Accuracy and trending ability of finger plethysmographic cardiac output monitoring in late pregnancy; [Précision et capacité au suivi de tendance du monitorage pléthysmographique au doigt du débit cardiaque en fin de grossesse]

Purpose: Individuals in late pregnancy are at risk of significant hemodynamic variations, especially during Cesarean delivery. Although non-invasive monitoring might enable the early detection of variations in cardiac output (CO), clinical validation is lacking. Methods: In a prospective, single-center study, we measured CO simultaneously with finger plethysmography and transthoracic echocardiography in 100 third-trimester pregnant individuals in the supine and left lateral decubitus (LLD) positions. Results: A Bland–Altman analysis revealed a mean (standard deviation) bias of 1.36 (1.04) L·min–1 in the supine position (95% limits of agreement, -0.68 to 3.4 L·min–1; percent error, 26.6%), indicating overestimation by finger plethysmography. The intra-class correlation coefficient was 0.43 (95% confidence interval [CI], 0.33 to 0.51). Regarding the changes in CO induced by the supine-to-LLD transition, the concordance rate in a four-quadrant plot was 98.3% (95% CI, 91.1 to 99.9%). Conclusion: Our study showed a poor reliability of finger plethysmography for static measurement of CO. Nevertheless, finger plethysmography had a reasonably high concordance rate for the detection of CO changes secondary to positional changes in late-pregnant individuals. Study registration date: www.clinicaltrials.gov (NCT03735043); registered 8 November 2018. © 2022, Canadian Anesthesiologists' Society.