Prognostic analysis of continuous lumbar cistern external drainage after aneurysmal subarachnoid hemorrhage / 国际外科学杂志

ABSTRACT

Objective:To analyze the influencing factors, prevention and treatment strategy of short-term poor prognosis of continuous lumbar cistern external drainage after aneurysms subarachnoid hemorrhage (aSAH).Methods:Used retrospective research method, the clinical data of 300 patients with aSAH combined with continuous lumbar cistern external drainage treated in Sanya Central Hospital (Hainan Third People′s Hospital) from March 2019 to March 2021 were selected as the training set. In addition, the clinical data of 144 patients with aSAH with continuous lumbar cistern external drainage treated in Sanya Central Hospital (Hainan Third People′s Hospital) from March 2017 to May 2019 were selected as the verification set. According to the results of postoperative follow-up, the patients in the training set were divided into two groups good prognosis group ( n=208) and poor prognosis group ( n=92). The demographic characteristics, past history, Hunt-Hess grade, modified Fisher grade, location of responsible aneurysm, postoperative complications, bone flap decompression and lumbar cistern drainage were compared between the two groups. The independent risk factors for prognosis of aSAH patients undergoing continuous lumbar cistern external drainage were screened by Cox proportional hazard regression model, and these factors were included and XGboost model was established. The prediction model was validated internally and externally in the training set and verification set AUROC(C-index) was used to verify the model differentiation; GiViTI calibration band and Hosmer-Lemeshow test were used to verify the model calibration; DCA curve was used to verify the clinical validity of the model. Results:Hunt-Hess grade, modified Fisher grade, drainage duration, average daily drainage volume, shunt-dependent hydrocephalus, aneurysm rebleeding, cerebral vasospasm and delayed cerebral ischemia were independent risk factors for poor prognosis in patients with aSAH who underwent continuous lumbar cistern external drainage( P<0.05). The XGboost model was successfully established by incorporating the above independent risk factors, and the internal and external verification of the XGboost model was carried out in the training set and verification set, respectively, the area under the curve of receiver operating characteristic was 0.882(95% CI 0.820-0.955) and 0.878(95% CI 0.774-0.928) respectively, and the model differentiation was good; the 80%-90% confidence interval of the GiViTI calibration curve did not cross the 45° angle bisector ( P>0.05). In the Hosmer-Lemeshow goodness-of-fit test, the P value were 0.581 and 0.716, respectively. The threshold probability value in the DCA curve was 30.4%. The clinical net benefit rate of the training set and verification set were 31% and 34%, respectively, indicating that the prediction model was clinically effective. Conclusions:The independent risk factors for poor prognosis of aSAH patients undergoing continuous lumbar cistern drainage are Hunt-Hess grade, modified Fisher grade, cerebral vasospasm, delayed cerebral ischemia and shunt-dependent hydrocephalus. The XGboost model constructed in this study can effectively predict the prognosis of patients with aSAH undergoing continuous lumbar cistern drainage, and provide reference for the formulation of follow-up treatment plans.