Introduction

The accurate assessment of the cardiac output (CO) is a prerequisite for the evaluation of various cardiac and pulmonary diseases. Invasive CO assessment with the thermodilution (TD) method or the direct Fick method (FM) are both considered as reference standards. As the measurement of oxygen consumption (V̇O2) is time consuming, indirect assessment of CO via estimation of V̇O2 with prediction equations (iFM) is preferred in clinical practice. However, discrepancies between the TD- and iFM-derived CO can be larger than 20%. We compared TD-derived CO with iFM using different formulas for V̇O2-estimation, among which the already widely implemented Krakau formula has yet to be validated. We further aimed to determine the predictors of a discrepancy between referent and predicted CO formulas.

Methods

This retrospective, observational, single center study, investigated 197 patients who underwent both right heart catheterization (RHC) and cardiac magnetic resonance imaging. The formulas used for calculation of iFM were proposed by Krakau (1), Lafarge (2), Dehmer (3) and Bergstra (4). TD-CO served as the reference standard and was compared with each method using the Wilcoxon signed-rank test and Pearson’s correlation coefficient (r). Predictors of a deviation > 20% from the reference TD-derived CO (ratio iFM_Krakau/TD-CO <0.8 OR ratio iFM_Krakau/TD-CO > 1.2) were determined in logistic regression analysis. The study was conducted in compliance with the Declaration of Helsinki. Ethical approval was waived by the local Ethics Committee.

Results

The mean age of the 197 analyzed patients was 62 (±14.9) years, and 29% were women. The mean ejection fraction was 50 % (30-54%) and the mean TD-derived Output was 5.4 l/min (4.4 - 6.5l/min).

When compared to TD-derived-CO and the other three equations for V̇O2 estimation and CO calculation, the patient number with CO-discrepancy >20% using the Krakau formula was lower (i.e. performed better) than Lafarge, was similar to Dehmer, and was higher compared to Bergstra. The formula of Lafarge underestimated, and the formula of Bergstra tended to overestimate CO-values. The formulas proposed by Krakau and Dehmer operated somewhere in between. However, none of the formulas tested showed good agreement with the TD method, i.e. percentage errors for all equations were >30%. The main predictors of a significant discrepancy between iFM_Krakau and TD-derived CO were aortic valve stenosis, higher pulmonary pressure, and smaller body height.

Discussion

Considering absolute deviations and percentage errors between TD-derived CO and the four prediction formulas, none of the equations seems reliable when applied to modern cohorts. The hitherto favoured formula for V̇O2 estimation proposed by Krakau performed within the range of the established formulas when the numbers of significant discrepancies from TD-derived CO were compared. Yet, none of formulas showed good agreement with TD-derived CO. Further studies with current patient cohorts are needed for the optimization of prediction formulas.

References

1: Krakau I. Thieme, 1999.

2: LaFarge CG et al. Cardiovasc Res. 1970;4(1):23-30.

3: Dehmer GJ et al. Clin Cardiol. 1982;5(8):436-40.

4: Bergstra A et al. Eur Heart J. 1995;16(5):698-703.