Abstract 112: Recrudescence of Neurological Deficits After Stroke: Clinical-imaging Phenotype, Triggers and Risk Factors

Abstract

Background: Re-emergence of prior stroke-related deficits or post-stroke recrudescence ( PSR ) has not been adequately studied. Methods: Using preliminary criteria for PSR (transient recurrence or worsening of residual post-stroke deficits; chronic brain infarct or hemorrhage but no acute vascular lesion on MRI including DWI; no severe ipsilateral cerebral artery stenosis; no evidence for seizure), we identified 153 patients (145 prior infarcts, 8 hypertensive ICH) with 164 admissions for PSR from 2000-2015. Clinical-imaging features of PSR were characterized. Triggers of PSR were identified by comparing PSR admissions to adjacent admissions without PSR (n=65). Risk factors for PSR were identified by comparing PSR cases to 1861 controls without PSR in our prospective institutional stroke registry. Results: PSR occurred 3.9±0.6y after stroke, lasted 18.4±20.4 hrs, and 69% resolved on Day 1. Mean age 67±16y, 60% women. Neuro-worsening was usually abrupt, mild, with motor-sensory or language dysfunction. No patient had isolated gaze paresis, hemianopia or neglect. During PSR the NIHSS worsened by mean 2.5±1.9 points; deficits were limited to a single subscale NIHSS item in 38%. The underlying chronic strokes were variably-sized, predominantly affected white-matter tracts, and 73% involved the MCA territory. Infection, hypotension, hyponatremia, insomnia/stress and benzodiazepine use were higher (p<0.05) during PSR as compared to adjacent admissions without PSR. As compared to the ischemic stroke registry controls, the subgroup of 145 cases with PSR after ischemic stroke had more women, African-American and Other races, diabetes, dyslipidemia, smoking, and more severe NIHSS scores at the time of index stroke (all p<0.05). The PSR group had more small-vessel infarcts and more strokes from “other definite” causes. Six patients with PSR received iv tPA without hemorrhagic complications. Conclusion: This detailed characterization of PSR should enable prompt diagnosis and distinguish PSR from mimics such as TIA, migraine, Todd’s paralysis, Uhthoff’s phenomenon, and others. This is the first large comprehensive study of PSR. Prospective studies are required to validate our proposed diagnostic criteria and decipher underlying mechanisms.