Abstract WMP93: Somatosensory Evoked Potentials Monitoring In Acute Ischemic Stroke (Promise) Study: A Biomarker Of Functional Recovery Prior To Mechanical Thrombectomy (clinicaltrials.gov: NCT04099615)

Abstract

Hypothesis: Somatosensory evoked potentials (SEPs), as surrogate of cerebral blood flow, may give substantial additional predictive value to that provided by clinical and imaging factors and optimize the selection of patients benefiting from mechanical thrombectomy (MT) Methods: Bilateral median nerve SEPs were recorded before and continuously during MT in consecutive patients with AIS and anterior LVO. N20 response ipsilateral to the ischemic hemisphere was measured. The adjusted predictive value of the N20 on functional Independence (modified Rankin Scale score ≤2) after MT was analyzed by binary logistic regression and its predictive value on the full range of disability by ordinal logistic regression. We constructed different regression models with other clinical and imaging predictors at the pre- and in- hospital setting to determine the independent predictive power of N20 for a potential treatment decision-making Results: A total of 223 patients were recruited. A blinded reading of SEPs recordings identified presence of N20 in 110 (49.3%), absence in 58 (26%) and not assessable in 55 patients due to radiofrequency interferences. Prior to MT, N20 had positive predictive value (PPV) of 93% (95%CI, 0.8 to 0.98) and increased 10 fold the likelihood of good functional outcome at 7 days (adjusted OR, 9.9; 95%CI, 3.1-44.6). Receiver operating curves showed that N20 had a higher capacity to predict good functional outcome (AUC 0.82) than models constructed with pre-hospital (age, sex, serum glucose, median blood pressure, NIHSS; AUC 0.75) or in-hospital (NIHSS, ASPECTS; AUC 0.72) factors. During MT, N20 had PPV of 100% (95% CI, 0.85-1) and was the unique independent factor associated with functional independence (adjusted OR, 2.74; 95%CI, 1.90-4.29). Perfusion CT or MR was available in a subgroup of 116 patients. Baseline N20 showed a higher capacity to predict good functional outcome (n=168, AUC 0.71) compared with ischemic core (n=127, AUC 0.66), ischemic penumbra (Tmax>6) (n=116, AUC 0.54) and collateral status (n=191, AUC 0.61) Conclusions: SEPs monitoring is a fast and bedside technique that could complement current clinical factors in terms of increasing the eligibility of AIS patients for MT and improving prognosis.