Diagnostic test accuracy of the initial electrocardiogram after resuscitation from cardiac arrest to indicate invasive coronary angiographic findings and attempted revascularization: A systematic review and meta-analysis

Aim: Conduct a diagnostic test accuracy systematic review and meta-analysis of the post-return of spontaneous circulation (ROSC) electrocardiogram (ECG) to indicate an acute-appearing coronary lesion and revascularization.

Methods: We searched PubMed, EMBASE, CINAHL, Cochrane Library, and Web of Science through February 18, 2020. Two investigators screened titles and abstracts, extracted data, and assessed risks of bias using QUADAS-2. We estimated sensitivity (Sn), specificity (Sp), and likelihood ratios (LR) for all reported ECG features to indicate all reported reference standards. Random-effects meta-analysis pooled comparable studies without critical risk of bias. GRADE methodology evaluated the certainty of evidence.

Results: Overall, 48 studies reported 94 combinations of ECG features and reference standards with wide variation in their definitions. Most studies had risks of bias from selection for coronary angiography and blinding to the ECG and/or reference standard. Meta-analysis combined 6 studies for STE and acute coronary lesion (Sn 0.70 [95% CI 0.54-0.82]; Sp 0.85 [95% CI 0.78-0.90]; LR + 4.7 [95% CI 3.3-6.7]; LR- 0.4 [95% CI 0.2-0.6]) and 4 studies for STE and revascularization (Sn 0.53 [95% CI 0.47-0.58]; Sp 0.86 [95% CI 0.80-0.91]; LR + 3.9 [95% CI 2.8-5.5]; LR- 0.5 [95% CI 0.5-0.6]). Overall certainty of evidence was low with substantial heterogeneity.

Conclusions: Based on low certainty evidence, STE had good classification for acute coronary lesion and fair classification for revascularization. STE was more specific than sensitive for these outcomes and no single ECG feature excluded them. Uniform definitions and terminology would greatly facilitate the interpretation of subsequent studies.