Clinical decision rules for termination of resuscitation during in-hospital cardiac arrest: A systematic review of diagnostic test accuracy studies

Aim: To assess whether any clinical decision rule for patients sustaining an in-hospital cardiac arrest (IHCA) can predict mortality or survival with poor neurological outcome. Methods: We searched online databases from inception through July 2020 for randomized controlled trials and non-randomized studies. Two reviewers assessed studies for inclusion. We followed PRISMA guidelines for Diagnostic Test Accuracy Studies, used the Quality Assessment of Diagnostic Accuracy Studies framework to evaluate risk of bias, and Grading of Recommendations Assessment, Development and Evaluation methodology to evaluate certainty of evidence. We assessed predictive values for no return of spontaneous circulation (ROSC), death before hospital discharge, and survival with unfavorable neurological outcome. Results: Out of 6436 studies, 92 studies were selected for full-text screening. We included 3 observational studies describing the derivation and external validation for the UN10 rule (Unwitnessed arrest; Nonshockable rhythm; 10 min of resuscitation without ROSC) amongst patients suffering from IHCA. No studies were identified for clinical implementation. Positive Predicted Values (PPV) for death before hospital discharge for the three studies were 100% (95% CI: 97.1%-100%), 98.9% (95% CI: 96.5%-99.7%), and 93.7% (95% CI: 93.3%-94.0%). One study reported a PPV for prediction of survival with unfavorable neurological outcome, 95.2% (95% CI: 94.9%-95.6%). The level of evidence was rated as very low certainty. Conclusions: We identified very low certainty evidence for one clinical decision rule (the UN-10 rule) that was unable to reliably predict mortality or survival with unfavorable neurological outcome for adults suffering from IHCA. We identified no evidence for children. PROSPERO CRD42020164091.