Postoperative midline shift and functional outcome after surgery for chronic subdural hematoma
Related Experiment Videos
Joel Abraham Velazquez-Castillo1, Victor Ramzes Chavez-Herrera2, Martin Paredes-Cruz3
1Faculty of Medicine, Universidad Nacional Autónoma de México (UNAM), Mexico City, Mexico; Department of Neurosurgery, Hospital de Especialidades Centro Médico Nacional Siglo XXI, IMSS, Mexico City, Mexico.
Background:
Chronic subdural hematoma (CSDH) lacks well-established postoperative radiological markers associated with long-term functional outcome. We evaluated the association between postoperative midline shift (MLS) and 6-month functional outcome using the modified Rankin Scale (mRS) after CSDH surgery.
Methods:
We retrospectively analyzed prospectively collected data from 172 surgical CSDH patients. The primary prognostic analysis was restricted to unilateral CSDH (n=122) because opposing mass effects alter MLS interpretation in bilateral hematomas. The primary outcome was unfavorable 6-month functional status (mRS 3-6). A parsimonious Firth penalized logistic regression model adjusting for age, admission Glasgow Coma Scale (GCS), and postoperative MLS (modeled continuously first, and as ≥5 vs. <5 mm secondarily) was fitted.
Results:
Among 122 unilateral CSDH patients, 22 (18.0%) had an unfavorable 6-month outcome. Across 106 re-evaluated postoperative CT scans, intraobserver and interobserver reliability was excellent for continuous MLS (ICC: 0.982 and 0.985) and high for the ≥5-mm threshold (κ: 0.902 and 0.862). In the primary Firth model, continuous postoperative MLS showed a positive but imprecisely estimated association with unfavorable outcome (OR 1.16 per 1-mm increase, 95% CI 0.983-1.36; p=0.079), with an apparent c-statistic of 0.704 (95% CI 0.564-0.844) and bootstrap optimism-corrected c-statistic of 0.674. In the secondary model, postoperative MLS ≥5 mm was associated with higher odds of unfavorable outcome (OR 2.73, 95% CI 1.003-7.45; p=0.049).
Conclusions:
Higher postoperative MLS showed a consistent direction of association with unfavorable 6-month outcome in unilateral CSDH, although the continuous estimate was imprecise. These findings support postoperative MLS as a potential prognostic marker rather than a therapeutic target, though its incremental prognostic value beyond pre-morbid functional status and preoperative radiological severity remains unknown. The 5-mm threshold does not justify imaging-driven reintervention.