| Background: Intracranial aneurysm is the leading cause of spontaneous subarachnoid hemorrhage,with potentially fatal and disabling consequences.Endovascular treatment has become the main treatment method for intracranial aneurysms.The outcomes and prognostic factors of patients undergoing surgery of intracranial aneurysms have been paid great attention by clinicians,patients and their families.Previous studies of prognostic factors focused on the aneurysm itself and patient factors,while the perioperative factors and anesthesia related factors are limited.Therefore,it is urgent to carry out research on the prognostic factors including anesthesia related factors on the outcome of these patients and to find out relevant factors to provide timely intervention,so as to improve the prognosis of patients.Objective: To investigate the prognostic factors of patients with intracranial aneurysms undergoing endovascular treatment,and to develop prediction models,early detection of high-risk patients,improvement of patient’s outcomes and the quality of life.Methods:1.We collected information of intracranial aneurysms cases hospitalized in the Department of Neurosurgery at Xijing Hospital from January 2011 to June 2020.The patients were 18 ～ 80 years old,diagnosed as intracranial aneurysms by DSA and underwent endovascular treatment were included.Patients with surgery for recurrent aneurysm,presence of severe cardiovascular disease(coronary disease,heart failure),malignancy,or disability or death within one year of surgery due to other disease or accident,or with significant missing data or lost to follow-up were excluded.The data were obtained by digital extraction,manual data entry and telephone follow-up.2.The prognosis of patients was evaluated according to the modified Rankin scale score at different time after surgery(30 days and 1 year after surgery).The cases were divided into poor prognosis group and good prognosis group according to the modified Rankin scale.The univariate analysis was applied to compare whether there were any influencing factors between the two groups.Then,the factors with P < 0.2 in univariate analysis were included in multivariate logistic regression analysis to obtain the independent influencing factors on the outcome of 30 days and 1 year after operation.3.The data of cases between January 2011 and June 2020 were charged as the training set,and cases between July 2020 and December 2021 were charged as the validation set.Variables with P < 0.05 in univariate analysis in the second step were included in the logistic model in R language to screen out the modeling variables,and the prediction models of prognosis in the 30 days and 1 year after operation were developed,and their nomograms were constructed;The ROC,calibration and DCA curve of the modeling set and validation set,respectively,were plotted in R language to test the models.Results:1.A total of 725 cases were included in this study,112 case had poor outcomes at 30 days after surgery,the incidence of poor outcomes was 15.45%.Of these,13 patients died,the case fatality rate was 1.79%.The univariate analysis showed that there were significant differences between the poor prognosis group and good prognosis group in age,timing of surgery,hypertension,number of aneurysms,preoperative aneurysm rupture,preoperative Hunt-Hess grade,preoperative Fisher grade,ASA status,whether awakening while leaving the operating room,whether extubation when leaving the operating room,duration of surgery,duration of anesthesia,and preoperative D-dimer level(P<0.05).The multivariate logistic regression analysis showed that preoperative Hunt-Hess grade(3 ～ 5)(OR2.601,95%CI: 1.522～4.445),preoperative Fisher grade(3～4)(OR 4.487,95%CI:2.085～9.656),ASA status(Ⅲ～Ⅴ)(OR 1.764,95%CI: 1.034～3.099),preoperative Ddimer level(>4.8mg/L)(OR 4.831,95%CI: 1.682～13.880),not awake at discharge from the operating room(OR 6.809,95%CI: 3.865～11.994)and duration of anesthesia(≥170min)(OR 2.453,95%CI: 1.156 ～ 5.209)play key roles in the prognosis after endovascular treatment 30 days.2.The total incidence of perioperative neurological complications(aneurysm rerupture,intracranial hematoma,thromboembolism,hydrocephalus and epilepsy)was3.03%.3.Ninety-two patients were with poor outcome and 25 patients died at 1 year after surgery.The incidence of poor outcome was 12.69% and case fatality rate was 3.45%.The univariate analysis showed that there were significant differences between the poor prognosis group and good prognosis group in age,timing of surgery,hypertension,diabetes,number of aneurysms,preoperative aneurysm rupture,preoperative Hunt-Hess grade,preoperative Fisher grade,surgical method,ASA status,use of etomidate,whether awakening while leaving the operating room,whether extubation when leaving the operating room,duration of surgery,use of sugammadex,duration of anesthesia,and preoperative D-dimer level(P<0.05).The multivariate logistic regression analysis showed that independent risk factors of poor outcome by 1 year after surgery included preoperative Hunt-Hess grade(3～5)(OR 2.067,95%CI: 1.084～3.943)、preoperative Fisher grade(2～4)(OR 2.866,95%CI: 1.231～6.670)、hypertension(OR 2.225,95%CI:1.134～4.365)、emergency of surgery(OR 2.068,95%CI: 1.024～4.177)、number of aneurysms(≥2)(OR 1.460,95%CI: 1.057 ～ 2.016)、 not awake at discharge from the operating room(OR 8.885,95%CI: 4.794 ～ 16.467)and the etomidate was used(OR0.279,95%CI: 0.097～0.805).4.The prediction model of outcome 30 days after surgery contained variables including preoperative Hunt-Hess grade,preoperative Fisher grade,ASA grade,preoperative D-dimer,failure to wake up at discharge from the operating room and duration of anesthesia.AUC value of the training set was 0.836(95% CI: 0.789～0.882).The mean absolute error of the calibration curve was 0.008.And the DCA curve showed benefit in the range of threshold probabilities 0.10～0.84;The validation set yielded an AUC of 0.846(95%CI: 0.768 ～ 0.924),with a mean absolute error of 0.039 for the calibration curve,and the DCA curve demonstrated benefit across the range of threshold probabilities 0.11～0.80.5.The predictor model at 1 year postoperatively contained variables including comorbid hypertension,preoperative Hunt-Hess grade,preoperative Fisher grade,timing of surgery,number of aneurysms,awakening at discharge from the operating room and use of etomidate.AUC value of the training set was 0.808(95%CI: 0.733～0.883),with a mean absolute error of 0.02 for the calibration curve,and the DCA curve showed benefit in the range of threshold probabilities 0.11～0.91.The validation set yielded an AUC of0.849(95% CI: 0.752～0.946),with a mean absolute error of 0.023 for the calibration curve,and the DCA curve demonstrated benefit in the threshold probability range of0.12～0.90.Conclusion: The rate of poor outcomes in patients with intracranial aneurysms undergoing endovascular treatment was 15.45%.Preoperative Hunt-Hess grade,preoperative Fisher grade,ASA grade,preoperative D-dimer level,not awakening at discharge from the operating room,and length of anesthesia were independent predictors of poor outcome by 30 days after surgery.The total incidence of perioperative neurological complications was 3.03%.The incidence of poor outcome by 1 year postoperatively in patients with intracranial aneurysms undergoing endovascular treatment was 12.69%.Preoperative Hunt-Hess grade,preoperative Fisher grade,hypertension,timing,number of aneurysms,failure to wake up at discharge from the operating room,and use of etomidate were independent predictors of poor outcome by 1 year after surgery.The established prediction model for poor outcomes 30 days and 1 year after surgery in patients with intracranial aneurysms undergoing endovascular treatment had a relatively good discrimination,calibration,and sensitivity.The models can potentially to be used to predict outcomes in intracranial aneurysms patients undergoing endovascular treatment. |
PDF Full Text Request