Computed tomography findings and in-hospital mortality in patients with rupture of type B aortic dissection

Abstract

Background

The relationships between the computed tomography (CT) findings and outcomes of patients with ruptured type B aortic dissection have not been clarified.

Purpose

To evaluate the initial CT findings of patients with ruptured type B aortic dissection and investigate the relationships between the initial CT findings and in-hospital mortality.

Material and Methods

This study was approved by the institutional review board. Thirty-three patients were diagnosed with ruptured Stanford type B aortic dissection at our hospital between 2007 and 2016 (21 men, 12 women; mean age = 76.1±10.7 years). We retrospectively evaluated the initial CT findings of ruptured type B aortic dissection and the relationships between clinical factors and in-hospital mortality using logistic regression analysis.

Results

Type B aortic dissections ruptured in the acute and chronic phases in 23 and 10 patients, respectively. The initial CT images showed various findings, including an open false lumen (58%), arch involvement (88%), hematomas in the pleural space (55%), hematomas in the pericardial space (18%), and the extravasation of vascular contrast material (12%). The mean maximum diameter of the affected aorta was 49.5 ± 16.1 mm. Among the 33 patients, 14 died at hospital. Female gender (hazard ratio = 10.284; 95% confidence interval [CI] = 1.61–65.54; P = 0.0136) and the presence of a hematoma in the pleural space (hazard ratio = 6.803; 95% CI = 1.07–43.24; P = 0.0421) were found to be predictors of in-hospital mortality.

Conclusion

Female gender and the presence of a hematoma in the pleural space are significant predictors of in-hospital mortality in patients with ruptured type B aortic dissection.

Keywords

Dissection aorta vascular computed tomography

Introduction

Aortic dissection (AD) and various related conditions are life-threatening and require immediate diagnosis and treatment. AD is a common cause of aortic emergencies. Death from type B AD is sometimes caused by aortic rupture. Despite recent advances in surgical and perioperative management techniques, emergency surgery for acute type B AD and related complications continues to be associated with high mortality rates (1 –4), particularly in patients who suffer aortic rupture.

Recently, multi-detector row computed tomography (MDCT) has been used for the detection and differential diagnosis of AD and its complications) (5). However, there are few reports about the CT findings of patients with ruptured type B AD. Moreover, the relationships between the CT findings and outcomes of patients with ruptured type B AD have not been clarified.

We retrospectively studied the cases of patients with type B AD who experienced aortic rupture. The purpose of this study is to evaluate the initial CT findings of such patients. We also investigated the relationships between the initial CT findings and in-hospital mortality of these patients.

Material and Methods

Patients

The ethics committee of our hospital approved this study and the need for informed consent was waived. The study group comprised 344 consecutive patients who were diagnosed with Stanford type B AD from January 2007 through 2016 at our hospital. Patients who presented with penetrating atherosclerotic ulcers or iatrogenic or traumatic AD were excluded.

Among the 344 patients with Stanford type B AD, 33 patients presented with aortic rupture (9.6%) between 2007 and 2016. The rupture of type B ADs was diagnosed based on the initial clinical and CT findings as follows: presence of acute chest pain; AD involving the aorta except for the ascending aorta with thrombosed or non-thrombosed false lumens; mediastinal, pericardial, or pleural fluid collection outside the adventitia of the affected aorta that exhibited hyper-attenuation on unenhanced CT (> 25 Hounsfield units on pre-contrast CT) (6 –8).

The study included 21 men and 12 women (age range = 50–91 years; mean age = 76.1 ± 10.7 years); 26 patients (79%) had a history of hypertension and none had Marfan syndrome.

Clinical variables were recorded on a standardized form, including information about the patients’ demographics, history, clinical symptoms, physical findings, and imaging results; the use of endovascular repair; the medical and surgical treatments administered; and outcomes (including mortality). Patients who did not respond to the initial treatment administered in the emergency room and had recurrent hypotension (systolic blood pressure <80 mmHg) despite the administration of adequate fluid were defined as being in shock (6).

The rupture of the AD occurred 0–3064 days after the onset of AD (mean = 708.1 ± 1212.9 days after the onset of AD). We divided the patients into two groups: the acute group and the chronic group. The acute group was defined as ruptures which occurred ≤2 months after the onset of AD (mean = 2.5 ± 5.2 days after the onset of AD). The chronic group was defined as ruptures which occurred >2 months after the onset of AD (mean = 2124.3 ± 730.7 days after the onset of AD).

In-hospital mortality was defined as death occurring during the initial hospital stay.

Image acquisition protocol

In all patients, CT scans were performed with and without contrast enhancement. Contrast-enhanced CT was performed with non-ionic contrast medium (300 mg/mL, oypalomin; Konica Minolta, Tokyo, Japan), which was delivered as a bolus injection (100 mL at 2–4 mL/s) using a power injector (Dual Shot GX; Nemoto Kyorindo, Tokyo, Japan).

CT was performed with a Somatom definition or definition flash scanner (Siemens Medical Systems, Erlagen, Germany) to generate axial images. Scanning was commenced at 30s and 120s (two phases) after the start of the contrast medium injection. The scanning parameters were as follows: 100–120 kVp and 140–200 mAs (effective); reconstruction was conducted at a 1-mm slice thickness and 1-mm intervals for both the unenhanced and contrast-enhanced images. Scanning was performed from the thoracic inlet to the pubic symphysis for both the unenhanced and contrast-enhanced images. Electrocardiographic gating was not used.

Image evaluation

The CT data were evaluated by creating axial and longitudinal interactive multiplanar reformation (MPR) images with a 1-mm slice thickness. The CT findings were assessed independently by two radiologists who had >20 years and 5 years of experience (ES and TO), with particular attention paid to the following items: (i) the presence and location of AD; (ii) the presence of closed or open false lumens; (iii) the maximal outer diameter of the affected aorta; (iv) the presence and location of hematomas (>25 HU on pre-contrast CT); (v) the presence and location of extravasation of vascular contrast material on contrast-enhanced CT; and (vi) the rupture site determined based on the location of hematomas, aortic wall irregularities, and/or the extravasation of contrast media.

Open false lumens were defined as when false lumens exhibited obvious contrast enhancement from entry to re-entry on contrast CT scans.

In this study, the aortic arch was defined as the segment between the brachiocephalic artery and the ligamentum arteriosum. The descending aorta was defined as the segment between the ligamentum arteriosum and the aortic hiatus of the diaphragm (9). Diameter measurements on cured MPR images were obtained using a dedicated workstation (Leonardo; Siemens Medical Solutions). All CT images for each patient were read in the same session, but in a blinded order, to ensure that comparisons across serial studies produced consistent results regarding the lesions and their locations.

Inter-observer agreement was assessed with weighted κ statistics for non-continuous variables and with the intraclass correlation coefficient (ICC) for continuous variables. Final decisions regarding the classification of lesions were made by the two observers reaching a consensus.

Statistical analysis

Categorical variables are described as absolute values and percentages and were compared with the χ² test or Fisher’s exact test, as appropriate, while the Mann–Whitney U-test was used to compare continuous variables. To determine predictors of in-hospital mortality, univariate analyses of potential prognostic factors were performed using the χ² or Fisher’s exact test. Multivariate analysis of the variables related to in-hospital mortality was conducted via logistic regression analysis, and the results are expressed as odds ratios (OR) and 95% confidence intervals (CI). Factors with P values < 0.1 in univariate analysis were adopted for further multivariate analysis. Two-sided P values <0.05 were considered to indicate significant differences. Statistical analyses were performed with the software JMP (SAS Institute, Cary, NC, USA).

Results

A high level of agreement between Observers 1 and 2 was seen during the detection and classification of the CT findings (k = 0.93).

The inter-observer agreement regarding the measurements was also high. The ICC for the maximum diameter of the aorta was 0.92.

Initial CT findings at rupture

The demographic characteristics, clinical profiles, and initial CT findings of the 33 patients are shown in Table 1. The mean ages of the men (72.8 ± 10.8) and women (81.8 ± 7.8) differed significantly (P = 0.0179).

Table 1.

Clinical features and initial CT findings in 33 cases with ruptured type B aortic dissection.

Parameter	Datum
Acute	23 (70)
Age (years)	76.1 ± 10.1
Male (years)	72.8 ± 10.8
Female (years)	81.8 ± 7.8
Female gender	12 (36)
Surgical intervention	14 (42)
In-hospital death	14 (42)
Clinical characteristics
Hypertension	26 (79)
Hemodialysis	4 (12)
Diabetes mellitus	6 (18)
Shock	4 (12)
Platelet counts <100,000/µL	7 (21)
CT findings at rupture
Non-closed false lumen	19 (58)
Arch involvement	29 (88)
Maximum diameter of affected aorta (mm)	49.5 ± 16.1
Hematoma in pleural space	18 (55)
Hematoma in the mediastinum	21 (64)
Hematoma in pericardial space	6 (18)
Extravasation of vascular contrast material	4 (12)
Rupture site
Aortic arch	11 (33)
Descending aorta	17 (52)
Abdominal aorta	5 (15)

Values are presented as n (%) or mean ± SD.

Surgical intervention including open surgery and endovascular aortic repair.

SD: standard deviation.

Open false lumens were identified in 19 patients (58%) and closed false lumens in the remaining 14 patients (42%) (Fig. 1). In spite of the presence of type B AD, hematomas were identified in the pericardial space in six patients (18%) (Fig. 2). All of these patients had mediastinal hematomas.

The rupture site was the aortic arch in 11 cases, the descending aorta in 17 cases, and the abdominal aorta in five cases. In four patients (12%), the extravasation of vascular contrast material was noted (aortic arch, one case; descending thoracic aorta, three cases) on contrast-enhanced CT (Fig. 3).

In-hospital mortality

Among the patients that suffered aortic rupture, 14 underwent emergency surgical interventions. Open surgery was performed in seven patients and thoracic endovascular aortic repair (TEVAR) was conducted in seven patients (Fig. 1). There were 19 patients who did not undergo any interventions. Eight patients or their families refused to undergo interventions, mainly due to their old age. In four patients, conservative treatment was indicated because CT images showed completely closed false lumens and no extravasation of contrast media. Two patients suffered refractory shock and died before surgery. Five patients were not indicated for surgery because of respiratory dysfunction or other complications (a previous severe cerebral infarction or previous heart failure).

The in-hospital mortality rate was 42.4% (14/33). Eight patients died due to hemorrhage caused by the initial rupture 0–3 days after the onset of the rupture. Four patients died of hemorrhage at sites other than the initial rupture site (descending aorta, two cases; abdominal aorta, two cases) 3–7 days after the onset of the rupture.

In four patients with conservative treatment, one patient died due to a cerebral infarction 11 days after the onset of the rupture and one patient died from pneumonia 47 days after the onset of the rupture.

Comparison of the patients’ characteristics and initial CT findings between the acute and chronic groups

Table 2 shows a comparison of the patients’ characteristics and CT findings between the acute and chronic groups. The incidence rates of hypertension (P = 0.0494) was significantly higher in the chronic group than in the acute group. The maximum diameter of the affected aorta (P = 0.0295) was also significantly larger in the chronic group than in the acute group. On the other hand, the incidence rate of arch involvement (P = 0.038) was significantly higher in the acute group than in the chronic group.

Table 2.

Univariate analysis of comparison of patient characteristics and CT findings between acute and chronic groups.

Parameter	Acute (n = 23)	Chronic (n = 10)	P value
In-hospital death	9(39)	5(50)	0.5615
Age (years)	74.8 ± 11.3	79.0 ± 8.7	0.4446
Female gender	8 (35)	4 (40)	0.7746
Surgical intervention	11 (48)	3 (30)	0.3410
Clinical characteristics
Hypertension	16 (70)	10 (100)	0.0494
Hemodialysis	3 (13)	1 (10)	0.8055
Diabetes mellitus	3 (13)	3 (30)	0.2458
Shock	2 (9)	2 (20)	0.3605
Platelet counts < 100,000 /μL	5 (22)	2 (20)	0.9106
Initial CT findings
Non-closed false lumen	14 (61)	5 (50)	0.5615
Arch involvement on initial CT	22 (96)	7 (70)	0.0380
Maximum diameter of affected aorta (mm)	45.1 ± 13.7	59.7 ± 17.3	0.0295
Hematoma in pleural space	14 (61)	4 (40)	0.2685
Hematoma in the mediastinum	15 (65)	6 (60)	0.7746
Hematoma in pericardial space	6 (26)	0 (0)	0.0742
Extravasation of vascular contrast material	3 (13)	1 (10)	0.8055
Rupture site at thoracic aorta	20 (87)	8 (80)	0.6085

Values are presented as n (%) or mean ± SD.

CT: computed tomography; SD: standard deviation.

Comparison of the patients’ characteristics and initial CT findings between groups with and without surgical intervention

We compared the patients’ characteristics and CT findings between groups with and without surgical intervention. The presence of hematoma in the mediastinum (P = 0.0323) was significantly higher in the group without surgical intervention than in the group with surgical intervention.

Predictors of in-hospital mortality among all patients

Univariate analysis revealed that female gender (P = 0.0042) and the presence of hematoma in the pleural space (P = 0.0173) were significant predictors of in-hospital mortality (Table 3). In the multivariate logistic regression analysis, female gender (hazard ratio [HR] = 10.285; 95% CI = 1.61–65.54; P = 0.0136) and the presence of hematoma in the pleural space (HR = 6.803; 95% CI = 1.07–43.2; P = 0.0421) were also identified as predictors of in-hospital mortality.

Table 3.

Univariate analysis of factors for in-hospital death.

Parameter	In-hospital death (n=14)	Alive (n=19)	P value
Acute	9 (64)	14 (74)	0.5615
Age > 70 years	10 (71)	14 (74)	0.8857
Female gender	9 (64)	3 (16)	0.0042
Surgical intervention	4 (29)	10 (53)	0.1669
Clinical characteristics
Hypertension	14 (74)	12 (86)	0.4035
Hemodialysis	1 (7)	3 (16)	0.4519
Diabetes mellitus	4 (29)	2 (11)	0.1841
Shock	2 (14)	2 (11)	0.7436
Platelet counts <100,000/µL	4 (29)	3 (16)	0.3747
Initial CT findings
Non-closed false lumen	8 (57)	11 (58)	0.9655
Arch involvement on initial CT	12 (86)	17 (89)	0.7436
Maximum diameter of affected aorta >40 mm	10 (71)	12 (63)	0.6184
Hematoma in pleural space	11 (79)	7 (37)	0.0173
Hematoma in the mediastinum	8 (57)	13 (68)	0.5056
Hematoma in pericardial space	1 (7)	5 (26)	0.1581
Extravasation of vascular contrast material	1 (7)	3 (16)	0.4519
Rupture site at thoracic aorta	13 (93)	15 (79)	0.2707

Values are presented as n (%).CT: computed tomography.

Predictors of in-hospital mortality in the acute group

Univariate analysis revealed that female gender (P = 0.01) and the presence of hematoma in the pleural space (P = 0.0273) were significant predictors of in-hospital mortality. However, in the multivariate logistic regression analysis, no significant predictors of in-hospital mortality were identified.

Predictors of in-hospital mortality in the chronic group

Univariate analysis revealed that the presence of diabetes mellitus (P = 0.0384) was the only significant predictor of in-hospital mortality. In the multivariate logistic regression analysis, no significant predictors of in-hospital mortality were identified.

Discussion

Aortic rupture is one of the most serious complication of type B AD (6, 10–13). Even in the most experienced hands, the mortality rate of surgery is high (14 –17).

Recently, the CT findings of various aortic diseases have been well documented and the relationships between the CT findings and outcomes of such patients have also been examined (5). However, there are few investigations regarding the CT findings of patients with ruptured type B AD. The relationships between CT findings and outcomes in patients with ruptured type B AD have not been yet clarified.

In this study, closed false lumens were identified on the initial CT scan obtained after the rupture of the AD in 14 patients (42%). The findings probably indicate that the false lumens might have been thrombosed soon after the rupture. Alternatively, the findings can be explained by intramural hematomas of the aorta. It is important to recognize that the frequency of closed false lumens is relatively high among patients whose ADs rupture.

In spite of the presence of type B AD, hematomas were identified in the pericardial space in six patients (18%). Although hematomas that arise in the pericardial space are indicative of Stanford type A AD, a few cases of hemopericardium or cardiac tamponade due to type B AD have been reported (18, 19). The exact mechanisms responsible for hemorrhage in the pericardial space have not been well-known; however, bleeding into the pericardial cavity from around the ductus arteriosum or cystic medial necrosis leading invisible dissection in the ascending aorta have been suspected (18, 19). We suppose that pericardial hematomas might be due to a leak of mediastinal hemorrhage through a pericardial defect, because all of our cases of pericardial hematomas were accompanied with mediastinal hemorrhage. Indirect bleeding into the pericardial space may not significantly increase the intrapericardial pressure; it can explain the fact that intrapericardial hematoma in type B AD were not associated with poor prognosis in most of the cases in this study. Clinicians should be aware that hematoma in the pericardial space is not rare and should be noted as a sign of rupture in type B AD.

In four patients (12%), the extravasation of vascular contrast material was seen during CT examinations. The mortality rate of patients with extravasation has been unknown. In fact, prompt treatment might be required if extravasation is seen on CT.

In this study, patients with type B ADs rupture were divided into acute and chronic groups. The incidence rates of hypertension and atherosclerotic disease were significantly higher in the chronic group than in the acute group. These results suggest that arteriosclerosis influenced the rupture of the AD to a greater extent in the chronic group than in the acute group. In previous studies, it was proposed that the chronic endothelialization of the aortic lumens formed by AD accelerates atheromatous changes in the luminal wall, but the mechanism responsible for and the clinical implications of such changes are not fully understood (20 –22). In this study, the maximum diameter of the aorta was significantly smaller in the acute group than in the chronic group. It is considered that the diameter of the aorta increased with time in the chronic group. It is important to know that a relatively small aorta can rupture in acute type B AD.

In this study, the in-hospital mortality rate was 42.4% (14/33). Aortic rupture is the most catastrophic complication of type B AD; previous studies shows that the mortality rate of open surgery is in the range of 29–50% (6, 14–17).

In the multivariate logistic regression analysis conducted in the current study, female gender was found to be a predictor of in-hospital mortality among all patients.

The International Registry of Acute Aortic Dissection (IRAD) study evaluated the differences in the clinical features, management strategies, and in-hospital outcomes between the genders in intact (non-ruptured) type A and B AD (23). The IRAD study examined 1078 patients, including 346 (32.1%) women. In the IRAD study, the women were significantly older than the men, i.e. 49.7% women and 28.5% men were aged >70 years. In addition, the women exhibited a significantly higher in-hospital mortality rate than the men (30.1% vs. 21.0%; P < 0.01) (23). Another study also showed that female gender was an independent predictor of in-hospital mortality in patients with type B AD (24). As they observed no significant variation in the surgical technique employed, the delay until surgery, or intraoperative hemodynamics between the genders, the differences in the postsurgical outcomes obtained for men and women might be attributable to the older age of the women and their inherently higher surgical risk (23, 24).

In this study, it is unclear why female gender was found to be a predictor of in-hospital mortality in cases of ruptured type B AD. Considering previous reports, these results might be explained by the women’s older age and inherently higher surgical risk because the mean ages of the men and women differed significantly even in our cases of ruptured type B AD.

In this study, multivariate logistic regression analysis also showed that the presence of a hematoma in the pleural space is a predictor of in-hospital mortality in all patients. In a previous study of autopsy cases, the most common site of fatal bleeding was the thoracic cavity, followed by the mediastinum and the retroperitoneal cavity (25). Our results support these findings. Bleeding into the thoracic cavity has a weak tamponade effect; therefore, such bleeding can be fatal.

Previous studies have revealed that age, aortic diameter, the presence of an open false lumen, and hypertension are significant predictors of prognosis in patients with AD (5). However, the present study did not identify these factors as significant prognostic predictors. The reason for these discrepancies is unclear, but the prognostic risk factors for intact AD might differ from those of ruptured AD.

In this study, the surgical interventions employed did not affect the patients’ prognosis. Our cases included patients who were out of surgical indication or refused surgery. There is a possibility that bias affected our results because the number of surgical cases was small. Thus, further investigations involving a greater number of cases are necessary. If the number of cases increase in the future, the superiority of surgical interventions may become clear.

This study had the following limitations. First, we diagnosed AD and aortic rupture using MDCT. Although we were not able to obtain corresponding pathological specimens for our patients, MDCT is currently regarded as the best modality for investigating aortic diseases (5). Second, the sample size was relatively small. Thus, our study might not have had sufficient power to detect some risk factors, so further larger scale studies are needed. Third, in this study, patients were divided into acute (≤2 months) and chronic (>2 months) groups because the number of patients was small. Moreover, in our country’s guidelines, the definition of acute and chronic is different from common classification. Therefore, we used this uncommon classification in our study. In the future, after increasing the number of patients, we would like to consider patients in three groups (acute, subacute, and chronic groups) using common classification. Fourth, it is possible that many patients in whom AD rupture die before they arrive at hospital. Therefore, selection bias might have affected our results. In this study, we only evaluated cases in which the patients were still alive at the time of their arrival at hospital, which might not have provided an accurate overall picture of AD. It is necessary to evaluate cases in which the patient dies before arriving at hospital in order to address this issue.

In conclusion, hematomas can arise in the pericardial space in cases of type B AD. In patients with ruptured type B AD, female gender and the presence of a hematoma in the pleural space were found to be significant predictors of in-hospital mortality. The development of strategies to identify and treat high-risk female patients with ruptured type B AD might improve the clinical outcomes of these patients.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by JSPS KAKENHI (Grant No. 15K09894).

References

Minami

Imoto

Uchida

, et al. Clinical outcomes of emergency surgery for acute type B aortic dissection with rupture. Eur J Cardiothorac Surg 2013; 44:360–364.

Umana

Miller

Mitchell

RS.

What is the best treatment for patients with acute type B aortic dissections: medical, surgical, or endovascular stent-grafting?

Ann Thorac Surg 2002; 74:S1840–1843.

Estrera

Miller

3rd Safi

, et al. Outcomes of medical management of acute type B aortic dissection. Circulation 2006; 114 (Suppl):I384–1389.

Nienaber

Eagle

KA.

Aortic dissection: new frontiers in diagnosis and management: part II: therapeutic management and follow-up.

Circulation 2003; 108:772–778.

Nienaber

Clough

RE.

Management of acute aortic dissection.

Lancet 2015; 385:800–811.

Faure

Canaud

Marty-Ané

, et al. Endovascular management of rupture in acute type B aortic dissections. Eur J Vasc Endovasc Surg 2015; 49:655–660.

Rakita

Newatia

Hines

, et al. Spectrum of CT findings in rupture and impending rupture of abdominal aortic aneurysms. Radiographics 2007; 27:497–507.

Castañer

Andreu

Gallardo

, et al. CT in nontraumatic acute thoracic aortic disease: typical and atypical features and complications. Radiographics 2003; 23:S93–S110.

Sueyoshi E, Sakamoto I, Hayashi K, et al. Growth rate of aortic diameter in patients with type B aortic dissection during the chronic phase. Circulation 2004;110(Suppl. 1):II256–261.

10.

Suzuki

Mehta

Ince

, et al. Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the International Registry of Aortic Dissection (IRAD). Circulation 2003; 108 (Suppl.1):II312–317.

11.

Trimarchi

Nienaber

Rampoldi

, et al. Role and results of surgery in acute type B aortic dissection: insights from the International Registry of Acute Aortic Dissection (IRAD). Circulation 2006; 114 (1 Suppl.):I357–364.

12.

Wilkinson

Patel

Williams

, et al. Early open and endovascular thoracic aortic repair for complicated type B aortic dissection. Ann Thorac Surg 2013; 96:23–30.

13.

Crawford

Svensson

Coselli

, et al. Aortic dissection and dissecting aortic aneurysma. Ann Surg 1988; 208:254–273.

14.

Hogendoorn

Hunink

Schlösser

, et al. Endovascular vs. open repair of complicated acute type B aortic dissection. J Endovasc Ther 2014; 21:503–514.

15.

Miller

Mitchell

Oyer

, et al. Independent determinants of operative mortality for patients with aortic dissection. Circulation 1984; 70 (Suppl. I):I153–164.

16.

Cambria

Brewster

Gertler

, et al. Vascular complications associated with spontaneous aortic dissection. J Vasc Surg 1988; 7:199–209.

17.

Minami

Imoto

Uchida

, et al. Clinical outcomes of emergency surgery for acute type B aortic dissection with rupture. Eur J Cardiothorac Surg 2013; 44:360–364.

18.

Takagi

Manabe

Sekino

, et al. Paradoxical hemopericardium in type B acute aortic dissection. J Thorac Cardiovasc Surg 2005; 129:939–340.

19.

Imanaka

Matsuoka

Abe

, et al. Acute Stanford type B dissection and cardiac tamponade: rupture from around ductus arteriosus. Ann Thorac Surg 2007; 84:278–280.

20.

Sueyoshi E, Nagayama H, Eishi K, et al. Rapidly progressive calcification of the false lumen in aortic dissection. J Vasc Surg 2015;62:1328–1329.

21.

Hachiya

Nitatori

Yoshino

, et al. CT of calcified chronic aortic dissection simulating atherosclerotic aneurysm. J Comput Assist Tomogr 1993; 17:374–378.

22.

Hachiya

Nitatori

Okada

, et al. Calcified false channel wall in aortic dissection. Nihon Igaku Hoshasen Gakkai Zasshi 1994; 54:258–263.

23.

Nienaber

Fattori

Mehta

, et al. Gender-related differences in acute aortic dissection. Circulation 2004; 109:3014–3021.

24.

Trimarchi

Tolenaar

Tsai

, et al. Influence of clinical presentation on the outcome of acute B aortic dissection: evidences from IRAD. J Cardiovasc Surg (Torino) 2012; 53:161–168.

25.

Nakashima

Kurozumi

Sueishi

, et al. Dissecting aneurysm: a clinicopathologic and histopathologic study of 111 autopsied cases. Hum Pathol 1990; 21:291–296.