Treatment indications for and outcome of endovascular repair of type B intramural aortic hematoma

Treatment indications for and outcome of endovascular repair of type B intramural aortic hematoma Moritz S. Bischoff, MD,a Katrin Meisenbacher, MD,a Michael Wehrmeister, MD,b Dittmar Böckler, MD,a and Drosos Kotelis, MD,c Heidelberg and Aachen, Germany

ABSTRACT Objective: The aim of this study was to analyze the outcome of thoracic endovascular aortic repair (TEVAR) and medical therapy in patients with aortic intramural hematoma type B (IMHB). Methods: Between January 2004 and January 2014, 41 IMHB patients were treated; 28 underwent TEVAR (16 male; median age, 69 years; group I) plus best medical therapy (BMT), whereas 13 had BMT alone (6 male; median age, 69 years; group II). Study end points were assessment of indications for TEVAR and BMT, clinical outcome, and evaluation of aortic morphology over time. Median follow-up was 32 months for group I and 40 months for group II. Results: In group I, TEVAR was immediately performed in 7 of 28 cases because of pain and imaging results (penetrating aortic ulcers, n ¼ 4; intramural blood pools, n ¼ 3). In 21 of 28 cases, TEVAR was undertaken because of clinical or radiologic signs of progression at a median of 10 days (range, 2-223 days). The median number of stent grafts implanted was two (range, 1-3). The median length of covered aorta was 15 cm (range, 9.5-33.4 cm). Technical success was achieved in 25 of 28. In-hospital mortality was 1 of 28 in group I and 0 of 13 in group II. Survival in group I was 81.5%, 77.8%, and 67% at 1, 2, and 4 years. There was no death in group II during follow-up. Aortic reinterventions were performed in 6 of 28 group I cases, including 2 open conversions for retrograde type A dissection. Aortic diameter decreased during follow-up in 10% in group I (vs 3% in group II; P ¼ .039). In group I, complete remodeling was seen in 7 of 27, regression in the remaining 20. In group II, complete remodeling was seen in 3 of 12; regression was seen in 9 of 12. No patient in group II required invasive treatment. Conclusions: BMT is justified in uncomplicated IMHB. However, IMHB becomes complicated in the majority of patients within 20 days. TEVAR in complicated IMHB is feasible but associated with a substantial aortic reintervention rate, reflecting technical challenges and fragile aortic wall conditions. (J Vasc Surg 2016;-:1-11.)

Aortic intramural hematoma type B (IMHB) differs from classic aortic dissection as it is defined as formation of thrombus in the aortic wall in the absence of a visible intimal tear or blood flow between the layers.1,2 Currently, treatment of IMHB is derived from studies on aortic dissection. Available guidelines recommend best medical therapy (BMT) in uncomplicated cases.3-5 However, there are numerous clinical (eg, refractory pain) and radiologic (eg, IMHB thickness) factors that are reported to be associated with disease progression, challenging the term uncomplicated in clinical practice.3,5 In fact, the natural course of IMHB is still unpredictable, reaching

From the Department of Vascular and Endovascular Surgerya and Department of Diagnostic and Interventional Radiology,b Heidelberg University Hospital, Heidelberg; and the Department of Vascular Surgery, RWTH Aachen Univer-

from complete IMHB resolution to aortic rupture. Current treatment algorithms for IMHB (ie, proposed by Evangelista et al) are based on retrospective data and are not yet proven in a prospective fashion.5 Regarding thoracic endovascular aortic repair (TEVAR), there is only limited literature on selection of patients, indications for treatment, and potential benefits of TEVAR on patient outcome. In contrast to classic dissection, there are no registry data or randomized controlled trials to compare the clinical course of IMH patients with one treatment or another. Therefore, the aim of this study was to provide data on the clinical and morphologic course of patients with IMHB. Furthermore, the indications leading to TEVAR as well as the achieved results are analyzed in light of the present treatment recommendations.

sity Hospital, Aachen.c Author conflict of interest: none.

METHODS

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

Study design and patient population. In this retrospective single-center analysis, a total of 41 patients with IMHB were treated between January 2004 and January 2014. Of those, 28 patients (16 male; median age, 69 years; range, 45-89 years) underwent TEVAR (group I). The demographic, procedural, and outcome data for all patients treated by TEVAR were extracted from a prospectively maintained database. Thirteen patients underwent BMT (group II; 6 male; median age, 69 years;

Correspondence: Moritz S. Bischoff, MD, Department of Vascular and Endovascular Surgery, Heidelberg University Hospital, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany (e-mail: [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 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.05.078

1

2

Journal of Vascular Surgery

Bischoff et al

---

2016

Table I. Patients’ characteristics All (N ¼ 41)

Group I, TEVAR (n ¼ 28)

Group II, BMT (n ¼ 13)

69 (43-89)

69.1 (45-89)

69 (43-81)

.74

22/19

16/12

6/7

.53

Heart failure

5 (12.2)

5 (17.9)

0

Hypertension

36 (87.8)

27 (96.4)

9 (69.2)

.069

Ischemic heart disease

9 (21.9)

7 (25)

2 (15.4)

.47

History of myocardial infarction

1 (2.4)

1 (3.6)

0

.32

Carotid stenosis

1 (2.4)

1 (3.6)

0

.32

Peripheral vascular disease

2 (4.9)

2 (7.1)

0

.16

COPD

7 (17.1)

6 (21.4)

1 (7.7)

.22

Age, years Gender, male/female ASA classification

Diabetes mellitus Obesity (BMI >30 kg/m2) Mean BMI, kg/m2 Renal insufficiency (creatinine level >1.2 mg/dL) History of smoking Previous aortic surgery

P

3 (2-4) .02

4 (9.8)

2 (7.1)

2 (15.4)

.48

10 (24.4)

7 (25)

3 (23.1)

.84

23.52

23.38

24.91

.70

6 (14.6)

5 (17.8)

1 (7.7)

.35

23 (56.1)

19 (67.7)

4 (30.8)

.02

4 (9.8)

3 (10.7)

1 (7.7)

.76

OAR/EVAR

3

2

1

Open thoracic surgery

1

1

0

IMH extent with respect to aortic zones 3-4

2

1

1

3-5

14

9

5

3-6

5

2

3

3-7

1

1

0

3-8

10

6

4

3-9

7

7

0

4-5

2

2

0

ASA, American Society of Anesthesiologists; BMI, body mass index; BMT, best medical therapy; COPD, chronic obstructive pulmonary disease; EVAR, endovascular aortic aneurysm repair; IMH, intramural hematoma; OAR, open aortic repair; TEVAR, thoracic endovascular aortic repair. Continuous data are expressed as median (range). Categorical data are presented as number (%).

range, 43-81 years). The conservative cohort treated at the authors’ institution was identified by individual case review of 1044 patients with the admission diagnosis aortic aneurysm or aortic dissection (International Statistical Classification of Diseases and Related Health Problems codes I71.1-I71.6 and I71.00-I71.07).6 As open repair for IMHB is not performed at the authors’ institution, this study does not contain data of patients treated by conventional surgical means. The comorbidities for groups I and II are displayed in Table I. In both groups, follow-up included medical history, physical examination, and computed tomography angiography (CTA) before discharge, at 6 months, at 1 year, and annually thereafter in uneventful cases. In complicated cases, follow-up was adjusted accordingly. The local Institutional Review Board approved the study (protocol No. S-594/2013). Individual patient consent was waived.

TEVAR at the time of presentation. Following common dissection protocols,7 patients presenting in an uncomplicated status are generally transferred to an intensive care unit for strict blood pressure control and pain management after initial CTA. Close CT monitoring is regularly performed on days 1, 4, and 10 after admission. In patients with persisting or recurrent pain despite medication or hemodynamic instability within the acute phase, the frequency of CT scans is adjusted accordingly. Depending on the clinical course and the imaging results, the patients are either scheduled for TEVAR or continue BMT under intensive care unit monitoring for 14 days. After an uneventful 14 days, the disease is considered stable and the patient is referred to a regular ward and assigned to follow-up. A reonset of symptoms or morphologic changes reinduce the surveillance process and eventual crossing over to TEVAR.

General institutional IMH treatment protocol. Patients in whom a complicated course is suspected by clinical and/or radiologic findings are treated with immediate

CTA data evaluation. CTA scans were performed following a standardized aortic protocol8 (native, arterial, and venous imaging phases). In group I, the initial

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

3

-

CTA scan at admission, the preoperative scan leading to the indication for TEVAR, the first postoperative in-hospital scan, and the first and last follow-up CTA scans were evaluated. In group II, the initial CTA scan at admission, the last in-hospital scan, and the last followup CTA scan were reviewed. Specific morphologic study parameters evaluated in both groups over time were maximum aortic diameter (MAD) and IMH thickness (IMHT) on corresponding levels on follow-up scans (centerline measurements; Supplementary Fig 1, online only), any secondary aortic disease (ie, penetrating aortic ulcer [PAU], aortic dissection, intramural blood pool) at admission or de novo formation between two scans, and IMH regression and aortic remodeling.9 CT data of all patients were interdisciplinary and consensually reviewed by an attending vascular surgeon and an attending radiologist in an independent fashion at the time of the study. Endovascular procedures. The implantation protocol has been previously published.10 Proximal and distal landing zones were intended to be at least 20 mm (according to the instructions for use of most of the devices) and were preoperatively evaluated using threedimensional reconstruction software and centerline measurements (iNtuition [TeraRecon, Foster City, Calif]; OsiriX PRO [acycan, Rochester, NY]). For IMHB, stent designs were preferably chosen without bare stents and maximal oversizing was 10%, related to centerline-based aortic diameter measurements at the proximal landing zones. The treatment goal was to cover the entire length of the IMH or, if this was not possible because of anatomic restrictions, to cover at least the most diseased part of the aorta. The protocols for left subclavian artery revascularization and cerebrospinal fluid drainage have been previously published.11,12 Further procedural details are displayed in Table II. Definitions and follow-up. IMH was defined as circular or crescentic thickening of the aortic wall without evidence of blood flow within the aortic wall on imaging.5 Localization was classified according to the Stanford classification used for aortic dissection.4 PAU was defined as an ulceration of an aortic atherosclerotic plaque through the internal elastic lamina into the aortic media (Supplementary Fig 2, online only).3,4 Ulcerlike projection (ULP) was defined as blood-filled pouches that protrude from the lumen into the thrombuscontaining aortic wall, which enhance to the same degree as the aortic lumen after administration of contrast material (Fig 1).13,14 Areas of focal contrast enhancement within the intramural hematoma were defined as intramural blood pools (Fig 2).15 A complicated course of IMHB was defined by imaging means in case of occurrence of MAD or IMHT increase, appearance of new or progressive concomitant pathologic

Table II. Technical procedural details Group I, TEVAR (n ¼ 28) Duration of procedure, minutes Contrast material load, mL CSFD No. of stent grafts

74 (35-285) 118 (30-240) 13 of 28 (46.4) 2 (1-3)

Device Gore CTAG

n ¼ 17

Gore TAG

n¼6

Gore CTAG þ TAG

n¼2

Medtronic Valiant

n¼2

Medtronic Talent

n¼1

Reversed trombone

3 of 28 (10.7)

Access Transfemoral

n ¼ 25

Iliac conduit

n¼1

Percutaneous

n¼1

Transfemoral þ transbrachial

n¼1

AIHA/rapid pacing

3 of 28 (10.7)/7 of 28 (25)

Covered LSA

10 of 28 (35.7)

Primary LSA revascularization

5 of 28 (17.9)

Covered aortic zones 2-4

3

2-5

6

3-4

5

3-5

9

4-5

5

AIHA, Adenosine-induced heart arrest; CSFD, cerebrospinal fluid drainage; LSA, left subclavian artery; TEVAR, thoracic endovascular aortic repair. Continuous data are expressed as median (range). Categorical data are presented as number (%).

changes (PAU, ULP, intramural blood pooling), transition to aortic dissection, or clinical deterioration (persisting or recurrent events of aortic pain and/or therapy-resistant hypertension). IMHB regression was defined as decrease in maximum IMHT. Complete aortic remodeling was defined as complete IMH regression along the thoracic descending aorta. Reporting standards for TEVAR, published in 2010 by the Society for Vascular Surgery, were applied.9 The median follow-up in group I was 32 months (range, 4 days-134 months; follow-up, 96% complete; n ¼ 1 lost to follow-up). In group II, the median follow-up was 40 months (range, 6 days to 103 months; follow-up, 92% complete; n ¼ 1 lost to follow-up). Statistics. Patient and disease characteristics are described as percentages or median (range). Continuous data were expressed using mean 6 standard deviation. PASW Statistics (version 18.0; IBM Corp, Somers, NY) was used for statistical analysis. Normal distribution was assessed using the Shapiro-Wilk test. Differences

4

Bischoff et al

Journal of Vascular Surgery ---

2016

Fig 1. Contrast-enhanced computed tomography angiography (CTA) of a 62-year-old man presenting with sudden thoracic pain who underwent thoracic endovascular aortic repair (TEVAR). CTA at admission (A) showed intramural hematoma type B (IMHB) of the descending thoracic aorta. The patient received antihypertensive treatment as well as pain management in an intensive care unit setting, and control scans during the hospital stay showed stable disease. Therefore, conservative treatment with a quadruple antihypertensive medication was continued. Planned control CTA 6 weeks after discharge (B, sagittal orientation; C, axial orientation; D, three-dimensional reconstruction) showed disease progression in terms of a new evolved ulcerlike projection (ULP) in the descending thoracic aorta; aortic diameter at this site progressed to 43 mm. Elective TEVAR was performed, with successful exclusion of the ULP (E). A, Anterior.

between group I and group II were evaluated using either Student t-test or Mann-Whitney U test. Differences in MAD and IMHT within groups I and II were analyzed using Wilcoxon signed rank test. Survival was calculated by Kaplan-Meier analysis. A P value of < .05 was considered to indicate statistical significance.

RESULTS Clinical presentation In group I, all but three patients presented symptomatic. The most common symptom was sudden thoracic pain (23/28), followed by abdominal pain and dyspnea (5/28 each). Ten of 28 patients had refractory pain. Nine of 28 had therapy-resistant blood pressure, and 3 of 28 cases showed signs of spinal cord ischemia (paraplegia in 2 and transient paraparesis in 1). In the three asymptomatic patients, IMHB was an incidental finding on CT scans performed for cardiac decompensation, trauma, and idiopathic lung fibrosis. In group II, 9 of 13 patients presented with thoracic and 4 of 13 patients with abdominal pain. In addition, one patient showed a transient paraparesis at the time of hospital admission. Radiologic findings and clinical course Group I. Median MAD at presentation was 42 mm (range, 30-71 mm). Preoperative median IMHT was

14 mm (range, 7-39 mm). IMH extent according to aortic zones is displayed in Table I. In addition to clinical signs (aortic pain, n ¼ 5; hemodynamic instability, n ¼ 2), 7 of 28 patients had concomitant pathologic changes on the initial CTA scan and therefore underwent TEVAR without further imaging (median, 2 days; range, 0-11 days). IMHB in conjunction with PAU was detected in 4 of 7 patients, and intramural blood pools were seen in 3 of 7 cases of those undergoing TEVAR at the time of presentation. In 3 of 7 patients, CT additionally showed bilateral pleural effusions. There was no case presenting with aortic rupture (Fig 3). The remaining 21 patients were kept under intensive care unit surveillance continuing BMT. Following the institutional IMH protocol, control CTA scans were performed. Hence, 21 of 28 patients with a complicated course (75%) had at least two preoperative scans (median, three; range, 2-8) before crossing over to TEVAR after a median of 10 days (range, 2-223 days). Clinical TEVAR indications in these 21 patients were therapyrefractory aortic pain in 10 and uncontrollable hypertension in nine. Furthermore, MAD increased in 16 of 21 cases by a median of 2 mm (range, 0.5-5 mm). IMHT increased in 10 of 21 cases by a median of 2.5 mm (range, 0.3-8.3 mm). In 6 of 21 patients, imaging revealed new or progressive formation of ULP, whereas in two additional patients, a de novo PAU was detected. In 10 of 21

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

5

-

Fig 2. Contrast-enhanced computed tomography angiography (CTA) of a 45-year-old woman presenting with lower back pain, dyspnea, and primary paraparesis who underwent thoracic endovascular aortic repair (TEVAR). CTA at admission (A) showed intramural hematoma type B (IMHB) of the descending thoracic aorta. As primary paraparesis completely resolved and there were no signs of end-organ ischemia, the patient was transferred to an intensive care unit for conservative treatment. Due to recurrent pain, a control scan was performed on the third in-hospital day, revealing numerous new intramural blood pools along the entire IMH (B, sagittal orientation; C, axial orientation). Therefore, TEVAR was performed 5 days after admission. Five-year follow-up displayed complete remodeling of the aorta (D, sagittal orientation; E, centerline measurement).

patients, newly developed or progressive intramural blood pools were detected. In one patient, IMHB progressed to type B aortic dissection. In 14 of 21 patients, progressive pleural effusions were documented (Fig 3). Group II. Initial median MAD was 37 mm (range, 27-51 mm), significantly lower compared with group I (P ¼ .016; Table III). IMHT was 11 mm (range, 4-20 mm), significantly lower compared with group I (P ¼ .036; Table III). IMH extent according to aortic zones is shown in Table I. Five of 13 patients had secondary pathologic findings on admission CTA showing concomitant intramural blood pools in four patients and ULP in one case. IMHT decreased during hospital stay in 12 of 13 cases (median, 2.5 mm; range, 11 to 1 mm), whereas in one patient, IMHT remained stable. MAD remained constant in all cases. There were no pleural or pericardial effusions. Clinically, pain as well as hemodynamics could be successfully treated in all cases. In-hospital mortality, morbidity, and secondary procedures Group I. The in-hospital mortality rate was 3.6% (1/28). A 77-year-old patient died 10 days after TEVAR of severe aortic valve insufficiency in association with a procedurerelated type A IMH recognized on postoperative CTA. The patient denied further treatment. Seven of 28

patients suffered from morbidity. Severe morbidity occurred in two patients, requiring secondary open interventions, both for retrograde type A dissection (7%; Table IV). In the first case, the dissection was detected on completion angiography, and urgent supracoronary replacement of the ascending aorta was performed. The second patient experienced recurrent aortic pain on the second postoperative day. CTA showed a retrograde type A dissection, which was treated by supracoronary ascending aorta and hemiarch replacement. Moderate and mild morbidity is summarized in the Supplementary Table (online only). After TEVAR, all patients experienced pain relief. Primary technical success was achieved in 25 of 28 cases. One patient had a primary type Ib endoleak. The patient was scheduled for endovascular repair but died of pneumonia 74 days after discharge before a scheduled reintervention. There were two cases of retrograde type A dissection after TEVAR, requiring open conversion (Table IV). The secondary technical end points of the TEVAR procedures are detailed in Table II. In total, the 30-day aortic reintervention rate was 21.4% (6/28). Secondary endovascular procedures during hospital stay were undertaken in two patients. In the first case, proximal endograft extension was performed on the third postoperative day because of IMH progression.

6

Bischoff et al

Journal of Vascular Surgery ---

2016

Fig 3. Indications leading to thoracic endovascular aortic repair (TEVAR; multiple selections possible). Upper figure, Initial TEVAR. Lower figure, Cases crossing over to TEVAR during surveillance. IMH, Intramural hematoma; IMHT, IMH thickness; MAD, maximum aortic diameter; PAU, penetrating aortic ulcer; RR, hypertension; progr., progressive; ULP, ulcerlike projection.

The second patient underwent distal endograft extension because of a newly developed aortic type B dissection below the stented aortic segment 21 days after the primary procedure. Two additional patients required secondary procedures within 30 days. One patient was retransferred to our institution 21 days after TEVAR with respiratory insufficiency. CTA revealed a hematothorax in conjunction with IMH progression suggestive of rupture at the site of the proximal landing zone. The patient underwent successful proximal endograft extension. A second patient presented with recurrent aortic pain 25 days after the TEVAR procedure. CTA showed new blood pools below the distal endograft end, and distal endograft extension was performed. The two patients who initially presented with paraplegia remained neurologically impaired, despite perioperative cerebrospinal fluid drainage (spinal cord ischemia grading system: 3a). Group II. In group II, there was no mortality during the hospital stay. In one patient, CTA on in-hospital day 14 showed a new intramural blood pool. The anticipated length of aortic coverage by TEVAR would have been 220 mm, starting from the orifice of the left subclavian artery, which was thought to carry a high risk for spinal

cord ischemia. Therefore, conservative treatment was continued. A control CT scan on day 22 after admission showed no further changes. Morphologic outcome during follow-up Group I. In group I, MAD decreased significantly in 24 of 27 patients (excluding the one patient lost to follow-up) by a median of 5 mm (range, 17 mm to þ5 mm; P ¼ .001, Wilcoxon signed rank test). In comparison to group II, MAD decreased significantly by absolute and relative means (P ¼ .027 and P ¼ .039, respectively, Mann-Whitney U test). Complete aortic remodeling was observed in 7 of 27 patients. Regression of IMHT was seen in 20 of 27 cases with a median regression of 7 mm (range, 14 to 1 mm). In 8 of these 20 patients, additional complete remodeling occurred in the stented area. The median IMHT on the last follow-up scan available was 5 mm (range, 0-34 mm), significantly smaller compared with CTA at admission (P ¼ .001, Wilcoxon signed rank test; Table III). Decrease in IMHT over time showed no difference between group I and group II, neither by absolute nor by relative means (P ¼ .97, t-test, vs P ¼ .113, Mann-Whitney U test; Table III). Three patients developed de novo ULP during follow-up, without need for reintervention.

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

7

-

Table III. Summary of morphologic results Group I, TEVAR MAD at presentation, mm MAD in last FU-CTA, mm MAD within groups over time, mm

Group II, BMT 37 (27-51)

.016

36 (25-71)

37 (26-49)

.842

CT1: 42 (30-71) FU-CTA: 36 (25-71)

.001 CT1: 37 (27-51) FU-CTA: 37 (26-49)

Decrease of MAD over time, mm Decrease of MAD over time, %

P

42 (30-71)

.161

5 (17 to þ5)

2 (7 to þ10)

.027

10 (35 to þ13)

3 (23 to þ33)

.039

Maximum IMHT at presentation, mm

14 (7-39)

11 (4-20)

.036

Maximum IMHT in last FU-CTA, mm

5 (0-34)

2 (0-4)

.029

IMHT within groups over time, mm

CT1: 14 (7-39) FU-CTA: 5 (0-34)

.001 CT1: 11 (4-20) FU-CT: 2 (0-4)

Decrease of IMHT over time, mm Decrease of IMHT over time, %

9 (19 to 0.3) 60 (100 to 2)

.001

9 (20 to 1)

.970

80 (100 to 12)

.113

BMT, Best medical therapy; CTA, computed tomography angiography; CT1, CTA scan leading to TEVAR (group I)/admission CTA (group II); FU-CTA, last available follow-up CTA scan in both groups; IMHT, intramural hematoma thickness; MAD, maximum aortic diameter; TEVAR, thoracic endovascular aortic repair. Data are expressed as median (range).

Group II. In group II, MAD decreased in 7 of 12 patients (excluding the one patient lost to follow-up) by 5 mm (range, 7 to 2 mm). In 4 of 12 patients, MAD remained stable; the remaining patient showed a progress in MAD of 10 mm. Median MAD decrease between the initial CTA scan and the last follow-up CTA scan was 2 mm (range, 7 to þ10 mm), whereas in contrast to group I, MAD decrease between these two scans showed no significance (P ¼ .161, Wilcoxon signed rank test; Table III). Complete aortic remodeling was observed in 3 of 12. IMH regression was seen in 9 of 12 with a median regression of 9 mm (6 to 16 mm). Median IMHT in the last follow-up scan was 2 mm (range, 0-4 mm), showing significant reduction in comparison to CTA at admission (P ¼ .001, Wilcoxon signed rank test). The absolute difference of IMHT over time was 9 mm (range, 20 to 1 mm), whereas the relative decrease was 80% (range, 100% to 12%; Table III). In group II, there was no need for surgical intervention during follow-up (Fig 4). Survival In group I, 10 patients died after discharge, yielding a total mortality rate of 40.7% (10/27) during a median followup of 32 months (range, 4 days to 134 months). The estimated survival rates in group I after 12, 24, and 48 months were 81.5%, 77.8%, and 67%, respectively. The causes of death were cancer (n ¼ 3), pneumonia (n ¼ 2), deadly trauma, and gastrointestinal bleeding (n ¼ 1 each). In three patients, the cause of death remained indeterminate. In group I, no death was found to be aorta or procedure related. In group II, there were no reported

deaths during a median follow-up of 40 months (range, 6 days to 103 months).

DISCUSSION In this study, the majority of patients (25/41) with IMHB showed signs of disease progression (clinically and/or on imaging). In 21 of 28 patients treated by TEVAR (group I), the indication was given due to dynamic changes displayed on consecutive CT scans, oftendbut not alwaysdin conjunction with clinical symptoms (15/21; Fig 3). After TEVAR, a significant reduction in both MAD and IMHT could be observed. Despite a rising awareness within the vascular community, the clinical workup of IMHB is insufficiently defined. Lacking high-evidence guidelines, classifications, indications for treatment, surgical interventions, and followup schedules of IMHB are derived from classic aortic dissection.16 Today, conservative treatment is advocated in uncomplicated cases of IMHB.3 However, the results reported herein challenge the term uncomplicated. Reported high-risk features of IMHB include therapyresistant arterial hypertension, persistent pain, MAD $40 mm, IMHT $10 mm, pleural effusion, and PAU.3-5,17 According to the available literature, a MAD $40 mm is the major predictor of IMH progression.5,18,19 IMHT has equally been identified as a strong risk factor.5,20 Proposed cutoff values range between 10 mm and 15 mm. In the TEVAR group analyzed herein, the mean initial aortic diameter was >40 mm, and an increase in aortic diameter was observed in more than half of cases (16/21) during the hospital stay. Almost half of the TEVAR

8

Journal of Vascular Surgery

Bischoff et al

---

2016

Table IV. Case-by-case presentation of secondary procedures (n ¼ 6)

No.

Gender/ age, years

IMH zones PLZ

TEVAR

Stentcovered zones

Oversizing of PLZ

Complication

1

F/50

3-8

3

Gore CTAG 31/ 31/150

3-4

7%

2

F/77

3-9

3

Gore TAG 31/31/ 150

3-5

20%

3

F/76

3-6

4

Gore CTAG 28/ 28/100

4-5

15%

Rupture

4

M/82

3-9

2

Gore CTAG 31/ 26/100, 28/ 28/150, 28/ 28/15 Subclaviancarotid bypass

2-5

10%

5

M/74

2-9

2

Gore TAG 40/ 40/150

2-4

6

M/67

3-7

2

Medtronic Valiant 36/ 36/150, 34/34/ 200, 34/34/ 100

2-5

Secondary procedure

Recurrent pain Distal endograft New blood extension pooling Gore CTAG 31/ 31/150 þ 31/31/ 100

Interval TEVARSP, days 25

FU Alive

3

Dead

Proximal endograft extension Gore CTAG 31/ 31/150

21

Dead

Type B dissection

Distal endograft extension Gore CTAG 31/ 31/150

21

Dead

9%

Retrograde type A dissection

Supracoronary ascending replacement

0

NA

17%

Retrograde type A dissection

Supracoronary ascending and hemiarch replacement

2

NA

Recurrent pain Proximal New blood endograft pooling extension Gore TAG 34/ 34/150 þ Medtronic Valiant 38/ 38/100 þ covered LSA

FU, Follow-up; IMH, intramural hematoma; LSA, left subclavian artery; NA, not applicable; PLZ, proximal landing zone; SP, secondary procedure; TEVAR, thoracic endovascular aortic repair.

patients (10/21) showed an increase in IMHT. Regarding the two treatment groups, both MAD and IMHT were significantly higher in the TEVAR cohort. Along with others, large or progressive pleural effusions are valued as an ominous sign.21,22 Progressive pleural effusions were seen in two-thirds of the TEVAR patients in this series. The presence of a concomitant PAU, regardless of whether it was found at the time of the initial evaluation or during follow-up, has been identified as another independent predictor for IMH progression.23 Greater PAU diameter ($20 mm) and depth ($10 mm) have been associated with an increased risk for complications in several studies.23-25 All patients with concomitant PAU underwent TEVAR after the first CTA scan (Supplementary Fig 2, online only). PAU is distinguished from ULP, which is not associated with atherosclerotic disease. However, differentiating between the two entities may be difficult in the absence of pathologic assessment.15 The correct terminology remains

controversial. Evangelista et al picture the existence of a PAU as being causative of IMH formation, in contrast to development of ULP as an early complication of IMH and therefore disease progression.26 ULPs should also be distinguished from another type of contrast enhancement in the aortic wall, the so-called intramural blood pools, which have no obvious connection to the aortic lumen.15 The prognostic significance of both, ULPs and intramural blood pools, remains uncertain. Sueyoshi et al described a 20% to 40% risk of ULP formation in IMH, representing a relevant factor in disease progression.27 TEVAR was proposed as a valuable therapeutic strategy.20,27,28 In our clinical experience, the CT finding of an intramural blood pool as well as new or progressive formation of PAU and ULP, respectively, was frequently associated with new-onset pain. Although evidence is lacking, we would classify these findings as a complicated course of IMHB and favor close surveillance. With regard to additional clinical symptoms, TEVAR may be indicated.

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

9

-

Fig 4. Contrast-enhanced computed tomography angiography (CTA) of an 81-year-old female nonsurgical patient. CTA at admission (A) displayed an intramural hematoma type B (IMHB) of the descending thoracic aorta. CTA after 14 days (B) showed stable disease. A follow-up scan obtained 13 months after presentation detected distinct IMH regression in axial (C) and sagittal (D) slices.

On the one hand, the available data show that the clinical and CT morphologic course of IMHB is more benign compared with classic type B dissection. As a result, current guidelines recommend medical therapy and repetitive imaging in uncomplicated cases.3-5,29 This study shows that IMHB patients can be treated medically with excellent midterm results regarding survival, significant IMH regression, or complete aortic remodeling over time (Fig 4). This is in line with prior reports.26,30,31 However, imaging is challenging the term uncomplicated. In retrospect, almost all of our patients showed risk factors for disease progression at some point. However, not all underwent TEVAR. In one case, there were hard criteria for endovascular repair (transformation of the IMH into uncomplicated type B aortic dissection in a patient in whom an intramural blood pool had been observed during the hospital stay). Because of the high risk for spinal cord ischemia, TEVAR was not performed, and the patient is in good health. On the other hand, data on TEVAR for IMHB are scant. Recent publications concluded that TEVAR for IMHB is safe and effective.32,33 Herein, TEVAR could be performed in a technically successful fashion in 25 of 28 patients. According to the literature, the reported mortality rates are about 5%, which compares favorably with our results.5 Lavingia et al reported that about 75% of the patients showed complete remodeling of the aorta (total aortic diameter/true lumen diameter ratio of 1) after

TEVAR (Fig 2).32 This is confirmed by this study since we observed either IMH regression (20/27) or complete remodeling (7/27) and significant reduction in maximum IMHT and aortic diameter during follow-up. Herein, we report an aortic reintervention rate in 6 of 28 TEVAR cases. Despite being in line with others reports,32-34 this represents a high number, requiring a ruthless discussion. All six secondary procedures (including two open conversions) were performed within 30 days after TEVAR. This underlines the highly fragile wall conditions in IMH patients (Table IV). As in complicated type B dissections, it has been hypothesized that IMH wall conditions become more stable after the acute phase because of fibrotic changes.5 Therefore, it seems appropriate to perform TEVAR in the subacute or chronic phase. Yet, the application of this concept is questionable in patients with undetermined radiologic findings but clinical symptoms. To minimize the risk of new lacerations, stent graft oversizing should not exceed 10% and balloon dilation of landing zones should not be performed. Some authors reported the risk of new intimal lesions or pseudoaneurysm formation due to endograft landing in IMH-affected zones.35,36 Therefore, Evangelista et al recently recommended that stent grafts should be anchored at least 15 mm within the intact aortic wall.5 At the authors’ institution, these three maxims are fully accepted. However, a detailed analysis of the patients in whom reinterventions

10

Journal of Vascular Surgery

Bischoff et al

---

were performed revealed that a deviation from these principles was present in some of the cases. In patient 1 (Table IV), the oversizing at the proximal landing zone was 7%. However, at the distal end, the endograft was landed in an area still affected by the IMH. As the endoprosthesis used was a nontapered device, the respective oversizing at the distal landing zone was 22%. In patient 2 (Table IV), the oversizing at the proximal landing zone was 20%, and the endograft was landed within the IMH. In patient 3 (Table IV), the endograft was equally proximally anchored in a diseased aortic wall with an oversizing of 15%. Therefore, the assumption seems appropriate that these complications were procedure induced. A major concern in IMHB patients is the risk for TEVAR-induced retrograde type A dissection.37 Undoubtedly, these procedures should be reserved for centers with circadian cardiothoracic surgery service. In one of our two patients with retrograde type A dissection (patient 6 [Table IV]), a stent graft-induced dissection cannot be ruled out as the oversizing at the proximal landing zone was 17%. These cases underline the necessity of careful sizing in IMH patients, equally with respect to the distal landing zone diameter. Clearly, landing zones should be nondiseased. Yet, the application of this concept may be limited in a huge portion of patients because of the linked necessity of head or visceral vessel revascularization. Herein, we observed proximal or distal endograft extension due to disease progression in four cases. In three of four cases, we did not cover the entire length of the IMH because of the vast IMH extent. Instead, spot stenting was performed to cover the most diseased part of the aorta. Therefore, the pathologic process was not fully excluded. In retrospect, this should be avoided whenever possible. However, the increased risk for spinal cord ischemia resulting from long-segment aortic coverage must be taken into account.38,39 This study is limited by its retrospective design and the underlying 10-year study period. IMHB represents an evolving field regarding clinical strategies and imaging, and decision making has changed over time. Therefore, it is not entirely possible to define in retrospect why some patients were not treated by TEVAR. A selection bias is inevitable. In addition, the BMT group does not represent a contemporary control group as they were retrospectively identified by database search and individual case review. Furthermore, the follow-up in the conservative group has been inconsistent during the last years as these patients were not in the main focus of follow-up. Therefore, clinical and radiologic changes in this cohort of patients may be under-reported. In light of the results of conservatively treated IMHB patients reported herein, our findings raise two questions: (1) What clinical and/or radiologic findings define the thresholds for a complicated course in IMHB patients? and (2) Which of these cases represent a justifiable indication for TEVAR? Both questions are unanswerable in

2016

a retrospective analysis but must be addressed by future prospective studies to generate data-driven decision algorithms.

CONCLUSIONS The clinical and morphologic data of the presented IMHB patients illustrate the contemporary perception of IMHB at the authors’ institution as a dynamic and somewhat incalculable aortic condition. Conservative treatment represents a valuable option in patients with uncomplicated IMHB with respect to aortic morphology and survival. However, the results suggest that IMHB shows a progressive course in the majority of patients. Yet, the significance on patient outcome remains ambiguous, notably with regard to discrete imaging findings. In progressive cases, TEVAR can be performed with sufficient technical feasibility and favorable results regarding aortic remodeling. However, there is a considerable morbidity and aortic reintervention rate. This reflects the underlying technical challenges and the fragile aortic wall conditions associated with IMHB. Despite published treatment recommendations, patient allocation to conservative or endovascular treatment remains individual and must be made by synopsis of clinical and radiologic findings. Prospective data are needed.

AUTHOR CONTRIBUTIONS Conception and design: MB, KM, DB, DK Analysis and interpretation: MB, KM, DB, DK Data collection: MB, KM, MW Writing the article: MB Critical revision of the article: MB, KM, MW, DB, DK Final approval of the article: MB, KM, MW, DB, DK Statistical analysis: MB, KM Obtained funding: Not applicable Overall responsibility: MB

REFERENCES 1. Krukenberg E. Beiträge zur Frage des Aneurysma dissecans. Beitr Pathol Anat Allg Pathol 1929;67:329-51. 2. Evangelista A, Mukherjee D, Mehta RH, O’Gara PT, Fattori R, Cooper JV, et al. Acute intramural hematoma of the aorta: a mystery in evolution. Circulation 2005;111:1063-70. 3. Erbel R, Aboyans V, Boileau C, Bossone E, Bartolomeo RD, Eggebrecht H, et al. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC). Eur Heart J 2014;35:2873-926. 4. Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE Jr, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/ SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology,

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

11

-

Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation 2010;121:e266-369. 5. Evangelista A, Czerny M, Nienaber C, Schepens M, Rousseau H, Cao P, et al. Interdisciplinary expert consensus on management of type B intramural haematoma and penetrating aortic ulcer. Eur J Cardiothorac Surg 2015;47:209-17. 6. International Statistical Classification of Diseases and Related Health Problems, 10th revision 2016. Available at: http://www.who.int/classifications/icd/en/. Accessed January 1, 2016. 7. Nienaber CA, Clough RE. Management of acute aortic dissection. Lancet 2015;385:800-11. 8. Bischoff MS, Geisbüsch P, Kotelis D, Müller-Eschner M, Hyhlik-Dürr A, Böckler D. Clinical significance of type II endoleaks after thoracic endovascular aortic repair. J Vasc Surg 2013;58:643-50. 9. Fillinger MF, Greenberg RK, McKinsey JF, Chaikof EL. Reporting standards for thoracic endovascular aortic repair (TEVAR). J Vasc Surg 2010;52:1022-33. 1033.e15. 10. Geisbüsch P, Kotelis D, Weber TF, Hyhlik-Dürr A, Kauczor HU, Böckler D. Early and midterm results after endovascular stent graft repair of penetrating aortic ulcers. J Vasc Surg 2008;48:1361-8. 11. Kotelis D, Bianchini C, Kovacs B, Müller T, Bischoff M, Böckler D. Early experience with automatic pressurecontrolled cerebrospinal fluid drainage during thoracic endovascular aortic repair. J Endovasc Ther 2015;22:368-72. 12. Kotelis D, Geisbüsch P, Hinz U, Hyhlik-Dürr A, von TenggKobligk H, Allenberg JR, et al. Short and midterm results after left subclavian artery coverage during endovascular repair of the thoracic aorta. J Vasc Surg 2009;50:1285-92. 13. Park GM, Ahn JM, Kim DH, Kang JW, Song JM, Kang DH, et al. Distal aortic intramural hematoma: clinical importance of focal contrast enhancement on CT images. Radiology 2011;259:100-8. 14. Sueyoshi E, Matsuoka Y, Imada T, Okimoto T, Sakamoto I, Hayashi K. New development of an ulcerlike projection in aortic intramural hematoma: CT evaluation. Radiology 2002;224:536-41. 15. Kruse MJ, Johnson PT, Fishman EK, Zimmerman SL. Aortic intramural hematoma: review of high-risk imaging features. J Cardiovasc Comput Tomogr 2013;7:267-72. 16. Böckler D, Hyhlik-Dürr A, Hakimi M, Weber TF, Geisbüsch P. Type B aortic dissections: treating the many to benefit the few? J Endovasc Ther 2009;16(Suppl 1):I80-90. 17. Sheikh AS, Ali K, Mazhar S. Acute aortic syndrome. Circulation 2013;128:1122-7. 18. Moizumi Y, Komatsu T, Motoyoshi N, Tabayashi K. Clinical features and long-term outcome of type A and type B intramural hematoma of the aorta. J Thorac Cardiovasc Surg 2004;127:421-7. 19. Sueyoshi E, Sakamoto I, Uetani M, Matsuoka Y. CT analysis of the growth rate of aortic diameter affected by acute type B intramural hematoma. AJR Am J Roentgenol 2006;186 (Suppl 2):S414-20. 20. Sueyoshi E, Imada T, Sakamoto I, Matsuoka Y, Hayashi K. Analysis of predictive factors for progression of type B aortic intramural hematoma with computed tomography. J Vasc Surg 2002;35:1179-83. 21. Evangelista A, Dominguez R, Sebastia C, Salas A, PermanyerMiralda G, Avegliano G, et al. Prognostic value of clinical and morphologic findings in short-term evolution of aortic intramural haematoma. Therapeutic implications. Eur Heart J 2004;25:81-7. 22. Sueyoshi E, Onitsuka H, Nagayama H, Sakamoto I, Uetani M. Endovascular repair of aortic dissection and intramural

23.

24.

25.

26.

27.

28. 29.

30.

31.

32.

33.

34. 35.

36.

37.

38.

39.

hematoma: indications and serial changes. Springerplus 2014;3:670. Ganaha F, Miller DC, Sugimoto K, Do YS, Minamiguchi H, Saito H, et al. Prognosis of aortic intramural hematoma with and without penetrating atherosclerotic ulcer: a clinical and radiological analysis. Circulation 2002;106:342-8. Schlatter T, Auriol J, Marcheix B, Lebbadi M, Marachet MA, Dang-Tran KD, et al. Type B intramural hematoma of the aorta: evolution and prognostic value of intimal erosion. J Vasc Interv Radiol 2011;22:533-41. Kitai T, Kaji S, Yamamuro A, Tani T, Kinoshita M, Ehara N, et al. Detection of intimal defect by 64-row multidetector computed tomography in patients with acute aortic intramural hematoma. Circulation 2011;124(Suppl):S174-8. Evangelista A, Dominguez R, Sebastia C, Salas A, PermanyerMiralda G, Avegliano G, et al. Long-term follow-up of aortic intramural hematoma: predictors of outcome. Circulation 2003;108:583-9. Sueyoshi E, Nagayama H, Hashizume K, Eishi K, Sakamoto I, Uetani M. Computed tomography evaluation of aortic remodeling after endovascular treatment for complicated ulcer-like projection in patients with type B aortic intramural hematoma. J Vasc Surg 2014;59:693-9. Song JK. Diagnosis of aortic intramural haematoma. Heart 2004;90:368-71. Tolenaar JL, Harris KM, Upchurch GR Jr, Evangelista A, Moll FL, di Eusanio M, et al. The differences and similarities between intramural hematoma of the descending aorta and acute type B dissection. J Vasc Surg 2013;58:1498-504. Nishigami K, Tsuchiya T, Shono H, Horibata Y, Honda T. Disappearance of aortic intramural hematoma and its significance to the prognosis. Circulation 2000;102(Suppl 3):III243-7. Attia R, Young C, Fallouh HB, Scarci M. In patients with acute aortic intramural haematoma is open surgical repair superior to conservative management? Interact Cardiovasc Thorac Surg 2009;9:868-71. Lavingia KS, Ahanchi SS, Redlinger RE, Udgiri NR, Panneton JM. Aortic remodeling after thoracic endovascular aortic repair for intramural hematoma. J Vasc Surg 2014;60: 929-35; discussion: 935-6. Felisaz A, Dufranc J, Heyndrickx M, Palcau L, Gouicem D, Berger L. Midterm results of type B intramural hematoma endovascular treatment. Ann Vasc Surg 2015;29:898-904. Mestres G, Capoccia L, Riambau V. Secondary procedures in thoracic aorta. J Cardiovasc Surg (Torino) 2014;55:731-40. Ponte M, Dias Ferreira N, Bettencourt N, Caeiro D, Fonseca M, Albuquerque A, et al. Aortic intramural hematoma: an unpredictable evolution. Rev Port Cardiol 2014;33:467.e1-7. Monnin-Bares V, Thony F, Rodiere M, Bach V, Hacini R, Blin D, et al. Endovascular stent-graft management of aortic intramural hematomas. J Vasc Interv Radiol 2009;20:713-21. Canaud L, Ozdemir BA, Patterson BO, Holt PJ, Loftus IM, Thompson MM. Retrograde aortic dissection after thoracic endovascular aortic repair. Ann Surg 2014;260:389-95. Bischoff MS, Peters AS, Kotelis D, Böckler D, Hyhlik-Dürr A. Strategische Überlegungen bei mit spinaler Ischämie assoziertem intramuralem Hämatom. Gefässchirurgie 2011;16:346-9. Bischoff MS, Di Luozzo G, Griepp EB, Griepp RB. Spinal cord preservation in thoracoabdominal aneurysm repair. Perspect Vasc Surg Endovasc Ther 2011;23:214-22.

Submitted Mar 29, 2016; accepted May 10, 2016.

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

11.e1

Bischoff et al

Journal of Vascular Surgery ---

Supplementary Fig 1 (online only). A, Centerline postprocessing of computed tomography angiography (CTA) scan showing intramural hematoma Stanford type B (IMHB). B, The perpendicular view shows an example of crescentic thickening of the aortic wall (arrow), demonstrating the evaluation of maximum aortic diameter (MAD) and intramural hematoma thickness (IMHT).

Supplementary Fig 2 (online only). Examples of penetrating aortic ulcer (PAU) showing a distinct correlation to atherosclerotic disease. A, PAU in a native-phase computed tomography (CT) scan, axial orientation. B, Contrast-enhanced CT angiography (CTA) at the same level, axial orientation. C, Sagittal view of a thoracic PAU.

2016

Journal of Vascular Surgery Volume

-,

Number

Bischoff et al

11.e2

-

Supplementary Table (online only). Morbidity in thoracic endovascular aortic repair (TEVAR) group Group I, TEVAR (n ¼ 28) Severe morbidity Retrograde type A dissection with open conversion

n¼2

Moderate morbidity Pneumonia with prolonged intubation

n¼2

Lymphatic fistula with surgical revision

n¼2

Mild morbidity Acute renal failure without need for hemodialysis

n¼1

Dissection of LSA with transient left hand paresthesia

n¼1

Postpuncture headache after CSFD

n¼1

CSFD, Cerebrospinal fluid drainage; LSA, left subclavian artery.