Article Text

PDF

Original research
Larger size ratio associated with the rupture of very small (≤3 mm) anterior communicating artery aneurysms
  1. Ting Xu1,
  2. Boli Lin1,
  3. Shuailiang Liu1,
  4. Xiaotong Shao1,
  5. Nengzhi Xia1,
  6. Yue Zhang1,
  7. Haoli Xu1,
  8. Yunjun Yang1,
  9. Ming Zhong2,
  10. Qichuan Zhuge2,
  11. Bing Zhao2,3,
  12. Weijian Chen1
  1. 1Department of Radiology, The first affiliated hospital of Wenzhou Medical University, Wenzhou, China
  2. 2Department of Neurosurgery, The first affiliated hospital of Wenzhou Medical University, Wenzhou, China
  3. 3Department of Neurosurgery, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Dr Weijian Chen, Department of Radiology, or Dr Bing Zhao, Department of Neurosurgery, The first affiliated Hospital of Wenzhou Medical University, Nanbai Xiang Town, Wenzhou 325000, China; wyyycwj{at}163.com, drzhaobing{at}yahoo.com

Abstract

Background Anterior communicating artery (AcoA) aneurysms have a high rupture risk, and ruptured AcoA aneurysms tend to be smaller than other intracranial aneurysms. We aimed to determine the incidence and morphologic predictors of aneurysm rupture of very small AcoA aneurysms.

Methods We conducted a retrospective analysis of 519 consecutive patients with single AcoA aneurysms between December 2007 and February 2015 in our hospital. Aneurysm morphologies were re-measured using CT angiography images. Very small aneurysms were defined as those with a maximum size ≤3 mm, and small aneurysms were defined as those with a maximum size ≤5 mm. Multivariate regression analyses were used to determine the association between aneurysm morphology and aneurysm rupture status.

Results Of the 474 ruptured AcoA aneurysms, 134 (28.3%) aneurysms were very small and 278 (58.6%) aneurysms were small. In the univariate analysis for very small aneurysms, larger aneurysm size (p=0.037), larger size ratio (p=0.002), higher aneurysm height (p=0.038), smaller vessel size (p=0.012), and dominant A1 segment configuration (p=0.011) were associated with aneurysm rupture. Multivariate analysis revealed that a larger size ratio was independently associated with the rupture status of the very small aneurysms (OR 3.69, 95% CI 1.5 to 9.0; p=0.004), and larger aneurysm size, larger size ratio, and dominant A1 segment configuration were associated with the rupture of small aneurysms.

Conclusions About one-third of ruptured AcoA aneurysms were very small. A larger size ratio, rather than other aneurysm morphologies, was independently associated with the rupture of very small AcoA aneurysms.

  • Aneurysm
  • CT Angiography

Statistics from Altmetric.com

Introduction

Despite improvements in the management of intracranial aneurysms (IAs), aneurysmal subarachnoid hemorrhage is still a devastating event, with high morbidity and mortality. Although some studies have shown that small IAs (<5 or 7 mm) have a very low risk of rupture,1–5 other observation studies have shown that there is a high proportion of small aneurysms among ruptured aneurysms.6–8 Anterior communicating artery (AcoA) aneurysms have a higher risk of rupture than those at other anterior circulation locations, and smaller ruptured aneurysms have been found in AcoA aneurysms.2 ,7–11 Most studies have reported risk factors for rupture in aneurysms >5 mm,2 ,4 ,5 ,9 ,10 and one study determined predictors for the rupture of IAs (<5 mm).8 However, no study has focused on the risk of rupture of very small AcoA aneurysms.

Current studies have shown that the morphologic characteristics of the aneurysm—such as aneurysm size, size ratio, and aneurysm lobulations—are important predictors of the rupture of aneurysms.8 ,10 ,12 ,13 These results suggest that aneurysm morphology may affect aneurysm rupture. In the present study we retrospectively collected data on the status of rupture in consecutive patients with AcoA aneurysms and re-measured aneurysm morphologies using three-dimensional (3D) CT angiography (CTA) images. We aimed to determine the incidence and the association between aneurysm morphologies and the rupture of very small AcoA aneurysms.

Methods

Patients and study design

This study was approved by the Institutional Review Board of our hospital. From December 2007 to February 2015, 574 consecutive patients with 584 AcoA aneurysms were treated and identified from the electronic medical record system. Details of characteristics of the Chinese patients with AcoA aneurysms have been described previously to determine aneurysm morphologies associated with hypertension or smoking.14 During the study all patients underwent CTA first when an intracranial aneurysm was suspected because CTA is a less invasive, cheaper, and less time-consuming examination. Aneurysms were confirmed by digital subtraction angiography (DSA). Of the 574 patients, 519 with single AcoA aneurysms were included and 55 patients were excluded (figure 1). According to the maximum aneurysm size, we defined very small aneurysms as those with a maximum aneurysm size of ≤3 mm and small aneurysms as those with a maximum aneurysm size of ≤5 mm.

Figure 1

Patient cohort and aneurysm size of ruptured aneurysms. (A) Study flow diagram. (B) Distribution of aneurysm size of ruptured aneurysms. CTA, CT angiography.

Definition of morphologic features

The morphologic features of aneurysms including aneurysm size, aneurysm height, perpendicular height, aspect ratio, size ratio, and aneurysm angle, flow angle, vessel angle, anterior cerebral artery (ACA) A1 segment configuration, and aneurysm dome projections were defined according to definitions in published studies.10 ,12–15 The maximum aneurysm size was defined by the largest cross-sectional diameter. The aneurysm height was measured between the center of the aneurysm neck and the greatest distance to the aneurysm dome. The size ratio was defined as the ratio of the maximum aneurysm height to the average vessel diameter of arteries (LACA1, LACA2, RACA1, RACA2, AcoA) associated with the aneurysms. The vessel diameter of a particular artery was measured by averaging the diameter of the cross-section of the artery proximal to the neck of the aneurysm (RACA1) and the diameter of the cross-section at 1.5×RACA11 from the neck of the aneurysm (RACA2) (figure 2).

Figure 2

Aneurysm morphology and rupture status. (A) Measurement of morphological features. (B) Unruptured aneurysm with a size ratio of 0.86. (C) Ruptured aneurysm with a size ratio of 1.85. ACA, anterior cerebral artery; AcoA, anterior communicating artery; D, maximum aneurysm size; Hmax, maximum aneurysm height; N, maximum aneurysm neck.

Measurement of morphologic features

All patients’ image data were independently collected by two neuroradiologists familiar with CTA image reconstructions and image measurements. Their average values were used for the analysis. A workstation (V.4.6; General Electric Medical Systems, Milwaukee, Wisconsin, USA) for post-processing of the images was used to reconstruct the 3D images of the aneurysms and surrounding vasculature and to measure the size, length, and angle of the aneurysms or vasculature. During the study period a 16-channel multidetector CT scanner (Lightspeed pro 16; General Electric Medical Systems) with a section thickness of 1.25 mm and a reconstruction interval of 0.625 mm was used to acquire the CTA images before June 2012, and a 64-channel multidetector CT scanner (Lightspeed VCT 64; General Electric Medical Systems) with a section thickness of 0.625 mm and a reconstruction interval of 0.625 mm and a 320-detector row CT scanner (Aquilion ONE; Toshiba Medical Systems, Japan) with a section thickness of 0.5 mm and a reconstruction interval of 0.5 mm were used to acquire the CTA images from July 2012.

Statistical analysis

Data collection and statistical analysis were performed using IBM SPSS V.22.0 (IBM SPSS, Armonk, New York, USA). Continuous variables were presented as mean±SD and categorical variables as frequencies (percentages). Univariate logistic regression analysis was used to assess the potential variables associated with aneurysm rupture. Clinical variables with a p<0.10 were entered into the multivariate analysis regression model to identify independent risk factors for the rupture of very small aneurysms using the backward method. The area under the receiver operating characteristic curves (AUC) was used to test the model's predictive ability. We also performed the same analysis for the rupture of small aneurysms. The adjusted OR and 95% CI were calculated. p<0.05 was considered statistically significant.

Results

Baseline characteristics

Of the included 519 patients with single aneurysms, 474 (91.3%) had ruptured aneurysms and 45 (8.7%) had unruptured aneurysms. The mean age was 56.2±12.3 years (range 23–88) and 281 (54.1%) patients were male. The mean aneurysm size was 5.1±2.5 mm (range 1.1–18.0); 319 (61.5%) aneurysms were ≤5 mm and 166 (32.0%) were ≤3 mm. The distribution of aneurysm size in the 474 patients with single ruptured aneurysms is shown in figure 1. The mean size of the ruptured aneurysms was 5.3±2.5 mm (range 1.1–18.0). A total of 278 (58.6%) of the ruptured AcoA aneurysms were small aneurysms and 134 (28.3%) were very small aneurysms.

Risk factors for the rupture of very small AcoA aneurysms

The baseline characteristics of 166 patients with 166 very small aneurysms are shown in table 1. Of the very small aneurysms, 134 (80.7%) were ruptured aneurysms. Ruptured aneurysms were associated with larger aneurysm size (2.6±0.5 mm vs 2.4±0.7 mm, p=0.037), larger size ratio (1.3±0.6 vs 0.9±0.5, p=0.002), greater aneurysm height (2.2±0.9 mm vs 1.9±0.8 mm, p=0.038), and smaller vessel size (1.9±0.5 vs 2.1±0.5, p=0.012). Ruptured aneurysms were more often found in the dominant A1 segment configuration (p=0.011). There were trends toward the ruptured status in younger patients (p=0.058) and those with greater perpendicular height (p=0.066) and larger aspect ratio (p=0.089). The multivariate logistic analysis showed that a larger size ratio was independently associated with aneurysm rupture in very small aneurysms (OR 3.69, 95% CI 1.5 to 9.0; p=0.004). The size ratio predicted the risk of rupture with an AUC of 0.69 (95% CI 0.58 to 0.80; p=0.001). The threshold of size ratio between the ruptured and unruptured aneurysms was 0.91, with a sensitivity of 71.6% and a specificity of 62.5%. Two illustrative cases are shown in figure 2.

Table 1

Univariate analysis for rupture of very small anterior communicating artery aneurysms

Risk factors for the rupture of small AcoA aneurysms

The baseline characteristics of 319 patients with small aneurysms are shown in table 2. Of the small aneurysms, 278 (87.1%) were ruptured aneurysms. Patients with ruptured aneurysms were younger than those with unruptured aneurysms (55.7±12.4 years vs 60.7±11.7 years; p=0.017). Ruptured aneurysms were associated with larger aneurysm size (3.6±1.1 mm vs 2.9±1.2 mm; p<0.01), larger size ratio (1.7±1.1 vs 1.1±0.6; p<0.001), greater aneurysm height (3.0±1.2 mm vs 2.3±1.2 mm; p=0.002), greater perpendicular height (2.6±1.1 mm vs 2.1±1.2 mm; p=0.012), smaller vessel size (1.9±0.5 mm vs 2.1±0.5 mm; p=0.005), and dominant A1 segment configuration (p=0.007). There were trends toward the ruptured status with larger aspect ratio (p=0.082) and larger aneurysm angle (p=0.076). The multivariate logistic analysis showed that aneurysm size (OR 1.7, 95% CI 1.0 to 2.9; p=0.043), size ratio (OR 4.5, 95% CI 1.5 to 14.7; p=0.008), and dominant A1 segment configuration (OR 3.22, 95% CI 1.4 to 7.7; p=0.008) were independently associated with aneurysm rupture in patients with small AcoA aneurysms. The mode—including size ratio, aneurysm size, and dominant A1 segment configuration—predicted the risk of rupture, with AUC of 0.75 (95% CI 0.67 to 0.84; p<0.001).

Table 2

Univariate analysis for rupture of small anterior communicating artery aneurysms

Discussion

Using a large number of consecutive patients with AcoA aneurysms, we found that most ruptured AcoA aneurysms were small and about one-third of ruptured aneurysms were very small. These results suggest that there is still a high risk of rupture in patients with small (and even very small) AcoA aneurysms. Moreover, the separate analyses of very small aneurysms and small aneurysms showed that the two aneurysm groups behave differently with regard to the performance of morphologic parameters to differentiate rupture status of the ACoA aneurysms. Our results provide statistical evidence that size ratio is a significant risk factor for the rupture of small AcoA aneurysms, in addition to aneurysm size, and the size ratio is a unique independent predictor of the rupture of very small AcoA aneurysms.

In this study, most ruptured aneurysms were small and nearly one-third of aneurysms were very small. Our results are in agreement with those of a previous study which showed that 94.4% of ruptured AcoA aneurysms were smaller than 10 mm and 43.6% of these aneurysms were smaller than 5 mm.16 These results also support those of a study on the natural course of unruptured cerebral aneurysms in a Japanese cohort2 which showed that, although the annual rupture rate associated with small unruptured aneurysms is quite low, aneurysms in the AcoA were associated with a relatively high risk of rupture. Previous studies have suggested that most aneurysms that bleed shortly after formation are never detected as unruptured aneurysms, and the critical size of aneurysm rupture is probably smaller if rupture occurs at the time of or soon after formation.17 ,18 However, a current study by Rahman et al19 has shown that none of the aneurysms had a significant decrease in size after aneurysm rupture, and their study suggests that unruptured aneurysms do not shrink after they rupture. In addition, Ohashi et al7 have suggested that aneurysm size at the time of rupture might be determined by the diameter of the parent arteries. The diameter of the AcoA is smaller than that of the middle cerebral artery and internal carotid artery. Therefore, aneurysms in the AcoA may bleed before they reach a large or giant size. These results suggest that a subgroup of smaller AcoA aneurysms still have a risk of rupture.

Although we focused on small and very small AcoA aneurysms, aneurysm size was still a significant risk factor for rupture of small AcoA aneurysms. Aneurysm size is a well-known risk factor for aneurysm rupture. More importantly, we found that the size ratio was significantly higher in ruptured aneurysms than in unruptured aneurysms. Aneurysm size, which was significant in the small subgroup, was no longer significant in the very small subgroup to differentiate rupture status. These results suggest that, although aneurysm size may be valid for small aneurysms, it may be potentially difficult for predicting differences in rupture status in very small aneurysms. On the other hand, the size ratio may be especially useful in differentiating the rupture status of very small AcoA aneurysms.

Kashiwazaki et al8 reported the clinical relationship between the size ratio and rupture in small IAs by analyzing a total of 854 patients and also found that the size ratio was an independent risk factor associated with aneurysm rupture. However, they grouped together all aneurysms and did not present data on very small aneurysms. Unlike their study, our study focused on very small AcoA aneurysms and showed that a large size ratio had a stronger correlation with aneurysm rupture than other aneurysm morphologies. A computational fluid dynamics study has demonstrated that a larger size ratio was associated with multiple vortices, complex flow patterns, and low aneurysmal wall shear stress, which may result in aneurysm rupture.20

We found that, in comparison with the A1 symmetric configuration, the A1 dominant configuration was significantly related to rupture of small and very small AcoA aneurysms, probably because of more complex flow patterns and stresses in the dominant A1 configuration. Hassan et al21 have reported that AcoA aneurysms with differences of ≥50% between the two A1 segments have high intra-aneurysmal blood flow and inflow zone shear stresses which lead to a higher risk of rupture in patients with AcoA aneurysms. However, the multivariate analysis showed that the A1 dominant configuration was only independently associated with the rupture of small AcoA aneurysms.

Study limitations

First, our study was conducted retrospectively. There may be heterogeneity in our analysis of the CTA images performed on different detector row scanners. Although DSA is a superior diagnostic modality compared with CTA, numerous studies have assessed the sensitivity and specificity of CTA for detecting IAs and suggest that CTA can provide accurate diagnostic and anatomic information on aneurysms, even small aneurysms.22 ,23 The primary examination may help in treatment decision-making before an invasive DSA examination. These features were re-measured using the standard protocol in a large number of consecutive patients. Second, we only focused on the morphologic features of aneurysms and did not examine other clinical risk factors that may affect aneurysm morphology. For example, hypertension was reported to be a significant risk factor for rupture in several studies.7 ,24 However, hypertension is a modifiable risk factor that can change over time. In our study the hypertensive status at the time of aneurysm diagnosis could not be accurately evaluated. Third, there were a relatively small number of patients with unruptured aneurysms in our institution during the study. This may reflect the fact that, in China, most patients with IAs have been diagnosed and treated after rupture. There may be a selection bias as this was a single-center retrospective cohort, which has limited generalization.

Conclusions

Our study showed that about one-third of ruptured aneurysms were very small aneurysms. Aneurysm size, size ratio, and the A1 dominant configuration were independently associated with the rupture of small AcoA aneurysms. However, a larger size ratio, rather than other aneurysm morphologies, was independently associated with aneurysm rupture of very small aneurysms. The size ratio may be useful in predicting the risk of rupture in very small AcoA aneurysms.

References

View Abstract

Footnotes

  • Contributors TX, BL, SL, XS, NX, YZ, and HX were involved in data measurement, collection and verification. BZ was involved in data analysis. YY, MZ, QZ, BZ, and WC were involved in data implementation. TX drafted the manuscript. YY, BZ, and WC critically revised the manuscript. BZ and WC had the idea for the study and protocol design. All authors read and approved the final manuscript.

  • Funding This study was supported by Zhenjiang Provincial Key Laboratory of Aging and Neurological Disorder Research, the Ministry of Science and Technology of China (grant 2012E10008; 2011-BAI08B06), and Wenzhou Bureau of Science and Technology (grant Y20140041).

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval The study was approved by the Institutional Review Board of the first affiliated hospital of Wenzhou Medical University.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.