Abstract
Background
Little is known about the initial dosing of loop diuretics among patients hospitalized for heart failure and its association with outcomes.
Methods
We identified patients admitted for heart failure at 24 hospitals across two health systems between January 1, 2017, and December 31, 2020. Initial diuretic dose was categorized relative to home dose in furosemide equivalents. The primary outcomes (length of stay) and secondary outcomes (rates of acute kidney injury, in‐hospital mortality, and 30‐day readmissions) were compared across categories of initial diuretic dose.
Results
Among 14 332 patients admitted for heart failure, the initial diuretic dose was lower‐than‐home dose in 1866 (13.0%) patients, equivalent‐to‐home dose in 3171 (22.1%) patients, and higher‐than‐home dose in 9295 (64.9%) patients. Compared with patients who received an equivalent or higher initial diuretic dose relative to home dose, risk‐adjusted length of stay was longer among patients receiving a lower dose (4.9 days versus 4.0 days versus 4.0 days, P<0.01). Compared with equivalent‐to‐home dose, a higher initial diuretic dose was associated with a higher risk‐adjusted rate of acute kidney injury (incident risk ratio [IRR], 1.17 [95% CI 1.04–1.31]; P=0.009) and a lower risk of 30‐day readmission (IRR, 0.85 [95% CI, 0.78–0.93]; P=0.005).
Conclusions
In a cohort of patients admitted for heart failure, more than 1 in 3 patients received an initial dose of intravenous loop diuretics that was lower than or equivalent to their home diuretic dose, and the initial dose was associated with length of hospital stay and acute kidney injury.
Keywords: diuretics, dose–response relationship, heart failure, hospitalization, length of stay
Nonstandard Abbreviations and Acronyms
- AKI
acute kidney injury
- AKIN
Acute Kidney Injury Network
- CV‐QUIC
Cardiovascular Quality Improvement and Care Innovation Consortium
- DOSE
Diuretic Strategies in Patients with Acute Decompensated Heart Failure
- ESCAPE
Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheter Effectiveness
- IRR
incident risk ratio
Clinical Perspective.
What Is New?
Despite decades of experience with intravenous loop diuretics in millions of patients hospitalized with heart failure, data to inform optimal dosing strategies are limited.
In a cohort of patients admitted for heart failure, more than 1 in 3 patients received an initial dose of intravenous loop diuretics that was lower than or equivalent to their home diuretic dose.
What Are the Clinical Implications?
An initial diuretic dose above home equivalent was associated with a shorter length of stay, increased risk of acute kidney injury, and lower risk of readmission.
In more than 90% of patients hospitalized for acute decompensated heart failure (HF), management includes the administration of an intravenous loop diuretic. 1 Despite decades of experience with intravenous loop diuretics in millions of patients hospitalized with HF, data to inform optimal dosing strategies from clinical trials and their applicability to routine clinical practice are limited. 2 The importance of optimal dosing of loop diuretics is suggested from a prior randomized trial that demonstrated higher bolus doses were associated with greater net fluid loss, weight loss, and relief from dyspnea, although no significant differences were noted in length of stay (LOS). 3 Informed by this trial, guidelines recommend the initial dose of intravenous loop diuretics be equal to or exceed a patient's oral daily home dose when requiring hospitalization for treatment of fluid overload. 4
Little is known about initial diuretic dosing approaches used in contemporary management of patients admitted with acute decompensated HF and the association with clinical outcomes. Lower doses may fail to achieve adequate diuresis and contribute to a longer LOS, although a prior single‐center observational study failed to demonstrate an association between initial diuretic dosing relative to home dose and LOS. 5 Higher doses of loop diuretics may contribute to activation of the renin–angiotensin and sympathetic nervous systems, electrolyte disturbances, and worsening of renal function. 2 In prior observational studies of patients hospitalized for HF, high doses of loop diuretics have been associated with increased risk of renal failure, worsening of HF, and death. 6 , 7 These prior studies did not account for the patient's home diuretic dose, which may reflect baseline HF severity, risk of adverse clinical events, and need for higher doses to achieve diuresis.
We sought to describe the initial loop diuretic dosing strategy relative to home dose in a population of patients hospitalized for acute HF and determine the association of hospital‐to‐home dosing ratio with clinical outcomes. We hypothesized that patients receiving a loop diuretic at a dose lower than their home dose would experience a longer length of hospital stay while patients receiving dosing regimens higher than their home dose would experience more acute kidney injury (AKI). Understanding current practice in the use of loop diuretics for patients with HF hospitalization may inform opportunities to shorten the length of hospitalization without worsening patient outcomes as part of high‐value care for this prevalent and costly condition.
METHODS
Study Setting, Patient Population, and Data Source
Primary Analysis
This analysis was conducted as part of a collaborative data project of the Cardiovascular Quality Improvement and Care Innovation Consortium (CV‐QUIC). 8 Data are maintained at the individual participating institutions and are not available to other researchers. Analytic methods and study materials are provided within the article. This study followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational cohort studies, and was approved by the institutional review committees of Allina Health and Atrium Health with waiver of informed consent granted.
We identified all patients admitted with a primary or secondary admission diagnosis of acute HF (International Classification of Diseases, Ninth Revision [ICD‐9] codes 428.21, 428.23, 428.31, 428.33, 428.41, 428.43, or International Classification of Diseases, Tenth Revision [ICD‐10] codes I50.21, I50.23, I50.31, I50.33, I50.41, I50.43) at 12 Allina Health hospitals in central Minnesota and western Wisconsin and 12 Atrium Health hospitals in North Carolina between January 1, 2017, and December 31, 2020. We excluded patients on observational status, no or unknown home loop diuretic dosing, no bolus dosed intravenous loop diuretic within 24 hours of admission, patients on a home thiazide‐type or thiazide‐like diuretic (eg, hydrochlorothiazide, chlorthalidone, metolazone, or indapamide) in addition to a loop diuretic given challenges in creating equivalent in‐patient dosing categorization, patients who received inotropes with or without vasopressors (ie, norepinephrine, phenylephrine, epinephrine, dopamine, vasopressin, dobutamine, or milrinone), and patients undergoing dialysis before admission or within 24 hours of admission (Figure). Our final analytic cohort included 14 332 patients.
Figure 1. Cohort identification.
Exposures
The primary exposure of interest was the first inpatient dose of intravenous loop diuretic among patients receiving bolus diuretic dosing. The initial diuretic dose was compared with the patient's home dose in furosemide equivalents using the DOSE (Diuretic Strategies in Patients With Acute Decompensated Heart Failure) trial logic. 3 In that trial, 40 mg of intravenous furosemide was considered equivalent to 40 mg of oral furosemide, 20 mg of oral torsemide, or 1 mg of oral bumetanide. In the bolus dosing arm of the DOSE trial, bolus loop diuretics were administered every 12 hours to achieve a dose equivalent‐to‐home dose (low dose) or 2.5 times the home dose (high dose). Informed by this trial, we categorized initial diuretic dosing into lower‐than‐home dose, equivalent‐to‐home dose, or higher‐than‐home dose as a function of 12‐hour intravenous dosing regimens (ie, a dose that was less than one‐half the patient's total daily home dose was categorized as lower‐than‐home dose).
Outcomes
The primary outcome of interest was length of hospital stay measured in days. Secondary outcomes included acute kidney injury (AKI), in‐hospital mortality, and 30‐day readmission. AKI was defined by the Acute Kidney Injury Network (AKIN) criteria and determined by the comparison of admission to peak serum creatinine level during hospitalization. AKI was categorized as: (1) present or absent; and (2) stage of AKI by the AKIN criteria 9 (stage 1: ≥0.3 mg/dL absolute or 1.5 to 2.0‐fold relative increase in serum creatinine; stage 2: >2‐ to 3‐fold increase in serum creatinine; stage 3: >3‐fold increase in serum creatinine or serum creatinine >4.0 mg/dL with an acute increase of >0.5 mg/dL; dialysis, dialysis before discharge).
Statistical Analysis
We summarized patient characteristics by category of initial diuretic dosing using counts (percentages) and medians (interquartile ranges) for categorical and continuous variables, respectively. The characteristics are compared among groups using the Pearson χ2 and the Kruskal–Wallis rank sum tests, as appropriate. For clinical outcomes, we compared median LOS and crude event rates of AKI, in‐hospital mortality, and 30‐day readmission by category of initial diuretic dosing. Finally, we estimated the patient‐level risk‐adjusted associations between initial diuretic dosing and the outcomes of interest from multivariable models with covariates as listed in Table 1 (see Supplemental Document for coding definitions) that were informed by clinical knowledge and data availability. For the hospital LOS, the estimates were derived using quantile regression; 95% CIs for the estimated parameters based on ranks and the corresponding P values are reported. The association with AKI, in‐hospital mortality, and 30‐day readmission were estimated from quasi‐Poisson regression models with a canonical link and robust estimators of variance (for the in‐hospital outcomes of mortality and AKI, the models were offset by the log of the LOS; for the 30‐day readmission, the LOS was used as a covariate). The estimated risk ratios (RRs), 95% CIs, and P values are reported. Exploratory subgroup analyses by age, sex, body mass index category, chronic obstructive pulmonary disease, diabetes, prior myocardial infarction, and estimated glomerular filtration rate categories are reported using the models adjusted for the same covariates. Subgroup analyses were not conducted on race given concerns this reflects a social rather than biological construct, or tobacco use, hypertension, or dyslipidemia given the lack of clinical rationale as to how these conditions could affect the association between diuretic dose and clinical outcomes. The estimated associations, their CIs, and interaction P values are reported. All statistical analyses were completed using R version 4.3.2 (R Core Team) in RStudio 2023.12 environment (Posit Software, PBC).
Table 1.
Patient Characteristics by Initial Diuretic Dosing Relative to Home Dose
| Overall | Initial diuretic dose relative to home dose | P value | |||
|---|---|---|---|---|---|
| Lower | Equivalent | Higher | |||
| N=14 332 | 1866 (13.0%) | 3171 (22.1%) | 9295 (64.9%) | ||
| Age, median (IQR), y | 73 (63–82) | 72 (61, 80) | 72 (62–82) | 74 (64–83) | <0.001 |
| White race, n (%) | 10 062 (70) | 1292 (69) | 2137 (67) | 6633 (72) | <0.001 |
| Men, n (%) | 7534 (53) | 1044 (56) | 1677 (53) | 4813 (52) | 0.004 |
| BMI, median (IQR) | 31 (26–37) | 33 (28–40) | 32 (26–38) | 30 (25–36) | <0.001 |
| Prior myocardial infarction, n (%) | 3569 (25) | 460 (25) | 837 (26) | 2272 (24) | 0.087 |
| Prior congestive HF, n (%) | 13 328 (93) | 1806 (97) | 3043 (96) | 8479 (91) | <0.001 |
| Hypertension, n (%) | 13 789 (96) | 1798 (96) | 3072 (97) | 8919 (96) | 0.059 |
| Diabetes, n (%) | 9074 (63) | 1374 (74) | 2159 (68) | 5541 (60) | <0.001 |
| Dyslipidemia, n (%) | 12 186 (85) | 1592 (85) | 2765 (87) | 7829 (84) | <0.001 |
| Cerebrovascular disease, n (%) | 5323 (37) | 653 (35) | 1213 (38) | 3457 (37) | 0.068 |
| COPD, n (%) | 7374 (51) | 1032 (55) | 1740 (55) | 4602 (50) | <0.001 |
| eGFR, mL/min per 1.73 m2, median (IQR) | 41 (29–51) | 38 (26–49) | 40 (29–50) | 41 (30–51) | <0.001 |
Kruskal–Wallis rank sum test; Pearson chi‐squared test.
BMI indicates body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimate glomerular filtration rate; HF, heart failure; and IQR, interquartile range.
Secondary Analysis
We utilized data within the electronic data warehouses of Allina Health and Atrium Health for our primary analysis. In a secondary analysis, we performed a limited chart review (conducted by authors NQ, EK, and SB) of a subset of 1010 patients admitted for HF in the Allina Health system to obtain outcomes data not readily available in the structured data within our electronic data warehouse. Patients were selected in order from the beginning of the primary cohort period of observation until abstraction resources were exhausted. Review was directly from the medical record and included net fluid output and weight change after the first day of hospitalization and presence or absence of an increase in diuretic dosing on day 2 of admission. If initial diuretic dose was associated with reduced LOS, we anticipated identifying a similar association between diuretic dosing and outcomes on the pathway toward shorter hospitalization (ie, more rapid diuresis). We conducted weekly internal audits and quality assurance throughout data abstraction to maintain high‐quality data abstraction.
RESULTS
Primary Analysis
Among 14 332 patients admitted for HF on home loop diuretics, 11 498 (80.2%) were on home furosemide, 2139 (14.9%) were on home torsemide, and 695 (4.8%) were on home bumetanide. The initial diuretic dose was lower‐than‐home dose in 1866 (13.0%) patients, equivalent‐to‐home dose in 3171 (22.1%) patients, and higher‐than‐home dose in 9295 (64.9%) patients, with a higher proportion of higher‐than‐home dose among patients on home furosemide (Table S1). Baseline characteristics by category of initial diuretic dose are presented in Table 1. Median LOS was longer in patients receiving a lower‐than‐home diuretic dose (4.9 days) compared with patients receiving an equivalent‐to‐home dose (4.0 days) or a higher‐than‐home dose (4.0 days) (P<0.001; Table 2). Rates of AKI were 15.3% in patients receiving lower‐than‐home dose, 10.9% for those receiving an equivalent‐to‐home dose, and 12.5% for patients receiving a higher‐than‐home dose (P<0.001; Table 2). In‐hospital mortality did not significantly differ by category of diuretic dosing while 30‐day readmission was lowest for patients receiving a higher‐than‐home dose (Table 2).
Table 2.
Patient Outcomes by Initial Diuretic Dosing Relative to Home Dose
| Outcome | Overall, N=14 332 | Initial diuretic dose relative to home dose | P value | ||
|---|---|---|---|---|---|
| Lower | Equivalent | Higher | |||
| 1866 (13.0%) | 3171 (22.1%) | 9295 (64.9%) | |||
| LOS, median (IQR), d | 4.0 (2.8–6.0) | 4.9 (3.0–7.0) | 4.0 (2.8–6.0) | 4.0 (2.3–6.0) | <0.001 |
| Acute kidney injury, n (%) | 1791 (12.5) | 285 (15.3) | 346 (10.9) | 1160 (12.5) | <0.001 |
| AKIN 1 | 1292 (9.0) | 205 (11) | 256 (8.1) | 831 (8.9) | 0.005 |
| AKIN 2 | 160 (1.1) | 23 (1.2) | 29 (0.9) | 108 (1.2) | |
| AKIN 3 | 185 (1.3) | 31 (1.7) | 31 (1.0) | 123 (1.3) | |
| New dialysis | 154 (1.1) | 26 (1.4) | 30 (0.9) | 98 (1.1) | |
| In‐hospital mortality, n (%) | 372 (2.6) | 56 (3.1) | 77 (2.5) | 239 (2.6) | 0.46 |
| 30‐d readmission, n (%) | 2348 (16.4) | 354 (19.0) | 569 (17.9) | 1425 (15.3) | <0.001 |
Kruskal–Wallis rank sum test; Pearson chi‐squared test.
AKIN indicates Acute Kidney Injury Network; IQR, interquartile range; and LOS, length of stay.
In risk‐adjusted analyses, compared with patients receiving an initial dose that was equivalent‐to‐home dose, patients who received a higher initial dose had a shorter LOS (−0.17 days [95% CI, −0.33 to −0.01 days]; P=0.04) and patients receiving a lower initial dose had a longer LOS (0.46 days [95% CI, 0.21–0.70 days]; P<0.001) (Table 3). Compared with patients receiving an initial dose that was equivalent‐to‐home dose, a higher‐than‐home initial dose was also associated with an increased risk of AKI (incident RR [IRR], 1.17 [95% CI, 1.04–1.31]; P=0.009) and a lower risk of 30‐day readmission (IRR, 0.85 [95% CI, 0.78–0.93]; P=0.005). Exploratory subgroup analyses are reported in the online Supplement (Figures S1 through S3). There were no apparent differences in the estimates between the subgroups (all P interaction>0.05).
Table 3.
Risk‐Adjusted Outcomes Relative to Initial Diuretic Dosing Equivalent‐to‐Home Dose
| Outcome | Initial diuretic dose relative to home dose | ||
|---|---|---|---|
| Equivalent | Lower | Higher | |
| 3171 (22.1%) | 1866 (13.0%) | 9295 (64.9%) | |
| Absolute LOS, d | Reference | 0.46 (0.21–0.70)* | −0.17 (−0.33 to −0.01)† |
| Relative risk of acute kidney injury | Reference | 1.14 (0.98–1.32) | 1.17 (1.04–1.31)* |
| Relative risk of in‐hospital mortality | Reference | 1.28 (0.90–1.80) | 1.06 (0.83–1.38) |
| Relative risk of 30‐d readmission | Reference | 1.05 (0.93–1.19) | 0.85 (0.78–0.93)* |
P<0.01.
P=0.04.
LOS indicates length of stay. Values are presented as incident risk ratio (95% CI).
Secondary Analysis
Among 1010 patients admitted for HF in whom limited chart review was conducted, initial diuretic dose was lower‐than‐home dose in 125 (12%) patients, equivalent‐to‐home dose in 206 (20%) patients, and higher‐than‐home dose in 679 (67%) patients. Baseline characteristics by category of initial diuretic dose are presented in Table 4. Patients with higher‐than‐home dose had larger net fluid loss and were less likely to have an increase in diuretic dose on day 2 of admission (Table 4).
Table 4.
Secondary Analysis Cohort of Initial Diuretic Dosing Relative to Home Dose
| Characteristic | Overall, N=1010 | Initial diuretic dose relative to home dose | Higher | P value* | |
|---|---|---|---|---|---|
| Lower | Equivalent | ||||
| 125 (12%) | 206 (20%) | 679 (67%) | |||
| Age, median (IQR), y | 78 (69–85) | 73 (62–82) | 79 (69–86) | 79 (70–85) | <0.001 |
| Women, n (%) | 462 (46) | 55 (44) | 90 (44) | 317 (47) | 0.69 |
| White race, n (%) | 903 (89) | 104 (83) | 184 (89) | 615 (91) | 0.10 |
| BMI, median (IQR) | 29 (25–35) | 31 (26–39) | 30 (25–35) | 29 (25–34) | 0.019 |
| Prior myocardial infarction, n (%) | 289 (29) | 27 (22) | 67 (33) | 195 (29) | 0.10 |
| Current smoker, n (%) | 75 (7.4) | 14 (11) | 13 (6.3) | 48 (7.1) | 0.20 |
| Hypertension, n (%) | 511 (51) | 71 (57) | 111 (54) | 329 (48) | 0.13 |
| Diabetes, n (%) | 511 (51) | 71 (57) | 111 (54) | 329 (48) | 0.13 |
| Dyslipidemia, n (%) | 829 (82) | 100 (80) | 177 (86) | 552 (81) | 0.26 |
| Cerebrovascular disease, n (%) | 343 (34) | 41 (33) | 69 (33) | 233 (34) | 0.94 |
| COPD, n (%) | 391 (39) | 58 (46) | 87 (42) | 246 (36) | 0.051 |
| eGFR <60, median (IQR), mL/min per 1.73 m2 | 40 (31–48) | 40 (30–47) | 39 (32–48) | 41 (31–48) | 0.77 |
| Outcomes | |||||
| LOS, median (IQR), d | 4.9 (2.9–7.1) | 4.8 (3.3–7.9) | 4.9 (2.9–7.5) | 4.3 (2.9–7.0) | 0.17 |
| Increase in diuretic dosing on day 2 of admission, n (%) | 112 (11) | 30 (24) | 33 (16) | 49 (7.3) | <0.001 |
| Net fluid loss, median (IQR), mL | 1060 (150–1950) | 862 (−79 to 1708) | 1004 (120–1685) | 1135 (209–2150) | 0.008 |
| Weight loss (day 2 compared with admission weight), median (IQR), kg | 1.5 (0.3–2.8) | 0.9 (0.1–2.6) | 1.3 (0.2–2.4) | 1.6 (0.4–2.9) | 0.051 |
| Acute kidney injury, n (%) | 233 (23) | 26 (21) | 50 (24) | 157 (23) | 0.87 |
| In‐hospital mortality, n (%) | |||||
| 30‐d readmission, n (%) | 175 (17) | 28 (22) | 39 (19) | 108 (16) | 0.17 |
BMI indicates body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimate glomerular filtration rate; HF, heart failure; IQR, interquartile range; and LOS, length of stay.
DISCUSSION
In a cohort of patients admitted for HF, more than 1 in 3 patients received an initial dose of intravenous loop diuretics equal to or lower than their home diuretic dose. Patients receiving an initial diuretic dose higher‐than‐home dose experienced a shorter stay with increased risk of AKI but no differences in readmission or in‐hospital mortality. These findings offer important insights into the use of diuretics in routine practice and suggest a potential opportunity to optimize diuretic dosing to shorten length of hospitalization.
The DOSE trial 3 was a randomized controlled study of different initial diuretic treatment strategies among patients admitted with acute HF. In the comparison of low‐dose and high‐dose bolus diuretics, patients receiving high‐dose diuretics had a significantly shorter length of hospital stay (4.7 versus 5.8 days), a key secondary end point of the trial. For the primary end point that was a composite of in‐hospital worsening HF, renal dysfunction, and death within 60 days, the trial did not find a difference between the diuretic strategies. In our study, the average LOS for the cohort was shorter than that observed in the DOSE trial. Similar to the DOSE trial, we observed a shorter length of hospital stay among patients with HF who received an initial diuretic dose that was higher than their home dosing regimen. In addition, our secondary chart abstraction analysis demonstrated that a higher‐than‐home dose regimen was associated with larger net fluid loss compared with patients taking an equivalent or lower‐than‐home dose. In addition, patients on lower dosing regimens were more likely to have their dose increased on day 2 of hospitalization. These findings suggest that insufficient initial diuretic dosing is associated with delays in diuresis and increased length of hospital stay.
Observational studies on the association between initial diuretic dosing and HF outcomes are limited. In a prior single‐center observational study that included 609 patients, no association was demonstrated between initial diuretic dosing relative to home dose and patient outcomes, including LOS. 5 However, LOS was numerically shorter in patients receiving a higher‐than‐home dose (5.0 versus 5.7) and the lack of statistical significance may reflect an insufficient sample size. A recent analysis of 3269 patients admitted for HF found that those taking a higher home loop diuretic dose were less likely to receive “optimal” initial intravenous diuretic dosing, defined as at least 2 times the home oral loop diuretic dose in intravenous formulation. 10 This analysis found no association between “optimal” versus suboptimal dosing and LOS or 30‐day readmission. 10 In a study of 191 patients with worsening renal function during hospitalization, daily doses of loop diuretic were significantly higher in patients with worsening renal function relative to controls (199±195 mg versus 143±119 mg, P<0.05). 6 However, patients with worsening renal function also had higher admission creatinine levels and comorbidities, suggesting that diuretic dosing in this study may reflect a risk factor rather than a mediator of outcomes. Our study does suggest that higher diuretic dosing is not without potential risk, given the higher risk‐adjusted rate of AKI among patients receiving an initial diuretic dose that was higher‐than‐home dose when compared with equivalent‐to‐home dose.
In a secondary analysis of the ESCAPE (Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheter Effectiveness) trial, among 395 patients who received diuretics during hospitalization, higher diuretic doses were associated with higher mortality. 7 However, this analysis did not account for the patient's home dosing regimen, which may reflect baseline HF severity, risk of adverse clinical events, and diuretic dosing required to achieve diuresis. Furthermore, the observed mortality rates suggested a nonlinear trend that may be oversimplified by the linear association tested in this study, as mortality declined with increasing doses up to 200 mg/d but increased at doses above 300 mg/d. This speaks to the complicated nature of the association between clinical presentation, comorbidities, diuretic dose, and clinical outcomes of patients hospitalized for HF.
Given the observational nature of our study, we cannot exclude the possibility of unmeasured confounding or bias by treatment indication. However, our secondary chart review analysis found patients taking a lower‐than‐home diuretic dose had lower overall fluid output and were 4 times more likely to have an increase in diuretic dosing on day 2 of hospitalization. These findings align with a pathway suggesting delays in adequate diuresis among patients receiving a lower‐than‐home equivalent dose of diuretics, which could contribute to LOS. We intentionally restricted our analysis to patients taking loop diuretics at home to avoid additional heterogeneity and complexity of categorizing diuretic dosing among patients naïve to loop diuretics. Similarly, we restricted our analysis to doses given on hospital wards to minimize heterogeneity related to management in clinic or emergency departments before admission. In addition, we noted the proportion of patients receiving higher‐than‐home equivalent dosing differed by home loop diuretic type, raising questions about the interplay of home loop diuretic and inpatient loop diuretic dosing. Future investigation of diuretic management before arrival on the hospital floor may yield additional insights. We did not evaluate for additional potentiating diuretics given concurrent to loop diuretics on admission as trials suggesting that these strategies may be helpful were not published before the period of investigation. 11 , 12 As assurance, analyses of rates of use of these adjunct diuretics in the first 24 hours of admission were low (1.7%) during the period of study. Relative to prior observational studies of diuretic dosing, our study included a larger sample size and number of hospitals. However, the study was conducted in two health systems and the clinical practice patterns in diuretic use and associated outcomes may not be representative of other hospitals.
CONCLUSIONS
In a multicenter cohort of patients hospitalized for HF, a higher‐than‐home initial diuretic dose was associated with shorter LOS and higher risk of AKI. These findings suggest that initial diuretic dosing plays an important role in determining clinical outcomes for acute HF. Further research to refine our understanding of appropriate initial diuretic dosing in patients hospitalized with HF may inform opportunities to improve the quality of care of this prevalent and costly condition.
Sources of Funding
Not applicable.
Disclosure
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.
Supporting information
Data S1
Table S1
Figures S1–S3
This manuscript was sent to Sakima Ahmad Smith, MD, MPH, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.124.038783
For Sources of Funding, see page 7.
References
- 1. Peacock WF, Costanzo MR, De Marco T, Lopatin M, Wynne J, Mills RM, Emerman CL. Impact of intravenous loop diuretics on outcomes of patients hospitalized with acute decompensated heart failure: insights from the ADHERE registry. Cardiology. 2009;113:12–19. doi: 10.1159/000164149 [DOI] [PubMed] [Google Scholar]
- 2. Ellison DH, Felker GM. Diuretic treatment in heart failure. N Engl J Med. 2017;377:1964–1975. doi: 10.1056/NEJMra1703100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Felker GM, Lee KL, Bull DA, Redfield MM, Stevenson LW, Goldsmith SR, LeWinter M, Deswal A, Rouleau JL, Ofili EO, et al. Diuretic strategies in patients with acute decompensated heart failure. N Engl J Med. 2011;364:797–805. doi: 10.1056/NEJMoa1005419 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE, Drazner MH, Fonarow GC, Geraci SA, Horwich T, Januzzi JL, et al. 2013 ACCF/AHA guideline for the management of heart failure: a report of the American College of Cardiology Foundation/American Heart Association task force on practice guidelines. J Am Coll Cardiol. 2013;62:e147–e239. doi: 10.1016/j.jacc.2013.05.019 [DOI] [PubMed] [Google Scholar]
- 5. Elashery AR, Aykent K, Kurdi H, Ibrahim M, He S, Petrini JR, Kramer HM. Association between loop diuretic dose administered in first 24 hours of heart failure admissions and length of hospital stay. J Community Hosp Intern Med Perspect. 2018;8:195–199. doi: 10.1080/20009666.2018.1503916 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Butler J, Forman DE, Abraham WT, Gottlieb SS, Loh E, Massie BM, O'Connor CM, Rich MW, Stevenson LW, Wang Y, et al. Relationship between heart failure treatment and development of worsening renal function among hospitalized patients. Am Heart J. 2004;147:331–338. doi: 10.1016/j.ahj.2003.08.012 [DOI] [PubMed] [Google Scholar]
- 7. Hasselblad V, Gattis Stough W, Shah MR, Lokhnygina Y, O'Connor CM, Califf RM, Adams KF. Relation between dose of loop diuretics and outcomes in a heart failure population: results of the ESCAPE trial. Eur J Heart Fail. 2007;9:1064–1069. doi: 10.1016/j.ejheart.2007.07.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. CV‐QUIC Collaborators , Bradley SM, Adusumalli S, Amin AP, Borden WB, Das SR, Downey WE, Ebinger JE, Gelbman J, Gluckman TJ, et al. The cardiovascular quality improvement and care innovation consortium: inception of a multicenter collaborative to improve cardiovascular care. Circ Cardiovasc Qual Outcomes. 2021;14:e006753. doi: 10.1161/CIRCOUTCOMES.120.006753 [DOI] [PubMed] [Google Scholar]
- 9. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A; the Acute Kidney Injury Network . Acute kidney injury network: report of an initiative to improve outcomes in acute kidney injury. Crit Care. 2007;11:R31. doi: 10.1186/cc5713 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Jacobs JA, Carter SJ, Bullock G, Carey JR, Pan IZ, Kinsey MS, Zheutlin AR, Kapelios CJ, Raju S, Fang JC, et al. Optimal initial intravenous loop diuretic dosing in acute decompensated heart failure. JACC: Advances. 2024;3:101250. doi: 10.1016/j.jacadv.2024.101250 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Mullens W, Dauw J, Martens P, Verbrugge FH, Nijst P, Meekers E, Tartaglia K, Chenot F, Moubayed S, Dierckx R, et al. Acetazolamide in acute decompensated heart failure with volume overload. N Engl J Med. 2022;387:1185–1195. doi: 10.1056/NEJMoa2203094 [DOI] [PubMed] [Google Scholar]
- 12. Trullàs JC, Morales‐Rull JL, Casado J, Carrera‐Izquierdo M, Sánchez‐Marteles M, Conde‐Martel A, Dávila‐Ramos MF, Llácer P, Salamanca‐Bautista P, Pérez‐Silvestre J, et al. Combining loop with thiazide diuretics for decompensated heart failure: the CLOROTIC trial. Eur Heart J. 2023;44:411–421. doi: 10.1093/eurheartj/ehac689 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1
Table S1
Figures S1–S3