Computed tomography angiography findings predict the risk factors for preoperative acute ischaemic stroke in patients with acute type A aortic dissection

Abstract

Open in new tabDownload slide

OBJECTIVES

Acute ischaemic stroke (AIS) is a highly dreaded complication of acute type A aortic dissection (ATAAD). Knowledge about independent predictors of preoperative AIS in patients with ATAAD remains unclear. The aim of this study was to identify the risk factors for preoperative AIS in patients with ATAAD by computed tomography angiography (CTA) findings.

METHODS

Between October 2014 and March 2017, 281 (217 male; mean age, 50 years) patients with ATAAD underwent aortic CTA and diffusion-weighted magnetic resonance imaging of the brain was used to confirm the results. The clinical data and CTA findings were evaluated retrospectively. Patients were divided into 2 groups depending on the presence or absence of preoperative AIS.

RESULTS

Preoperative AIS was detected in 103 (36.7%) of the patients with ATAAD. Univariable analysis of the clinical characteristics and CTA findings revealed that age, aortic valve insufficiency (moderate or severe), the ratio of the diameter of the true lumen of the ascending aorta to the diameter of the involved ascending aorta, intimal flap plaque, dissection of the common carotid artery (CCA), the lower density of the unilateral internal carotid artery, the CCA originating from the false lumen and dissection of the subclavian artery were implicated in patients with ATAAD with AIS. Multivariable analysis further showed that aortic valve insufficiency (moderate or severe) [odds ratio (OR) 2.033, 95% confidence interval (CI) 1.052–3.931; P = 0.035], 2 CTA findings including the ratio of the diameters (OR 0.074, 95% CI 0.011–0.516; P = 0.009) and dissection of the CCA (OR 2.422, 95% CI 1.389–4.224; P = 0.002) were independent risk predictors for preoperative AIS in patients with ATAAD. The lower density in the false lumen, the same enhancement in the true and false lumen with re-entry and the stenosis of the true lumen without re-entry significantly increased the risk of preoperative AIS in CCA dissection.

CONCLUSIONS

Aortic valve insufficiency (moderate or severe), the ratio of the diameters of the true and false lumens and CCA dissection are independent predictors of preoperative AIS in patients with ATAAD. The specific carotid and aortic CTA findings may help to predict the risk factors for preoperative AIS in patients with ATAAD.

Clinical registration number

20120216-4.

INTRODUCTION

Acute ischaemic stroke (AIS) is a highly dreaded complication of acute type A aortic dissection (ATAAD). International Registry of Acute Aortic Dissection data showed that stroke occurred in 6% of patients with ATAAD and significantly increased the number of in-hospital deaths [1]. Information on preoperative AIS may be useful to establish the ideal therapeutic approach; however, adequate risk stratification for ATAAD and neurological dysfunction in the preoperative algorithm are difficult to determine, due to atypical symptoms or to chest and/or back pain.

Diffusion-weighted magnetic resonance imaging (MRI) has been shown to be the most sensitive technique for identifying AIS due to its ability to detect rapid shifts in the ratio of extracellular to intracellular water content in the brain [2]; however, the use of this modality for patients with ATAAD in an emergency situation is often limited for practical reasons. Although dissection of the common carotid artery (CCA) detected by ultrasound has been helpful in the preoperative diagnosis of AIS in patients with ATAAD, the evidence is still indirect and limited [3–5]. Computed tomography angiography (CTA) is a highly accurate and established imaging modality with a proven track record of sensitivity and specificity approaching 100% for the detection of acute aortic disease [6]. Aortic CTA is paramount in accurately confirming or excluding critical aortic lesions, defining location and extent, and describing associated complications [7]. According the International Registry of Acute Aortic Dissection, CTA may now be the imaging modality of choice when aortic dissection is suspected [8]. Our previous study demonstrated that CTA findings may identify the risk of postoperative temporary and permanent neurological dysfunction in patients with ATAAD [9]. However, knowledge about independent predictors of preoperative AIS in patients with ATAAD remains scarce. Detailed CTA information correlated with preoperative AIS in patients with ATAAD has not been reported. Our goal was to explore findings related to the risk of CTA in predicting preoperative AIS in patients with ATAAD.

METHODS

Study population and definitions

This retrospective study complied with the Helsinki Declaration (2000) and was approved by the institutional review board of Xijing Hospital affiliated with the Fourth Military Medical University (20120216-4). Informed consent was obtained from each patient or legal representative. From October 2014 to March 2017, all consecutive patients with ATAAD underwent both aortic CTA and brain diffusion-weighted MRI of the brain when neurological deficits were suspected. There were 384 patients in the initial database. Exclusion criteria included (i) non-ATAAD; (ii) a previous aortic operation; (iii) a history of cerebral infarction or haemorrhagic stroke; (iv) unevaluable image quality; and (v) malignant tumours, blood diseases and severe diseases of the heart, liver, lung and kidney systems. A total of 281 patients were included in this study. For each enrolled patient with ATAAD, age, sex, medical history, presented signs and symptoms and the diagnostic results of cardiac Doppler ultrasound in the emergency department were recorded. The diagnosis of type A aortic dissection was confirmed with CTA. ATAAD was defined as any non-traumatic dissection of the aorta proximal to the left subclavian artery presenting within 2–7 days after symptom onset [10]. AIS was defined as 1 or more non-contiguous hyperintense lesions on diffusion-weighted images.

Acquisition and post-processing of computed tomography data

All examinations were performed using second-generation dual-source CT equipment (Somatom Definition Flash, Siemens Healthcare, Forchheim, Germany) with a high-pitch spiral scan mode. Non-electrocardiogram-gated high-pitch dual-source CTA of the whole aorta was used to evaluate the aortic diseases. Patients underwent combined CTA imaging of the neck and aorta in the cranio-caudal direction. The scanning parameters were as follows: tube voltage of 100 kV, pitch of 3.0, slice collimation of 2 mm × 128 mm × 0.6 mm by means of a z-flying focal spot and a reference tube current of 300 mAs per rotation. For all scans, patients were in a supine position with both arms raised. Each patient received an injection of 70 ml of iopromide (Ultravist 370, 370 mg I/ml, Bayer Schering Pharma, Berlin, Germany) at a flow rate of 5 ml/s, followed by 40 ml of saline solution. Bolus tracking was performed in the suprarenal descending aorta with an attenuation threshold of 100 HU. The raw data were transferred to an external workstation (syngo MMWP VE 36 A, Siemens Healthcare, Forchheim, Germany) for further processing.

Computed tomography angiography findings, definitions and parameters

We focused on the carotid arteries, the ascending aorta (aAO) and the aortic arch. The carotid arteries including the CCA and the internal carotid artery are directly and the aAO is indirectly associated with the cerebral artery and cerebral blood supply. The diameter of the true lumen was defined as the minimum diameter of the true lumen in the section orthogonal to the centreline; the diameter of the aAO was measured at the same level. The ratio of the diameters was calculated by dividing the diameter of the true lumen by that of the aAO. Intimal flap plaque was defined as atherosclerotic plaque that was detected in the intimal flap of the aAO or the aortic arch. A thrombosed false lumen in an aAO referred to a low-attenuation haematoma present in the false lumen of the compromised aAO. A retrograde aAO dissection was a primary intimal tear in the aortic arch or descending aorta in which the intimal flap extended to the aAO in a retrograde manner without re-entry. Lower density of a unilateral internal carotid artery/vertebral artery was defined as the mean intraluminal CT value of the unilateral internal carotid artery/vertebral artery at the level of the C1 vertebra lower than that in the offside artery for more than 100 Hu.

Analysis of detailed computed tomography angiography characteristics of the involved common carotid artery correlated with acute ischaemic stroke

Although previous studies have described the dissection of the CCA as an important risk factor for AIS [11, 12], an evaluation of the detailed impaired status of a compromised CCA has not been reported in the literature. Moreover, in our clinical experience, AIS is not present in all patients with ATAAD with CCA dissection. We thus further analysed the detailed characteristics of the CCA dissection, including the location, extent, the true and false lumen and with or without re-entry in the impaired CCA. Two types were analysed for CCA dissection with re-entry: (i) lower density in the false lumen and (ii) the same density in both the true and false lumens. In addition, 2 types were analysed for CCA dissection without re-entry: (i) severe stenosis of true lumen and (ii) mild or moderate stenosis of the true lumen. Severe stenosis (>75%) of the true lumen was defined as a minimum diameter of the true lumen that was <25% of the diameter of the CCA in the same section [1].

Statistical analyses

Statistical analyses were performed using statistical software SPSS 20.0 (IBM SPSS Statistics for Windows, Version 20.0, 2011; IBM Corp., Armonk, NY, USA). Summary statistics are presented as counts with percentages for categorical values, as mean and standard deviation for normally distributed continuous variables and as median with quartiles for non-normally distributed continuous variables. Data distribution was checked using the Shapiro–Wilk test. Continuous variables with normal data distributions were compared using unpaired t-tests; for data with skewed distributions, non-parametric Mann–Whitney U-tests were used; categorical variables were compared using the χ² statistics; and the Fisher’s exact test was used if observed frequencies were <5. To identify the predictors for preoperative AIS in patients with ATAAD, multiple logistic regression analysis was used, and variables with a univariable value of P-value <0.05 were incorporated into the multivariable model. The Hosmer–Lemeshow test was performed to quantitatively assess the model fit in logistic regression. Receiver operating curves were analysed to assess the best cut-off value of the continuous variables to predict preoperative AIS with maximal accuracy using MedCalc Software. The optimal cut-off value was calculated using the Youden index. All analyses were 2-tailed, and P-values <0.05 were considered statistically significant.

RESULTS

Patient characteristics

The demographics and clinical characteristics of all patients with ATAAD with and without AIS are shown in Table 1. Among 281patients with ATAAD, 103 patients had AIS (of whom 86 were Debakey type I and 17 were Debakey type II) and 178 patients had no AIS (of whom 152 were Debakey type I and 26 were Debakey type II). The non-parametric Mann–Whitney test indicated that the age of subjects with AIS was significantly older than those without AIS (50 vs 52 years; P = 0.021). A significantly higher percentage of aortic valve insufficiency was found in the AIS group (26.2%) than in group without AIS (14.6%; P = 0.017). There were no significant differences in terms of sex, medical history, presenting symptoms and signs and the interval from symptoms to presentation/diagnosis between the 2 groups (P > 0.05; for all comparisons).

Computed tomography angiography findings of preoperative acute ischaemic stroke in patients with acute type a aortic dissection

Detailed CTA findings that were considered to represent possible risk factors for preoperative AIS are presented in Table 2. The non-parametric Mann–Whitney test indicated a significantly smaller diameter of the true lumen in the aAO and of the ratio of the diameters in the AIS group (0.25; 0.14–0.34) compared to that in the cohort without AIS (0.33; 0.20–0.44) (P < 0.001), although there was no significant difference in the diameters of the aAO between the 2 groups (P = 0.72). Moreover, patients with ATAAD with preoperative AIS had a higher incidence of intimal flap plaque (P = 0.006); CCA dissection (P = 0.000); a lower density of the unilateral internal carotid artery (P = 0.001); CCA originating from the false lumen (P = 0.013); subclavian artery dissection (P = 0.046); and right subclavian artery dissection (P = 0.002). Between the 2 groups, there were no significant differences in any of the following: the thrombosed false lumen of the involved aAO (P = 0.69); retrograde aAO dissection (P = 0.82); entry tear in the aortic arch (P = 0.71); and lower density of the unilateral vertebral artery (P = 0.10), vertebral artery dissection (P = 0.63), the vagus subclavian artery (P = 0.28), the subclavian artery originating from the false lumen (P = 0.28), the vertebral artery originating from the false lumen (P = 0.37) and the vertebral artery originating from the aortic arch (P = 0.87).

Risk predictors for preoperative acute ischaemic stroke in patients with acute type A aortic dissection

Univariable analysis of clinical characteristics and CTA findings revealed that age, aortic valve insufficiency (moderate or severe), the ratio of the diameter of the true lumen of the aAO to the diameter of the involved aAO, intimal flap plaque, CCA dissection, lower density of the unilateral internal carotid artery, CCA originating from the false lumen and subclavian artery dissection were implicated in patients with ATAAD with AIS. Multivariable analysis further showed that aortic valve insufficiency (moderate or severe) [odds ratio (OR) 2.033, 95% confidence interval (CI) 1.052–3.931; P = 0.035], 2 CTA findings including the ratio of the diameter of the true lumen of the aAO to that of the diameter of the involved aAO (OR 0.074, 95% CI 0.011–0.516; P = 0.009) (Fig. 1A–C) and CCA dissection (OR 2.422, 95% CI 1.389–4.224; P = 0.002) were independent risk predictors for preoperative AIS in patients with ATAAD (Table 3). Additionally, the Hosmer–Lemeshow test yielded a P-value of 0.245 without statistical significance, suggesting a favourable model fit in logistic regression. We used a receiver operating characteristic curve to find the best parameter and the cut-off point of continuous variables (the ratio of the diameters) for predicting the occurrence of AIS. The results revealed that the area under the receiver operating characteristic curve of the ratio of the diameters to predict preoperative AIS was 0.787 (CI 0.720–0.853), and the optimal cut-off value of the ratio of the diameters was 0.33 (Fig. 2).

Computed tomography angiography findings of dissection of the common carotid artery related to preoperative acute ischaemic stroke in patients with acute type A aortic dissection

The detailed CTA characteristics of CCA dissections related to AIS are shown in Table 4. Of the 281 patients with ATAAD initially studied, 118 patients (42.0%) with CCA dissections were detected in the present study. Of those 118 patients, 61 patients (51.7%) had preoperative AIS. There was no significant difference in the location of the dissection of the CCA on the right side, left side or on both sides between the 2 groups (all P > 0.05). When the dissection involved the CCA, re-entry in the CCA combined with the lower density in the false lumen (P = 0.000) (Fig. 3A and C) and no re-entry in the CCA combined with severe stenosis of true lumen (P = 0.000) (Fig. 3B and D) significantly increased the risk of AIS in patients with ATAAD.

DISCUSSION

AIS is a highly dreaded complication of ATAAD. The underlying mechanisms of AIS in patients with ATAAD are brain tissue ischaemia from hypotension and direct compromise of cerebral circulation [13]. AIS was associated with increased in-hospital morbidity and mortality in patients with ATAAD [1]. Patients with ATAAD complicated with AIS do not often complain of the classical chest or back pain; only about half of the patients with AIS have such a complaint [14, 15], and it is easy to delay diagnosis and appropriate surgical treatment.

The presence or absence of AIS can be used to predict the preoperative risk factors in patients with ATAAD. Diffusion-weighted MRI is the most accurate method for diagnosing AIS, but performing MRI for all patients with ATAAD is impractical in emergency situations. CTA was the most commonly used imaging modality for aortic dissection; its diagnostic sensitivity and specificity approach 100% [16–19]. CTA permits the early recognition and characterization of aortic dissection as well as of the critical findings related to important complications, which are essential for optimizing treatment and improving clinical outcomes [20, 21]. Therefore, using CTA to detect risks related to preoperative AIS may allow the confident prediction of AIS and enable prompt appropriate treatment of ATAAD, especially in patients for whom there is not enough time to undergo brain diffusion-weighted MRI in the emergency department.

In our study, the incidence of ATAAD complicated with AIS was 36.7%, which was significantly higher than the 6–22% reported in the literature [1, 22, 23]. This difference may mainly relate to all patients with ATAAD enrolled in this study with suspected AIS who underwent diffusion-weighted MRI. Our results also confirmed that patients with ATAAD with AIS were older than those without AIS, although age was not independent risk factor for AIS. This result is similar to that from another published report [23], which stated that the age of patients with ATAAD with stroke was significantly older than those without stroke. In addition, there were significantly more patients with aortic valve insufficiency in the AIS group than in the group without AIS. More importantly, our data indicated that aortic valve insufficiency was one of the risks for preoperative AIS in patients with ATAAD. It is reasonable to presume that massive regurgitation due to aortic valve insufficiency leads to decreased blood flow in the tearing aAO, aortic arch and supra-aortic branches, resulting in subsequent cerebral ischaemic infarction.

The diameter of the true lumen of the aAO was perhaps noteworthy because the ratio of the respective diameters was another independent CTA risk predictor for preoperative AIS in patients with ATAAD. The ratio of the diameters, that is, of the diameter of the minimum true lumen divided by the diameter of the aAO from the same slice, may accurately predict the risk of preoperative AIS when the value is <0.33. It can be imagined that a severely stenotic true lumen will affect the forwards flow of blood in the aAO and indirectly lead to cerebral hypoperfusion.

As described previously [21], CCA dissection was common in patients with ATAAD complicated with AIS. In this study, the incidence of CCA dissection was as high as 42% and an independent CTA risk predictor of preoperative AIS. Approximately 59.2% of patients with ATAAD patients with AIS had dissection of the CCA, indicating that involvement of the CCA may cause cerebral hypoperfusion. A previous study also indicated that dissection of the CCA may correlate with cerebral malperfusion, which underscores the strong potential for AIS in patients with ATAAD [24]. However, AIS was not caused by extension of the dissection towards the supra-aortic vessels, whereas AIS was caused by aetiological mechanisms such as thromboembolism or severe hypotension.

When the CCA is involved in the dissection, different detailed findings may lead to different outcomes in regard to AIS in patients with ATAAD. We found that patients with ATAAD with AIS had a significantly longer extension of the CCA dissection than those without AIS. Moreover, AIS was significantly more common when CCA dissection involved re-entry and a lower density of the false lumen or when the CCA dissection had no re-entry but had severe stenosis of the true lumen. The results suggested that both emboli from the false lumen flowing into the cranial arteries via the re-entry and the decreased antegrade flow due to the severely narrowed true lumen may result in preoperative AIS in patients with ATAAD. This valuable information about the characteristics of CCA dissection discernible with CTA may apprise the surgeon of critical procedural details.

As we all know, patients with ATAAD complicated with preoperative AIS have adverse outcomes, but the imaging data of the preoperative AIS reported in patients with ATAAD are limited. In this study, we were extremely fortunate that all 103 patients with preoperative AIS underwent CTA and diffusion-weighted MRI examinations because such data are rare. CTA and diffusion-weighted MRI imaging can help us to objectively diagnose preoperative AIS in patients with ATAAD and more clearly understand the mechanism, such as CCA dissection, that may lead to cerebral malperfusion and retrograde dissection of the aAO combined with thrombus in the false lumen, which are likely to lead to embolic cerebral infarction. These findings are critical for guiding future clinical interventions and preventing adverse outcomes.

Limitations of the study

This study only reflected a single-centre experience with a limited number of patients. Because of the small sample size and the limited number of events, variables for multivariable models were also limited. Larger, multicentre studies are needed to determine the real risk factors for preoperative AIS of ATAAD identifiable via CTA.

CONCLUSIONS

This study assessed the predictive risk factors identifiable using CTA imaging of preoperative AIS in patients with ATAAD. The results revealed that aortic valve insufficiency (moderate or severe) and 2 CTA findings including the ratio of the diameter of the true lumen of the aAO to the diameter of the involved aAO and CCA dissection are independent risk predictors for preoperative AIS in patients with ATAAD. The specific carotid and aortic CTA findings may help predict individual risk for preoperative AIS in patients with ATAAD.

Funding

This work was supported by the Subject Boosting Project of Xijing Hospital [XJZT18ML20] and the National Key Research and Development Program of China [2016YFC1301900].

Conflict of interest: none declared.

Author contributions

Hongliang Zhao: Formal analysis; Writing—original draft. Wanling Ma: Formal analysis; Writing—original draft. Didi Wen: Data curation; Investigation. Weixun Duan: Data curation; Investigation. Minwen Zheng: Conceptualization; Project administration; Supervision.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• aAO

  Ascending aorta

 
• AIS

  Acute ischaemic stroke

 
• ATAAD

  Acute type A aortic dissection

 
• CCA

  Common carotid artery

 
• CI

  Confidence interval

 
• CTA

  Computed tomography angiography

 
• MRI

  Magnetic resonance imaging

 
• OR

  Odds ratio

Author notes