Article Text
PDF
Abstract
Background and purpose The risk factors for intraprocedural rupture (IPR) of unruptured intracranial aneurysms (UIAs) and the outcomes of IPR itself are unclear. This study was performed to identify the independent risk factors for and outcomes of IPR.
Materials and methods We retrospectively evaluated the medical records and radiologic data of 1375 patients (1406 UIAs) who underwent coil embolization from January 2001 to October 2016.
Results IPR occurred in 20 aneurysms of 20 patients (1.4%). Univariate analyses showed that the rate of IPR was significantly higher in the treatment of aneurysms with a small dome size, aneurysms in the anterior communicating artery (AcomA) (6.6%), and patients with a medical history of dyslipidemia. Multivariate analyses showed that a small dome size and aneurysms in the AcomA were independently associated with IPR (p=0.0096 and p=0.0001, respectively). IPR induced by a microcatheter was associated with a higher risk of severe subarachnoid hemorrhage than other causes of IPR (57% vs 0%, respectively). Thromboembolic complications occurred in seven (35%) patients with IPR. Six (30%) patients required external ventricular drainage placement after developing symptoms of acute hydrocephalus. The overall morbidity and mortality rates from IPR were 0.22% and 0.15%, respectively.
Conclusions Aneurysms in the AcomA and with a small dome size are likely to be risk factors for IPR. IPR induced by microcatheters can result in poor outcomes. The rate of IPR-associated thromboembolic complications is high, and IPR itself is associated with acute hydrocephalus. If managed appropriately, however, most patients with IPR can survive without neurological deterioration.
- Aneurysm
- Coil
- Complication
- Hemorrhage
- Technique
Statistics from Altmetric.com
Introduction
Endovascular treatment of unruptured intracranial aneurysms (UIAs) has become widespread with increasing clinical experience and technologic improvements. However, endovascular treatment of UIAs still has inherent risks of morbidity and mortality.1 Intraprocedural rupture (IPR) is one of the most feared complications of endovascular treatment.2–4
The predictors, technical considerations, management, and outcomes of IPR have been sporadically reported, but such data are limited due to the small size of most case series.4–7 Additionally, these case series included ruptured aneurysms.3 4 8 The treatment outcome in patients with ruptured aneurysms depends on the preoperative symptoms and therefore does not reflect the outcome of IPR itself. Additionally, the treatment strategies and management protocols involving antiplatelet medications differ between ruptured and unruptured aneurysms. We excluded patients with ruptured aneurysms from the present study to accurately assess the outcomes of IPR itself and risk factors for IPR in patients with UIAs. The purpose of this study was to identify the independent risk factors for IPR in patients with UIAs and the outcomes of IPR itself.
Materials and methods
Study population
From January 2001 to October 2016, 1603 patients with 1639 UIAs underwent endovascular treatment in our hospital. Patients with dissecting or fusiform aneurysms and aneurysms treated with parent artery sacrifice or only stenting, such as that performed with a flow diverter, were excluded. Finally, 1406 UIAs in 1375 patients were evaluated in this study. The indications for treating aneurysms with a small dome were the presence of a family history, enlargement of the aneurysm size, irregularities of the aneurysm profile, a medical history of subarachnoid hemorrhage (SAH), and patient preference.
We retrospectively reviewed the medical records and radiologic data for these patients. The following data were collected: age, sex, medical history, aneurysm location and size, characteristics and management of IPR, requirement for temporary external ventricular drain (EVD) placement, other complications, and modified Rankin scale (mRS) score on admission and at last follow-up. The medical histories were obtained by detailed inquiry.
The permanent morbidity and mortality rates associated with the treatment were evaluated at 1 month. Morbidity was defined as an mRS score of 2–5. When the perioperative mRS score was >1, morbidity was defined as any increase in the mRS score.
Endovascular procedures
All patients were pretreated with 100 mg aspirin and 75 mg clopidogrel daily for >5 days. Most patients underwent the endovascular procedures under local anesthesia (n=1317, 96%); aged or restless patients underwent general anesthesia (n=58, 4%). After femoral puncture, a bolus of 4000–5000 U heparin was administered intravenously, followed by intermittent intravenous infusion of 1000–2000 U to maintain the activated clotting time at twice the patient’s baseline throughout the procedure. Wide-necked aneurysms were coiled with the aid of either a remodeling balloon or a stent. We defined a balloon-assisted technique as that in which the balloon was inflated. We always prepared a balloon catheter for IPR even when the balloon-assisted technique was not used.
Diagnosis of IPR
IPR was suspected on the basis of clinical symptoms and diagnosed when extravasation of contrast material was visualized on digital subtraction angiography.
Statistical analysis
The χ2 test and Fisher’s exact test were performed as appropriate. Univariate analyses were performed to identify potential variables associated with IPR. Clinical variables with a p value <0.05 in the univariate analysis were entered into a multivariate logistic regression model. The analysis of dyslipidemia was performed slightly differently; the number of patients with dyslipidemia was counted in the univariate analysis, while the number of aneurysms with dyslipidemia was counted in the multivariate analysis. However, the difference was ignored because it was slight compared with the overall population. A p value <0.05 was considered to indicate statistical significance. All statistical analyses were performed with JMP software, version 10.0 (SAS Institute, Cary, North Carolina, USA).
Results
Demographics
In total, 1406 aneurysms in 1375 patients were analyzed, 20 (1.4%) of which developed IPR. The characteristics of the patients are summarized in table 1. The median patient age was 60 years and 78% were female. The numbers of patients with a medical history of hypertension, diabetes mellitus, and dyslipidemia were 668 (49%), 74 (5%), and 360 (26%), respectively. A history of smoking and a family history of UIA were confirmed in 337 (25%) and 190 (14%) patients, respectively. A total of 412 (30%) patients had multiple aneurysms. Univariate analyses indicated that a medical history of dyslipidemia was significantly associated with IPR (p=0.036).
Characteristics of patients
Details about the aneurysm size, location, and treatment are summarized in table 2. Fifty-eight (4%) patients underwent general anesthesia. In total, 340 (24%), 468 (33%), and 598 (43%) patients were treated with a coil only, stent-assisted technique, and balloon-assisted technique, respectively. The median size of the aneurysmal dome was 6.0 mm (IQR 4.7–8.0). Univariate analyses showed that aneurysms with a small dome size (p=0.0028) and those located in the anterior communicating artery (AcomA) (p=0.0019) were associated with IPR. Multivariate analyses showed that aneurysms with a small dome size (p=0.0096) and those located in the AcomA (p=0.0001) were independently associated with IPR (table 3). The median size of the aneurysms located in the AcomA was 6.0 mm (IQR 4.8–7.7). Among them, the aneurysmal size was significantly smaller in patients with IPR than in those without IPR (4.5 vs 6.8 mm, respectively; p=0.0016). Fifteen aneurysms were <3 mm; however, IPR did not occur among them. A receiver operating characteristic curve analysis for the aneurysmal dome size was performed to determine the cut-off value for the occurrence of IPR. The result revealed that the aneurysm size threshold separating the IPR and no-IPR groups was 6.98 mm with a sensitivity of 100% and specificity of 36% (area under curve=0.70).
Aneurysm size, location, and treatment
Multivariate analysis of risk factors for intraprocedural rupture
Timing and characteristics of IPR
The characteristics, management, and outcomes of IPR are summarized in table 4. IPR occurred during coil placement in 14 aneurysms and during aneurysm access in four aneurysms. Additionally, IPR was induced by coils in 10 aneurysms, a microguidewire in one aneurysm, and microcatheter navigation in seven aneurysms. IPR was induced by different factors in another two aneurysms (stent deployment in one case and movement of the embolized coils from the intended aneurysm to the middle cerebral artery with subsequent aneurysm rupture in the other case). Among the 10 aneurysms with IPR induced by coils, three ruptures occurred in the framing stage, two occurred in the filling stage, and five occurred in the finishing stage such as near the occlusion. Three of the four aneurysms with IPR that occurred during aneurysm access occurred during treatment of an AcomA aneurysm.
Characteristics, management, and outcomes of IPR (n=20)
Immediately after rupture induced by coils, microguidewires, and microcatheters, headache occurred in seven (70%), one (100%), and three (43%) patients, respectively, and disturbance of consciousness occurred in 0 (0%), 0 (0%), and four (57%) patients, respectively. Thus, severe SAH occurred only in the aneurysms perforated by a microcatheter. Consequently, poor outcomes were observed in one (10%) aneurysm perforated by coils and three (43%) aneurysms perforated by microcatheters. The detailed characteristics of IPR are shown in table 5.
Outcomes of perforation of different causes following endovascular coil embolization
Management of IPR
To prevent continued aneurysmal rupture, protamine sulfate was given for heparin reversal in 18 cases; protamine sulfate was not given in two cases because they were side wall type aneurysms treated with a balloon-assisted coil embolization technique and bleeding was immediately controlled by inflating the balloon. Additional coils were deployed for dome protection in 13 cases and parent artery occlusion was performed in one case; additional coils could not be deployed in six cases. In four of these six cases, the rupture occurred in the finishing stage and additional coils could not be deployed because of tight packing. In the other two cases, the microcatheter could not be advanced into the aneurysm. Clipping was performed in these six cases to control the bleeding. In three of the six cases, rescue clipping was urgently performed because the bleeding could not be controlled by endovascular management. In the other three cases, safety clipping was performed because the rupture point was either unclear or near the aneurysmal neck and additional coils could not be deployed.
Complications and outcomes
Six (30%) patients required EVD placement after symptoms of acute hydrocephalus (table 3). Thromboembolic complications occurred in seven (35%) cases; five (71%) complications resulted in transient neurological deficits and two (29%) resulted in permanent disability. Fifteen (75%) cases had an mRS score of 0 at the immediate and last follow-up examinations. One case involving coil perforation had an mRS score of 3 due to a thromboembolic complication, and three cases involving microcatheter perforation had mRS scores of 2, 2, and 6, respectively. Another case involving perforation by distal coil migration had an mRS score of 6. Consequently, the morbidity and mortality rates were 0.22% and 0.15%, respectively.
Discussion
In this study we investigated the outcomes of IPR itself and the risk factors for IPR in patients with UIA with adjustment for the patients’ demographic characteristics and the clinical characteristics of the aneurysm and IPR. Patients with ruptured aneurysms were excluded to accurately determine the outcomes of IPR itself and the risk factors for IPR in patients with UIA. The results showed that aneurysms with a small dome size and those located in the AcomA were independent predictors of IPR in patients with UIA. In contrast, the treatment technique was not associated with IPR. IPR induced by a microcatheter resulted in severe SAH and poor outcomes, and the rates of acute hydrocephalus and thromboembolic complications were high after IPR.
Several studies have been performed to investigate IPR; however, most of them involved patients with ruptured aneurysms, performed general anesthesia, and were designed as small case series.5 6 8 9 The first advantage of this study is that our series is the largest size to date. The second advantage is that almost all endovascular treatments were performed under local anesthesia; therefore, no cases of IPR were missed due to masking of symptom manifestation. Treatment in previous studies was performed under general anesthesia.5 6 8 In such studies, patients with an endovascular device outside the aneurysms but no hemodynamic response or contrast in the subarachnoid space were analyzed as having no IPR event without postoperative image evaluation. In previous studies, the rate of IPR in patients with UIA ranged from 0.9% to 2.4%.1 4–6 8 We observed a comparable rate of 1.4% in the present study. The rates of IPR and clinical outcomes from major series involving treatment of UIA are summarized in table 6.
Review of IPR in unruptured aneurysms in recent studies
Predictors of IPR
In a previous prospective multicenter study of 700 patients with UIA, the univariate analysis showed that the rate of IPR was significantly different according to the size of the aneurysms (1–6 mm: 3.7%, 7–15 mm: 0.7%).7 Oishi et al 5 reported a trend toward a higher incidence of rupture of aneurysms located in the AcomA. In the present study, the multivariate analysis showed that aneurysms with a small dome size (p=0.0022) and those located in the AcomA (p=0.0002) were independent predictors of IPR. To our knowledge, this is the first report to indicate that localization in the AcomA and a small dome size are independent predictors of IPR in patients with UIA. This may reflect the difficult anatomic characteristics of endovascular coil embolization in the AcomA, such as an unfavorable dome–neck ratio and acute internal carotid artery–anterior cerebral artery angle.10 The UCAS Japan investigators reported that UIAs in the AcomA, including small UIAs, had a high risk of rupture,11 which may also be associated with the higher rate of IPR of UIAs located in the AcomA.
Some studies have shown that the balloon-assisted technique is not associated with an increased incidence of IPR,4 while others have demonstrated an increase in IPR compared with conventional coil embolization.12 In our series there was no relationship between the balloon-assisted technique and IPR.
In a retrospective study of 189 patients with SAH, the rate of IPR was significantly higher under local anesthesia than general anesthesia.13 This was a result of unexpected patient motion and occurred in accordance with the unexpected movement. In a previous study of IPR in patients with UIA, the patients were treated under general anesthesia.5 In the present study there was no relationship between IPR and either local or general anesthesia. Furthermore, no cases of IPR were associated with unexpected patient motion, and all cases of IPR could be quickly identified and managed because of the patients’ clinical manifestations. If the patient’s clinical condition allows for selection of either local or general anesthesia, local anesthesia may be useful, especially when IPR occurs.
Causes of IPR
Immediately after IPR, disturbance of consciousness occurred in five (25%) cases; four (80%) were induced by microcatheters. McDougall et al 14 reported four IPRs during embolization of 200 aneurysms, and the only death was the result of perforation of the aneurysm by a microcatheter. Santillan et al 9 reported that two of 12 IPRs were induced by microcatheter perforation. Of these two patients, one patient died and the other required parent artery occlusion. Hemorrhage is difficult to control if the rupture is caused by microcatheter perforation of the dome because of the increased size of the defect.14 In the present study, clinical outcome depended on the cause of the IPR. The rates of good clinical outcomes of IPR induced by a coil, microguidewire, and microcatheter were 90%, 100%, and 57%, respectively.
Complications and outcomes
In the ATENA study, the IPR-associated morbidity and mortality rates were 28% and 17%, respectively.7 In the present study, the IPR-associated morbidity and mortality rates were 15% and 10%, respectively, and comparable with those in previous studies.
The overall morbidity and mortality rates associated with thromboembolic events in this series were 0.7% (10/1375) and 0.0% (0/1375), respectively. Of the 20 patients with IPR, two (10%) had permanent disability caused by thromboembolic complications. Thus, the rate of IPR-associated thromboembolic complications is high and likely to be associated with heparin reversal and inflation of the balloon catheter.
Stapleton et al 3 reported that aneurysmal re-rupture during coil embolization was associated with the need for temporary EVD placement in cases of aneurysmal SAH. In one large series of aneurysmal SAH, CT revealed acute hydrocephalus in 15% of patients, 40% of whom were symptomatic.15 In the present study, six patients required EVD placement after symptoms of acute hydrocephalus. Without operation cases, six of 14 (43%) patients required EVD placement, and this rate was much higher than in patients with aneurysmal SAH. This result suggests that IPR itself is a risk factor for acute hydrocephalus regardless of the severity of SAH. No studies have been performed to evaluate the effect of radiographic dyes within the subarachnoid space and the mechanism of cerebrospinal fluid absorption. However, contrast material is more viscous than cerebrospinal fluid and blood, and previous researchers have found that its presence in high concentrations in the subarachnoid space may reduce cerebrospinal fluid resorption through the arachnoid villi, leading to hydrocephalus.3 Our results supported these findings. After IPR we should pay attention to the potential development of acute hydrocephalus.
Limitations of study
There are several limitations in this study. First, this was a single-center retrospective study. Compared with the total number of patients who underwent endovascular treatments, the number of patients with IPR in our case series is quite small. Therefore, although the results of several evaluations in our study were statistically significant, such observations should be validated in larger populations and across multiple centers. In our hospital, middle cerebral artery aneurysms are mainly treated by surgical clipping; therefore, we could not evaluate the risk of IPR during endovascular treatment of middle cerebral artery aneurysms in the present study. Finally, no aneurysm morphological parameters such as irregularity were included.
Conclusions
Aneurysms in the AcomA and aneurysms with a small dome are independent risk factors for IPR. Microcatheter-induced IPR can result in poor outcomes. The rate of IPR-associated thromboembolic complications is high, and IPR itself is associated with acute hydrocephalus regardless of the severity of SAH. If managed appropriately, however, most patients with IPR can survive without complications. Further studies of the risk factors for and outcomes of IPR are required to improve our understanding and reduce morbidity and mortality rates.
References
Footnotes
Contributors Acquisition of data: all authors. Analysis and interpretation of data: SK and HI. Drafting of the article: SK and HI. Critical revision of the article: all authors.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent This study was retrospective and approved by ethics committee.
Ethics approval Ethics committee of Kobe City Medical Center General Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement All available data can be obtained by contacting the corresponding author.