This was a retrospective single-center study. A total of 1494 patients with ruptured intracranial aneurysms who underwent endovascular interventional embolization between January 2010 and December 2019 at the department of neurosurgery in our hospital were included in the study. The criteria for inclusion were as follows: SAH confirmed by head computed tomography or magnetic resonance imaging and if necessary by lumbar puncture, ruptured intracranial aneurysms confirmed by computed tomography angiography (CTA) or digital subtraction angiography (DSA), receipt of endovascular interventional embolization. The patients excluded from this study were those with cerebral arteriovenous malformations, intracranial vascular abnormalities and coagulation dysfunction. According to whether IPR had occurred, patients enrolled in this study were categorized into IPR and non-IPR groups.

All endovascular treatments and evaluations were performed by one of two proficient endovascular neurosurgeons each with more than 10 years of experience. One neurosurgeons performed 735 cases and the other performed 759 cases. There was no significant individual difference between them. All patients underwent the endovascular procedures under general anesthesia within 5 days after the onset of SAH. After the right or left femoral artery was successfully punctured and cathetered by Seldinger's technique, 5,000 U heparin was intravenously administered, and followed by intravenous infusion of 1,000 U per hour to maintain the activated clotting time at twice the normal level throughout the process. Angiography was performed to determine the best working angle, and the most appropriate microcatheter and microwire were selected for each patient according to the characteristics of aneurysms. After the microcatheter was placed in the appropriate position, the micro-coils were used to tamponade aneurysms. The stents and balloons were selected on a case-by-case basis.

For patients with cerebral vasospasm after SAH, nimodipine was infused intra-arterially before coiling and continued throughout the procedure at a rate of 1–2 mg per hour. The dosage of nimodipine was adjusted according to blood pressure to ensure that the decrease of systolic blood pressure did not exceed 15 mmHg. Acute treatments of vasospasm also included nitroglycerin and/or verapamil. Once an IPR occurs, delivery of coils was continued, and the packing was completed as rapidly as possible. Heparin was stopped immediately and reversed with protamine. Craniotomy and/or external ventricular drainage were performed if necessary. For patients with poor clinical condition, mild therapeutic hypothermia was adopted if needed.⁹

The following information was collected for all enrolled patients: 1) baseline characteristics, including age, gender, past medical history, as well as clinical condition before and after coil embolization; 2) characteristics of aneurysms: location, multiple aneurysms, aneurysm size, aneurysm shape, wide neck; 3) treatment related factors: time from symptom onset to intervention, treatment modality, cerebral vasospasm during aneurysm embolization.

Clinical condition before coil embolization was assessed according to preoperative Hunt and Hess Grade, which was dichotomized into good clinical condition (Hunt and Hess Grade I–III) and poor clinical condition (Hunt and Hess Grade IV–V). Clinical condition after coil embolization was assessed according to Glasgow Outcome Scale (GOS) at discharge, which was dichotomized into good outcome (GOS 4 or 5) and poor outcome (GOS 1–3). IPR was highly suspected based on clinical symptoms, such as a sudden increase in blood pressure and/or a sudden decrease in heart rate, and even pupil dilatation. It was diagnosed when extravasation of contrast material, microguidewire or microcatheter perforation, and the protruding of coils from the confines of the aneurysm wall were visualized on DSA. Aneurysm size was measured by the senior neurointerventional surgeons on the basis of precise measurements made on three-dimensional angiography, defined as the largest diameter from the neck to the dome of the cavity at the optimal projectile angle. Wide-neck aneurysm was defined as an absolute neck width ≥ 4 mm or a dome to neck ratio ≤ 2. Irregular morphology of aneurysm was defined as the presence of small bleb(s) or daughter or secondary aneurysm(s) protruding from the aneurysm fundus, or other irregular shapes, such as bilobed or multilobed structure.¹⁰ Cerebral vasospasm after SAH was defined using both clinical and radiological criteria.¹¹ Objective imaging methods, such as transcranial doppler, CTA and DSA, were used to diagnose intracranial atherosclerosis. Intracranial atherosclerosis was diagnosed as varying degrees of luminal stenotic lesions involving intracranial arteries, (including anterior cerebral artery, middle cerebral artery, posterior cerebral artery, intracranial vertebral artery, basilar artery and intracranial internal carotid artery [ICA]), and accompanied by dyslipidemia, especially elevated low-density lipoprotein cholesterol (> 3.64 mmol/L). In addition, elevated high sensitivity C-reactive protein, homocysteine and lipoprotein (a) could be used as an auxiliary index for the diagnosis of intracranial atherosclerosis. Intracranial stenosis was defined as a localized narrowing of the lumen by 50% or more that can affect the parent arteries.¹² ICA atherosclerosis was defined as present when the ICA stenosis was more than 30%, and categorized into 3 grades according to the degree of stenosis: mild (< 50%), moderate (50%–70%), and severe (> 70%).

All statistical analyses were performed using SPSS 20.0 for Windows (SPSS Inc, Chicago, IL, USA). Continuous variables were expressed as the mean ± standard deviation, and categorical variables were expressed as frequencies (%). The differences between two groups were tested with t-test for continuous variables, and χ² test for categorical variables. Fisher's exact test was performed if sample size was too small. Univariate analysis and multivariate logistic regression analysis was performed to identify the risk factors for IPR in patients with ruptured aneurysms. The receiver operating characteristic (ROC) analysis was performed to determine the cutoff values (with the highest sum of specificity and sensitivity) of continuous variables for IPR during coil embolization. A value of P < 0.05 was considered to indicate statistical significance.

This study was conducted in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The research protocol was approved by the Medical Ethics Board of the Renmin Hospital of Wuhan University (approval No. WDRM201912-036). Informed consent was obtained from all patients enrolled in this study or their next of kin.

A total of 1,494 patients with ruptured intracranial aneurysms were recruited in this study: 658 men and 836 women, with an average age of 51.7 ± 12.0 years (ranging from 23 to 77 years), and the aneurysm size ranged from 1.2 to 38.3 mm, with an average size of 7.4 ± 4.8 mm. The patients’ descriptive statistics were listed in Table 1. There was no significant difference in age, gender, past medical history, clinical condition before coil embolization, characteristics of aneurysms except for the size and shape of intracranial aneurysms, and the treatment strategies between the two groups. Compared to non-IPR group, the IPR group scored significantly higher on the rate of poor outcome, cerebral atherosclerosis, aneurysms with irregular morphology and vasospasm during embolization. The mean size of aneurysms in IPR group was significantly smaller than that in non-IPR group. The time from symptom onset to intervention was significantly shorter in non-IPR group.

There were 41 cases suffered from IPR, showing that the incidence rate was 2.7%. The aneurysms in patients with IPR were mainly located in the distal segments of the main branches of the ICA. Totally, 15 patients of IPR (36.6%) presented with cerebral vasospasm during embolization. Among them, 12 cases presented with cerebral vasospasm caused by SAH before endovascular treatment, and 3 patients suffered from cerebral vasospasm caused by stimulation of arterial wall induced by microguidewire and/or microcatheter in the process of conveying. The clinical features of patients with IPR were summarized in Table 2.

The timing of IPR was as follows: 29 occurred during coil placement, 7 during super selection, and 5 during the other processes, such as angiography. Of the 36 patients with IPR during coiling or microcatheter positioning, there were 7 cases of berry aneurysm (19.4%), and 23 cases of small aneurysm (63.9%). Among the 19 aneurysms with IPR induced by coils, 8 ruptures occurred in the framing stage, 5 occurred in the filling stage, and 6 occurred in the finishing stage. IPR occurred in the dome of aneurysms in 8 patients, of which 5 happened in the framing stage and 3 in the filling stage; IPR happened in aneurysm wall in 2 cases, 1 of which occurred during the framing stage and 1 during the filling stage; IPR occurred in the neck of aneurysms in 4 cases, all of which happened in the finishing stage. The other 5 patients lacked detailed information regarding the site of IPR. After immediate salvage treatment, 32 patients survived without neurological impairment, 5 patients died of extensive SAH, and 4 patients suffered from different neurological deficits: 1 patient suffer from persistent vegetative state, 1 presented with global aphasia, 1 with numbness and weakness in the left arm and 1 with a severe right paralysis and slurring of speech. The mortality and disability rates of IPR were 12.2% and 9.8%, respectively.

Five factors identified by univariate analysis were further analyzed by multivariate logistic regression model and 4 risk factors for IPR were identified, including aneurysm size, aneurysms with irregular morphology, time from symptom onset to intervention and vasospasm during embolization (Table 3). The ROC curve analysis indicated that the area under the curve (AUC) for aneurysm size was 0.697 (95% confidence interval [CI], 0.616–0.778; P < 0.001). By applying the cutoff value of 3.4 mm, the sensitivity and specificity of the aneurysm size in predicting IPR in patients with ruptured aneurysms were obtained as 78.8% and 53.7%, respectively (Fig. 1A). The AUC for the time from symptom onset to intervention was 0.659 (95% CI, 0.577–0.741; P = 0.001); the cutoff value of the time from symptom onset to intervention was 2 days, with a sensitivity of 78.0% and a specificity of 45.2% (Fig. 1B).