Abstract
Background
Approaching the congestive patient is a complex task that requires the combination of different assessment methods. The Strauss formula uses variations in haemoglobin and haematocrit to estimate plasma volume variations (PVV) and haemoconcentration. However, this formula was only validated in outpatients followed with chronic heart failure. We aimed to assess the applicability of this formula to hospitalized patients for acute heart failure (AHF).
Methods
We conducted a single-centre, retrospective, observational study of 302 patients who were admitted to our hospital for AHF during 2016 and were discharged alive. Baseline clinical, laboratory and demographic characteristics were evaluated at admission and the Strauss formula was applied, as PVV (%) = 100 x [(Hb A / Hb D) x (1 − Hct D) / (1 − Hct A)] − 100), where A = admission and D = discharge. At discharge, we considered that a positive change (≥0%) in PV regarding the admission was linked to an increase in PV (haemodilution); a negative change (<0%) correlated to a decrease in the PV (haemoconcentration). The primary endpoint was a composite of cardiovascular death (CV-death) and HF readmission at 3-months.
Results
Mean age was 76±11 years and 57% were male. At baseline, 92% were on clinical-haemodynamic profile B, with a median NT-proBNP of 2157 (IQR 1161–4242) pg/dL, a mean of glomerular filtration rate (GFR) of 63±57 mL/min/m2, a mean haemoglobin of 12±2 g/dL and a mean haematocrit of 38±6%. At discharge, the median plasma volume variation was −1.1% (IQR – 9.6 to 7.8) and the distribution of PVV values in the histogram reveals that a large proportion of patients (44%) increased or maintained plasma volume (PVV ≥0% – haemodilution). The group of patients who decreased plasma volume at discharge was slightly younger (75 vs 78 years, p=0.044), showing higher numerical decreases in NT-proBNP, gamma-glutamyl transferase (gGT) and bilirubin at discharge. A positive change in PV (PVV >0%) during admission almost doubled the risk for readmission and CV-death at 3-months [OR 1.9 (95% CI: 1.1 to 3.1, p=0.026], after adjusting for age and sex.
Conclusions
In this work, we demonstrate that PVV, as calculated by the Strauss formula, increases or is unchanged in 44% of patients admitted with AHF and is strongly associated with a composite of 3-months CV death and HF readmission. Tools to guide the management of residual congestion are of great importance to assess the optimal discharge timing.
Funding Acknowledgement
Type of funding source: None
Discordance between estimated and measured changes in plasma volume among patients with acute heart failure
