Presenting cerebrovascular reactivity as a determinant of direct and indirect surgical revascularization success in North American patients with moyamoya vasculopathy.

Abstract

Objective: Both direct and indirect surgical revascularization techniques are commonly applied for the treatment of moyamoya disease and syndrome; however, responses can be heterogeneous and efficacy in the context of ischemic disease is not yet formally known from randomized clinical trials. Here, a prospective, longitudinal interventional study was performed to test the hypothesis that presenting 1) parenchymal cerebrovascular reactivity (CVR) and 2) CVR response times portend hemodynamic improvements after direct and indirect revascularization.

Methods: Catheter angiography and hypercapnic blood oxygenation-weighted 3-T MRI (spatial resolution 3.5 × 3.5 × 3.5 mm, repetition time 2000 msec) were acquired before and 11.0 ± 7.9 months and 12.6 ± 6.9 months after surgery, respectively. In response to a 5% fixed-inspired CO2 respiratory challenge, time regression analyses were utilized to quantify maximal cerebrovascular reactivity (CVRmax) and time to reach maximal cerebrovascular reactivity (CVRdelay) to test the overarching hypothesis that presurgical measures predicted postsurgical CVRmax increases and CVRdelay reductions. Age, sex, surgical type, and preoperative impairment were considered as relevant explanatory variables in the regression analysis (significance criterion p < 0.05).

Results: A total of 47 operative hemispheres (32 indirect-only and 15 direct or combined direct-indirect revascularization) from 30 adult patients (median [range] age 43 [20-59] years) were evaluated. Direct/combined versus indirect revascularized brain hemispheres were matched for age (44.1 ± 11.1 vs 44.7 ± 13.8 years, p = 0.864), prior infarct (92.9% vs 92.6%, p = 0.976), and Suzuki stage within 1 stage on the 6-point staging scale (4.1 ± 0.7 vs 3.4 ± 0.6). Across all hemispheres and surgical procedures, CVRmax increased (p = 0.022) and CVRdelay decreased (p = 0.009) after surgery; however, responses varied considerably across hemispheres and surgical procedures. On multiple regression analysis, extent of preoperative impairment, quantified as preoperative CVRmax and moderated by the type of surgery performed, was an indicator of intervention-induced outcome in hemodynamics (p = 0.015). No effect of preoperative CVRdelay or age was found for outcomes.

Conclusions: The findings confirm heterogeneous CVR responses approximately 1 year after revascularization across patients, albeit moderated by type of revascularization. Of the variables considered, lower presurgical CVR provided the most significant indicator of the likelihood of postsurgical hemodynamic improvement.