Interchangeability of cardiac output measurements between non-invasive photoplethysmography and bolus thermodilution: A systematic review and individual patient data meta-analysis.

Abstract

BACKGROUND\nContinuous non-invasive cardiac output devices using digital photoplethysmography (PPG) are widely available for bedside use, but their interchangeability with reference methods has not yet been evaluated in a systematic review and patient data meta-analysis.\n\n\nMETHODS\nA systematic review and meta-analysis of studies comparing non-invasive cardiac output monitoring using PPG with the invasive bolus thermodilution method was performed. With ethical approval, all published studies from the PUBMED, Embase, Scopus, Web of Science, and Google Scholar databases from January 1, 2010 to January 1, 2018 were included. From these analysed studies, individual patient data were interpreted using the interchangeability methods for both absolute values and changes in cardiac output measurements.\n\n\nRESULTS\nTen studies comparing PPG and bolus thermodilution in the operating room and intensive care settings were included. The interchangeability rate (95% CI) was 37% (24-48) (n=1350 pairs of measurements). The interchangeability rate was poorer with the CNAP device (CNSystems, Graz, Austria) [18% (17-20)] than with the Clearsight (Edwards Lifesciences, Irvine, CA) device [33% (31-34), P<0.0001], for patients receiving norepinephrine [19% (17-20) vs. 33% (32-34), P<0.0001], and for patients with low mean arterial pressure (<65mmHg) [26% (23-29) vs. 30% (29-31), P<0.0001]. Among the 1009 comparisons of the changes in cardiac output between both methods, 561 (56%) were interpretable with a trend interchangeability rate at 24% (12-36).\n\n\nCONCLUSIONS\nCardiac output measurements using PPG were not interchangeable with bolus thermodilution in regard to both absolute values and changes in cardiac output measurements, and should be used with caution in clinical practice.\n\n\nTRIAL REGISTRATION\nPROSPERO ID CRD42018089513.