2):227–32. 10.1161/CIRCHEARTFAILURE.112.000152 [DOI] [PubMed] [Google Scholar]
  • 19.Nakayama H, Koyama S, Kuragaichi T, Shiba M, Fujiwara H, Takatsu Y, et al. Prognostic Value of Rising Serum Albumin During Hospitalization in Patients With Acute Heart Failure. Am J Cardiol. 2016;117(8):1305–9. 10.1016/j.amjcard.2016.01.030 [DOI] [PubMed] [Google Scholar]
  • 20.Uchino Y, Watanabe M, Takata M, Amiya E, Tsushima K, Adachi T, et al. Effect of Oral Branched-Chain Amino Acids on Serum Albumin Concentration in Heart Failure Patients with Hypoalbuminemia: Results of a Preliminary Study. American journal of cardiovascular drugs: drugs, devices, and other interventions. 2018;18(4):327–32. 10.1007/s40256-018-0269-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Tanada Y, Shioi T, Kato T, Kawamoto A, Okuda J, Kimura T. Branched-chain amino acids ameliorate heart failure with cardiac cachexia in rats. Life sciences. 2015;137:20–7. 10.1016/j.lfs.2015.06.021 [DOI] [PubMed] [Google Scholar]
  • 22.Regalbuto R, Maurer MS, Chapel D, Mendez J, Shaffer JA. Joint Commission requirements for discharge instructions in patients with heart failure: is understanding important for preventing readmissions? Journal of cardiac failure. 2014;20(9):641–9. 10.1016/j.cardfail.2014.06.358 [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Vincenzo Lionetti

19 Oct 2020

PONE-D-20-30330

Association of an increase in serum albumin levels with positive 1-year outcomes in acute decompensated heart failure: A cohort study

PLOS ONE

Dear Dr. Kato,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

 All issues raised by expert reviewers are required.

Please submit your revised manuscript by Dec 03 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • 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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Vincenzo Lionetti, M.D., PhD

Academic Editor

PLOS ONE

Journal Requirements:

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

1. 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. In your Methods section, please provide additional information about the statistical analysis performed, for example by describing any correction for multiple comparisons performed.

3. Please amend your list of authors on the manuscript to ensure that each author is linked to an affiliation. Authors’ affiliations should reflect the institution where the work was done (if authors moved subsequently, you can also list the new affiliation stating “current affiliation:….” as necessary).

[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: Yes

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: The present paper is an analysis of the Kyoto Congestive Heart Failure registry evaluating 3160 patients who were 1) discharged after an hospitalization for acute heart failure (HF) and 2) had serum albumin dosed at admission and at discharge. The Authors evaluated the relationship between any increase in serum albumin and all-cause death or HF hospitalization over a median 475 day follow-up. Patients with increased albumin (34%) had a lower risk of this outcome, even after adjusting for baseline albumin. The prognostic impact of increased albumin seemed greater in patients in the lower quartiles of baseline albumin. While there are no major flaws in the analysis and the results are reasonable, I have some doubts about the evaluation of the prognostic impact of "any increase in albumin", instead of an increase that is above the intra-individual variability of this biomarker. Please find below some other comments.

Abstract: "An intervention to increase the albumin levels in the treatment for ADHF needs to be investigated". I would delete this statement in the Abstract.

Introduction: please expand on the causes and prognostic impact of hypoalbuminemia.

Methods:

- "Death was regarded as cardiovascular in origin unless obvious noncardiovascular

causes could be identified". I would prefer to adjudicate as cardiovascular deaths only the cases with clear evidence of cardiovascular disease as the underlying cause, non-cardiovascular deaths those where there were other obvious cases, and classify as "unclear" the cause of the death in the remaining cases.

- Your definitions of "chronic kidney disease" and particularly "renal dysfunction" are questionable.

- "Interquartile range" would be better replaced by "interquartile interval".

- Page 15: "HF hospitalization associated with acute coronary syndrome". ACS events, possibly complicated by HF, should be clearly differentiated from acute HF.

Results:

- Table 1: BNP would be better expressed as ng/L.

- The percent changes in albumin in the 2 groups (any increase vs. no increase) should be reported. You may consider stratifying patients based on certain thresholds of delta % albumin.

- Follow-up duration could be converted in months.

- In your survival analysis, you should take into account competing risks (cardiovascular death vs. non-cardiovascular death, HF hospitalization vs. all-cause death).

- Splitting the population into two broad categories (any increase vs. no increase) instead of evaluating delta % changes in albumin results in a loss of information (doi: 10.1002/sim.2331). You should consider evaluating the prognostic value of delta % changes, possibly adjusting for disease severity, the duration of hospital admission, and other potential confounders.

Reviewer #2: In this paper, Kato et al. aimed at assessing the prognostic implication of a change in albumin during hospitalization for acute decompensated heart failure (ADHF) on prognosis.

The data of this study are derived from the Kyoto Congestive Heart Failure registry, a physician-initiated, prospective, observational, multicenter cohort study that enrolled consecutive patients hospitalized for ADHF October 2014 and March 2016.

This allowed the inclusion of an important number of patients (3160 patients).

The study is well written, and the results are interesting.

Nevertheless, I have some concerns that need to be addressed.

1) The authors state that patients were divided into 2 groups according to the increase or the absence of increase in albumin levels before discharge.

How the increase in albumin level was defined. A simple increase in 1 unit of albumin was sufficient to define increase, or a stricter definition was used?

2) The paragraph on Ethics is too long. For the publication, I think it is necessary to indicate that “The investigation conformed with the principles outlined in the Declaration of Helsinki and approved by the ethical committees of the participating hospitals”. A detailed list of these hospitals and corresponding approbation numbers should be provided as supplementary material.

3) The paragraph “Definition” doesn’t identify the content of the paragraph itself. I suggest indicating the title of this paragraph as “Outcome”.

4) The “detailed definition of patients characteristics should be inserted in the paragraph “Population”.

5) The authors should add some data about albumin assessment. The fact that they consider variation in albumin concentration during hospitalization is indicated in the introduction, but this should be clearly indicated in the method. They should indicate when Albumine was assessed at admission and before discharge. Any increase in albumin level (even the smallest one) corresponded to albumin increase? This should be defined in detail.

6) Can the author indicate the cause of hospital admission?

7) In table 1, the p-value for HFmEF and HFpEF is not indicated and should be added

8) In table 1, the definition “Albumin, g/dl *☨” should be replaced by “Increase in albumin levels”.

9) In table 1, the p-value for the different quartiles of albumin is not indicated

10) Were patients taking also non-MRA diuretics at discharge?

11) Table 2 is difficult to read. I think the authors should simply indicate unadjusted an unadjusted HR in this table and maybe add the comparison of events among groups in Table 1. A similar comment can be applied to Table 3.

12) From a clinical point of view, were all patients discharged in a condition of euvolemia? Did the study include patients admitted for cardiogenic shock?

13) In the conclusion the authors state that “Independent of baseline albumin levels, an increase of albumin was associated with a lower 1-year risk for the composite of all-cause death or HF hospitalization in patients hospitalized for ADHF”. P-value for the adjusted risk of HF is >0.05, which means that an increase in albumin level is not associated with a significant reduction in HF hospitalization. The statement should be corrected and these results should put in perspective in the discussion.

14) I think that all the survival curves should indicate the adjusted comparison of survival/hospitalization curve and not unadjusted data. Otherwise, the results are largely misleading.

15) Were the authors able to identify parameters associated with the increase in albumin levels during hospitalization?

**********

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.

PLoS One. 2020 Dec 28;15(12):e0243818. doi: 10.1371/journal.pone.0243818.r002

Author response to Decision Letter 0

23 Nov 2020

We thank the reviewers for careful assessment and positive comments.

Reviewer #1: The present paper is an analysis of the Kyoto Congestive Heart Failure registry evaluating 3160 patients who were 1) discharged after an hospitalization for acute heart failure (HF) and 2) had serum albumin dosed at admission and at discharge. The Authors evaluated the relationship between any increase in serum albumin and all-cause death or HF hospitalization over a median 475 day follow-up. Patients with increased albumin (34%) had a lower risk of this outcome, even after adjusting for baseline albumin. The prognostic impact of increased albumin seemed greater in patients in the lower quartiles of baseline albumin. While there are no major flaws in the analysis and the results are reasonable, I have some doubts about the evaluation of the prognostic impact of "any increase in albumin", instead of an increase that is above the intra-individual variability of this biomarker.

Response:

We thank the reviewer for very important comments. In order to evaluate the prognostic impact of "any increase in albumin", instead of an increase that is above the intra-individual variability, we also included the sensitivity analysis on the percent change of albumin levels. We assessed the outcomes according to the quartiles of the % change and showed that the higher and the highest quartiles were significantly associated with the lower risk relative to the lowest quartile of the % change.

“We compared albumin levels at admission/hospital presentation (anytime of the day) and at the nearest time to discharge in the morning. Any increase in albumin level corresponded to albumin increase.” (Methods, page 5, lines 19-21)

“Sensitivity analysis based on the percent change of albumin levels

We stratified patients into 4 groups based on the quartiles of the percent change of albumin levels: the lowest quartile (≤ -11.1%), the lower quartile (> -11.1% and ≤ -3.0%), the higher quartile (> -3.0% and ≤ 5.3%), and the highest quartile (> 5.3%) (S4 Table). The cumulative 1-year incidence of the primary outcome measure (a composite endpoint of all-cause death or hospitalization for HF) was significantly lower in the higher and the highest quartiles than that in the lowest quartile (S5 Figure). The lower risk of the higher and the highest percent change groups relative to the lowest percent change group was significant after adjustment (HR: 0.70, 95% CI: 0.58-0.83, P <0.0001) (S6 Table)”. (Results, page 16)

“We could not assess the intra-individual variability. In order not only to evaluate the prognostic impact of any increase in albumin but also to evaluate the increase that is above the intra-individual variability, we added the sensitivity analysis on the percent change of albumin levels. From a prognostic point of view, the cut-off point might exist around -3.0%; however, this value may be validated by further studies”. (Discussion, page 21, lines 12-16)

Please find below some other comments.

Abstract: "An intervention to increase the albumin levels in the treatment for ADHF needs to be investigated". I would delete this statement in the Abstract.

Response:

Thank you for your suggestions. I have removed this sentence in the Abstract.

Introduction: please expand on the causes and prognostic impact of hypoalbuminemia.

Response:

We appreciate your comments. Now we have expanded the introduction section as follows:

“This prognostic value probably refers primarily to the syndrome of malnutrition-inflammation and the severity of comorbidities (5). Albumin decreased by the malnutrition, hemodilution, renal loss, and shortened half-life due to severe illness such as inflammation (6). Serum albumin has many physiological properties, including in particular antioxidant, anti-inflammatory, anticoagulant and anti-platelet aggregation activity. It also plays an essential role in the exchange of fluids across the capillary membrane (6). In addition to that, hypoalbuminemia may be a potentially modifiable risk factor through the disruption of the protective roles and the exacerbation of peripheral congestion and pulmonary edema (6).” (Introduction, page 4, lines 6-14)

Methods:

- "Death was regarded as cardiovascular in origin unless obvious noncardiovascular

causes could be identified". I would prefer to adjudicate as cardiovascular deaths only the cases with clear evidence of cardiovascular disease as the underlying cause, non-cardiovascular deaths those where there were other obvious cases, and classify as "unclear" the cause of the death in the remaining cases.

Response:

We thank the reviewer for the important points. The definition of the death in the present study was in accordance with the previous studies in this registry. The clear identification of death is actually difficult when the patients with heart failure often accompany the infection such as pneumonia, although we rigorously collected the data; thus, we adopted the all cause death (and heart failure hospitalization) as the primary outcome in the main analysis.

- Your definitions of "chronic kidney disease" and particularly "renal dysfunction" are questionable.

Response:

We appreciate your comments. We deleted the unclear or questionable definitions and use the term eGFR<30 ml/min/1.73m2.

- "Interquartile range" would be better replaced by "interquartile interval".

Response:

We have replaced interquartile range by interquartile interval.

- Page 15: "HF hospitalization associated with acute coronary syndrome". ACS events, possibly complicated by HF, should be clearly differentiated from acute HF.

Response:

We thank the reviewer for the careful assessment. We included the patients hospitalized with acute decompensated heart failure with any cause. As the reviewer pointed out, ACS is one of the causes of acute decompensation and should be differentiated. Thus, we have added the subgroup analysis stratified by ACS.

“Subgroup analysis based on the acute coronary syndrome

When stratified by the acute coronary syndrome, the lower risk of the increase group relative to the no-increase group for the primary outcomes and all-cause death was consistently seen in patients with or without acute coronary syndrome. The effect was directionally significant in patients with acute coronary syndrome (S7 Table).” (Result, page 18)

Results:

- Table 1: BNP would be better expressed as ng/L.

Response:

We have changed the expression.

- The percent changes in albumin in the 2 groups (any increase vs. no increase) should be reported. You may consider stratifying patients based on certain thresholds of delta % albumin.

Response:

We also included the % change in the 2 groups. We also included the analysis stratified by the quartile based on the % change as mentioned above.

- Follow-up duration could be converted in months.

Response:

We have converted the follow-up duration in months.

- In your survival analysis, you should take into account competing risks (cardiovascular death vs. non-cardiovascular death, HF hospitalization vs. all-cause death).

Response:

We have modified the survival analysis with competing risks regarding CV death vs. non-CV death and HF hospitalization vs. all-cause death. We have added the Fine and Grays’ method in the method section.

- Splitting the population into two broad categories (any increase vs. no increase) instead of evaluating delta % changes in albumin results in a loss of information (doi: 10.1002/sim.2331). You should consider evaluating the prognostic value of delta % changes, possibly adjusting for disease severity, the duration of hospital admission, and other potential confounders.

Response:

We thank the reviewer for very important comments. As a sensitivity analysis, we included the % change of albumin levels and assess the outcomes according to the quartiles of the % change. We used the variables for this analysis same as the main analysis adding the duration of hospital stay. The results are mostly consistent with the main analysis. (Results, page 16)

 

Reviewer #2: In this paper, Kato et al. aimed at assessing the prognostic implication of a change in albumin during hospitalization for acute decompensated heart failure (ADHF) on prognosis.

The data of this study are derived from the Kyoto Congestive Heart Failure registry, a physician-initiated, prospective, observational, multicenter cohort study that enrolled consecutive patients hospitalized for ADHF October 2014 and March 2016.

This allowed the inclusion of an important number of patients (3160 patients).

The study is well written, and the results are interesting.

Nevertheless, I have some concerns that need to be addressed.

Response:

We thank the reviewers for careful assessment and positive comments.

1) The authors state that patients were divided into 2 groups according to the increase or the absence of increase in albumin levels before discharge.

How the increase in albumin level was defined. A simple increase in 1 unit of albumin was sufficient to define increase, or a stricter definition was used?

Response:

We thank the reviewer for the very important questions. The increase was defined as an increase in less than 1 unit. As the reviewer 1 pointed out, there may be an intra-individual variety; thus, we have added a sensitivity analysis using the % change and assess the outcomes according to the quartiles of the % change. The results were almost consistent with the main analysis. In the analysis, there seems to be a difference in outcomes between Q2 and Q3.

However, the purpose of the present study is not seeking the threshold of albumin changes. Instead, we adopted the simpler definition i.e. increase or non-increase in the main analysis.

“Sensitivity analysis based on the percent change of albumin levels

We stratified patients into 4 groups based on the quartiles of the percent change of albumin levels: the lowest quartile (≤ -11.1%), the lower quartile (> -11.1% and ≤ -3.0%), the higher quartile (> -3.0% and ≤ 5.3%), and the highest quartile (> 5.3%) (S4 Table). The cumulative 1-year incidence of the primary outcome measure (a composite endpoint of all-cause death or hospitalization for HF) was significantly lower in the higher and the highest quartiles than that in the lowest quartile (S5 Figure). The lower risk of the higher and the highest percent change groups relative to the lowest percent change group was significant after adjustment (HR: 0.70, 95% CI: 0.58-0.83, P <0.0001) (S6 Table)”. (Results, page 16)

“We could not assess the intra-individual variability. In order not only to evaluate the prognostic impact of any increase in albumin but also to evaluate the increase that is above the intra-individual variability, we added the sensitivity analysis on the percent change of albumin levels. From a prognostic point of view, the cut-off point might exist around -3.0%; however, this value may be validated by further studies”. (Discussion, page 21, lines 12-16)

2) The paragraph on Ethics is too long. For the publication, I think it is necessary to]: indicate that “The investigation conformed with the principles outlined in the Declaration of Helsinki and approved by the ethical committees of the participating hospitals”. A detailed list of these hospitals and corresponding approbation numbers should be provided as supplementary material.

Response:

We have included the details of ethics in the supplementary material.

3) The paragraph “Definition” doesn’t identify the content of the paragraph itself. I suggest indicating the title of this paragraph as “Outcome”.

Response:

We thank the reviewer for the comments. I have changed the title of the paragraph.

4) The “detailed definition of patients characteristics should be inserted in the paragraph “Population”.

Response:

We have included the definition of patient’s characteristics should be inserted in the paragraph “Population”.

5) The authors should add some data about albumin assessment. The fact that they consider variation in albumin concentration during hospitalization is indicated in the introduction, but this should be clearly indicated in the method. They should indicate when albumin was assessed at admission and before discharge. Any increase in albumin level (even the smallest one) corresponded to albumin increase? This should be defined in detail.

Response:

We have included the data about albumin assessment including the timing of assessment in the method section. We clarify any increase in albumin level defined as an increase in albumin.

“We compared albumin levels at admission/hospital presentation (anytime of the day) and at the nearest time to discharge in the morning. Any increase in albumin level corresponded to albumin increase.” (Methods, page 5, lines 19-21)

In order not only to evaluate the prognostic impact of any increase in albumin but also to evaluate the increase that is above the intra-individual variability, we added the sensitivity analysis on the percent change of albumin levels as mentioned above.

6) Can the author indicate the cause of hospital admission?

Response:

We appreciate your comments. Unfortunately, the precipitating factors for the acute decompensation could not be clearly identified and not collected in the present study.

7) In table 1, the p-value for HFmEF and HFpEF is not indicated and should be added.

Response:

We have added the p-value in the table.

8) In table 1, the definition “Albumin, g/dl” should be replaced by “Increase in albumin levels”.

Response:

We have added the increase in albumin levels and the definition in Table 1.

9) In table 1, the p-value for the different quartiles of albumin is not indicated.

Response:

We have added the p-value in the table.

10) Were patients taking also non-MRA diuretics at discharge?

Response:

We have added the diuretics at discharge in the table 1. We thank the reviewer for the careful assessment.

11) Table 2 is difficult to read. I think the authors should simply indicate unadjusted an unadjusted HR in this table and maybe add the comparison of events among groups in Table 1. A similar comment can be applied to Table 3.

Response:

Thank you for your kind suggestion. In table 2 and 3, comparisons between the two groups was based on the time-considering method, such as proportional hazard method. In addition, it would be clearly presented how the unadjusted hazard ratio increases or decreases after the adjustment for confounders. Thus, we would like to keep the Table 2 and 3 in the present form in the revised manuscript.

12) From a clinical point of view, were all patients discharged in a condition of euvolemia? Did the study include patients admitted for cardiogenic shock?

Response:

We appreciate your very important comments. It is very difficult to say all patients discharge were in a condition of euvolemia. Instead, we have added the BNP levels at discharge and discussions about the volume status. We included all patients who had acute decompensated heart failure and underwent heart failure-specific treatment involving intravenous drugs within 24 hours of hospital presentation. Thus, patients with low blood pressure were included in the present study. We included the systolic blood pressure <90mmHg into the adjusting variables.

“In addition, patients with no increase in albumin levels had a higher BNP levels at discharge, although the difference was not large. From a clinical point of view, the assessment of intravascular volume status is crucial when we interpret the change of albumin levels.”(Page 22, lines 8-11)

13) In the conclusion the authors state that “Independent of baseline albumin levels, an increase of albumin was associated with a lower 1-year risk for the composite of all-cause death or HF hospitalization in patients hospitalized for ADHF”. P-value for the adjusted risk of HF is >0.05, which means that an increase in albumin level is not associated with a significant reduction in HF hospitalization. The statement should be corrected and these results should put in perspective in the discussion.

Response:

We thank the reviewer for the valuable comments. Primary outcome is a composite outcome but the differences in HF hospitalization were not statistically significant. As the reviewer pointed out, we have specified the outcomes to avoid the unclearness in the conclusion section, and added the perspectives in the discussion section.

“whether the increase of albumin can be associated with a lower risk for HF hospitalization should be determined by further studies.” (Discussion, page 22, lines 16-18)

“Conclusion Independent of baseline albumin levels, an increase of albumin was associated with a lower 1-year risk for all-cause death in patients hospitalized for ADHF.” (Page 23, lines 8-10)

14) I think that all the survival curves should indicate the adjusted comparison of survival/hospitalization curve and not unadjusted data. Otherwise, the results are largely misleading.

Response:

We appreciate your comments. We described the cumulative incidence of each group in the text and main figures consistently. We totally agree on the reviewer’s suggestion; thus, we have added the adjusting survival curves in the supplementary figures. If the editors and reviewers strongly suggests that this result should be included in the main figure, we will be glad to add the data in the main figures.

15) Were the authors able to identify parameters associated with the increase in albumin levels during hospitalization?

Response:

We appreciate your comments. We have included the multivariate analysis regarding the baseline factors associated with the increase in albumin levels.

“To analyze the baseline factors associated with high LVMI, we used a multivariable logistic regression model involving the following potential independent clinically relevant variable. The adjusted odds ratios and 95% CIs were calculated.” (Methods, page 9, lines 8-10)

“Baseline factors associated with the increase in albumin levels

According to the multivariable logistic regression analysis, the albumin levels at admission < 3 g/dL and anemia were independent positive factors associated with the increase in albumin levels, while age ≥80 years, eGFR<30 ml/min/1.73m2, body mass index<22kg/m2, NYHA III/IV at presentation, and the history of diabetes mellitus were independent negative factors associated with the increase in albumin levels (Table 4).” (Results, page 19-20)

Attachment

Submitted filename: Review comments.docx

Click here for additional data file. (26.7KB, docx)

Decision Letter 1

Vincenzo Lionetti

27 Nov 2020

Association of an increase in serum albumin levels with positive 1-year outcomes in acute decompensated heart failure: A cohort study

PONE-D-20-30330R1

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,

Vincenzo Lionetti, M.D., PhD

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: The Authors have modified their manuscript according to my suggestions. I have no further comments.

Reviewer #2: The authors dealt with al the comments raised by the reviewed. They added supplementary statistical analysis. The current version of the manuscript is significantly improved and I do not propose further modifications. I think that the current version of the article is suitable for publication.

**********

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

Acceptance letter

Vincenzo Lionetti

11 Dec 2020

PONE-D-20-30330R1

Association of an increase in serum albumin levels with positive 1-year outcomes in acute decompensated heart failure: A cohort study 

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. Vincenzo Lionetti

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Checklist. STROBE statement—checklist of items that should be included in reports of cohort studies.

    (DOCX)

    Click here for additional data file. (32.7KB, docx)
    S1 File. Ethical approval of other participating centers.

    (DOCX)

    Click here for additional data file. (18.1KB, docx)
    S1 Fig. Adjusted survival curves.

    (A) A composite of all cause death and hospitalization for heart failure. (B) All-cause death. (C) Heart failure hospitalization. We adjusted outcomes for the same variables in COX hazard model.

    (DOCX)

    Click here for additional data file. (66.7KB, docx)
    S2 Fig. Kaplan Meier curves for the primary outcome measure stratified by quartiles of the percent change of albumin levels.

    (DOCX)

    Click here for additional data file. (140.8KB, docx)
    S1 Table. Baseline characteristics of the patients with versus without albumin data.

    (DOCX)

    Click here for additional data file. (22KB, docx)
    S2 Table. Patient characteristics based on the quartiles of the percent change of albumin levels.

    (DOCX)

    Click here for additional data file. (26.8KB, docx)
    S3 Table. Outcomes based on the quartiles of the percent change of albumin levels.

    (DOCX)

    Click here for additional data file. (23.8KB, docx)
    S4 Table. Subgroup analysis with or without acute coronary syndrome.

    (DOCX)

    Click here for additional data file. (25.2KB, docx)
    Attachment

    Submitted filename: Review comments.docx

    Click here for additional data file. (26.7KB, docx)

    Data Availability Statement

    The minimal data set is ethically restricted by the Institutional Review Board of Kyoto University Hospital. This is because the secondary use of the data was to be reviewed by the Ethics Commission at the time of the initial application. Data are available from the Ethics Committee (contact via TK or directly to ethcom@kuhp.kyoto-u.ac.jp) for researchers who meet the criteria for access to confidential data.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES