Doppler ultrasound peak systolic velocity versus end tidal carbon dioxide during pulse checks in cardiac arrest.

Abstract

An accurate, non-invasive measure of return of spontaneous circulation (ROSC) is needed to improve management of cardiac arrest patients. During a pulse check in Emergency Department (ED) cardiac arrest patients, we compared the correlation between 1) end tidal carbon dioxide (ETCO₂) and systolic blood pressure (SBP), and 2) Doppler ultrasound peak systolic velocity (PSV) and SBP. Additionally, we assessed the accuracy of PSV ≥ 20 cm/sec in comparison to previously suggested ETCO₂ ≥ 20 or ≥ 25 mmHg thresholds to predict ROSC with SBP ≥ 60 mmHg. This was a secondary analysis of a previously published prospective observational study of ED cardiac arrest patients with an advanced airway and femoral arterial line in place. During each pulse check, highest SBP, highest PSV, and ETCO₂ at the end of the pulse check were recorded. Spearman correlation coefficients were calculated and compared using a Fisher Z-transformation. Accuracy of previously determined PSV and ETCO₂ thresholds for detecting ROSC with SBP ≥ 60 mmHg were compared using McNemar's tests. Based on data from 35 patients with 111 pulse checks, we found a higher correlation between PSV and SBP than ETCO₂ and SBP (0.71 vs 0.31; p < 0.001). Diagnostic accuracy of PSV ≥ 20 cm/sec for detecting ROSC with SBP ≥ 60 mmHg was 89% (95% CI: 82%-94%) versus 59% (95% CI: 49%-68%) and 58% (95% CI: 48%-67%) for ETCO₂ ≥ 20 and ≥ 25 mmHg, respectively. During a pulse check, Doppler ultrasound PSV outperformed ETCO₂ for correlation with SBP and accuracy in detecting ROSC with SBP ≥ 60 mmHg.