MAD-score: a Prognostic Model for Malignant Middle Cerebral Artery Infarction Patients Treated with Decompressive Craniectomy

Malignant middle cerebral artery infarction (MMI) is a severe condition caused by space-occupying brain oedema due to ischaemia. While decompressive hemicraniectomy (DC) improves outcome, many survivors experience lasting disabilities, and tools to predict outcome are lacking. We aimed to identify independent predictors of outcome following DC in MMI and to create a pragmatic, bedside tool for patient prognostication. This retrospective, multicentre study analysed 318 MMI-patients treated with DC at four Swedish university hospitals (2008–2022). Demographic, preoperative, and postoperative variables were collected from medical records. Six-month modified Rankin Scale (mRS) was the primary endpoint, with favourable outcome defined as mRS 0–3. Uni- and multivariable ordinal logistic regressions were employed to identify independent predictors of mRS. Significant independent preoperative variables were included in the MMI-DC (MAD) score. Median age was 56 (IQR: 48–61) years; 75% were male. At 6 months, median mRS was 4 (4–5), 24% achieved favourable outcome, and 17% were deceased. MAD-score assigned 1 point to each independent preoperative predictor of worse outcome: higher age (≥60), female sex, lower GCS M (<4), and impaired pupillary reactivity (all p≤0.03). MAD-stratified outcome, ordered from 0 to 4 points, was 32%, 25%, 15%, 0%, and 0% (favourable outcome) and 9%, 16%, 23%, 40%, and 100% (mortality). A large proportion of MMI-patients recover unfavourably following DC. The presented MAD-score was an internally reliable prognostic tool for MMI-patients treated with DC. External validity of MAD and predictive value of female sex should be further explored.