Timing of Worsening Renal Function in Patients Hospitalized for Heart Failure Exacerbation Who Were Being Treated With Intravenous Diuretic Therapy

Abstract

Background: This retrospective observational study investigated the incidence of worsening renal function (WRF) in patients hospitalized for heart failure (HF) and treated with intravenous diuretics in Japan.

Methods and Results: Associations between WRF at any point and HF treatments, and the effects of WRF on outcomes were evaluated (Diagnosis Procedure Combination database). Of 1,788 patients analyzed (mean [±SD] age 80.5±10.2 years; 54.4% male), 641 (35.9%) had WRF during a course of hospitalization for worsening HF: 208 (32.4%) presented with WRF before admission (BA-WRF; estimated glomerular filtration rate decreased by ≥25% from baseline at least once between 30 days prior to admission and admission); 44 (6.9%) had WRF that persisted before and after admission (P-WRF); and 389 (60.7%) had WRF develop after admission (AA-WRF). Delayed initial diuretic administration, higher maximum doses of intravenous diuretics during hospitalization, and diuretic readministration during hospitalization were associated with a significantly higher incidence of AA-WRF. Patients with WRF at any time point were at higher risk of death during hospitalization compared with patients without WRF, with adjusted hazard ratios of 3.56 (95% confidence interval [CI] 2.23–5.69) for BA-WRF, 3.23 (95% CI 2.21–4.71) for AA-WRF, and 13.16 (95% CI 8.19–21.15) for P-WRF (all P<0.0001).

Conclusions: Forty percent of WRF occurred before admission for acute HF; there was no difference in mortality between patients with BA-WRF and AA-WRF.

Worldwide, approximately 64 million people have heart failure (HF).¹ In Japan, the number of patients with HF is increasing rapidly because of the aging population.² According to the Japanese Circulation Society and results of the Japanese Registry Of All cardiac and vascular Diseases (JROAD), the number of patients with HF hospitalization increased from 212,793 in 2012 to 260,157 in 2016, indicating an increase in the number of HF cases by over 10,000 patients per year, and there are 3.5 times the number of hospitalizations for HF than for acute myocardial infarction.³ Furthermore, the costs associated with hospitalizations for HF are higher than the costs for any other acute cardiovascular disease in Japan.⁴

The management of acute congestive HF aims to alleviate fluid overload, which is often achieved through the intravenous (IV) administration of pharmacologic agents such as diuretics, particularly loop diuretics. Despite their essential role in the treatment of congestive HF, the excessive use of loop diuretics has been associated with a decline in renal function and a poorer patient prognosis.^5–7 Considering this, the European Society of Cardiology has provided clinical practice guidelines that outline a specific timeline for administering diuretics to relieve signs and symptoms, prevent organ injury, and improve the prognosis of patients with HF.⁸

Worsening renal function (WRF) is a common complication in patients with congestive HF, with an incidence of around 30% during hospitalization.^5,9,10 The relationship between WRF and the prognosis of HF patients is a contentious topic in the literature. Although some studies found an association between WRF and adverse outcomes and increased mortality in the acute and post-hospitalization phases of HF,^5,11–14 others did not;^15–17 thus, no conclusion has been reached. Previous reports are also limited in that they only evaluated WRF after admission,¹⁸ which may not capture the full picture of WRF in a clinical course of acute exacerbation of HF. Reportedly, renal function gradually worsens with worsening HF from approximately 30 days before admission.¹⁹

The aims of this study were to: (1) estimate the incidence of WRF in patients hospitalized for HF and treated with IV diuretics (e.g., furosemide); (2) determine whether treatment after admission was associated with the development of WRF; (3) investigate the association between the presence or absence of WRF before admission (from 30 days prior to the day of admission) and WRF after admission; and (4) clarify whether there are differences in patient prognosis between WRF presenting before or after admission.

Methods

Study Design

This was an observational study targeting patients with HF identified by International Classification of Diseases, 10th Revision codes I09.9, I11.0, I13.9, and I50.0–I50.9 who were hospitalized at 40 facilities between January 1, 2018, and November 30, 2020, who received IV diuretics (furosemide) within the period from the day before admission to 2 days after admission, in line with the recommendations for acute exacerbations of congestive HF,⁸ and for whom laboratory data were available.

The study protocol and related documents were reviewed and approved by the Otsuka Pharmaceutical Co., Ltd. Research and Development Research Ethics Committee (Reference no. 210127). The study was conducted in compliance with the protocol and the ethical principles that have their origin in the Declaration of Helsinki and the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.²⁰ The study was registered with the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (https://www.umin.ac.jp/) under the identifier UMIN000046005.

Patients

The inclusion criteria were as follows: (1) patients hospitalized for HF within the inclusion period (in case of multiple hospitalizations during the study period, only data for the first hospitalization were included); (2) patients prescribed an IV diuretic during the hospitalization (from the day before admission for HF to 2 days after admission); and (3) patients whose estimated glomerular filtration rate (eGFR) was measured 30 days before admission, on the day of admission, and after admission.

Patients were excluded from the study if they were <20 years of age on the day of admission for HF, had a history of at least 1 dialysis treatment before admission due to HF, and were hospitalized for >30 days.

Study Measures and Outcomes

To determine WRF before admission, eGFR measured from 1 year to 30 days before admission (with the value closest to 30 days before admission) was used as the baseline. eGFR values were calculated as follows:

eGFR = 194 × (serum creatinine)^−1.094 × age^−0.287 in males

eGFR = 194 × (serum creatinine)^−1.094 × age^−0.287 × 0.739 in females

WRF before admission was defined as a decrease in eGFR by ≥25% from the baseline at least once in the period from 30 days before admission to the day of admission. Patients were categorized into the 4 groups based on when they experienced WRF: WRF before admission only (BA-WRF), WRF after admission only (AA-WRF), persistent WRF (P-WRF) before and after admission, and no WRF (non-WRF).

To determine WRF after admission, eGFR measured on the day of hospitalization was used as the baseline. WRF after admission was defined as a decrease in eGFR by ≥25% from baseline at least once from the day after admission to the day of discharge, which differs slightly from the definition of WRF from the Heart Failure Association of the European Society of Cardiology (e.g., a 25% decrease in GFR within 48 h).¹⁸

This study evaluated the following parameters: the incidence of WRF in patients hospitalized for HF and administered IV diuretics; the characteristics of the use of diuretics (e.g., the duration from admission to the initial dose, initial dose, duration from admission to the maximum dose, maximum dose, and associations with the development of WRF); the association between the presence or absence of WRF before admission (from 30 days before admission to the day of admission) and after admission; and in-hospital deaths within 30 days after admission.

Data Source and Collection

Data were obtained from the Diagnosis Procedure Combination (DPC) entries in the database curated by Medical Data Vision Co., Ltd. (MDV; Tokyo, Japan). The database contains data from 37.42 million patients treated in 451 acute care hospitals, amounting to approximately 26% of Japanese hospitals after adoption of the DPC on September 30, 2021. Data were anonymized; thus, no consent was required from patients included in the database.

Statistical Analysis

Descriptive statistics are used for baseline characteristics; categorical variables are presented as the number and percentage of total patients, whereas continuous variables are presented as the mean±SD or median with interquartile range (IQR).

The incidence rate of WRF was plotted as Kaplan-Meier curves, and the cumulative incidence rate at each time point (2, 5, 10, 15, and 30 days) was calculated for the time until the onset of WRF according to the administration of IV diuretics; the groups were compared using log-rank tests. Mortality was plotted using a Kaplan-Meier curve for time to in-hospital death according to the presence or absence of WRF before and after admission and according to tolvaptan use (yes/no) during hospitalization. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated and tested using a Cox proportional hazards model adjusted for sex and age or adjusted for sex, age, and oral diuretic use before/after admission. The same analysis was performed with the following adjustments for sensitivity analysis: sex; age; eGFR at 30 days before admission; and hypertension, dyslipidemia, and activities of daily living (ADL) score at the time of admission.

No imputation was used for missing data. A 2-sided α of 0.05 was used to establish statistical significance using t-tests, χ² tests, Kruskal-Wallis tests, and analysis of variance, and data were calculated with 95% CIs. No adjustments were made for multiplicity because of the exploratory nature of the tests. Statistical Analysis Software version 9.4 (SAS Institute, Inc., Cary, NC, USA) was used for the statistical analysis.

Results

Patients and WRF

Of the 96,579 patients with HF enrolled in the DPC database, 13,684 were admitted for HF during the observation period. The number of patients prescribed IV diuretics from the day before admission to 2 days after admission was 9,958. After excluding 305 patients, mainly for missing data or loss to follow-up, and removing data from 7,865 patients missing eGFR data measured 30 days before admission, on the day of admission, or after admission, data for 1,788 patients were included in the analysis (Figure 1).

Figure 1.

Patient disposition. eGFR, estimated glomerular filtration rate; HF, heart failure; IV, intravenous; MDV, Medical Data Vision.

Of the 1,788 patients analyzed, 252 (14.1%) presented with WRF within 30 days prior to admission. Of these 252 patients, the renal function of 44 (17.5%) worsened further after admission (P-WRF), whereas that of 208 (82.5%) remained stable (BA-WRF). Of the 1,536 patients with stable renal function before admission, 389 (25.3%) presented with WRF after admission (AA-WRF) and 1,147 (74.7%) did not (non-WRF).

Table 1 summarizes the baseline characteristics of patients at the time of hospitalization by the presence of WRF before admission and the development or further worsening of renal function after admission. At baseline, there were few between-group differences in comorbidities (e.g., hypertension, dyslipidemia, diabetes) and risk factors, except for age and chronic kidney disease. However, those with AA-WRF tended to use fewer angiotensin-converting enzyme inhibitors (P=0.526) and β-blockers (P=0.018). Loop diuretics were used with a similar frequency by patients with P-WRF and patients with AA-WRF and non-WRF. Tolvaptan was the most frequently used concomitant medication among patients with WRF, regardless of its timing. In addition, potassium levels tended to be higher in P-WRF than non-WRF patients. Patients with AA-WRF tended to have a higher systolic blood pressure than non-WRF patients.

Table 1.

Baseline Characteristics According to WRF Status Before and After Admission

Variable	Study population (n=1,788)	WRF 30 days before admission				P value
		Yes		No
		WRF after admission		WRF after admission
		Yes (n=44)	No (n=208)	Yes (n=389)	No (n=1,147)
Sex
Male	991 (55.4)	27 (61.4)	110 (52.9)	193 (49.6)	661 (57.6)	0.033A
Female	797 (44.6)	17 (38.6)	98 (47.1)	196 (50.4)	486 (42.4)	–
Age at time of hospitalization (years)	80.5±10.2	79.6±11.2	79.4±10.1	81.8±9.6	80.3±10.4	0.025B
Age ≥85 years	690 (38.6)	15 (34.1)	68 (32.7)	165 (42.4)	442 (38.5)	–
AHF hospitalization within 1 year	301 (16.8)	5 (11.4)	36 (17.3)	52 (13.4)	208 (18.1)	0.127A
Period between prior and current hospitalization (days)	107.0 [49.0–222.0]	32.0 [21.0–57.0]	123.5 [64.5–240.0]	78.0 [42.0–187.5]	115.5 [51.0–230.0]	–
SBP at admission (mmHg)						<0.001A
<100	120 (8.2)	6 (15.4)	27 (16.8)	23 (7.4)	64 (6.7)
100 to <140	770 (52.3)	21 (53.8)	85 (52.8)	134 (43.2)	530 (55.2)
≥140	581 (39.5)	12 (30.8)	49 (30.4)	153 (49.4)	367 (38.2)
BMI (kg/m2)	n=1,730	n=41	n=201	n=373	n=1,115	0.547B
Mean±SD	23.22±4.49	22.77±4.66	22.85±4.34	23.24±4.81	23.30±4.40
Minimum–maximum	13.2–50.0	13.8–35.3	14.6–42.5	14.3–47.4	13.2–50.0
Comorbidities at admission
Hypertension	1,330 (74.4)	30 (68.2)	150 (72.1)	290 (74.6)	860 (75.0)	–
Diabetes	755 (42.2)	16 (36.4)	101 (48.6)	173 (44.5)	465 (40.5)	–
Ischemic heart disease	702 (39.3)	13 (29.5)	72 (34.6)	147 (37.8)	470 (41.0)	–
Dyslipidemia	694 (38.8)	11 (25.0)	76 (36.5)	145 (37.3)	462 (40.3)	–
CKD	535 (29.9)	17 (38.6)	55 (26.4)	109 (28.0)	354 (30.9)	–
Arrhythmia	903 (50.5)	21 (47.7)	95 (45.7)	183 (47.0)	604 (52.7)	–
Valvular disease	90 (5.0)	3 (6.8)	6 (2.9)	15 (3.9)	66 (5.8)	–
COPD	95 (5.3)	0 (0.0)	12 (5.8)	12 (3.1)	71 (6.2)	–
Charlson Comorbidity Index	3.9±2.4	4.4±2.3	3.9±2.3	3.8±2.4	3.9±2.4	0.403B
Loop diuretics at admission	1,211 (67.7)	31 (70.5)	149 (71.6)	239 (61.4)	792 (69.0)	0.028A
Other concomitant medications at admission
β-blocker	921 (51.5)	24 (54.5)	98 (47.1)	178 (45.8)	621 (54.1)	0.018A
ACE inhibitor	292 (16.3)	7 (15.9)	35 (16.8)	54 (13.9)	196 (17.1)	0.526A
ARB	680 (38.0)	17 (38.6)	80 (38.5)	156 (40.1)	427 (37.2)	0.790A
Aldosterone antagonist	473 (26.5)	12 (27.3)	65 (31.3)	103 (26.5)	293 (25.5)	0.398A
Tolvaptan	318 (17.8)	7 (15.9)	46 (22.1)	59 (15.2)	206 (18.0)	0.261A
Thiazide diuretic	150 (8.4)	1 (2.3)	15 (7.2)	37 (9.5)	97 (8.5)	0.367A
SGLT2 inhibitor	92 (5.1)	2 (4.5)	8 (3.8)	17 (4.4)	65 (5.7)	0.600A
Laboratory measures
Sodium (mEq/L)	139.66±4.51	139.65±5.47	138.30±5.16	139.58±4.78	139.94±4.19	<0.001B
Potassium (mEq/L)	4.299±0.698	4.611±0.885	4.573±0.786	4.185±0.714	4.276±0.652	<0.001B
Creatinine (mg/dL)	1.546±1.142	2.254±1.211	2.050±1.704	1.380±0.918	1.485±1.044	<0.001B
BUN (mg/dL)	30.09±17.63	48.97±27.29	41.89±22.77	26.67±14.02	28.39±15.83	<0.001B
eGFR (mL/min/1.73 m2)	38.98 [5.92–54.93]	26.01 [15.82–32.00]	29.65 [17.64–39.78]	41.59 [28.15–58.05]	40.77 [27.20–56.22]	<0.001B
eGFR <60 mL/min/1.73 m2	1,436 (80.3)	44 (100.0)	198 (94.2)	296 (76.1)	900 (78.5)
BNP (pg/mL)	633.70 [309.00– 1,107.20]	787.20 [416.20– 971.00]	909.90 [519.60– 1,533.40]	551.80 [313.60– 1,015.90]	588.30 [290.30– 1,052.50]	<0.001B
NT-proBNP (pg/mL)	5,452.00 [2,602.00– 10,522.00]	3,790.00 [2,730.00– 16,411.00]	6,159.00 [3,474.00– 14,675.00]	7,059.00 [4,043.00– 11,463.00]	5,186.00 [2,388.00– 10,120.00]	0.012B
Albumin (g/dL)	3.480±0.510	3.302±0.600	3.429±0.544	3.448±0.511	3.507±0.497	0.009B
NYHA functional class						0.107A
1	9 (2.5)	1 (12.5)	2 (8.3)	3 (3.4)	3 (1.2)
2	66 (18.1)	0 (0.0)	3 (12.5)	15 (17.2)	48 (19.5)
3	158 (43.3)	2 (25.0)	12 (50.0)	34 (39.1)	110 (44.7)
4	132 (36.2)	5 (62.5)	7 (29.2)	35 (40.2)	85 (34.6)
ADL score at admission	55.8±39.7	34.9±42.7	48.1±39.6	48.4±40.1	60.3±38.8	<0.001B

Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). ^AP value calculated using χ² tests. ^BP value calculated using analysis of variance. ACE, angiotensin-converting enzyme; ADL, activities of daily living; AHF, acute heart failure; ARB, angiotensin II receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; BUN, blood urea nitrogen; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; NT-proBNP, N-terminal pro B-type natriuretic peptide; NYHA, New York Heart Association; SBP, systolic blood pressure; SGLT2, sodium-glucose cotransporter 2; WRF, worsening renal function.

Occurrence of WRF After Admission

Supplementary Figure 1 shows the occurrence of WRF after admission, which increased with the number of days after admission. The cumulative incidence of WRF at 30 days after admission was 35.8% (95% CI 32.0–40.0%).

IV Diuretic Use During Hospitalization and Effects on WRF Occurrence

Table 2 summarizes the use of IV diuretics. Overall, 89.9% of the study population received IV diuretics on Day 1 after admission. However, on Day 1, IV diuretics tended to be prescribed less often to patients with P-WRF (77.3%) and with AA-WRF (87.7%) than to non-WRF patients (91.5%). A higher proportion of patients with P-WRF was prescribed an intermediate dose of loop diuretics (>20–≤40 and >40–≤80 mg/day) immediately before admission, whereas patients with AA-WRF or non-WRF tended to receive the lowest doses (≤20 mg/day) more often. Maximum IV diuretic doses were generally prescribed on Day 1 to patients without WRF before admission. A higher proportion of patients with P-WRF received the highest doses of IV diuretics daily (≥80 mg/day), experienced increases in the IV diuretic dose after admission (mainly during the first or second day after admission), and had IV diuretics readministered after completion of initial treatment and after at least 1 day had elapsed from the end of their initial treatment to readministration. There were significant differences between patients with WRF 30 days before admission and those without for all parameters tested, namely: duration from admission to the initial dose of IV diuretic (P=0.0019); initial IV diuretic dose after admission (P<0.0001); duration from admission to the maximum dose of IV diuretic (P<0.0001); maximum dose of IV diuretics after admission (P<0.0001); increase in IV diuretic dose after admission (P<0.0001); time from admission to increase in IV diuretics after admission (P=0.0386); and readministration of IV diuretics (P<0.0001).

Table 2.

Actual Use of Intravenous Diuretics Before and After Admission

		Study population (n=1,788)	WRF 30 days before admission				P value
			Yes		No
			WRF after admission		WRF after admission
			Yes (n=44)	No (n=208)	Yes (n=389)	No (n=1,147)
Duration from admission to the initial dose of IV diuretic	1 day	1,607 (89.9)	34 (77.3)	182 (87.5)	341 (87.7)	1,050 (91.5)	0.002A
	2 days	144 (8.1)	7 (15.9)	15 (7.2)	44 (11.3)	78 (6.8)
	3 days	37 (2.1)	3 (6.8)	11 (5.3)	4 (1.0)	19 (1.7)
Initial daily IV diuretic dose after admission	≤20 mg/day	1,240 (69.4)	22 (50.0)	140 (67.3)	244 (62.7)	834 (72.7)	<0.001A
	≥20–<40 mg/day	365 (20.4)	12 (27.3)	38 (18.3)	93 (23.9)	222 (19.4)
	≥40–<80 mg/day	95 (5.3)	3 (6.8)	18 (8.7)	23 (5.9)	51 (4.4)
	≥80 mg/day	88 (4.9)	7 (15.9)	12 (5.8)	29 (7.5)	40 (3.5)
Duration from admission to the maximum dose of IV diuretic	1 day	1,239 (69.3)	16 (36.4)	137 (65.9)	250 (64.3)	836 (72.9)	<0.001A
	2 days	345 (19.3)	14 (31.8)	41 (19.7)	82 (21.1)	208 (18.1)
	3 days	204 (11.4)	14 (31.8)	30 (14.4)	57 (14.7)	103 (9.0)
Maximum dose of IV diuretics after admission (maximum daily dose)	≤20 mg/day	949 (53.1)	11 (25.0)	98 (47.1)	175 (45.0)	665 (58.0)	<0.001A
	≥20–<40 mg/day	518 (29.0)	9 (20.5)	60 (28.8)	126 (32.4)	323 (28.2)
	≥40–≤80 mg/day	184 (10.3)	8 (18.2)	30 (14.4)	44 (11.3)	102 (8.9)
	≥80 mg/day	137 (7.7)	16 (36.4)	20 (9.6)	44 (11.3)	57 (5.0)
Increase in IV diuretic dose after admission	Yes	392 (21.9)	20 (45.5)	49 (23.6)	99 (25.4)	224 (19.5)	<0.001B
	No	1,396 (78.1)	24 (54.5)	159 (76.4)	290 (74.6)	923 (80.5)
Time from admission to increase in IV diuretics after admission	1–2 days	262 (66.8)	13 (65.0)	38 (77.6)	56 (56.6)	155 (69.2)	0.039A
	3–4 days	84 (21.4)	4 (20.0)	6 (12.2)	27 (27.3)	47 (21.0)
	5–6 days	21 (5.4)	2 (10.0)	2 (4.1)	8 (8.1)	9 (4.0)
	≥7 days	25 (6.4)	1 (5.0)	3 (6.1)	8 (8.1)	13 (5.8)
Readministration of IV diuretics	Yes	246 (13.8)	14 (31.8)	34 (16.3)	70 (18.0)	128 (11.2)	<0.001B
	No	1,542 (86.2)	30 (68.2)	174 (83.7)	319 (82.0)	1,019 (88.8)

^AP value calculated using the Kruskal-Wallis test. ^BP value calculated using the χ² test. IV, intravenous; WRF, worsening renal function.

Patients who received the maximum diuretic dose ≥2 days after admission had a significantly higher cumulative incidence of WRF at all time points compared with those receiving the maximum diuretic dose at or by Day 1 of hospitalization (log-rank test P=0.0001; Figure 2A). Similarly, the cumulative incidence of WRF increased significantly with an increased time between admission and the first IV diuretic administration from 0% on admission day to 34.2% on Day 30 after admission (log-rank test P=0.0203; Supplementary Figure 2A).

Figure 2.

Occurrence of worsening renal function by (A) duration from admission to the maximum dose and (B) maximum dose.

Compared with patients receiving maximum IV diuretic doses ≤20 mg, the cumulative incidence of WRF was higher at all time points among those receiving IV diuretic doses from >20 to <40 mg/day (log-rank test P=0.0057) and ≥40 mg/day (P<0.0001; Figure 2B).

Similarly, patients receiving intermediate (≥20–<40 mg/day) and higher (>40 mg/day) doses of IV diuretics as initial treatment had a significantly higher cumulative incidence of WRF at all time points compared with those receiving the lower IV diuretic doses (≤20 mg/day; log-rank test P=0.0044 and P<0.0001, respectively; Supplementary Figure 2B). Diuretic readministration during hospitalization was associated with a significantly higher cumulative incidence of WRF (log-rank test P=0.0203; Supplementary Figure 2C). To provide context for these findings, baseline characteristics were compared by IV diuretic maximum daily dose. There was a significant difference in the following baseline characteristics according to IV diuretic maximum daily dose categories (≤20, >20 to ≤40, and >40 mg): age, Charlson Comorbidity Index (CCI), tolvaptan use, potassium, creatinine, blood urea nitrogen, N-terminal pro B-type natriuretic peptide, eGFR, albumin, history of hospitalization for acute HF within 1 year, and ADL score at admission (Table 3).

Table 3.

Baseline Characteristics by Maximum Daily Dose of Intravenous Diuretic

Variable	Study population (n=1,788)	Maximum daily dose of IV diuretic (mg)			P value
		≤20 (n=949)	>20 to ≤40 (n=518)	>40 (n=321)
Sex
Male	991 (55.4)	510 (53.7)	295 (56.9)	186 (57.9)	0.301A
Female	797 (44.6)	439 (46.3)	223 (43.1)	135 (42.1)	–
Age at time of hospitalization (years)	80.5±10.2	81.3±10.0	79.9±10.5	78.8±10.3	<0.001B
Age ≥85 years	690 (38.6)	398 (41.9)	186 (35.9)	106 (33.0)	–
AHF hospitalization within 1 year	301 (16.8)	179 (18.9)	82 (15.8)	40 (12.5)	0.023A
Period between the prior and current hospitalizations (days)	107.0 [49.0–222.0]	91.0 [49.0–217.0]	144.0 [57.0–248.0]	104.5 [38.5–239.0]	–
SBP at admission (mmHg)					0.151A
<100	120 (8.2)	62 (8.1)	27 (6.2)	31 (11.6)
100 to <140	770 (52.3)	403 (52.7)	231 (52.6)	136 (50.7)
≥140	581 (39.5)	299 (39.1)	181 (41.2)	101 (37.7)
BMI (kg/m2)	n=1,730	n=926	n=505	n=299	0.377B
Mean±SD	23.22±4.49	23.12±4.20	23.22±4.47	23.53±5.31
Minimum–maximum	13.2–50.0	13.2–45.1	14.3–47.4	13.8–50.0
Comorbidities at the time of admission
Hypertension	1,330 (74.4)	699 (73.7)	386 (74.5)	245 (76.3)	–
Diabetes	755 (42.2)	387 (40.8)	216 (41.7)	152 (47.4)	–
Ischemic heart disease	702 (39.3)	373 (39.3)	195 (37.6)	134 (41.7)	–
Dyslipidemia	694 (38.8)	364 (38.4)	199 (38.4)	131 (40.8)	–
CKD	535 (29.9)	240 (25.3)	173 (33.4)	122 (38.0)	–
Arrhythmia	903 (50.5)	505 (53.2)	243 (46.9)	155 (48.3)	–
Valvular disease	90 (5.0)	34 (3.6)	30 (5.8)	26 (8.1)	–
COPD	95 (5.3)	54 (5.7)	27 (5.2)	14 (4.4)	–
Charlson Comorbidity Index	3.9±2.4	3.8±2.4	3.9±2.3	4.2±2.4	0.017B
Loop diuretics at admission	1,211 (67.7)	633 (66.7)	344 (66.4)	234 (72.9)	0.091A
Other concomitant medications at admission
β-blocker	921 (51.5)	498 (52.5)	251 (48.5)	172 (53.6)	0.242A
ACE inhibitor	292 (16.3)	157 (16.5)	83 (16.0)	52 (16.2)	0.965A
ARB	680 (38.0)	350 (36.9)	208 (40.2)	122 (38.0)	0.467A
Aldosterone antagonist	473 (26.5)	271 (28.6)	121 (23.4)	81 (25.2)	0.084A
Tolvaptan	318 (17.8)	159 (16.8)	79 (15.3)	80 (24.9)	0.001A
Thiazide diuretic	150 (8.4)	67 (7.1)	48 (9.3)	35 (10.9)	0.069A
SGLT2 inhibitor	92 (5.1)	57 (6.0)	18 (3.5)	17 (5.3)	0.110A
Laboratory measures
Sodium (mEq/L)	139.66±4.51	139.85±4.30	139.32±4.62	139.65±4.87	0.098B
Potassium (mEq/L)	4.299±0.698	4.252±0.651	4.329±0.715	4.388±0.792	0.005B
Creatinine (mg/dL)	1.546±1.142	1.370±1.018	1.622±1.113	1.944±1.396	<0.001B
BUN (mg/dL)	30.09±17.63	27.05±15.38	31.42±17.27	36.65±21.74	<0.001B
eGFR (mL/min/1.73 m2)	38.98 [5.92–54.93]	41.90 [29.50–57.06]	37.65 [23.63–54.19]	32.60 [18.02–47.41]	0.003B
eGFR <60 mL/min/1.73 m2	1,436 (80.3)	735 (77.5)	428 (82.7)	273 (85.1)
BNP (pg/mL)	633.70 [309.00–1,107.20]	534.10 [266.00–1,065.30]	719.25 [419.20–1,139.90]	731.65 [336.10–1,111.45]	0.080B
NT-proBNP (pg/mL)	5,452.00 [2,602.00–10,522.00]	4,733.50 [2,099.00–9,018.00]	5,702.50 [2,730.00–10,353.00]	6,964.50 [3,516.00–16,797.00]	0.005B
Albumin (g/dL)	3.480±0.510	3.516±0.497	3.466±0.491	3.398±0.563	0.002B
NYHA functional class					0.643A
1	9 (2.5)	5 (2.4)	3 (3.0)	1 (1.7)
2	66 (18.1)	34 (16.3)	21 (21.2)	11 (19.0)
3	158 (43.3)	91 (43.8)	46 (46.5)	21 (36.2)
4	132 (36.2)	78 (37.5)	29 (29.3)	25 (43.1)
ADL score at admission	55.8±39.7	59.3±39.2	54.4±38.9	47.5±41.4	0.001B

Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). ^AP value calculated using χ² tests. ^BP value calculated using analysis of variance. ACE, angiotensin-converting enzyme; ADL, activities of daily living; AHF, acute heart failure; ARB, angiotensin II receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; BUN, blood urea nitrogen; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; NT-proBNP, N-terminal pro B-type natriuretic peptide; NYHA, New York Heart Association; SBP, systolic blood pressure; SGLT2, sodium-glucose cotransporter 2; WRF, worsening renal function.

Outcomes of Patients With HF and P-WRF, BA-WRF, AA-WRF, and Non-WRF

Among patients with P-WRF, BA-WRF, AA-WRF, and non-WRF, 27 of 44, 28 of 208, 64 of 389, and 47 of 1,147 patients, respectively, died during hospitalization (Figure 3). In a Cox proportional hazards model adjusted for age and sex, compared with non-WRF patients (reference group), the HR for clinical outcomes among P-WRF patients was 13.16 (95% CI 8.19–21.15; P<0.0001), that for BA-WRF patients was 3.56 (95% CI 2.23–5.69; P<0.0001), and that for AA-WRF patients was 3.23 (95% CI 2.21–4.71; P<0.0001). After adjusting for sex, age, eGFR at 30 days before admission, and hypertension, dyslipidemia, and ADL score at the time of admission, the HR for clinical outcomes compared with non-WRF patients was 9.69 (95% CI 5.54–16.93; P<0.0001) among P-WRF patients, 3.06 (95% CI 1.81–5.17; P<0.0001) for BA-WRF patients, and 2.56 (95% CI 1.63–4.01; P<0.0001) for AA-WRF patients (Supplementary Figure 3). Results of multivariate analysis with oral loop diuretic use (yes/no) before and after admission as additional factors (Table 4) were similar to the original multivariate analysis results (Figure 3). Analyses were also conducted with patients stratified by tolvaptan use during hospitalization; HRs were highest for patients in the P-WRF group, regardless of the use of tolvaptan during admission (Figure 4). HRs for BA-WRF and AA-WRF patients in both stratification groups were similar to those reported in the original (i.e., unstratified) analysis (Figure 3).

Figure 3.

Survival rate of patients during hospitalization according to worsening renal function (WRF) status: no WRF (non-WRF), WRF after admission only (AA-WRF), WRF before admission only (BA-WRF), and WRF before and after admission (P-WRF). ^aThe reference group for the hazard ratios (HR) is the non-WRF group. CI, confidence interval.

Table 4.

Multivariate Analyses of the Incidence of In-Hospital Mortality Using the Covariates of Sex, Age, and Oral Loop Diuretics Use Before or After Admission

	Hazard ratio (95% CI)	P value
Covariates: sex, age, and oral loop diuretics use before admission
AA-WRF vs. non-WRF	3.73 (1.49–9.31)	0.0049
BA-WRF vs. non-WRF	3.75 (1.79–7.87)	0.0005
P-WRF vs. non-WRF	16.05 (7.09–36.34)	<0.0001
Covariates: sex, age, and oral loop diuretics use after admission
AA-WRF vs. non-WRF	3.52 (1.40–8.82)	0.0074
BA-WRF vs. non-WRF	3.22 (1.54–6.76)	0.0005
P-WRF vs. non-WRF	8.80 (3.61–21.46)	<0.0001

AA-WRF, worsening renal function (WRF) after admission only; BA-WRF, WRF before admission only; CI, confidence interval; non-WRF, no WRF; P-WRF, WRF before and after admission.

Figure 4.

Survival rate of patients during hospitalization (A) with or (B) without tolvaptan use after hospital admission according to worsening renal function (WRF) status: no WRF (non-WRF), WRF after admission only (AA-WRF), WRF before admission only (BA-WRF), and WRF before and after admission (P-WRF). ^aThe reference group for the hazard ratios (HR) is the non-WRF group. CI, confidence interval.

Discussion

This retrospective study, based on electronic data records, aimed to clarify the association between acute HF and WRF among hospitalized patients receiving IV diuretics and to evaluate their prognosis. For this, we analyzed patient background, actual treatment, and in-hospital mortality according to the timing of WRF (i.e., before and after admission) to evaluate the impact of WRF before admission. Our data showed that of the 1,788 patients with available eGFR data, 14.1% already had WRF from 30 days before admission to the time of admission, and 25.3% developed WRF after admission. Notably, there were no clear differences in patient characteristics and in-hospital mortality between patients with BA-WRF and AA-WRF; furthermore, those with P-WRF had the worst prognosis. Importantly, our findings regarding WRF prior to hospitalization are novel. Previous studies have discussed WRF in the course of admission among patients with HF but have primarily focused on WRF after admission. Without information on the baseline renal function that is not influenced by acute congestion, it is not theoretically possible to accurately assess changes in renal function during acute management of HF. Similar results were obtained both by adding oral loop diuretics before and after hospitalization in a multivariate analysis and by stratified analysis with and without tolvaptan after hospitalization, suggesting that similar results are obtained with and without other concomitant diuretics.

The relationship between changes in renal function and HF is complicated. In the acute setting of decompensated HF, both WRF and improved renal function (IRF) have been associated with similar hemodynamic derangements.¹⁸ It is often assumed that the prognosis is good in patients with IRF after hospitalization; however, this prognosis may not be relevant for patients whose renal function deteriorated before hospitalization, as they had a poor prognosis. Recent studies have shown that IRF is associated with adverse outcomes and increased mortality compared with stable renal function or WRF.^21–23 Although the underlying mechanisms of this paradoxical association remain poorly understood, IRF may indicate inadequate decongestion in some patients or may reflect disease severity in others, suggesting that WRF and IRF may represent a continuum of renal dysfunction depending on baseline kidney function. Our data suggest that IRF encompasses WRF before admission (BA-WRF), and IRF may, in part, represent a stage of apparent WRF recovery. In our study, there was no difference in mortality between the BA-WRF and AA-WRF subgroups, which would further support this hypothesis.^24,25

In the present study, patients with P-WRF had the most severe ADL scores at the time of admission, as well as the worst prognosis, but patients with P-WRF as well as those with BA-WRF were likely more severely ill than those with non-WRF. These findings reflect the severely hemodynamic derangements already present in these patients before admission, which may interact with other factors, such as treatments administered during admission and treatment response,^18,19 which can lead to further WRF. In the case of WRF before admission, treatment should be considered more carefully; if renal function worsens further after admission, in-hospital prognosis may be much worse, and the length of hospital stay may be prolonged.

Both before and after admission, the maximum diuretic dose was higher in patients with WRF, and the time from admission to administration of the initial and maximum diuretic doses was longer. In addition, these patients required higher doses and dose increases, and many required readministration of IV diuretics after completion of the initial treatment. These findings suggest that diuretic responsiveness was not good, and maximum doses were needed, as well as additional administration of IV diuretics. A possible explanation is that in Japan the initial dose of IV diuretics may be lower than needed, even for patients with poor baseline renal function.²⁶ Based on clinical experience, it seems that the approach is to administer a low-dose IV diuretic, monitor the reaction, and gradually increase the dose, which may be reflected in the present observations. However, it seems important to set an appropriate initial dose based on renal function.²⁷ We note that patients with higher maximum diuretic doses had higher CCI scores, along with lower eGFR and higher blood urea nitrogen values, at baseline. This highlights the potential association between the maximum diuretic dose and the severity of HF, which includes renal function.

In addition, we found that the time to IV diuretic treatment plays an important role in WRF development during hospitalization. The cumulative incidence of WRF was significantly higher at all time points for patients receiving initial diuretic dosing on Day 2 or later of hospitalization compared with those receiving the initial diuretic dosing at or by Day 1. A previous study found that venous congestion is the most important factor contributing to WRF in decompensated patients with advanced HF.²⁸ Thus, early and appropriate administration of IV diuretics to improve congestion may be a key factor in avoiding WRF.

This study has some limitations, such as those inherent to the study design and data source (MDV DPC database), limiting the variables that could be evaluated; deaths and cardiovascular mortality estimates were limited to those during the hospital stay, and data on echocardiographic parameters were limited or unavailable. The data used in this study (DPC data) are from MDV, which has the largest DPC database in Japan, and do not include data from DPC hospitals that do not provide data to MDV or from facilities that have not been approved as DPC hospitals by the Ministry of Health, Labour and Welfare. Therefore, the findings of this study may not be generalizable to all Japanese patients. We adjusted for other possible confounders; however, not all confounders could be adjusted for. Finally, this study only evaluated patients who had eGFR data 30 days prior to admission, at admission, and after admission.

A certain number of patients with acute exacerbation of HF have deterioration of renal function not only after admission, but also before admission. We estimated the incidence of WRF in patients with HF, and found that in this population those with P-WRF admitted to hospital with acute HF had the worst prognosis, followed by patients with AA-WRF. In addition, among patients who did not have WRF after admission, those with BA-WRF were found to have as poor a prognosis as those with AA-WRF. We also identified some differences in background factors among these patients. Delayed introduction of IV diuretics, even from Day 2 of hospitalization, and higher doses of IV diuretics were associated with a higher incidence of WRF after admission, regardless of the presence or absence of prior WRF. Additional prospective studies are needed to further elucidate the pathophysiological causal relationship by examining the association with the timing of eGFR decline, as well as urinary markers of renal tubular injury and clinical outcomes, including long-term prognosis.

Acknowledgments

The authors thank Keyra Martinez Dunn, MD, of Edanz (www.edanz.com) for providing medical writing support, which was funded by Otsuka Pharmaceutical Co., Ltd, in accordance with Good Publication Practice (GPP 2022) guidelines (https://www.ismpp.org/gpp-2022).

Sources of Funding

This research was funded by Otsuka Pharmaceutical Co., Ltd.

Disclosures

Y.S. has received lecture fees from Otsuka Pharmaceutical Co., Ltd., and Ono Pharmaceutical Co., Ltd. Y.K., H.M., K.O., and M.M. are employees of Otsuka Pharmaceutical Co., Ltd.

IRB Information

The protocol and related documents were reviewed and approved by the Otsuka Pharmaceutical Co., Ltd. Research and Development Research Ethics Committee (Reference no. 210127).

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-23-0440

References

• 1.   GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990–2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018; 392: 1789–1858.
• 2.   Yasuda S, Miyamoto Y, Ogawa H. Current status of cardiovascular medicine in the aging society of Japan. Circulation 2018; 138: 965–967.
• 3.   Komuro I, Kaneko H, Morita H, Isobe M, Nakayama H, Minematsu K, et al. Nationwide actions against heart failure pandemic in Japan: What should we do from academia? Circ J 2019; 83: 1819–1821.
• 4.   Kanaoka K, Okayama S, Nakai M, Sumita Y, Nishimura K, Kawakami R, et al. Hospitalization costs for patients with acute congestive heart failure in Japan. Circ J 2019; 83: 1025–1031.
• 5.   Metra M, Nodari S, Parrinello G, Bordonali T, Bugatti S, Danesi R, et al. Worsening renal function in patients hospitalised for acute heart failure: Clinical implications and prognostic significance. Eur J Heart Fail 2008; 10: 188–195.
• 6.   Butler J, Forman DE, Abraham WT, Gottlieb SS, Loh E, Massie BM, et al. Relationship between heart failure treatment and development of worsening renal function among hospitalized patients. Am Heart J 2004; 147: 331–338.
• 7.   Freda BJ, Slawsky M, Mallidi J, Braden GL. Decongestive treatment of acute decompensated heart failure: Cardiorenal implications of ultrafiltration and diuretics. Am J Kidney Dis 2011; 58: 1005–1017.
• 8.   Mullens W, Damman K, Harjola VP, Mebazaa A, Brunner-La Rocca HP, Martens P, et al. The use of diuretics in heart failure with congestion: A position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 2019; 21: 137–155.
• 9.   Yaku H, Ozasa N, Morimoto T, Inuzuka Y, Tamaki Y, Yamamoto E, et al; KCHF Study Investigators. 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. Circ J 2018; 82: 2811–2819.
• 10.   Krumholz HM, Chen YT, Vaccarino V, Wang Y, Radford MJ, Bradford WD, et al. Correlates and impact on outcomes of worsening renal function in patients > or =65 years of age with heart failure. Am J Cardiol 2000; 85: 1110–1113.
• 11.   George LK, Koshy SKG, Molnar MZ, Thomas F, Lu JL, Kalantar-Zadeh K, et al. Heart failure increases the risk of adverse renal outcomes in patients with normal kidney function. Circ Heart Fail 2017; 10: e003825.
• 12.   Roy AK, Mc Gorrian C, Treacy C, Kavanaugh E, Brennan A, Mahon NG, et al. A comparison of traditional and novel definitions (RIFLE, AKIN, and KDIGO) of acute kidney injury for the prediction of outcomes in acute decompensated heart failure. Cardiorenal Med 2013; 3: 26–37.
• 13.   Logeart D, Tabet JY, Hittinger L, Thabut G, Jourdain P, Maison P, et al. Transient worsening of renal function during hospitalization for acute heart failure alters outcome. Int J Cardiol 2008; 127: 228–232.
• 14.   Klein L, Massie BM, Leimberger JD, O’Connor CM, Piña IL, Adams KF Jr, et al. Heart failure predict postdischarge survival: Results from the outcomes of a prospective trial of intravenous milrinone for exacerbations of chronic heart failure (OPTIME-CHF). Circ Heart Fail 2008; 1: 25–33.
• 15.   Metra M, Davison B, Bettari L, Sun H, Edwards C, Lazzarini V, et al. Is worsening renal function an ominous prognostic sign in patients with acute heart failure?: The role of congestion and its interaction with renal function. Circ Heart Fail 2012; 5: 54–62.
• 16.   Ahmad T, Jackson K, Rao VS, Tang WHW, Brisco-Bacik MA, Chen HH, et al. Worsening renal function in patients with acute heart failure undergoing aggressive diuresis is not associated with tubular injury. Circulation 2018; 137: 2016–2028.
• 17.   Cowie MR, Komajda M, Murray-Thomas T, Underwood J, Ticho B; POSH Investigators. Prevalence and impact of worsening renal function in patients hospitalized with decompensated heart failure: Results of the prospective outcomes study in heart failure (POSH). Eur Heart J 2006; 27: 1216–1222.
• 18.   Mullens W, Damman K, Testani JM, Martens P, Mueller C, Lassus J, et al. Evaluation of kidney function throughout the heart failure trajectory: A position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 2020; 22: 584–603.
• 19.   Gottlieb SS, Abraham W, Butler J, Forman DE, Loh E, Massie BM, et al. The prognostic importance of different definitions of worsening renal function in congestive heart failure. J Card Fail 2002; 8: 136–141.
• 20.   von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. J Clin Epidemiol 2008; 61: 344–349.
• 21.   Natov PS, Ivey-Miranda JB, Cox ZL, Moreno-Villagomez J, Maulion C, Bellumkonda L, et al. Improvement in renal function during the treatment of acute decompensated heart failure: Relationship with markers of renal tubular injury and prognostic importance. Circ Heart Fail 2023; 16: e009776.
• 22.   Wettersten N, Horiuchi Y, van Veldhuisen DJ, Ix JH, Mueller C, Filippatos G, et al. Decongestion discriminates risk for one-year mortality in patients with improving renal function in acute heart failure. Eur J Heart Fail 2021; 23: 1122–1130.
• 23.   Testani JM, McCauley BD, Chen J, Coca SG, Cappola TP, Kimmel SE, et al. Clinical characteristics and outcomes of patients with improvement in renal function during the treatment of decompensated heart failure. J Card Fail 2011; 17: 993–1000.
• 24.   Beldhuis IE, Streng KW, van der Meer P, Ter Maaten JM, O’Connor CM, Metra M, et al. Trajectories of changes in renal function in patients with acute heart failure. J Card Fail 2019; 25: 866–874.
• 25.   Beldhuis IE, Ter Maaten JM, Figarska SM, Damman K, Pang PS, Greenberg B, et al. Disconnect between the effects of serelaxin on renal function and outcome in acute heart failure. Clin Res Cardiol 2023; 112: 901–910.
• 26.   Shiraishi Y, Kurita Y, Matsukawa M, Mori H. Real-world intravenous diuretic use to treat congestion in patients with heart failure: An observational study using a research database. Circ Rep 2023; 5: 27–37.
• 27.   Felker GM, Lee KL, Bull DA, Redfield MM, Stevenson LW, Goldsmith SR, et al. Diuretic strategies in patients with acute decompensated heart failure. N Engl J Med 2011; 364: 797–805.
• 28.   Mullens W, Abrahams Z, Francis GS, Sokos G, Taylor DO, Starling RC, et al. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol 2009; 53: 589–596.