Interaction Between Door to Diuretic Time and Dosing As A Predictor of Mortality and Readmission

Introduction A delayed door-to-diuretic time (D2D), defined as the amount of time from hospital presentation to the administration of intravenous diuretic, is a predictor of serious adverse events among patients with acute heart failure (AHF). In addition to timing, the DOSE Trial demonstrated under dosing was also associated with adverse events. Previous investigations have not examined an interaction effect between timing and dosing. We hypothesized that those with the longest delays and lowest dose would have higher 30-day mortality and readmission compared to those with the shortest delays and highest dose. Methods We performed a secondary analysis of existing medical record data from AHF patients who presented to the emergency department between 2020-2021. Inclusion criteria were an ICD-10 code for AHF, positive Framingham Criteria, and the use of intravenous bolus furosemide. Exclusion criteria included ventricular assist devices, dialysis, and ultrafiltration therapy. Data collected included demographics, signs/symptoms, medications and outcomes. Individual hospitalizations were treated as independent events. The composite endpoint was 30-day all-cause mortality or readmission. D2D was statistically split into quartiles, and the dose was split into low dose (<40mg), moderate dose (40mg), moderate-high dose (41-79mg), and high dose (>80mg). Descriptive statistics, t-tests/ANOVA or Mann Whitney U/Kruskal-Wallis and generalized linear models without and with age, sex, race, NYHA class, and Charlson Comorbidity Index as covariates are reported. Results We examined 258 hospitalizations among 164 individuals (median age=71 IQR 19 years; creatinine 1.4 IQR 1; NTpro-BNP 5387 IQR 9778; Charlson Comorbidity Index 5 IQR 3; 54% male; 76% White; 85% NYHA III-IV; 53% n=136 with LVEF <30%). Approximately 42% (n=108) of patients experienced the endpoint. Nearly half (44%) of doses were 40mg (median=40, IQR 20), and the median D2D was 4.5 hours IQR 6.2 hours despite significant variance in symptomology; there were no differences in D2D and dosing across each other. Univariate analysis, D2D (OR=0.94 95%CI 0.76-1.12), dose (OR=0.88 95%CI 0.69-1.13) nor the interaction term (OR=0.94 95%CI 0.85-1.04) were significant. Using D2D (OR=0.92, 95%CI 0.73-1.14) nor dose (OR=0.85 95% CI 0.66-1.10) were significant predictors of the composite endpoint after covariate correction but the interaction D2D*Dose (OR=0.91 95% CI0.82-1.01 p=0.09) was approaching significance. When examining just the index hospitalization (n=164), while D2D (OR=0.85, 95%CI 0.63-1.15) and DD (OR=0.83, 95%CI 0.59-1.20) were not significant predictors, D2D*Dose was significant (OR=0.81 95%CI 0.67-0.97 p=0.02). Conclusions In conclusion, our preliminary data suggest an interaction between D2D and diuretic dosing on the composite endpoint of 30-day all-cause mortality and readmission. However, given that a large majority of doses were 40mg, there was poor and unequal variance in dosing. Secondary analysis of the DOSE Trial, which randomized patients to low dose vs. high dose, may be helpful for understanding the interaction between D2D and dose.