Author + information
- Jin Joo Park, MDa,
- Sun-Hwa Kim, PhDa,
- Il-Young Oh, MDa,
- Dong-Ju Choi, MDa,∗ (, )
- Hyun-Ah Park, MDb,
- Hyun-Jai Cho, MDc,
- Hae-Young Lee, MDc,
- Jae-Yeong Cho, MDd,
- Kye Hun Kim, MDd,
- Jung-Woo Son, MDe,
- Byung-Su Yoo, MDe,
- Jaewon Oh, MDf,
- Seok-Min Kang, MDf,
- Sang Hong Baek, MDg,
- Ga Yeon Lee, MDh,
- Jin Oh Choi, MDh,
- Eun-Seok Jeon, MDh,
- Sang Eun Lee, MDi,
- Jae-Joong Kim, MDi,
- Ju-Hee Lee, MDj,
- Myeong-Chan Cho, MDj,
- Se Yong Jang, MDk,
- Shung Chull Chae, MDk and
- Byung-Hee Oh, MDc
- aCardiovascular Center, Division of Cardiology, Seoul National University Bundang Hospital, Seongnam, Republic of Korea
- bDepartment of Family Medicine, Inje University Seoul Paik Hospital, Seoul, Republic of Korea
- cDepartment of Internal Medicine, Seoul National University Hospital, Seoul, Republic of Korea
- dHeart Research Center of Chonnam National University, Gwangju, Republic of Korea
- eYonsei University Wonju College of Medicine, Wonju, Republic of Korea
- fYonsei University College of Medicine, Seoul, Republic of Korea
- gDepartment of Internal Medicine, the Catholic University of Korea, Seoul, Republic of Korea
- hDepartment of Internal Medicine, Sungkyunkwan University College of Medicine, Seoul, Republic of Korea
- iDivision of Cardiology, Asan Medical Center, Seoul, Republic of Korea
- jChungbuk National University College of Medicine, Cheongju, Republic of Korea
- kKyungpook National University College of Medicine, Daegu, Republic of Korea
- ↵∗Address for correspondence:
Dr. Dong-Ju Choi, Cardiovascular Center, Seoul National University Bundang Hospital, College of Medicine, Seoul National University, Gumiro 166, Bundang, Seongnam, Gyeonggi-do, Republic of Korea.
Objectives This study sought to examine the impact of door-to-diuretic (D2D) time on mortality in patients with acute heart failure (AHF) who were presenting to an emergency department (ED).
Background Most patients with AHF present with congestion. Early decongestion with diuretic agents could improve their clinical outcomes.
Methods The Korea Acute Heart Failure registry enrolled 5,625 consecutive patients hospitalized for AHF. For this analysis, the study included patients who received intravenous diuretic agents within 24 h after ED arrival. Early and delayed groups were defined as D2D time ≤60 min and D2D time >60 min, respectively. The primary outcomes were in-hospital death and post-discharge death at 1 month and 1 year on the basis of D2D time.
Results A total of 2,761 patients met the inclusion criteria. The median D2D time was 128 min (interquartile range: 63 to 243 min), and 663 (24%) patients belonged to the early group. The baseline characteristics were similar between the groups. The rate of in-hospital death did not differ between the groups (5.0% vs. 5.1%; p > 0.999), nor did the post-discharge 1-month (4.0% vs. 3.0%; log-rank p = 0.246) and 1-year (20.6% vs. 19.3%; log-rank p = 0.458) mortality rates. Get With the Guidelines-Heart Failure risk score was calculated for each patient. In multivariate analyses with adjustment for Get With the Guidelines-Heart Failure risk score and other significant clinical covariates and propensity-matched analyses, D2D time was not associated with clinical outcomes.
Conclusions The D2D time was not associated with clinical outcomes in a large prospective cohort of patients with AHF who were presenting to an ED. (Registry [Prospective Cohort] for Heart Failure in Korea [KorAHF]; NCT01389843)
Acute heart failure (AHF) is a life-threatening condition with high morbidity and mortality, which requires immediate medical intervention (1). Most patients with AHF present with signs and symptoms of congestion, but few present with symptoms of low cardiac output (2). Thus, early decongestion with diuretic agents is 1 of the cornerstones for the treatment for patients with AHF (1).
In patients with ST-segment elevation myocardial infarction (STEMI), the door-to-balloon time correlates with the extent of myocardial injury and patients’ outcomes (3,4); accordingly, the current practice guideline recommends a door-to-balloon time of <90 min (5). Similarly, the concept of “time to therapy” has been introduced in patients with AHF, by emphasizing the rapid initiation of diagnosis and therapeutic intervention. Recently, Matsue et al. (6) reported that patients with AHF receiving intravenous loop diuretic agents within 60 min after emergency department (ED) arrival had lower in-hospital mortality rates.
In contrast to patients with STEMI, where acute occlusion of a coronary artery and the following ischemia lead to irreversible myocardial injury within a short time period (“time is myocardium”), AHF is a “subacute” process with a remote trigger and successive decompensation leading to hemodynamic and clinical congestion. Considering the reversibility of congestion and the relatively long time period between the trigger and full-blown AHF, the door-to-diuretic (D2D) time may have limited clinical impact on clinical outcomes.
In this study, we sought to examine the effect of D2D time on the in-hospital and post-discharge clinical outcomes in a large, prospective cohort of patients with AHF presenting to an ED.
The Korea Acute Heart Failure (KorAHF) registry was a prospective, multicenter cohort study that consecutively enrolled 5,625 patients hospitalized for AHF syndrome from 10 well-known tertiary university hospitals throughout Korea between March 2011 and December 2014 (NCT01389843). Detailed information on the study design and its results have been previously reported (7,8). Patients with signs or symptoms of heart failure (HF) and lung congestion, objective findings of left ventricular systolic dysfunction, or structural heart disease were eligible for the study. All patients were scheduled for follow-up at least 3 years after the index hospitalization. The mortality data for patients who were lost to follow-up were collected from the National Insurance data or National Death Records.
In this study, we included only patients who were admitted to an ED and received intravenous furosemide within 24 h after ED arrival. Patients who received the first dose of furosemide 24 h after ED arrival were excluded because they may have received intravenous furosemide for worsening HF during the hospital admission. We also excluded patients who were undergoing long-term dialysis therapy before hospital admission and patients who received oral furosemide in ED. The study protocol was approved by the ethics committee or Institutional Review Board at each hospital. Written informed consent was waived by the ethics committee or Institutional Review Board. The study complied with the Declaration of Helsinki.
Study variables and definitions
The ED arrival time and intravenous furosemide administration time were documented for each patient, and the D2D time was defined as the time interval between ED arrival and the first intravenous administration of furosemide. Early and delayed treatment groups were defined as D2D time <60 min and ≥60 min, respectively. The cutoff time of 60 min was decided on the basis of earlier reports, which concluded that a D2D time <60 min was reported to be associated with a 61% lower in-hospital mortality rate (6). The primary outcomes were the rates of in-hospital death and 1-month and 1-year post-discharge all-cause death. Secondary outcomes included the use of catecholamines, mechanical circulatory support devices, and mechanical ventilation.
We planned 4 exploratory subgroup analyses according to the following: 1) ejection fraction (EF); 2) HF onset (de novo vs. acute decompensation); 3) renal function (chronic kidney disease vs. no chronic kidney disease); and 4) New York Heart Association (NYHA) functional class (II to III vs. IV). Patients were classified as having HF with reduced, midrange, and preserved EF with a left ventricular ejection fraction cutoff of 40% and 50%, respectively. Chronic kidney disease was defined as an initial glomerular filtration rate <60 ml/min/1.72 m2.
A sample size of 2,750 (650 in the early group and 2,100 in the delayed group) would achieve a 90% power to detect a difference between the group proportions of 4%. The in-hospital death rate in the early group is assumed to be 2% (6). The test statistic used is the 2-sided z-test with pooled variance. The significance level of the test was targeted at 5%.
Data are presented as numbers and frequencies for categorical variables and as mean ± SD or median with interquartile ranges for continuous variables. For comparisons among groups, the chi-square test (or the Fisher exact test when any expected count was <5 for a 2 × 2 table) was used for categorical variables, and the unpaired Student's t-test or Mann-Whitney U test was used for continuous variables.
The Get With the Guidelines–Heart Failure (GWTG–HF) score was calculated for each patient (9) and was used for estimation of risk. Kaplan-Meier curves were plotted and compared, using the log-rank test for evaluation of post-discharge outcomes. A multivariable logistic regression and Cox proportional hazards regression models were used to determine the independent effect of D2D time on in-hospital and post-discharge outcomes, respectively. Variables found to be statistically significant (p < 0.10) in the univariate analysis were included in the multivariable model, except for variables with >10% missing values or variables that were closely related to other clinical variables. As a sensitivity analysis, we developed models with adjustment with GWTG–HF score and propensity score matching. Propensity score matching was done for comparisons between the 2 groups (early group vs. delayed group). The propensity score was estimated using multivariable logistic regression analysis with the variables listed in the second adjustment model. A propensity score-matched cohort was created using the nearest neighbor method without replacement, in a 1:2 ratio. The “MatchIt” package of R programming (The R Foundation for Statistical Computing, Vienna, Austria) was used for the matching.
A 2-sided probability value <0.05 was considered indicative of a statistically significant difference. Statistical tests were performed using SPSS software version V.22 (IBM, Armonk, New York), R programming version 3.1.0, and PASS 11 (NCSS, LLC, Kaysville, Utah). All the analyses were performed by a professional statistician (S.-H.K.).
The KorAHF registry included 5,625 patients at 10 tertiary academic centers in Korea. In this study, 8 centers with 4,524 patients participated. Of them, 766 patients did not visit an ED, 129 were undergoing long-term dialysis, and 868 did not receive intravenous furosemide within 24 h, so the data of 2,761 patients were available for the final analysis (Figure 1). The mean age was 70 years, 49% were male, 55% had de novo HF, 37% had diabetes, and 62% had hypertension.
The D2D time was non-nominally distributed with a median of 128 min (interquartile range: 63 to 243 min). According to the definition, 663 (24%) patients were classified as belonging in the early group and 2,098 (76%) were classified as belonging in the delayed group. The baseline characteristics were similar between the groups, except that patients in the early group had higher mean blood pressure, whereas those in the delayed group had a higher rate of atrial fibrillation. More NYHA functional class IV patients were in the early group, whereas patients in the delayed group had higher GWTG–HF scores with a marginal significance (p = 0.058) (Table 1). In a logistic regression analysis, the presence of ischemic heart disease and a higher GWTG score were independently associated with delayed D2D time, whereas atrial fibrillation, chronic obstructive pulmonary disease, advanced NYHA functional class, and a higher heart rate were associated with early D2D time. The participating center was not associated with D2D time (Table 2).
Regarding intravenous catecholamine and vasodilator use, there was no difference between the groups. Patients in the early group required mechanical ventilation more often (19% vs. 13%; p < 0.001) than those in the delayed group (Table 3).
During hospital admission, 140 (5.1%) patients died. The rate of in-hospital mortality did not differ between the groups (Figure 2). When stratified according to the GWTG–HF score, the in-hospital death rate increased with GWTG–HF score tertiles, but without significant difference between the early and delayed groups (Online Table 1).
On logistic regression analysis, delayed treatment was not associated with in-hospital death in univariate analysis (odds ratio: 1.03; 95% confidence interval: 0.69 to 1.53), after adjustment for GWTG–HF score (odds ratio: 1.00; 95% confidence interval: 0.66 to 1.52) and significant covariates (odds ratio: 1.05; 95% confidence interval: 0.64 to 1.70) (Table 4).
During follow-up after hospital discharge, 85 (3.1%) patients died at 1 month, and 514 (18.6%) patients at 1 year. In Kaplan-Meier survival curves, the rates of post-discharge 1-year all-cause death and hospitalization for worsening HF did not differ between the groups (Figures 3A and 3B).
Delayed treatment was not associated with worse outcomes in univariate analysis, after adjustment for GWTG–HF score, and in multivariable analysis (Table 4).
A total of 1,959 patients were matched on the basis of the propensity score. The baseline characteristics of the cohort after matching were well balanced (Online Table 2). In the matched population, the rate of in-hospital mortality and the post-discharge 1-month and 1-year mortality rates did not differ between the early and delayed groups (Online Figures 1A and 1B).
We performed 4 exploratory subgroup analyses including left ventricular EF, HF onset, renal function, and NYHA functional class. There was no significant interaction between D2D time and the subgroups, and D2D time was not associated with in-hospital and post-discharge outcomes across all subgroups (Online Figure 2).
Relationship between D2D time and in-hospital mortality rate
Most patients received intravenous diuretic agents within 200 min, and the number of patients with D2D time >200 min decreased rapidly. In a restricted cubic spline modeling with 4 knots, the association between D2D time and predicted in-hospital mortality rate was not linear; there seems to be an initial decrease in the first 100 min, followed by a gradual increase afterward. However, because of the wide confidence interval, a clear association cannot be made (Figure 4).
Akin to treatment of acute myocardial infarction, the concept of “time to therapy” has also been introduced in patients with AHF, by emphasizing the rapid initiation of diagnosis and therapeutic intervention. Because most patients with AHF present with congestion, current AHF treatment guidelines also advocate the timely initiation of diuretic therapy (2). Early decongestion improves symptoms and possibly patients’ survival (6). In this study of a large prospective cohort of patients with AHF presenting to EDs, we examined the impact of D2D time on outcomes, and showed that D2D time was not associated with either in-hospital or post-discharge outcomes.
D2D time varied notably among patients with a median of 128 min and an interquartile range of 63 to 243 min. Because this was not a randomized controlled study, but rather a prospective cohort study, the observed D2D time reflected real clinical practice: the time to diagnosis of HF and the time to the decision to initiate diuretic agents varied considerably from patient to patient.
A total of 24% of the patients received intravenous diuretic agents within 60 min after ED arrival. Although the baseline characteristics were similar between the groups, patients in the early group had more NYHA functional class IV dyspnea and lung congestion visible on chest radiography, as expected. Symptoms indicative of advanced congestion may have alerted the clinicians to earlier diagnosis and initiation of diuretic agents. However, patients in the early group also had more favorable risk profiles such as higher blood pressure and lower GWTG–HF score than those in the delayed group, with a marginal significance.
The main finding of this study was that there was no difference in the in-hospital and post-discharge outcomes between patients in the early and delayed groups. The neutral association between D2D and the clinical outcomes was consistent in the univariate and multivariate analyses with adjustment for GWTG–HF score and significant clinical covariates, as well as in the propensity score–matched analysis. Our study findings contradict those of Matsue et al. (6), who reported that D2D time <60 min was associated with a 61% reduced risk for in-hospital mortality in 1,291 patients with AHF presenting at ED in Japan. Although there are differences in patients’ characteristics between the 2 studies, this finding is significant because the study populations of both studies were East Asians with comparable risk profiles (Online Table 3).
The concept of “time to therapy” in patients with AHF includes the timely initiation of diuretic therapy, inferring from the experiences in patients with acute myocardial infarction who benefit from early recanalization therapy. Although both “acute” myocardial infarction and “acute” HF seem to be “acute” processes, there are important differences in the pathophysiologic aspects. In patients with STEMI, sudden occlusion of a coronary artery is the trigger that causes ischemia and myocardial cell deaths, which begin instantly (“time is myocardium”) (10). Rapid reperfusion, which specifically targets the underlying pathophysiology, preserves the myocardium and heart function and consequently improves the prognosis. In contrast, AHF is a heterogeneous disease, which is a syndrome rather than a single disease entity with various triggers (e.g., infection or noncompliance), followed by successive hemodynamic and clinical congestion. Adamson et al. (11), using an implantable hemodynamic monitor, showed that the pressure increases 4 days before the exacerbations requiring hospitalization in patients with HF. Similarly, in the CHAMPION (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients) study, the daily measurement of pulmonary artery pressures and adjustment of medication with diuretic agents and vasodilators reduced the rehospitalization rate, thus demonstrating the subacute nature of AHF (12). AHF is a “subacute” process with a remote trigger, followed by successive hemodynamic and clinical congestion, which can be reversed with diuretic therapy. Currently there is no targeted therapy for the pathophysiology of AHF (13) and there is a relatively long time interval between the trigger and full-blown AHF. Therefore, the length of D2D time on clinical outcomes in patients with AHF has a limited role compared with door-to-balloon time in patients with STEMI.
In the RELAX-AHF (Relaxin in Acute Heart Failure) trial (14), treatment with serelaxin was associated with substantial improvement of dyspnea and biomarkers of organ injury and reduced 180-day mortality rates, thereby supporting the hypothesis that myocardial damage is a progressive phenomenon in patients with AHF, and early treatment mitigating this organ damage could improve outcomes. In contrast, larger studies designed to evaluate the effect of early administration of nesiritide in the ASCEND-HF (the Acute Study of Clinical Effectiveness of Nesiritide in Decompensated Heart Failure) trial (15) and ularitide in the TRUE-AHF (Trial of Ularitide’s Efficacy and Safety in patients with Acute Heart Failure) trial (16) failed to show survival benefit despite rapid improvement of congestion.
First, this was not a randomized controlled study that was specifically designed to evaluate the impact of D2D time on clinical outcomes in patients with AHF. Therefore, the possibility that confounding factors may have influenced the study results cannot be ruled out, although we performed several sensitivity analyses including multivariable adjustment and propensity score matching. Nevertheless, the large sample size with sufficient power makes the neutral association between D2D time and clinical outcomes less prone to type I error. Indeed, our study is a large prospective cohort study that investigated the effect of D2D time on in-hospital and post-discharge outcomes. Second, the timing and dose of initial intravenous furosemide were not uniform, and the diuretic effect was not monitored. Third, because we included only East Asians who had different patients’ characteristics compared with representative AHF cohorts from the United States (17) and Europe (18), the study results cannot be generalized to all patients with HF. Most importantly, given the nature of the study design, the study results are at best hypothesis generating, and the effect of D2D time on clinical outcomes must be confirmed in successive confirmatory studies.
D2D time was not associated with clinical outcomes in this large cohort of patients with AHF.
COMPETENCY IN MEDICAL KNOWLEDGE: In patients with AHF who were presenting to an ED, D2D time was not associated with either in-hospital or post-discharge outcomes.
TRANSLATIONAL OUTLOOK: Given the observational nature of this study, the definitive effect of D2D time on mortality should be investigated in successive clinical studies.
This work was supported by Research of Korea Centers for Disease Control and Prevention (grant nos. 2010-E63003-00, 2011-E63002-00, 2012-E63005-00, 2013-E63003-00, 2013-E63003-01, 2013-E63003-02, and 2016-ER6303-00) and by the SNUBH Research Fund (grant nos. 14-2015-029, 16-2017-003). All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute heart failure
- emergency department
- ejection fraction
- Get With the Guidelines–Heart Failure
- heart failure
- New York Heart Association
- ST-segment elevation myocardial infarction
- Received December 4, 2017.
- Accepted December 28, 2017.
- 2018 The Authors
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Mebazaa A.,
- Yilmaz M.B.,
- Levy P.,
- et al.
- McNamara R.L.,
- Wang Y.,
- Herrin J.,
- et al.
- O’Gara P.T.,
- Kushner F.G.,
- Ascheim D.D.,
- et al.
- Matsue Y.,
- Damman K.,
- Voors A.A.,
- et al.
- Lee S.E.,
- Lee H.Y.,
- Cho H.J.,
- et al.
- Peterson P.N.,
- Rumsfeld J.S.,
- Liang L.,
- et al.
- Gibson C.M.
- Adamson P.B.,
- Magalski A.,
- Braunschweig F.,
- et al.
- Januzzi J.L. Jr..,
- Felker G.M.
- Packer M.,
- O’Connor C.,
- McMurray J.J.V.,
- et al.
- Fonarow G.C.,
- Heywood J.T.,
- Heidenreich P.A.,
- et al.