Clinical CVST and selectively detected radiological venous sinus thrombosis after vestibular schwannoma resection: a single-centre retrospective cohort

Introduction Post-operative cerebral venous sinus thrombosis (CVST) is increasingly recognised after lateral skull-base surgery. When centres use routine early venography, asymptomatic thrombi are commonly detected; clinically symptomatic events are less frequent. A single-centre experience applying a sinus-sparing strategy (Bill’s Island during TL exposure, avoidance of fixed venous-sinus retraction in all cases) with selective lumbar drainage (LD) reports observed CVST rates with selective venography. Methods Retrospective cohort of 54 consecutive vestibular schwannoma resections (translabyrinthine [TL] or retrosigmoid [RS]) at a tertiary neurosurgical centre from February 2016 to December 2023. Venography was selective, using CT venography (CTV) for clinical triggers. The primary outcome was clinical CVST, defined as symptomatic CVST with imaging confirmation. Secondary outcomes were radiological CVST detected under selective venography, longitudinal functional outcomes measured by the Modified Rankin Scale (mRS) and Barthel Index (BI), cerebrospinal-fluid (CSF) complications, lumbar-drain use, and other postoperative events. Functional outcomes were assessed at 3, 6, 12 and 24 months. Results Clinical CVST was 0/54 (one-sided 95% upper confidence limit 5.40%). Post-operative CTV was performed selectively in 15/54 patients (27.8%) at a median of postoperative day 2. Radiological CVST occurred in 1/15 (6.7%) among those imaged; this corresponded to 1 event in the full cohort (1/54, 1.9%; 95% CI 0.047%–9.89%), although this should not be interpreted as the true radiological incidence because venography was selective rather than protocolised. The single radiological CVST arose adjacent to a minor intra-operative sigmoid-sinus wall injury, was asymptomatic, and occurred in a case without prophylactic LD; anticoagulation was not used. Patients who did not undergo venography did not develop symptoms prompting imaging or meet criteria for clinical CVST during the recorded postoperative course. Median mRS improved from 2 (3–6 months) to 1 (12–24 months); BI median was 100 at all time-points. Conclusion In this selective-venography series using a sinus-sparing, retraction-avoiding operative strategy, no clinical CVST and a low rate of radiological thrombosis were observed. Because imaging was selective, the true burden of silent radiological thrombosis cannot be determined from this study. Prospective multicentre studies with protocolised venography and harmonised endpoints are warranted.