Abstract
Objective:
The effects of treatment modality (endovascular coiling or surgical clipping) on incidence of cerebral vasospasm and infarction following aneurysmal subarachnoid haemorrhage (aSAH) remain controversial. This study compared outcomes between endovascular coiling versus surgical clipping to treat patients with acute (< 72 h) aSAH.
Methods:
Patients with aSAH were randomized to receive endovascular or surgical treatment. All patients underwent clinical assessments, angiography and brain computed tomography.
Results:
Data from 186 patients were analysed: 94 in the endovascular group and 92 in the surgical group. Demographics and severity of aSAH were comparable between the groups. Incidence of symptomatic vasospasm, cerebral infarction and complete occlusion were significantly lower in the endovascular coiling group than in the surgical clipping group. Good clinical recovery 12 months after aSAH was seen in 75.0% and 69.7% of surviving patients in the endovascular versus coiling group, respectively.
Conclusions:
In this study, coiling yielded less symptomatic vasospasm, cerebral infarction and complete occlusion than surgical clipping, with no between-group differences in clinical outcome.
Keywords
Introduction
Subarachnoid haemorrhage (SAH) caused by the rupture of intracranial aneurysms remains a serious healthcare problem 1 with approximately one half of survivors sustaining irreversible brain damage. 2 Although a few cases of unruptured intracranial aneurysm have been reported to resolve spontaneously, direct treatment is usually recommended. In cases involving ruptured aneurysm, patients should be treated routinely.3,4 Early treatment (< 72 h) is recommended for ruptured aneurysms as ∼20% of patients experience another rupture within the first 2 weeks following SAH. 5
Despite developments in the treatment of SAH, the case fatality rate remains stable, 6 with cerebral vasospasm and infarction being the two major causes of death and disability.3,7 The treatment of aneurysmal SAH (aSAH) continues to be a matter of debate, with one study suggesting that surgical treatment itself could produce neurological morbidity, cognitive impairment and increased mortality. 8 Treating recently ruptured cerebral aneurysms with endovascular coiling avoids the difficulties encountered during early surgical clipping on swollen brain tissue, and might carry lower risk of morbidity and mortality. 9 Few comparative studies of the long-term outcome of surgical versus endovascular treatment in acute aSAH have been reported.
The present study evaluated differences in terms of clinical, angiographical and computed tomography (CT) scan outcomes between treatment with endovascular coiling or surgical clipping, in patients with acute aSAH.
Patients and methods
Study Population
Consecutive patients with acute aSAH, admitted to the Department of Neurosurgery, Fengxian District Central Hospital (Branch Hospital of Shanghai Sixth People's Hospital), Shanghai Jiaotong University, Shanghai, China, between April 2005 and February 2009, were enrolled into the study and assigned (according to a computer-generated randomization schedule) to undergo either endovascular coiling or surgical clipping treatment. Each patient's age, sex and aneurysm site was recorded. The clinical status of the patient was rated using the Hunt and Hess scale. 10 In order to diagnose SAH, CT scans were performed at the following times: on admission; on the day (24 – 48 h) after treatment; on the day of discharge. To evaluate the occurrence of angiographic vasospasm, digital subtraction angiography was performed at a mean ± SD day 9 ± 2 (between day 4 and 14) 11 following aSAH, or earlier when clinical or transcranial doppler ultrasonography data suggested the presence of vasospasm. If angiographic vasospasm was confirmed, hypervolaemic hypertensive haemodilution therapy was initiated.
The study protocol was approved by the Fengxian Central Hospital Ethics Committee (No. FH2005032203), and signed informed consent was obtained from each patient or family before treatment.
Assessment of Symptomatic Vasospasm and Cerebral Infarction
All imaging material was evaluated by two independent blinded reviewers. Symptomatic vasospasm was defined as follows: (i) new neurological deficit occurring between days 4 and 14 following aSAH; (ii) no other identifiable cause for neurological deterioration as demonstrated on CT scan, rebleeding, acute or worsening hydrocephalus, electrolyte disturbances, hypoxia or seizures; (iii) vasospasm on cerebral angiography.
Cerebral infarction was diagnosed on the discharge CT scan as a newly detected focal hypodense region, when compared with the 24 – 48 h postaneurysm treatment CT scan, that was not attributable to intracerebral haemorrhage, ventricular drain placement or another nonvascular aetiology. In this way, lesions related to the initial bleeding or to the intervention could be excluded.
Follow-up
Follow-up imaging was performed by digital subtraction angiography, CT angiography to evaluate the occurrence of angiographic vasospasm or CT for detection of infarction. Following endovascular coil treatment, imaging follow-up was routinely performed at 3 and 12 months. In cases of increasing coil compaction, angiographic follow-up was performed more frequently. Following surgical clipping, imaging follow-up was performed if the neurosurgeon considered it necessary.
Clinical follow-up was performed in both groups during outpatient clinic visits. A structured telephone interview was performed with outpatients or family who missed the clinic visits; a close relative was contacted in cases where the patient was unavailable. Patients with a follow-up period of < 1 year were considered to be lost to follow-up. Clinical outcome status was assessed with use of the modified Rankin Scale (score 0 – 6) 12 at 12 months. Poor outcome was defined as a modified Rankin Scale score of ≥ 3. Earlier follow-up visits were scheduled if changes in the morphological features of the aneurysm were noted.
Statistical Analyses
All statistical procedures were performed with SPSS® software, version 13.0 (SPSS Inc., Chicago, IL, USA) for Windows®. Quantitative data were presented as mean ± SD, and qualitative data as percentages or absolute numbers. Between-group comparisons were performed using Student's t-test for quantitative data and Pearson's χ2-test for qualitative variables. A P-value < 0.05 was considered to be statistically significant.
Results
Of the 192 patients entered in the study, 96 were randomized to receive endovascular treatment and 96 to receive surgical treatment. In both groups, all surgeries were carried out by the same team, which was experienced in performing both surgical procedures. Two patients in the endovascular treatment group and four patients in the surgical treatment group were not treated for their ruptured aneurysm. Thus, baseline characteristics were available for 186 treated patients. There were no statistically significant between-group differences in terms of age, sex, Hunt and Hess scale grades, 10 location of target aneurysm or time interval between aSAH and treatment procedure (Table 1).
Baseline characteristics of patients treated with endovascular coiling or surgical clipping following acute aneurysmal subarachnoid haemorrhage (aSAH)
Data presented as n (%) patients or mean ± SD.
There were no statistically significant between-group differences (P ≥ 0.05); Student's t-test and Pearon's χ2-test. ICA, internal carotid artery; MCA, middle cerebral artery; ACA, anterior cerebral artery; AComA, anterior communicating artery; BA, basilar artery; PCoA, Posterior communicating artery; aSAH, acute aneurysmal subarachnoid haemorrhage.
Symptomatic vasospasm was documented in 22/94 (23.4%) of the endovascular coiling patients and 34/92 (37.0%) of the surgical clipping patients (P < 0.05). Surgical clipping increased the risk of symptomatic vasospasm compared with coiling (odds ratio 1.24, 95% confidence intervals 1.01, 1.51). There were significantly fewer new cerebral infarctions in the endovascular coiling group compared with the surgical clipping group (12/94 [12.8%] versus 20/92 [21.7%], respectively, P < 0.05). Of the patients experiencing new cerebral infarctions, the incidence of symptomatic vasospasm-related infarctions was significantly lower in those undergoing endovascular treatment (eight of 12 patients [66.7%]), compared with surgical treatment (17/20 patients [85.0%]) (P < 0.05).
Imaging follow-up was performed in all patients. In those who had received surgical clipping, imaging showed complete occlusion in 77/92 (83.7%), a residual neck in 11/92 (12.0%), and a residual aneurysm in four of 92 patients (4.3%). In those who had received endovascular coiling, complete occlusion was demonstrated in 61/94 (64.9%), a residual neck in 21/94 (22.3%), and a residual aneurysm in 12/94 patients (12.8%). The differences between the two groups were statistically significant for residual aneurysm only (P < 0.05). Incidence of complete aneurysm occlusion was significantly lower following endovascular coiling compared with surgical treatment (P < 0.05). Rebleeding occurred in three of 94 (3.2%) patients following endovascular coiling treatment, and three of 92 (3.3%) patients following surgical treatment, with no significant between-group difference.
At the end of the 1-year follow-up, 10/94 (10.6%) and 14/92 (15.2%) patients died after endovascular and surgical treatment, respectively, with no significant between-group difference. Clinical follow-up was completed in all 162 surviving patients: 63/84 (75.0%) and 53/78 (67.9%) surviving patients in the endovascular and surgical treatment groups, respectively, achieved a good outcome (modified Rankin Scale score 12 0 – 2); no significant between-group differences were observed.
Discussion
The primary goal in treating ruptured aneurysms is to prevent haemorrhage, thereby decreasing the rate of mortality and dependency. Endovascular coiling has become an effective treatment option, with a good safety profile for ruptured intracranial aneurysms, since Guglielmi et al. 13 revolutionized this technique with the introduction of electrolytically detachable coils. The present study investigated coiling and clipping procedures to treat intracranial aneurysms. After a follow-up period of 12 months, death rates of 10.6% (coiled patients) and 15.2% (clipped patients), were similar to those reported by the International Subarachnoid Aneurysm Trial after 5 years of follow-up. 14 One-year outcome did not differ significantly between the groups, with 67.9 – 75.0% of surviving patients achieving a good recovery (modified Rankin Scale score 0 – 2). These results were in accordance with another study describing good clinical outcomes (modified Rankin Scale score 1 – 2) in 79% and 75% of the surviving patients, (coiled or clipped, respectively) treated within 3 days following SAH. 15 Symptomatic vasospasm, requirement for permanent shunt creation, size of the ruptured aneurysm and Hunt and Hess grade were shown to be independent predictors of clinical outcome. 15 Though medical and surgical advances have reduced the impact of cerebral vasospasm on outcome following SAH, it remains a major cause of morbidity and mortality.16,17 Whether the coiling or clipping procedure increases the risk of symptomatic vasospasm is debatable.3,10,18 One meta-analysis compared the rate of symptomatic vasospasm after coiling or clipping and suggested a trend towards less symptomatic vasospasm after coiling than after clipping. 19 The studies included were, however, limited by their retrospective nature, differences in study design, lack of angiographic diagnosis of vasospasm and varying definitions of symptomatic vasospasm. In addition, patients treated by coiling in this analysis were older than those treated with clipping, with a poor clinical prognosis and an aneurysm more often located in the posterior circulation, so the two treatment groups were not comparable; in addition, only one study was prospective and randomized. 19
Patient characteristics in the present prospective randomized study were similar in both groups (including age, sex, severity of subarachnoid bleeding, clinical grade and aneurysm location). The study suggested that the type of procedure used to treat a ruptured aneurysm was significantly associated with the risk of symptomatic vasospasm or cerebral infarction, and provides supportive evidence for the current practice of electing to perform endovascular coiling, rather than surgical clipping, to treat a ruptured aneurysm.
In accordance with other research, 20 the present study revealed an association between symptomatic vasospasm and cerebral infarction. Cerebral infarctions due to symptomatic vasospasm significantly influence morbidity and mortality following aSAH, and lead to poor clinical outcomes. 21 The incidence of symptomatic vasospasm-related infarctions, out of the total infarction rate in the present study, was significantly lower in the coiled patients than in the clipped patients (66.7 versus 85%, respectively).
Several studies have reported that the rate of incomplete occlusion is higher in coiling than in clipping treatments, but it remains unclear how incomplete coil occlusion affects the bleeding rate.3,14,22 Coiling resulted in a significantly lower incidence of complete occlusion compared with clipping in the present study. Post-treatment imaging revealed a significant difference in residual aneurysm in the coiled patients (12.8%) compared with the clipped patients (4.3%). Rebleeding occurred with approximately equal frequency in both treatment groups. Published studies suggest that rebleeding rates depended on the occlusion rates and on the follow-up period after clipping or coiling.14,22,23 Following ruptured aneurysms, retreatment rates were similar in patients with a neck remnant, and in those with complete occlusion. 24 A perfect angiographic result was not strived for in the present study, which may have increased the complication rate in attempting to achieve the ideal occlusion.
The present study was a randomized controlled trial, in which basic between-group patient characteristics were comparable. These enabled the study of differences in symptomatic vasospasm between coiled and clipped patients independent of confounding factors. Angiographic vasospasm or ischaemic lesions revealed on the CT scan were used to define symptomatic vasospasm. Moreover, no surviving patients missed follow-up. These factors strengthen the theory that between-group differences in occurrence of symptomatic vasospasm, infarction, rebleeding and clinical outcome were due to the mode of treatment.
The present study was limited by the small sample size and, as a result, outcome events could have been the result of chance. In the clipped patients, surgery-related damage might have been diagnosed as symptomatic vasospasm, which may have led to different diagnoses of symptomatic vasospasm between different physicians. More post-treatment CT scans (which were used to detect cerebral infarcts) were performed in the coiling group in the present study, which may have affected the results (data not shown).
In conclusion, this randomized, prospective study revealed that patients who received endovascular coiling following aSAH experienced significantly lower incidences of symptomatic vasospasm, cerebral infarction and complete occlusion, compared with patients who received surgical clipping. In addition, coiled patients showed modest (although not significant) improvements in clinical outcome, compared with clipped patients.
Footnotes
Conflicts of interest: The authors had no conflicts of interest to declare in relation to this article.