Neurological outcome predictors after extracorporeal cardiopulmonary resuscitation: a systematic review

To consolidate current evidence on predictors of neurological outcome following extracorporeal cardiopulmonary resuscitation (eCPR) in patients with cardiac arrest. We conducted a systematic review of the literature across databases including PubMed, MEDLINE, Embase, CINAHL, the Cochrane Library, and Web of Science. Studies assessing neurological outcomes post-eCPR were identified, with a total of 10 studies eligible for individual assessment of which 8 comprising 4353 patients allowed to perform collective statistical analysis. Favorable neurological outcomes were associated with age < 65 years (OR = 6.17), shockable rhythm at extracorporeal membrane oxygenation initiation (OR = 6.67) or hospital arrival (OR = 3.68), and initial pH ≥ 7.0 (OR = 2.01). Other factors involved the presence of any life sign (gasping, positive pupillary light reaction, or increased level of consciousness before or throughout cardiopulmonary resuscitation) (OR = 9.63; Se 0.89, Sp 0.46, PPV 0.22, NPV 0.96), transient return of spontaneous circulation, non-hypoxic mechanism of occurred hepatitis, public location, and hypothermic etiology of cardiac arrest; however, each of those findings was supported by only one study. Unfavorable outcomes were linked to hypoxic brain injury on computed tomography (OR = 12.40; Se 0.366, Sp 0.955, PPV 0.767, NPV 0.787) and elevated serum creatinine (OR = 2.22). The TiPS65 scale showed high predictive accuracy in two studies when the cut-off point was set at 4 points (88.4% and 88.6%; Se 0.172, Sp 0.971, PPV 0.423, and NPV 0.906, and Se 0.193, Sp 0.985, PPV 0.646, and NPV 0.896, respectively). Some predictors, like call-to-hospital time and bystander cardiopulmonary resuscitation, had mixed results across studies. Neurological prognostication in eCPR patients is a complex problem requiring the consideration of multiple variables regarding patient's and cardiac arrest characteristics. Future research should focus on the determination of outcome-affecting factors and assessment of their applicability in clinical settings. New knowledge on this ground will help to create recommendations for eCPR initiation and termination, consequently contributing to treatment results improvement. PROSPERO CRD42024530305.