Fig 1.
Validation cohort flow diagram.
AHF = Acute Heart Failure.
Table 1.
Clinical characteristics of patients in the validation cohort.
Fig 2.
Hemodynamic profiles by cardiac index vs. vascular tone (systemic vascular resistance index, SVRI).
Profiles were numbered similarly to facilitate between cohort comparisons: 1 (purple)—lowest SVRI and highest cardiac index, 2 (gold)—highest SVRI and lowest cardiac index, 3 (black) cardiac index and SVRI between profiles 1 and 2. (A) Profiling in the validation cohort (VC) alone. (B) The VC patients and their profiles are overlayed with the derivation cohort (DC). Patients in the DC had acute heart failure and were monitored in the emergency department like the VC, but were enrolled in a prior study (external cohort). Few patients classified in a particular profile in the VC would have been classified differently in the DC. (C) The VC overlayed with the control cohort (CC). The CC included patients enrolled in the same study as the DC, but who had sepsis rather than AHF. Profiling in the CC differed from VC, with several VC patients who would have been classified in a different profile by profiling of the CC (and visa versa).
Table 2.
Study outcomes overall and by hemodynamic profile.
Fig 3.
Comparison of cardiac index vs. vascular tone hemodynamic profiles by 30-day adverse events, emergency department (ED) characteristics, and ED disposition.
(A) Hemodynamic profiles 1–3 in the validation cohort (inset) are compared by individual components of the composite 30-day secondary outcome. Compared to profile 2 (gold in inset), profile 3 (black) and profile 1 (purple) had greater rates of any outcome in the composite. (B) The were no statistically significant differences between profiles in actual ED disposition decisions (ICU, or discharge from ED), ED treatments administered, or the presence of unstable vital signs or need for supplemental oxygen.
Fig 4.
Survival to primary and secondary study outcomes by hemodynamic profile.
Kaplan Meier curves for three hemodynamic profiles by cardiac index and vascular tone (see inset) in the validation cohort, through 90-day follow-up. (A) Primary outcome: All-cause mortality or cardiac arrest. Profile 3 (black) has significantly worse survival compared to profiles 1 or 2 (purple and gold, respectively) through each of 30, 60 and 90 days. (B) Secondary outcome: A composite of invasive cardiac procedure, new or emergent dialysis, intubation, mechanical cardiac support or transplant, and death or cardiac arrest. Profile 2 has significantly better event-free survival compared to profiles 1 or 3 through each of 30, 60 and 90 days.