Abstrak
Latar Belakang: Perdarahan intrakranial simtomatik merupakan salah satu komplikasi yang fatal pada pasien stroke iskemik akut yang mendapat terapi trombolisis. Penelitian ini bertujuan untuk membuat sistem skoring prediksi perdarahan intrakranial pasca trombolisis. Metode: Penelitian ini dilakukan dengan desain cohort retrospective dari registri stroke trombolisis di Rumah Sakt Pusat Otak Nasional periode Januari 2018 hingga Desember 2022. Variabel luaran adalah perdarahan intrakranial simtomatik berdasarkan kriteria European Cooperative Acute Stroke Study III (ECASS III). Analisis regresi Cox dengan constant time dilakukan untuk mendapatkan risiko relatif (RR) pada determinan terpilih dan dilakukan konversi dari koefesien beta menjadi sistem poin. Hasil : Dari 655 pasien, terdapat 26 (3,97%) mengalami perdarahan intrakranial simtomatik. Model skoring dibuat berdasarkan rentang skor 0-4. Prediktor pada perdarahan intrakranial simtomatik adalah: Glucose >200 mg/dL (11 poin), Early infarct sign (10 poin), NIHSS>10 (12 poin), atrium fibrilasi (13 poin) menunjukkan nilai diskriminasi AUC ROC 0,745 (95%CI 0,656-0,841) dengan nilai p Hosmer–Lemeshow (HL) sebesar 0,6483. Kesimpulan : Model skoring baru memberikan nilai prediksi dalam meningkatkan diagnostik risiko perdarahan intrakranial simtomatik pasca trombolisis intravena.
Background: Symptomatic intracranial hemorrhage is a fatal complication in acute ischemic stroke patients receiving thrombolysis therapy. This study aims to create a risk scoring system for post-thrombolysis intracranial hemorraghe. Methods: This study was conducted with a retrospective cohort design from the thrombolysis stroke registry at the National Brain Center Hospital for the period January 2018 to December 2022. The outcome variable was symptomatic intracranial hemorrhage based on the European Cooperative Acute Stroke Study III (ECASS III) criteria. Cox regression analysis with constant time was performed to obtain the relative risk (RR) of the selected determinants and the beta coefficient was converted to a point system. Results: Of the 655 patients, 26 (3.97%) had symptomatic intracranial hemorrhage. The scoring model is made based on the score range 0-4. Predictors of symptomatic intracranial hemorrhage were: Glucose >200 mg/dL (11 point), Early infarct sign (10 point), NIHSS>10 (12 point), atrium fibrilasi (13 point) score showed an AUC ROC discrimination value of 0,751 (95%CI 0,656-0,841) with a Hosmer–Lemeshow (HL) p value of 0,6483. Conclusion: The new scoring model provides a predictive value for the risk of symptomatic intracranial hemorrgahe after intravenous thrombolysis.