The Two-Gasometry Indirect Fick Method: An Inconsistent Method for Measuring Cardiac Output in Critical Patients

Background: The two-gasometry indirect Fick method (iFM) is commonly applied to estimate cardiac output (CO) and cardiac index (CI) in low-resource settings. Validation against other accessible methods such as transthoracic echocardiography (TTE) is lacking. Objective: To determine agreement and clinical interchangeability of CO and CI measurements from TTE (reference test) and two-gasometry iFM (index test) in critical care patients from a low-resource hospital. Methods: Cross-sectional study including critical patients. CO and CI were measured by intensive care medicine residents with TTE (3 per patient-event) and the two-gasometry iFM (1 per patient-event). Agreement was evaluated using mean absolute difference, linear mixed-effects models, and Bland-Altman analysis for repeated measures with 95% limits of agreement (LoA). Precision, bias, and variability were quantified with coefficients of variation, error, and least significant change (LSC), with 95% confidence intervals (95%CI) using bootstrapping. Results: A total 243 measurements were performed across 81 time points in 52 participants. Bland-Altman analysis and linear mixed-effects models revealed poor correlation for both CO (ICC = 0.09) and CI (ICC = -0.09). The mean bias between TTE and iFM was -1.59 L/min (95 LoA: -8.1 to 4.92) for CO and -1 L/min/m2 (95% LoA: -4.8 to 2.8) for CI. The mean absolute percentage error of the two-gasometry iFM method compared to TTE was 57.5% (95%CI: 45.5 to 74.6). The LSC was 11.9% for TTE and 80.4% for two-gasometry iFM. Conclusions: The two-gasometry indirect Fick method performs poorly compared to TTE and should not be used for decision making in critical patients.