Treatment strategy based on the natural course of the disease for patients with spontaneous isolated abdominal aortic dissection

From the Society for Vascular Surgery

Treatment strategy based on the natural course of the disease for patients with spontaneous isolated abdominal aortic dissection Ji-Hee Kang, MD,a Young-Wook Kim, MD, PhD,a Seon-Hee Heo, MD,a Shin-Young Woo, RN, MPH,a Yang-Jin Park, MD, PhD,a Dong-Ik Kim, MD, PhD,a and Duk-Kyung Kim, MD, PhD,b Seoul, Korea

ABSTRACT Objective: This study aimed to determine the natural history of spontaneous isolated abdominal aortic dissection (SIAAD) and to establish an optimal management strategy for patients with SIAAD. Methods: We searched the database of thoracoabdominal computed tomography (CT) performed at a single institution from January 2003 to July 2016 using the keywords “aortic dissection” and “dissection AND aorta.” Once a diagnosis of SIAAD was made, we investigated the initial clinical and morphologic features and aorta-related events for all patients and morphologic changes of the aortic dissection (AD) during the follow-up period for the patients who underwent follow-up CT scans. We compared characteristics of the patients, frequencies of clinical events (aortic rupture, intervention, death), and morphologic changes (false lumen enlargement, progression of AD, remodeling of AD, and involvement of iliac or visceral artery) during the follow-up period according to the location of AD (infrarenal vs suprarenal), symptom status (symptomatic vs asymptomatic), and gender. Results: There were 210 (10.7%) patients (median age, 69.4 years [interquartile range, 61.3-74.7]; male, 73.3%) who were diagnosed with SIAAD among 1958 patients with AD. SIAAD was most frequently located at the infrarenal aorta (86.2%), extended to the iliac (12.4%) or visceral artery (2.9%), and was symptomatic in 13.3% of patients. During the study period, aortic rupture developed in two patients (0.9%), aortic intervention was required in five (2.4%), and aorta-related deaths were identified in three (1.4%). Among 138 (65.7%) patients who underwent follow-up CT scans, 81.9% showed no morphologic change or remodeling during the follow-up period (median, 25 months; range, 1-158 months; interquartile range, 12.3-49.1 months). In the meantime, false lumen enlargement and longitudinal progression of AD developed in 8.7% and 6.5% of patients, respectively. However, newly developed visceral artery extension was not found in any of the patents. When characteristics of the patients and frequencies of clinical events or late morphologic changes of AD were compared on the basis of the location of AD, symptom status, and gender, female gender, presence of symptoms, and suprarenal SIAAD were more frequently associated with aorta-related mortality. False lumen enlargement was more frequent in the suprarenal AD group than in the infrarenal AD group. Conclusions: Based on our observation, the majority of symptomatic and asymptomatic SIAAD patients can be managed conservatively unless they present with aortic rupture, concomitant large aortic aneurysm, or underlying connective tissue disease. However, a more proactive management strategy may be required for female, symptomatic patients or those with suprarenal SIAAD. (J Vasc Surg 2017;-:1-11.)

From the Division of Vascular Surgerya and Division of Cardiology,b Heart, Stroke and Vascular Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine. Author conflict of Interest: none. Presented at the 2016 Vascular Annual Meeting of the Society for Vascular Surgery, National Harbor, Md, June 8-11, 2016. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: Young-Wook Kim, MD, PhD, Division of Vascular Surgery, Department of Surgery, Heart, Stroke and Vascular Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, Irwon-ro 81, Gangnam-gu, Seoul 06351, Korea (e-mail: [email protected]; [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.03.435

Aortic dissection (AD) often affects the thoracic aorta and has received much attention from cardiologists and aortic surgeons. Isolated abdominal aortic dissection (IAAD) is rarely reported as having a potential risk of aortic rupture, acute visceral or leg ischemia, or neurologic symptoms such as paraplegia.1 Most of the previous reports on IAAD were retrospective reviews of a small number of symptomatic patients or case series regarding successful surgical or endovascular treatment. In the International Registry of Acute Aortic Dissection (IRAD), 18 (1.3%) patients with isolated acute abdominal aortic dissection (IAAAD) were reported among 1417 patients with acute AD.2 On the basis of IRAD data, the patients with IAAAD showed abdominal pain, mesenteric ischemia, and limb ischemia more frequently compared with patients with type B dissection. Accordingly, continued surveillance appears to be mandatory, 1

2

Journal of Vascular Surgery

Kang et al

---

and aggressive surgical or endovascular management seems justified to improve the life expectancy of patients with IAAAD. Clinical features, expansion rate, predictors of aortic growth or aortic rupture, and optimal treatment strategies have been well described for patients with thoracic AD.3-5 On the other hand, IAAD has been reported much less frequently, and its natural history or optimal management has not yet been well established. When iatrogenic or traumatic AD is excluded from the IAAD, the entity is termed spontaneous IAAD (SIAAD). There has been no debate on the treatment of IAAD with aortic rupture or visceral or leg ischemic symptoms. However, there have been various suggestions for management of noncomplicated, asymptomatic SIAAD ranging from early prophylactic interventions to conservative management.6,7 Advocates for early prophylactic interventions are concerned about the risk of aortic rupture, visceral ischemia, or leg ischemia in SIAAD patients.8-10 There is not sufficient evidence to support either early prophylactic intervention or conservative treatment for patients with noncomplicated SIAAD. In this study, we sought to determine the clinical characteristics and natural history of SIAAD and to establish an optimal treatment strategy for such patients based on the natural history of the disease.

2017

ARTICLE HIGHLIGHTS d

d

d

Type of Research: Retrospective cohort study Take home message: Of 210 patients with spontaneous isolated abdominal aortic dissection, 81.9% showed no change during a median follow-up of 25 months (1-158 months). Rupture, connective tissue disease, and large aneurysms were associated with mortality; female gender, symptoms, and suprarenal locations predicted increased risk. Recommendation: The authors recommend that patients with asymptomatic infrarenal spontaneous isolated abdominal aortic dissection can be observed and that we use the same surgical indications as for degenerative abdominal aortic aneurysms.

METHODS Enrollment of patients. SIAAD was defined as a spontaneous AD confined to the abdominal aorta regardless of its extension to a visceral or iliac artery and not due to a traumatic or iatrogenic cause. This study was approved by the Institutional Review Board of Samsung Medical Center, and obtaining informed consent from patients was waived for this retrospective review. We searched SIAAD patients from the database of patients who underwent contrast-enhanced thoracoabdominal computed tomography (CT) scans at a single institution from January 2003 to July 2016 using keywords of “aortic dissection” and “dissection AND aorta.” After excluding thoracic ADs and traumatic or iatrogenic ADs (Fig 1), we reviewed all CT images of the remaining patients to confirm the presence of a typical AD flap in the abdominal aorta. For a diagnosis of AD to be made, a typical double lumen on the axial view of contrastenhanced aortic CT images was identified. We excluded patients showing intramural hematoma (defined as the presence of a regionally thickened aortic wall in the absence of a double lumen and intimal flap) or penetrating atherosclerotic ulcers (defined as atherosclerotic plaques that ulcerate and disrupt the internal elastic lamina).11,12 For patients thought to have connective tissue disease (CTD), a diagnosis of Marfan syndrome was made on the basis of the revised Ghent nosology with or without presence of a bonafide FBN1 mutation.13

Fig 1. Enrollment of patients with spontaneous isolated abdominal aortic dissection (SIAAD). AD, Aortic dissection; CT, computed tomography; IAAD, isolated abdominal aortic dissection.

If Ehlers-Danlos syndrome was suspected on the basis of the clinical and radiologic findings, diagnosis was confirmed with genomic DNA extracted from peripheral blood leukocytes with the Wizard Genomic DNA Purification Kit (Promega, Madison, Wisc).14 Data collection. Study patients’ demographics, clinical features, symptom status, coexisting disease or atherosclerotic risk factors, and AD-related events were retrieved from the electronic medical records. To compare characteristics of the patients, frequencies of clinical events, and late morphologic changes of AD, we classified the patients by symptom status (symptomatic vs asymptomatic), location of AD at initial presentation (infrarenal vs suprarenal), and gender.

Journal of Vascular Surgery Volume

-,

Number

Kang et al

Table I. Demographic and clinical features of 210 patients with spontaneous isolated abdominal aortic dissection (SIAAD) at initial presentation Feature Age, years, median (IQR) Sex, male

69.4 (61.3-74.7) 154 (73.3)

Symptomatic

28 (13.3) 28

Back pain

9

Chest pain

1

Foot coldness

1

Visceral malperfusion

1

Coexisting disease or risk factors Hypertension

132 (62.9)

Renal cyst

85 (40.5)

Current smoker or ex-smoker

83 (39.5)

Diabetes mellitus

51 (24.3)

Coronary arterial disease

50 (23.8)

AAA

35 (16.7)

Chronic obstructive pulmonary disease

16 (7.6)

Chronic kidney disease Vasculitis CTDc

Table II. Morphologic features at the initial presentation on computed tomography (CT) images in 210 patients with spontaneous isolated abdominal aortic dissection (SIAAD) Feature Length of AD, mm, median

a

Abdominal pain

b

3

-

5 (2.4) 5 (2.4) 4 (1.9)

AAA, Abdominal aortic aneurysm; CTD, connective tissue disease; IQR, interquartile range. Values are reported as number (%) unless otherwise indicated. Definitions: Hypertension is defined as blood pressure >140 mm Hg systolic or 90 mm Hg diastolic on at least two occasions or current use of antihypertensive pharmacologic therapy. Diabetes mellitus is defined as a history of diabetes, need for antidiabetic agents, fasting blood glucose concentration >126 mg/dL or hemoglobin A1c level >6.5%, 2-hour plasma glucose concentration $200 mg/dL during an oral glucose tolerance test, or random plasma glucose concentration >200 mg/dL. Coronary arterial disease indicates when patients underwent any type of coronary artery reconstruction or with documented history of acute myocardial infarction or angina pectoris. Chronic kidney disease is defined as serum creatinine concentration >2.0 mg/dL or patients who underwent kidney transplantation or dialysis. a The numbers of symptomatic patients can overlap. b Includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. c Includes two patients with Marfan syndrome and two patients with type IV Ehlers-Danlos syndrome.

Death and cause of death of patients with SIAAD were confirmed from the Causes of Death Statistics in Statistics Korea. On the initial CT images, we determined the length, diameter, and location of the AD; involvement of the visceral branches; degree of aortic wall calcification; and direction of entry site of AD on aortic circumference. To classify the location of AD, infrarenal AD was defined as AD not involving the renal artery orifice; paravisceral AD was defined as AD involving the renal, superior mesenteric, or celiac artery; and supraceliac AD was defined as abdominal AD extended to the supraceliac segment of the aorta not involving thoracic aorta. Morphologic changes of the AD during the follow-up period were assessed by comparing the initial CT image

Location of AD Supraceliac aorta Paravisceral aorta Infrarenal aorta Iliac artery involvement Visceral artery involvement

18 (range, 6-290; IQR, 10-30.5) No. (%) 18 (8.6) 8 (3.8) 184 (87.6) 26 (12.4) 5 (2.4)a

Superior mesenteric artery

3 (1.4)

Renal arteries

3 (1.4)

Degree of aortic wall calcification None

66 (31.4)

<25% of the circumference

113 (53.8)

25%-50% of the circumference

23 (11.0)

>50% of the circumference

8 (3.8)

Direction of the entry site of AD 12 to 3 o’clock

51 (24.3)

3 to 6 o’clock

50 (23.8)

6 to 9 o’clock

53 (25.2)

9 to 12 o’clock

59 (28.1)

AD, Aortic dissection; IQR, interquartile range. a One patient had both renal and superior mesenteric artery involvement. Paravisceral AD was defined as AD that extended to the renal or superior mesenteric artery beyond its ostium; supraceliac AD was defined as AD involving supraceliac segment of aorta below the diaphragm.

with the last follow-up CT image in terms of false lumen enlargement, distal or proximal progression of the AD, false lumen thrombosis, remodeling of the dissected aorta, and extension to the visceral or iliac artery. The length of AD was calculated on the axial view of CT images by multiplying the number of slices showing a double lumen by the slice thickness (2 or 3 mm). To assess enlargement of the false lumen, the maximal orthogonal diameter of the largest portion of AD was measured on an axial view of a CT image using an electronic caliper, and then we compared the diameters in the initial and last CT images. Complete remodeling was defined if there was no dissection or aortic luminal narrowing found on the follow-up CT images. We analyzed characteristics of the patients, frequencies of clinical events (aortic rupture, aortic intervention, allcause and aorta-related mortality), and morphologic changes (false lumen enlargement, progression, remodeling of AD, and involvement of iliac or visceral artery) during the follow-up period according to the location of AD (infrarenal vs suprarenal), symptom status (symptomatic vs asymptomatic), and gender.

4

Journal of Vascular Surgery

Kang et al

---

2017

Table III. Comparison of patients’ characteristics and frequencies of late clinical events or morphologic changes by the location of aortic dissection (AD) Location of AD Infrarenal

P

Suprarenal

Characteristic at initial presentation (N ¼ 210) No. of patients Age, years, median (IQR)

184

26

69.2 (61-74.6)

70.5 (62-74.6)

.948a <.001b

Sex, female

40 (21.7)

16 (61.5)

Symptomatic

20 (10.9)

8 (30.8)

.011c

Coexisting AAA

29 (15.8)

6 (23.1)

.349c

History of vasculitis

4 (2.2)

1 (3.8)

.487c

Coexisting CTD

2 (1.1)

2 (7.7)

.076c

Visceral artery involvement Iliac artery involvement

0 20 (10.9)

4 (15.4)

<.001c

6 (23.1)

.105c

Clinical events during the follow-up period (N ¼ 210; median, 29 months; range, 1-158 months; IQR, 29.1-59.9 months) 3 (1.6)

Aortic rupture

2 (1.6)

0

1.000c

29 (15.8)

6 (23.1)

.398c

1 (0.5)

2 (7.7)

.041c

All-cause mortality Aorta-related mortality

2 (7.7)

.117c

Required aortic intervention

Morphologic changes during the follow-up period (n ¼ 138; median, 25 months; range, 1-158 months; IQR, 12.3-49.1 months) No. of patients

124

14

102 (82.3)

9 (64.3)

.149c

False lumen enlargement

8 (6.5)

4 (28.6)

.021c

Longitudinal progression

8 (6.5)

1 (7.1)

Proximal progression

5 (4.0)

Distal progression

3 (2.4)

1 (7.1)

7 (5.6)

1 (7.1)

Partial

4 (3.2)

1 (7.1)

Total

3 (3.4)

0

Complete remodeling

2 (1.6)

0

1.000c

Extend to iliac artery

1 (0.8)

0

1.000c

No change

False lumen thrombosis

1.000c

0 .585c

AAA, Abdominal aortic aneurysm; CTD, connective tissue disease; IQR, interquartile range. Values are reported as number (%) unless otherwise indicated. Suprarenal AD includes paravisceral and supraceliac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. CTD includes two patients with Marfan syndrome and two patients with type IV Ehlers-Danlos syndrome. a Mann-Whitney test. b 2 c test. c Fisher exact test.

Statistical analysis. Data are expressed as frequencies and percentages, means 6 standard deviations, or medians with interquartile ranges (IQRs) and compared using the Mann-Whitney test, c2 test, and Fisher exact test between two groups. Kaplan-Meier survival curves were used for demonstrating the cumulative incidence of clinical events and adverse morphologic changes. In all statistical tests, significant variance was determined at a P < .05 level using SPSS software (PC version 23.0; IBM Corp, Armonk, NY).

RESULTS During the study period from January 2003 to July 2016, there were 1958 patients with AD identified with

keyword searching from the database of thoracoabdominal CT image interpretations. Excluding thoracic AD (n ¼ 1742) and traumatic or iatrogenic abdominal AD (n ¼ 6), 210 (10.7%) SIAAD patients were enrolled for further analysis. Among 210 SIAAD patients, follow-up CT images were available for 138 (65.7%) patients during the median follow-up period of 25 months (range, 1-158 months; IQR, 12.3-49.1 months; Fig 1). Patients who underwent follow-up studies with some other imaging modality, such as duplex ultrasound or non-contrastenhanced CT, were excluded from the analysis. The demographic and clinical features of SIAAD patients are shown in Table I. The median age of the patients was 69.4 years (IQR, 61.3-74.7 years), and the

Journal of Vascular Surgery Volume

-,

Number

Kang et al

5

-

Table IV. Comparison of characteristics of the patients and frequencies of late clinical events or morphologic changes by symptom status Symptom status Asymptomatic

P

Symptomatic

Characteristic at the initial presentation (N ¼ 210) No. of patients Age, years, median (IQR)

182

28

70.4 (61.8-75.5)

62.7 (51.1-71.1)

.005a

Sex, female

38 (20.9)

18 (64.3)

<.001b

Suprarenal AD

18 (9.9)

8 (28.6)

.011c

Coexisting AAA

28 (15.4)

7 (25.0)

.272c

History of vasculitis

5 (2.7)

0

.375c

Coexisting CTD

1 (0.5)

3 (10.7)

.008c

Visceral artery involvement

1 (0.5)

3 (10.7)

.008c

20 (11.0)

6 (21.4)

.127c

Iliac artery involvement

Clinical events during the follow-up period (N ¼ 210; median, 29 months; range, 1-158 months; IQR, 29.1-59.9 months) Required aortic intervention Aortic rupture All-cause mortality Aorta-related mortality

4 (2.2) 0 32 (17.6) 1 (0.5)

1 (3.6)

.515c

2 (7.1)

1.000c

3 (10.7)

.585c

2 (7.1)

.047c

Morphologic changes during the follow-up period (n ¼ 138; median, 25 months; range, 1-158 months; IQR, 12.3-49.1 months) No. of patients

120

18

No change

98 (81.7)

13 (72.2)

.348c

False lumen enlargement

10 (8.3)

2 (11.1)

.657c

Longitudinal progression

8 (6.7)

1 (5.6)

1.000c

Proximal progression

4 (3.3)

1 (5.6)

Distal progression

4 (3.3)

0

7 (5.8)

1 (5.6)

False lumen thrombosis

1.000c

Partial

5 (5.8)

0

Total

2 (1.7)

1 (5.6)

Complete remodeling

1 (0.8)

1 (5.6)

.245c

Extend to iliac artery

1 (0.8)

0

1.000c

AAA, Abdominal aortic aneurysm; AD, aortic dissection; CTD, connective tissue disease; IQR, interquartile range. Values are reported as number (%) unless otherwise indicated. Suprarenal AD includes paravisceral and supraceliac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. CTD includes two patients with Marfan syndrome and two patients with type IV Ehlers-Danlos syndrome. a Mann-Whitney test. b 2 c test. c Fisher exact test.

majority were male (n ¼ 154 [73.3%]). Of all SIAAD patients, 28 (13.3%) were symptomatic at initial presentation, including sudden-onset abdominal pain in all patients with back pain (n ¼ 9), chest pain (n ¼ 1), foot coldness (n ¼ 1), or findings of multiorgan failure due to visceral artery malperfusion (n ¼ 1). All symptomatic patients in this study showed acute onset of symptoms within 2 weeks. The identified coexisting disease or atherosclerotic risk factors were hypertension, renal cysts, and abdominal aortic aneurysm (AAA) in 62.9%, 40.5%, and 16.7%, respectively. Five patients had past history of vasculitis, which included Takayasu arteritis (n ¼ 4) and Kawasaki disease (n ¼ 1). Four patients had CTD that included Marfan syndrome (n ¼ 2) and type IV Ehlers-

Danlos syndrome (n ¼ 2; Table I). All four patients with CTD were female, and two patients (50%) had suprarenal AD. Also, three patients (75%) were symptomatic at the initial presentation. Morphologic features of SIAAD at the initial presentation are shown in Table II. The median length of SIAAD was 18 mm (range, 6-290 mm; IQR, 10-30.5 mm). The SIAAD was most often located at the infrarenal aorta (87.6%), followed by the supraceliac (8.6%) and paravisceral (3.8%) aorta. On the initial CT images, AD extended to the iliac artery (12.4%) and to the visceral arteries in 2.4%, including three (1.4%) superior mesenteric and three (1.4%) renal arteries. Aortic wall calcification around the AD lesion

6

Journal of Vascular Surgery

Kang et al

---

2017

Table V. Comparison of characteristics of the patients and frequencies of late clinical events or morphologic changes by gender Gender Male

P

Female

Characteristic at the initial presentation (N ¼ 210) No. of patients Age, years, median (IQR)

154

56

71.2 (63-75.5)

61.5 (53-71.8)

<.001a

Symptomatic

10 (6.5)

18 (32.1)

<.001b

Suprarenal AD

10 (6.5)

16 (28.6)

<.001b

Coexisting AAA

26 (16.9)

9 (16.1)

1.000b

History of vasculitis

0

5 (8.9)

.001c

Coexisting CTD

0

4 (7.1%)

.005c

3 (5.4)

.059c

13 (23.2)

.004b

Visceral artery involvement Iliac artery involvement

1 (0.6) 13 (8.4)

Clinical events during the follow-up period (N ¼ 210; median, 29 months; range, 1-158 months; IQR, 29.1-59.9 months) Required aortic intervention Aortic rupture All-cause mortality Aorta-related mortality

4 (2.6) 0 25 (16.2) 0

1 (1.8)

1.000c

2 (3.6)

.070c

10 (17.9)

.780b

3 (5.4)

.018c

Morphologic changes during the follow-up period (n ¼ 138; median, 25 months; range, 1-158 months; IQR, 12.3-49.1 months) No. of patients

103

35

86 (83.5)

25 (71.4)

.120b

False lumen enlargement

7 (6.8)

5 (14.3)

.180c

Longitudinal progression

6 (5.8)

3 (8.6)

.692c

Proximal progression

2 (1.9)

3 (8.6)

Distal progression

4 (3.9)

No change

False lumen thrombosis

0 1.000c

6 (5.8)

2 (5.7)

Partial

3 (2.9)

2 (5.7)

Total

3 (2.9)

0

Complete remodeling

1 (1.0)

1 (2.9)

.444c

Extend to iliac artery

1 (1.0)

0

1.000c

AAA, Abdominal aortic aneurysm; AD, aortic dissection; CTD, connective tissue disease; IQR, interquartile range. Values are reported as number (%) unless otherwise indicated. Suprarenal AD includes paravisceral and supraceliac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. CTD includes two patients with Marfan syndrome and two patients with type IV Ehlers-Danlos syndrome. a Mann-Whitney test. b 2 c test. c Fisher exact test.

was <50% of the aortic wall circumference in 96% of patients. Regarding pathogenesis of SIAAD, we hypothesized that aortic wall calcification affects initiation or protection of the AD. However, we found no significant association between SIAAD and amount of aortic wall calcification. The direction of the entry site of AD did not show any predilection around the aortic wall circumference (Table II). The results regarding comparison of patients’ characteristics and frequencies of clinical events or morphologic changes during the follow-up period by location of AD are shown in Table III. Female gender, presence of symptoms, visceral artery involvement, aorta-related mortality, and false lumen enlargement were more

common in the suprarenal AD group than in the infrarenal AD group (Table III). When the variables were compared between symptomatic and asymptomatic groups, we found that younger age, female gender, CTD, visceral artery involvement, and aorta-related mortality were more frequently associated with symptomatic SIAAD than its asymptomatic counterpart (Table IV). On the comparison by gender, the results showed that younger age, suprarenal AD, history of vasculitis, CTD, visceral artery involvement, and aorta-related mortality were more frequently associated with female patients than with male patients (Table V). By analyzing the patients excluding those with CTD, we found differences in the patients’ characteristics on the

Journal of Vascular Surgery Volume

-,

Number

Kang et al

Event - free rate (%)

1.00

0.95 Rupture Intervention Aorta related mortality

0.90

0.85

0.80 0

24

48

72

96

120

144

168 (month)

Number at risk Rupture

210

120

71

33

20

10

Intervention

210

116

69

32

19

9

2

Aorta related mortality

210

120

71

33

20

10

2

2

Fig 2. Cumulative clinical event-free rate (calculated with Kaplan-Meier estimates).

Cumulative incidence

1.0

Progression of AD False lumen enlargement Extent to iliac artery Total event

0.8 0.6 0.4 0.2 0.0 0

24

48

72

96

120 144 168 (month)

Number at risk Progression of AD

7

-

138

101

61

30

16

9

2

False lumen enlargement 138

101

58

29

17

8

2

Extent to iliac artery

138

101

61

30

16

9

2

Total event

138

101

58

29

16

8

2

Fig 3. Cumulative incidence of adverse morphologic changes of spontaneous isolated abdominal aortic dissection (SIAAD; calculated with Kaplan-Meier estimates). AD, Aortic dissection.

comparison by location of AD, symptom status, and gender difference. However, there was no significant difference in frequencies of clinical events or late morphologic changes (Supplementary Tables I-III, online only). Fig 2 shows a Kaplan-Meier curve demonstrating the cumulative incidence of clinical events including aortic rupture, aorta-related mortality, and aortic intervention during the follow-up period. Among 138 patients with follow-up CT images, morphologic changes of AD showed no change or remodeling in 81.9%, false lumen enlargement in 8.7%, progression of AD in 6.5%, and iliac extension in 0.8%. However, there was no newly developed visceral artery extension during the follow-up period. Interestingly, there was complete remodeling of AD in 1.4% of SIAAD patients. Fig 3 shows a Kaplan-Meier curve of cumulative incidence of adverse

morphologic changes of AD including false lumen enlargement, progression of AD, and extension to the iliac artery during the follow-up period. Comparison of patients’ characteristics and late outcomes by gender is shown in Table V. We found that female SIAAD patients were younger and were more likely to have presence of symptoms, vasculitis, CTD, suprarenal AD, iliac artery involvement, and higher aorta-related mortality than male patients. However, there was no significant difference in the late morphologic changes between the two groups. Even though we compared those two groups after excluding four CTD patients from the female group, the results remained the same, although the female group no longer showed an association with aortic rupture or aorta-related mortality. During the study period, five patients with SIAAD underwent elective AAA repair (one open repair, four endovascular repairs) because of the large diameter (>5 cm) of concurrent AAA at initial presentation. Among 12 (8.7%) patients showing false lumen enlargement on follow-up CT examination, 1 developed aortic rupture resulting in death, 1 underwent aortic open repair, and the remaining 10 patients are under surveillance. Among nine (6.5%) patients showing longitudinal progression (four distal and five proximal), one patient was a rupture case and the remaining eight patients are under surveillance so far. One patient who developed aortic rupture showed both false lumen enlargement and longitudinal progression of AD. During the follow-up period, 35 (16.7%) deaths were identified from the Causes of Death Statistics in Statistics Korea. Among them, three aorta-related deaths were attributed to two aortic ruptures and one multiple organ failure due to acute visceral organ malperfusion. The other causes of death during the follow-up period were malignant disease (n ¼ 25), cardiacrelated death (n ¼ 2), respiratory failure (n ¼ 2), gastrointestinal bleeding due to gastric cancer (n ¼ 1), septic shock of unknown origin (n ¼ 1), and other unidentified cause (n ¼ 1).

DISCUSSION SIAAD is a rare disease entity with a reported incidence of <2% of all patients with AD.10 The IRAD reported 18 (1.3%) acute isolated abdominal ADs among 1417 patients with acute (#14 days of symptom onset) AD.2 However, in this registry report, five (28%) patients had previously undergone AAA open repair. Moreover, 2 (11%) patients with iatrogenic abdominal ADs due to cardiac catheterization were included, whereas only 11 (0.78%) patients had a spontaneous cause. Although the IRAD includes only acute AD, many previous reports of IAAD included both acute and chronic, symptomatic and asymptomatic, and spontaneous and traumatic or iatrogenic origin IAADs. Therefore, it is difficult to ascertain the true

8

Journal of Vascular Surgery

Kang et al

---

2017

Table VI. Demographic and clinical features of spontaneous isolated abdominal aortic dissection (SIAAD) patients reported in previous studies Mean age, years

Acute or chronic

Farber et al,16 2002 (10)

62

NR

0

NR

70

Mozes et al,9 2002 (41)

58

Acute and chronic

0

NR

>70

Author, year (No.)

2

CTD, %

Concomitant AAA

Symptomatic, %

Leg ischemia, % 10 NR

Trimarchi et al, 2007 (18)

68

Acute

0

28

100

25

Kalko et al,17 2008 (8)

63

Acute

0

100

63

25

Mantelas et al,18 2009 (6)

62

NR

NR

NR

100

17

Kouvelos et al,6 2013 (14)

65

Acute

NR

NR

100

0 NR

8

Bockler et al, 2016 (18)

58

NR

6

18

45

Vermeersch et al,20 2015 (5)

60

Chronic

0

NR

0

0

Zhu et al,19 2015 (28)

57

Acute and chronic

NR

75

NR

Faries et al,15 2016 (37)

69

Chronic

NR

49

32

11

31

7.1

Giribono et al, 2016 (9)

67

NR

NR

0

22

11

Jonker et al,27 2009 (92)

59

Acute and chronic

1.1

42

83

17

Guangqi et al,32 2009 (121)

57

Acute and chronic

0.8

NR

NR

Jawadi et al,29 2014 (21)

72

Acute and chronic

5

19

62

5

17

62.

Acute

0

38

50

25

61

NA

6.4

47

68

12.1

Kalko et al, Sum (436)

2008 (8)

8.2

AAA, Abdominal aortic aneurysm; CTD, connective tissue disease; NA, not applicable; NR, not reported.

incidence and natural course of SIAAD from the prior reports. In our literature review,2,6,8,10,15-26 symptomatic patients numbered 68%, and AAA or CTD coexisted in 47% and 6.4% of patients, respectively. Aortic rupture developed in 6.9% of the patients, and the iliac or visceral artery was involved in 48.4% and 11%, respectively (Table VI). However, other authors reported a higher frequency of asymptomatic patients.8,15 The proportion of symptomatic patients or frequency of aortic rupture or visceral artery ischemia can vary, depending on the method of patient enrollment. Contrary to the previous reports, we found that a smaller portion (13.3%) of SIAAD patients were symptomatic at initial presentation, with a lower prevalence of coexisting AAA (16.7%) or visceral organ malperfusion (0.5%). This difference may be attributed to a large number of asymptomatic patients identified by keyword searching. As in our study, 8.7% of patients with abdominal AD were found to have pre-existing infrarenal AAA in a recent study from China.19 Mozes et al9 reviewed 41 English studies and reported that 14% of SIAAD patients presented with aortic rupture. They recommended elective repair of SIAAD, considering the high mortality rate in case of aortic rupture. Also in a meta-analysis of 92 patients including 73 (79%) SIAAD patients, Jonker et al27 reported a 10% incidence of aortic rupture and higher mortality and complication rates in patients receiving conservative treatment than in patients who underwent interventional treatment with either open or endovascular repair. However, there were many authors who recommended selective surgical or endovascular treatment in cases of

aortic rupture, presence of refractory pain or symptoms associated with visceral or lower limb ischemia, or aneurysmal change.1,28 Bockler et al8 reported treatment of 18 patients with SIAAD including 10 asymptomatic patients who underwent conservative management. According to them, the majority of patients with SIAAD eventually require invasive treatment; endovascular treatment can be a preferable treatment option today. Because aortic enlargement resulting in rupture is the most worrisome complication of SIAAD, the morphologic changes of AD during long-term follow-up have been the focus of follow-up examinations even in patients with focal and small-diameter AD. Considering the possibility of sudden development of morphologic changes or clinical events, we included all SIAAD patients who underwent clinical follow-up and CT examinations regardless of the duration of follow-up. In our series, we found that clinical or significant morphologic changes developed only in a small number of patients during the follow-up period. Currently, there are several reports describing good long-term results after endovascular treatment of SIAAD.6,15,19,29 According to these studies, the treatment result of SIAAD is superior in patients with chronic, asymptomatic AD compared with patients with acute or symptomatic AD. Symptomatic patients naturally seek interventional treatment. However, it is still not known how to identify good candidates for earlier, more proactive, interventional treatment of patients with chronic, asymptomatic SIAAD. To find any difference in clinical features or frequency of morphologic changes, we compared infrarenal vs

Journal of Vascular Surgery Volume

-,

Number

Kang et al

9

-

Table VI. Continued. Aortic rupture, %

Extend to iliac artery, %

Extend to visceral artery, %

Endovascular treatment, %

Open surgical repair, %

Medical treatment, %

Aorta-related mortality, %

10

30

NR

10

30

50

0

14

50

NR

NR

NR

0

20

6

NR

16

5.6

27.8

66.6

25

100

NR

0

100

0

0

17

NR

NR

50

50

0

16.7

0

NR

NR

100

0

0

0

NR

6

NR

NR

22.2

22.2

50

0

0

NR

NR

0

0

100

0

NR

NR

NR

100

0

0

3.6

3

43

5.4

27.0

NR

2.7

0

66

0

44

0

56

9.8

63

10.8

21

50

27

4.3

4.1

NR

NR

100

0

0

4.1

0

NR

NR

38

25

100

NR

0

11

54

6.9

48.4

suprarenal ADs, asymptomatic vs symptomatic ADs, and male vs female patients. As we anticipated, visceral artery involvement is more likely in patients with suprarenal AD compared with infrarenal AD. Furthermore, surgical or endovascular treatment is more challenging in patients with suprarenal AD. When the patients’ initial characteristics were compared between infrarenal and suprarenal AD groups, we found that female gender, presence of symptoms, and visceral artery involvement were more common in the suprarenal AD group. They also showed higher aorta-related mortality compared with those in the infrarenal AD group. Based on the comparison of frequencies of clinical events or late morphologic changes between infrarenal and suprarenal SIAAD, false lumen enlargement was more frequent in suprarenal SIAAD than in the infrarenal group before and after exclusion of CTD patients (5.7% vs 30.7%; P ¼ .012; Table III; Supplementary Table I, online only). Zhu et al19 reported different characteristics of patients among supraceliac, paravisceral, and infrarenal AD groups. Specifically, they reported that the mean age of patients in the supraceliac and paravisceral AD groups was younger than in the infrarenal group. However, in our study, there was no significant difference in age between infrarenal and suprarenal AD groups (median age, 70.5 vs 69.2 years; P ¼ .948). In our study, three aorta-related deaths occurred because of visceral malperfusion (n ¼ 1) at the initial presentation and aortic rupture in two patients with type IV Ehlers-Danlos syndrome during the study period. Aorta-related mortality was more common in symptomatic patients (n ¼ 28; 7.1% vs 0.5%; P ¼ .047). Among

2.7

4.7

0

62

0

100

0

0

20

14.2

4.8

symptomatic SIAAD patients, only one patient underwent elective endovascular aortic aneurysm repair because of unrelieved persistent back pain when the aortic diameter was 38 mm. The initial symptoms disappeared in a few days in the majority of patients. Although symptomatic SIAAD seemed more likely to be associated with younger age, female gender, CTD, visceral artery involvement, and aorta-related mortality, this finding was not consistent when we excluded the patients with CTD (Supplementary Table II, online only). We think not all symptomatic patients require early aortic intervention. It has been generally accepted that aortic aneurysm or AD secondary to Marfan syndrome requires earlier surgical treatment than for patients with other degenerative aneurysms or AD.30 The two aortic ruptures in this study developed only in patients with type IV Ehlers-Danlos syndrome at the time of aortic diameter of 28 mm and 21 mm. Because of the risk of aortic rupture for such patients, early intervention may be acceptable even in patients with small aortic diameter. However, because high surgical risk and poor prognosis of both open surgical and endovascular treatment are well known for patients with Ehlers-Danlos syndrome, we cannot currently suggest a proper management strategy for those patients. The limitations of our study are its retrospective design and low follow-up rate. In addition, our cohort included patients with both acute and chronic abdominal AD. The major problem of this study is that no power analysis was performed because of the small numbers of patients and events. Because of these limitations, we cannot

10

Kang et al

Journal of Vascular Surgery ---

suggest a specific indication for elective aortic intervention or appropriate treatment method in this study.

CONCLUSIONS In SIAAD patients, visceral or iliac artery involvement was uncommon, and the majority of SIAAD patients showed no significant change in clinical or morphologic features during follow-up, with low (1.4%) disease-specific mortality. On the basis of our observations, we recommend that asymptomatic infrarenal SIAAD be conservatively managed with the same surgical indications as for nonruptured degenerative AAA unless it is associated with CTD. However, a more proactive management strategy may be required in female, symptomatic patients with SIAAD, in particular those with suprarenal AD. Risk vs benefit of the therapeutic intervention should be clarified before treatment is decided for patients with SIAAD.

AUTHOR CONTRIBUTIONS Conception and design: JH, YW, SH Analysis and interpretation: JH, YW, SH Data collection: JH, YW, SH, SY, YJ, DI, DK Writing the article: JH, YW Critical revision of the article: JH, YW, SH, SY, YJ, DI, DK Final approval of the article: JH, YW, SH, SY, YJ, DI, DK Statistical analysis: JH, SY Obtained funding: Not applicable Overall responsibility: YW

REFERENCES 1. Nandeesh BN, Mahadevan A, Santosh V, Yasha TC, Shankar SK. Acute aortic dissection presenting as painful paraplegia. Clin Neurol Neurosurg 2007;109:531-4. 2. Trimarchi S, Tsai T, Eagle KA, Isselbacher EM, Froehlich J, Cooper JV, et al. Acute abdominal aortic dissection: insight from the International Registry of Acute Aortic Dissection (IRAD). J Vasc Surg 2007;46:913-9. 3. Jonker FH, Trimarchi S, Rampoldi V, Patel HJ, O’Gara P, Peterson MD, et al. Aortic expansion after acute type B aortic dissection. Ann Thorac Surg 2012;94:1223-9. 4. van Bogerijen GH, Tolenaar JL, Rampoldi V, Moll FL, van Herwaarden JA, Jonker FH, et al. Predictors of aortic growth in uncomplicated type B aortic dissection. J Vasc Surg 2014;59:1134-43. 5. Nozdrzykowski M, Etz CD, Luehr M, Garbade J, Misfeld M, Borger MA, et al. Optimal treatment for patients with chronic Stanford type B aortic dissection: endovascularly, surgically or both? Eur J Cardiothorac Surg 2013;44:e165-74; discussion: e174. 6. Kouvelos GN, Vourliotakis G, Arnaoutoglou E, Papa N, Avgos S, Peroulis M, et al. Endovascular treatment for isolated acute abdominal aortic dissection. J Vasc Surg 2013;58: 1505-11. 7. Klonaris C, Lioudaki S, Katsargyris A, Psathas E, Kouvelos G, Doulaptsis M, et al. Late open conversion after failed endovascular aortic aneurysm repair. J Vasc Surg 2014;59:291-7. 8. Bockler D, Bianchini Massoni C, Geisbusch P, Hakimi M, von Tengg-Kobligk H, Hyhlik-Durr A. Single-center experience in the management of spontaneous isolated abdominal aortic dissection. Langenbecks Arch Surg 2016;401:249-54.

2017

9. Mozes G, Gloviczki P, Park WM, Schultz HL, Andrews JC. Spontaneous dissection of the infrarenal abdominal aorta. Semin Vasc Surg 2002;15:128-36. 10. Borioni R, Garofalo M, De Paulis R, Nardi P, Scaffa R, Chiariello L. Abdominal aortic dissections: anatomic and clinical features and therapeutic options. Tex Heart Inst J 2005;32:70-3. 11. Valente T, Rossi G, Lassandro F, Rea G, Marino M, Muto M, et al. MDCT evaluation of acute aortic syndrome (AAS). Br J Radiol 2016;89:20150825. 12. Cambria RP, Morse S, August D, Gusberg R. Acute dissection originating in the abdominal aorta. J Vasc Surg 1987;5:495-7. 13. Loeys BL, Dietz HC, Braverman AC, Callewaert BL, De Backer J, Devereux RB, et al. The revised Ghent nosology for the Marfan syndrome. J Med Genet 2010;47:476-85. 14. Yang JH, Lee ST, Kim JA, Kim SH, Jang SY, Ki CS, et al. Genetic analysis of three Korean patients with clinical features of Ehlers-Danlos syndrome type IV. J Korean Med Sci 2007;22:698-705. 15. Faries CM, Tadros RO, Lajos PS, Vouyouka AG, Faries PL, Marin ML. Contemporary management of isolated chronic infrarenal abdominal aortic dissections. J Vasc Surg 2016;64: 1246-50. 16. Farber A, Wagner WH, Cossman DV, Cohen JL, Walsh DB, Fillinger MF, et al. Isolated dissection of the abdominal aorta: clinical presentation and therapeutic options. J Vasc Surg 2002;36:205-10. 17. Kalko Y, Kafa U, Basaran M, Kosker T, Ozcaliskan O, Yucel E, et al. Surgical experiences in acute spontaneous dissection of the infrarenal abdominal aorta. Anadolu Kardiyol Derg 2008;8:286-90. 18. Mantelas M, Antonitsis P, Kaitzis D, Hatzibaloglou A, Moros I. Spontaneous isolated dissection of the abdominal aorta: single-center experience. Interact Cardiovasc Thorac Surg 2009;8:398-401. 19. Zhu QQ, Li DL, Lai MC, Chen XD, Jin W, Zhang HK, et al. Endovascular treatment of isolated abdominal aortic dissection and postoperative aortic remodeling. J Vasc Surg 2015;61:1424-31. 20. Vermeersch N, Hendriks JM, Lauwers P, Salgado R, Van Schil PE. Aortic dissection limited to abdominal aorta: an underrecognized entity? Acta Chir Belg 2015;115:293-8. 21. Carreno JA, Llaneza JM, Alvarez LJ, Menendez MA, Rodriguez-Olay J, Vallina M, et al. Spontaneous acute abdominal aortic dissection. Ann Vasc Surg 1998;12:373-8. 22. Bello SO, Kouerinis I, Pillay W. Spontaneous supraceliac isolated abdominal aortic dissection sparing major visceral and renal vessels and presenting as chronic limb ischemia. Int J Vasc Med 2011;2011:890204. 23. Ivkosic A, Budincevic H, Krstonijevic Z, Baric M, Lojo N, Trajbar D, et al. Spontaneous isolated dissection of the abdominal aorta. Coll Antropol 2013;37:1361-3. 24. Holper P, Hyhlik-Durr A, Kotelis D, von Tengg-Kobligk H, Bockler D. Paraplegia after spontaneous dissection of the abdominal aorta. Vasa 2009;38:254-8. 25. Handa N, Nishina T, Nishio I, Asano M, Noda K, Ueno Y. Endovascular stent-graft repair for spontaneous dissection of infra-renal abdominal aorta. Ann Vasc Surg 2010;24:955. e1-4. 26. Beigi AA, Samani RE. Acute spontaneous isolated dissection of abdominal aorta. J Res Med Sci 2009;14:323-5. 27. Jonker FH, Schlosser FJ, Moll FL, Muhs BE. Dissection of the abdominal aorta. Current evidence and implications for treatment strategies: a review and meta-analysis of 92 patients. J Endovasc Ther 2009;16:71-80. 28. Zink JN, Maness MM, Bogey WM, Stoner MC. Spontaneous isolated abdominal aortic dissection involving the celiac,

Journal of Vascular Surgery Volume

-,

Number

Kang et al

11

-

superior mesenteric, inferior mesenteric, right renal, left iliac, and right superficial femoral arteries. J Vasc Surg 2015; 61:1605. 29. Jawadi N, Bisdas T, Torsello G, Stavroulakis K, Donas KP. Endovascular treatment of isolated abdominal aortic dissections: long-term results. J Endovasc Ther 2014;21:324-8. 30. Schoenhoff FS, Jungi S, Czerny M, Roost E, Reineke D, Matyas G, et al. Acute aortic dissection determines the fate of initially untreated aortic segments in Marfan syndrome. Circulation 2013;127:1569-75. 31. Giribono AM, Ferrara D, Spalla F, Narese D, Bracale U, Pecoraro F, et al. Endovascular treatment of spontaneous

isolated abdominal aortic dissection. Acta Radiol Open 2016;5. 2058460116681042. 32. Guangqi C, Xiaoxi L, Wei C, Songqi L, Chen Y, Zilun L, et al. Endovascular repair of Stanford type B aortic dissection: early and mid-term outcomes of 121 cases. Eur J Vasc Endovasc Surg 2009;38:422-6. Submitted Dec 20, 2016; accepted Mar 22, 2017.

Additional material for this article may be found online at www.jvascsurg.org.

11.e1

Journal of Vascular Surgery

Kang et al

---

2017

Supplementary Table I (online only). Comparison of characteristics of the patients, clinical events, and frequencies of morphologic changes by the location of aortic dissection (AD) Location of AD Infrarenal

P

Suprarenal

Characteristic at initial presentation (N ¼ 206) No. of patients Age, years, median (IQR)

182

24

69.4 (61.4-74.7)

71.1 (61.9-75.4)

.699a

Sex, female

38 (20.9%)

14 (58.3)

<.001b

Symptomatic

19 (10.4%)

6 (25.0)

.087c

Coexisting AAA

29 (15.9%)

5 (20.8)

.560c

1 (4.2)

.465c

0

3 (12.5)

.001c

19 (10.4%)

5 (20.8)

.169c

History of vasculitis Visceral artery involvement Iliac artery involvement

4 (2.2)

Clinical events during the follow-up period (N ¼ 206; median, 29 months; range, 1-158 months; IQR, 10.5-59.3 months) Required aortic intervention Aortic rupture All-cause mortality Aorta-related mortality

3 (1.6) 0

2 (8.3)

.105c

0

NA

28 (15.4%)

5 (20.8)

.553c

0

1 (4.2)

.117c

Morphologic changes during the follow-up period (n ¼ 135; median, 41.1 months; range, 1-158 months; IQR, 21.2-64.8 months) No. of patients No change

122 101 (82.8)

13 8 (61.5%)

.130c

False lumen enlargement

7 (5.7)

4 (30.8)

.012c

Longitudinal progression

7 (5.7)

1 (7.7)

.565c

Proximal progression

4 (3.3)

0

Distal progression

3 (2.5)

1 (7.7)

7 (5.7)

1 (7.7)

4 (3.3)

1 (7.7)

False lumen thrombosis Partial Total

.565c

3 (2.5)

0

Complete remodeling

2 (1.6)

0

1.000c

Extend to iliac artery

1 (0.8)

0

1.000c

AAA, Abdominal aortic aneurysm; CTD, connective tissue disease; IQR, interquartile range; NA, not applicable. Values are reported as number (%) unless otherwise indicated. Excluded patients with CTD. Suprarenal AD includes paravisceral and supraceliac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. a Mann-Whitney test. b 2 c test. c Fisher exact test.

Journal of Vascular Surgery Volume

-,

Number

Kang et al

11.e2

-

Supplementary Table II (online only). Comparison of characteristics of the patients, clinical events, and frequencies of morphologic changes by symptom status Symptom status Asymptomatic

P

Symptomatic

Characteristic at initial presentation (N ¼ 206) No. of patients

181

25

70.4 (62-75.5)

64.7 (52.4-71.1)

Sex, female

37 (20.4)

15 (60.0)

< .001b

Suprarenal AD

18 (9.9)

6 (24.0)

.087c

Coexisting AAA

28 (15.5)

6 (24.0)

.263c

History of vasculitis

5 (2.8)

0

1.000c

Visceral artery involvement

1 (0.6)

Age, years, median (IQR)

Iliac artery involvement

19 (10.5)

.017a

2 (8.0)

.039c

5 (20.0)

.182c

Clinical events during the follow-up period (N ¼ 206; median, 29 months; range, 1-158 months; IQR, 10.5-59.3 months) Required aortic intervention Aortic rupture All-cause mortality Aorta-related mortality

4 (2.2) 0 31 (17.1) 0

1 (4.0) 0

.480c NA

2 (8.0)

.383c

1 (4.0)

.121c

Morphologic changes during the follow-up period (n ¼ 135; median, 41.1 months; range, 1-158 months; IQR, 21.2-64.8 months) No. of patients No change

119 98 (82.4)

16 11 (68.8%)

.193c

False lumen enlargement

9 (7.6)

2 (12.5)

.620c

Longitudinal progression

7 (5.9)

1 (6.3)

1.000c

Proximal progression

3 (2.5)

1 (6.3)

Distal progression

4 (3.4)

False lumen thrombosis

7 (5.9)

Partial

5 (4.2)

Total

0 1 (6.3)

1.000c

0

2 (1.7)

1 (6.3)

Complete remodeling

1 (0.8)

1 (6.3)

Extend to iliac artery

1 (0.8)

0

.224c 1.000c

AAA, Abdominal aortic aneurysm; AD, aortic dissection; CTD, connective tissue disease; IQR, interquartile range; NA, not applicable. Values are reported as number (%) unless otherwise indicated. Excluded patients with connective tissue disease. Suprarenal AD includes paravisceral and supra-celiac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. a Mann-Whitney test. b 2 c test. c Fisher exact test.

11.e3

Journal of Vascular Surgery

Kang et al

---

2017

Supplementary Table III (online only). Comparison of characteristics of the patients, clinical events, and frequencies of morphologic changes by gender Gender Male

P

Female

Characteristic at initial presentation (N ¼ 206) No. of patients Age, years, median (IQR)

154

52

71.2 (63-75.5)

61.8 (53.6-72.6)

<.001a

Symptomatic

10 (6.5)

15 (28.8)

<.001b

Suprarenal AD

10 (6.5)

14 (26.9)

<.001b

Coexisting AAA

26 (16.9)

8 (15.4)

.801b

0

5 (9.6)

.001c

1 (0.6)

2 (3.8)

.158c

11 (21.2)

.013b

History of vasculitis Visceral artery involvement Iliac artery involvement

13 (8.4)

Clinical events during the follow-up period (N ¼ 206; median, 29 months; range, 1-158 months; IQR, 10.5-59.3 months) Required aortic intervention Aortic rupture All-cause mortality Aorta-related mortality

4 (2.6) 0 25 (16.2) 0

1 (1.9)

1.000c

0

NA

8 (15.4)

.885b

1 (1.9)

.252c

Morphologic changes during the follow-up period (n ¼ 135; median, 41.1 months; range, 1-158 months; IQR, 21.2-64.8 months) No. of patients

103

32

86 (83.5)

23 (71.9)

.145b

False lumen enlargement

7 (6.8)

4 (12.5)

.290c

Longitudinal progression

6 (5.8)

2 (6.3)

1.000c

Proximal progression

2 (1.9)

2 (6.3)

Distal progression

4 (3.9)

0

6 (5.8)

2 (6.3)

Partial

3 (2.9)

2 (6.3)

Total

No change

False lumen thrombosis

3 (2.9)

0

Complete remodeling

1 (1.0)

1 (3.1)

Extend to iliac artery

1 (1.0)

0

1.000c

.419c 1.000c

AAA, Abdominal aortic aneurysm; AD, aortic dissection; CTD, connective tissue disease; IQR, interquartile range; NA, not applicable. Values are reported as number (%) unless otherwise indicated. Excluded patients with connective tissue disease. Suprarenal AD includes paravisceral and supraceliac AD. Vasculitis includes four patients with Takayasu arteritis and one patient with past history of Kawasaki disease. a Mann-Whitney test. b 2 c test. c Fisher exact test.