Association between Cerebral Autoregulation-Derived Mean Arterial Pressure Deviation and Neurological Prognosis in NICU Patients: A Prospective, Exploratory, Observational Study

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background and Purpose Cerebral autoregulation (CA) is a core physiological mechanism that maintains adequate cerebral blood flow (CBF) despite fluctuations in cerebral perfusion pressure. Patients in the neurological intensive care unit (NICU) often exhibit varying degrees of CA impairment, rendering the brain vulnerable to hypoperfusion and insufficient oxygen delivery even at blood pressure levels recommended by population-based guidelines. CA-derived indices enable the assessment of individualized mean arterial pressure (MAP) within specific time windows. This study aimed to investigate whether personalized CA-derived MAP deviation is associated with neurological prognosis in NICU patients. Methods We conducted a retrospective analysis of prospective data from adult patients (≥ 18 years) admitted to the NICU of a neurocenter between June 2025 and October 2025. Near-infrared spectroscopy (NIRS) was used to synchronously monitor the dynamic linear correlation between cerebral tissue oxygen saturation (rSO₂) and MAP, and the cerebral oximetry index (COx) reflecting CA function was calculated. A multi-window weighting algorithm was applied to determine the time-varying optimal MAP (MAPₒₚₜ) for each patient, and the median of the NIRS-MAP reference interval (NMM) within the target time window was identified. The NIRS-MAP Median Deviation (NMMD) was defined as the absolute difference between NMM and the mean MAP (MM) within 24h, 48h, and 72h respectively. Unfavorable neurological outcomes were defined as a higher modified Rankin Scale (mRS) score at discharge compared with admission. Multivariate binary logistic regression models adjusted for age, gender, and underlying diseases were used to examine the association between NMMD (in different time windows and under the condition of MM < NMM, i.e., toward the lower limit of autoregulation [LLA]) and unfavorable neurological outcomes. Results Among the 34 enrolled patients, 23 (67.6%) had favorable outcomes and 11 (32.4%) had unfavorable outcomes. NMMD at 24h and 48h was significantly associated with unfavorable neurological outcomes, with odds ratios (OR) ranging from 0.729 to 0.852 (p ≤ 0.033). Notably, the association remained significant for 24h NMMD when MM < NMM (OR: 0.776, 95% confidence interval [CI]: 0.612–0.985, p = 0.037), and the corresponding time burden parameter (tNMMD) also showed statistical significance (OR: 0.766, 95% CI: 0.592–0.989, p = 0.041). Receiver operating characteristic (ROC) analysis demonstrated that 24h NMMD had the strongest discriminative ability (area under the curve [AUC] = 0.84). Conclusion In NICU patients, a larger NMMD during hospitalization, especially within the first 24h, is independently associated with a reduced risk of unfavorable neurological outcomes, and this association persists even when MAP deviates toward the LLA. This finding suggests that a moderate degree of MAP deviation may be beneficial for preserving CA function and exerting its compensatory effects, highlighting the independent importance of early CA assessment and individualized blood pressure control within 24h for neurological prognosis. Future studies with larger sample sizes are needed to further validate its clinical significance.

Article activity feed