- Email: [email protected]
Treatment strategy based on the natural course for patients with spontaneous isolated superior mesenteric artery dissection
From the Society for Vascular Surgery
Treatment strategy based on the natural course for patients with spontaneous isolated superior mesenteric artery dissection Seon-Hee Heo, MD,a Young-Wook Kim, MD, PhD,a Shin-Young Woo, RN, MPH,a Yang-Jin Park, MD, PhD,a Kwang-Bo Park, MD, PhD,b and Duk-Kyung Kim, MD, PhD,c Seoul, Republic of Korea
ABSTRACT Objective: The optimal treatment of spontaneous isolated superior mesenteric artery dissection (SISMAD) is still not known, and we sought to determine an optimal treatment strategy for patients with SISMAD based on its natural clinical course. Methods: We retrospectively reviewed consecutive patients with SISMAD treated from 2001 through 2016. Diagnosis and angiographic type of SISMAD were determined with contrast-enhanced computed tomography (CT) scan, and the clinical features were obtained using a fixed form questionnaire. All patients were treated conservatively, except for five who unselectively underwent primary interventional treatment. For the follow-up examinations, clinical features and morphologic changes of superior mesenteric artery dissection were examined with CT angiography every 6 to 12 months. Results: During the past 15 years, 116 patients with SISMAD (male, 92%; mean age, 54.7 6 10.8 years; symptomatic, 76%) were encountered. Clinical features and morphologic changes on CT examinations were available in 100% and 88% of the patients, respectively, during the mean follow-up of 53 6 39 months (range, 1-173 months). Of 83 symptomatic patients managed conservatively, 96% achieved pain resolution; 4% experienced prolonged pain, including one patient with bowel gangrene. After pain resolution, 20% of patients developed late recurrence of abdominal pain, which was relieved with conservative management, whereas two patients (12%) required surgery to treat bowel stricture. Follow-up examinations (n ¼ 102) by CT angiography revealed no change in 34%, partial or complete remodeling in 63%, aneurysmal change in 2%, and dissection progression in 1% of the patients. Antithrombotic therapy offered no beneficial effects on either clinical or morphologic outcomes. Conclusions: With conservative treatment, the majority of patients with SISMAD showed clinical improvement and no morphologic changes during long-term follow-up. We thus recommend a conservative management strategy as the first-line treatment for patients with SISMAD, regardless of angiographic type. (J Vasc Surg 2016;-:1-10.)
Spontaneous isolated superior mesenteric artery dissection (SISMAD) is a rare vascular disorder. A classic autopsy series1 reported an incidence of superior mesenteric artery (SMA) dissection of 0.08%. However, this autopsy series included many stillborn and neonatal
From the Division of Vascular Surgery,a Department of Radiology,b and Department of Cardiology,c Samsung Medical Center, Sungkyunkwan University School of Medicine. Author conflict of interest: none. Presented at the International Forum at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Correspondence: Young-Wook Kim, MD, PhD, Division of Vascular Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Irwon-ro 81, Gangnam-gu, Seoul 06351, Korea (e-mail: young52.kim@ samsung.com; [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.2016.10.109
deaths, and if stillbirths and neonates were excluded, its prevalence was 0.09%. Bauersfeld2 reported the first case series of SISMAD in 1947. Since then, an increasing number of case reports and case series have described this rare vascular disorder.3-5 Current advanced imaging techniques, such as abdominal computed tomography (CT), have facilitated the detection of SISMAD.6 Previous reports7,8 have indicated that SISMAD is more prevalent in men in their 50s. Moreover, SISMAD can be detected in asymptomatic patients8 on contrastenhanced CT, with arterial dissection usually starting at the proximal SMA.9 Although the precise mechanism underlying this observation is unknown, SISMAD is more frequently reported in Asian countries.7-14 Currently, open surgical treatment is rarely attempted for SISMAD, but whether primary SMA stenting or conservative treatment is the optimal initial management strategy of SISMAD remains controversial. Some authors7,13,15 have proposed conservative treatment, whereas others16,17 have recommended primary SMA 1
2
Journal of Vascular Surgery
Heo et al
---
stenting. No prospective study has yet been performed to examine whether conservative or endovascular treatment is more efficacious or longer lasting free of a poor outcome. Moreover, previous reports regarding an optimal treatment for SISMAD have been based on short-term results with relatively small numbers of patients. In this study, we investigated the natural course of patients with SISMAD by reviewing the long-term results of conservative management. In addition, we aimed to identify angiographic features predictive of late disease outcomes.
METHODS This study was approved by the Institutional Review Board of Samsung Medical Center. An informed consent was waived for this retrospective review.
ARTICLE HIGHLIGHTS d
d
d
d
d
d
Study subjects. After approval by the Institutional Review Board, we reviewed a prospectively collected database of the patients with SISMAD. Information was collected from 2001 through 2016. Diagnosis and collection of patient information. SISMAD was diagnosed according to the typical duallumen signs of SMA on axial views of contrastenhanced CT or dissected segment of the SMA on CT angiography. Catheter-based angiography was performed selectively for the purposes of SMA stenting or confirmation of mesenteric collateral circulation. When the true and false SMA lumens were completely occluded, diagnosis of SISMAD was made when the SMA dissection was confirmed on another slice or previous CT images. Patients with concomitant aortic and SMA dissection or iatrogenic SMA dissection were excluded from this study. For angiographic classification of SISMAD, we modified our previous classification system8 by adding type Ib (presence of the outflow branch from the false lumen in patients with type I SISMAD). Angiographic type is based on the patency of true and false lumens, presence of entry and re-entry sites, and presence of the outflow branch from the false lumen of the dissected SMA segment (Fig 1). To determine the SISMAD angiographic type, various CT images were reviewed, including axial, coronal, and sagittal views. Reconstructed images were also reviewed. At the initial presentation of each patient with SISMAD, we obtained clinical features and coexisting diseases with a fixed form questionnaire (Table I). We measured the pain severity score on a 10-point visual analog scale and investigated the duration of abdominal pain before pain resolution. Pain resolution was defined as pain relief requiring no more analgesics and oral intake without postprandial pain. Treatment of SISMAD. For treatment of SISMAD, all patients were treated conservatively, with the exception of
2016
Significance: This study examines outcomes of conservative treatment for patients with spontaneous isolated superior mesenteric artery dissection (SISMAD). Type of Research: Retrospective cohort study Take Home Message: Conservative treatment of 83 symptomatic patients with SISMAD resulted in pain resolution in 96%. Late recurrence occurred in 20% of patients and was successfully managed conservatively in 88%. Recommendation: The authors suggest conservative treatment as initial therapy for patients with SISMAD, regardless of angiographic type. Strength of Recommendation: 2. Weak Level of Evidence: B. Moderate
five patients who unselectively underwent primary interventional treatment (open surgery, three; primary SMA stenting, two) early in the study period. Conservative treatment included close observation of patients for development of bowel ischemia, fasting, intravenous fluid therapy or total parenteral nutrition therapy until oral intake was available, and pain control with intravenous nonsteroidal anti-inflammatory drugs or narcotic analgesics. After pain relief, patients were allowed to progress from a sip of water to regular food step by step. Regarding antithrombotic therapy, we routinely administered anticoagulation therapy with intravenous heparin or low-molecular-weight heparin or antiplatelet agent until 2009, after which we selectively used antithrombotic therapy only for patients showing an occlusion (type III) or severe compromised blood flow (type IIb) to the SMA. In selecting the anticoagulation agent or antiplatelet agent, the presence of atherosclerotic change and the use of an antiplatelet agent before the onset of SISMAD were considered. Antithrombotic therapy was discontinued after the acute phase of the SMA dissection. This change was motivated by the concern that the risk of intramural hemorrhage at the dissected segment of the SMA would be increased by antithrombotic therapy.18 SMA intervention (either surgical or endovascular) was considered only when abdominal pain persisted for a week or longer with conservative treatment, when the patient experienced pain suggestive of bowel gangrene, or in symptomatic patients with short life expectancy, such as patients with coexisting advanced cancer. Follow-up examinations. Once abdominal pain subsided with conservative treatment and the patients were able to tolerate oral intake, they were periodically followed up at an outpatient clinic to assess the recurrence of abdominal symptoms and morphologic
Journal of Vascular Surgery Volume
-,
Number
Heo et al
3
-
Fig 1. Angiographic classification and subtype frequencies of spontaneous isolated superior mesenteric artery dissection (SISMAD) according to lumen patency, superior mesenteric artery (SMA) entry or re-entry sites, and presence of distal outflow from the false lumen.
changes of the SMA lesion at 6-month intervals. If no symptom recurrence or interval changes on CT images occurred after 1 year, CT angiography was performed annually. During the follow-up period, the frequency of persistent or recurrent abdominal symptoms (pain or discomfort), requirement of surgical or endovascular intervention, and morphologic changes on follow-up CT images were reviewed. Follow-up results were compared by angiographic type. For follow-up of morphologic changes in the SMA lesion, multidetector CT angiography was performed to
examine the extent of distal or proximal progression of the arterial dissection, occlusion or remodeling of the false or true lumen, aneurysmal changes of the false lumen, or development of another visceral artery dissection. Complete remodeling was defined as the absence of residual arterial dissection and the absence of arterial narrowing on the follow-up CT image. Partial remodeling was defined as improved luminal patency of the SMA but with luminal narrowing or arterial dissection on the follow-up CT image. Aneurysmal dilation was defined as focal SMA dilation >150% of the normal adjacent SMA.13
4
Journal of Vascular Surgery
Heo et al
---
Table I. History-taking sheet at initial presentation for patients with spontaneous isolated superior mesenteric artery dissection (SISMAD) Name Patient ID No. Date of birth
(yy/mm/dd)
Sex
Male/female
Abdominal symptoms
Absent Abdominal/back pain Nausea/vomiting Bloody stool
Symptom onset
Sudden/gradual
Pain location
Epigastric or upper abdominal Whole abdomen Back
Pain nature
Steady/intermittent
Pain aggravation by eating
Yes/no
Pain severity score
( ) on a 10-point visual analog scale
Duration of initial pain
( ) days ( ) hours
Abdominal tenderness
Absent/present
Coexisting disease or risk
Hypertension Ex or current smoker Connective tissue disease Intra-abdominal cancer
ID, Identification.
Statistical analysis. Continuous data are reported as mean 6 standard deviation (range), whereas categorical data are reported as number (%). Data were compared with the Kruskal-Wallis test or Fisher exact test. All statistical tests were conducted at the .05 level using SPSS software version 23 (IBM Corp, Armonk, NY).
RESULTS During the past 15 years, 116 patients with SISMAD (mean age, 54.7 6 10.8 years; male, 92%; symptomatic, n ¼ 88 [76%]) were encountered at a single institution. All the patients received contrast-enhanced CT scans or CT angiography. Among them, six patients (5.3%) underwent selective catheter-based angiography for the purpose of SMA stenting (n ¼ 3), confirmation of mesenteric collateral circulation due to recurrent abdominal pain (n ¼ 1), and preoperative evaluation before rectal cancer surgery (n ¼ 1). During the followup period (mean, 53 6 39 months; range, 1-173 months), clinical features and characteristics were obtained from all patients. These included 78 patients (67%) who visited outpatient clinics and 38 patients (33%) who completed telephone surveys. Follow-up CT angiographic examinations were conducted in 102 patients (88%). In the 88 symptomatic patients, the mean abdominal pain severity score was 8 on a 10-point visual analog scale. The mode of pain onset was sudden in 88% of
2016
the patients and aggravated by food intake in 20% of the patients. After five patients who unselectively underwent interventional treatment were excluded, 83 symptomatic patients (94%) were treated with conservative management. Eighty patients (96%) showed pain resolution within a mean period of 3.7 days (range, 1-12 days) after pain onset, whereas abdominal pain was incompletely relieved in three patients (4%) during a mean period of 40 days (range, 19-61 days) with conservative treatment. Of the three patients with prolonged, unrelieved pain, one was treated with SMA stenting, one was treated with open surgery, and one refused further treatment (Fig 2). Of the 80 patients who showed pain resolution with conservative treatment, 16 (20%) developed recurrent abdominal pain during the follow-up period. The time at onset of the recurrent abdominal pain varied from 4 days to 943 days (mean, 190 days; median, 103 days) after the initial resolution of abdominal pain. The mean number of pain recurrences was 1.7 6 0.9 (range, 1-3) in 16 patients with recurrent abdominal symptoms; among them, 5 patients were hospitalized for pain control and nutrition support, whereas the other 11 patients could be treated as outpatients. The recurrent abdominal pain was relieved with conservative management in 14 patients (88%); 2 patients (12%) required bowel resection for treatment of small bowel stricture. These patients showed a very short segmental narrowing of the small bowel. Symptomatic patients were more common in type IIa (90%), IIb (85%), and III (100%) angiographic types than in type Ia (42%) or Ib (50%; P < .001). However, no significant differences were observed between the angiographic types regarding pain severity, characteristics, or duration at initial presentation (Table II). Comparison of clinical outcomes in symptomatic patients during the follow-up period between angiographic types did not identify any significant differences in recurrent abdominal symptoms, requirement for SMA intervention, or requirement for bowel resection between groups (Table III). Moreover, antithrombotic therapy did not yield any beneficial effects with respect to either clinical or morphologic outcomes of SISMAD (Table IV). During the follow-up period, 102 (88%) patients underwent follow-up CT angiography. On the follow-up CT angiography scans, 64 (63%) remodeled dissected lesions (partially remodeled, 37%; completely remodeled, 26%), 35 (34%) lesions categorized as no change, 2 (2%) lesions with aneurysmal changes, and 1 (1%) lesion with SMA dissection progression were observed. Mean time of duration of complete remodeling was 16 6 16 months (range, 3-63 months) after initial pain relief. It was shown that 61% of complete remodeling was found within 12 months and 81% of complete remodeling was found within 24 months after initial pain relief. No SMA ruptures occurred during the follow-up period. Comparison of the
Journal of Vascular Surgery Volume
-,
Number
Heo et al
5
-
Fig 2. Treatments of spontaneous isolated superior mesenteric artery dissection (SISMAD) and their long-term results. SMA, Superior mesenteric artery.
morphologic changes between angiographic types in symptomatic patients with SISMAD revealed that SMA lesion remodeling was more frequent in type III lesions, whereas “no change” was more frequent in type I lesions (Table III). Fig 3 illustrates typical SMA lesion morphologic changes as observed by follow-up CT angiography. Among all patients with SISMAD, six (5%) underwent laparotomy. Four patients underwent laparotomy within 1 month, whereas two patients underwent laparotomy at 1 month after symptom onset. The six laparotomies included three retrograde aorto-SMA bypasses for symptomatic patients, which were performed early in the study on an unselective basis (Fig 4). Among these three patients, one (0.8%) showed thrombotic SMA occlusion and small bowel gangrene, whereas the other two showed no evidence of bowel gangrene. For the patient with bowel gangrene, we performed segmental resection of the gangrenous small bowel and SMA revascularization. Two (1.7%) delayed laparotomies were required because of recurrent abdominal pain at 1.5 months and
3.8 months after initial pain resolution, respectively. At the laparotomies, we found short segmental stricture of the small bowel but no evidence of bowel ischemia. We performed segmental resection of the small bowel without SMA intervention for those patients. There were 28 asymptomatic patients in whom SISMAD was incidentally detected during the workup for abdominal cancer (n ¼ 8), health care screening (n ¼ 2), and workup for abdominal aortic aneurysm (n ¼ 3) or other benign diseases (n ¼ 15). When we reviewed angiographic features in 28 asymptomatic patients, we found that they consisted of 17 (61%) type I and 11 (39%) type II. There was no patient with type III. We found that type I SISMAD was more common in asymptomatic patients. During the follow-up period, none of the asymptomatic patients developed recurrent abdominal symptoms or required late interventional treatment. When the treatment results of symptomatic and asymptomatic patients were assessed together, 95% of all patients with SISMAD showed pain resolution
6
Journal of Vascular Surgery
Heo et al
---
2016
Table II. Comparison of demographic and clinical features of spontaneous isolated superior mesenteric artery dissection (SISMAD) patients by angiographic type (N ¼ 116) Type I Type
Type Ia
No. of patients
19 (16)
Age, years
62 (33-80)
Male
Type II Type Ib
Type IIa
Type IIb
Type III
12 (10)
10 (9)
67 (58)
8 (7) 57.5 (40-85)
.054a
61 (91)
8 (100)
.554b
9 (90)
57 (85)
8 (100)
<.001b
6 (50)
1 (10)
10 (15)
0
d
7 (5-10)
8 (3-10)
8 (6-10)
7 (88)
6 (100)
9 (100)
0
1 (17)
2 (22)
1 (1-14)
1.5 (1-8)
2 (1-4)
0
0
0
50.5 (39-78)
49.5 (45-80)
16 (84)
12 (100)
10 (100)
Symptomatic patients
8 (42)
6 (50)
Asymptomatic patients
11 (58)
Pain severity score Sudden onset of pain Pain after eating Duration of pain, days Prolonged pain >1 week
P NA
50 (36-82)
9 (2-10)
.804a
50 (88)
5 (63)
.179b
10/49 (20)
3/7 (43)
.364b
2.5 (1-11)
.681a
2 (3)
1 (12.5)
.453b
8 (1-10)
1 (1-19)
Coexisting disease Hypertension
7 (37)
3 (25)
6 (60)
24 (36)
0
.094b
Ex or current smoker
6 (32)
5 (42)
5 (50)
27 (40)
4 (50)
.846b
Connective tissue disease
1 (5)
0
0
0
0
.422b
Intra-abdominal cancer
1 (5)
4 (33)
0
7 (10)
0
.099b
NA, Not applicable. Pain severity score was measured by patients on a 10-point visual analog scale. Hypertension was defined as systolic blood pressure >140 mm Hg, diastolic blood pressure >90 mm Hg, or use of antihypertensive medication. Continuous data are shown as median (range); categorical data are expressed as number (%). a Kruskal-Wallis test. b Fisher exact test.
Table III. Long-term follow-up results after conservative treatment for symptomatic spontaneous isolated superior mesenteric artery dissection (SISMAD) patients (n ¼ 83) by angiographic type Angiographic type, No. (%) Ia (n ¼ 8)
Ib (n ¼ 6)
IIa (n ¼ 9)
IIb (n ¼ 52)
III (n ¼ 8)
Initial pain relief
6 (75)
3 (50)
7 (78)
40 (80)
5 (71)
.570a
Recurrent pain
2 (25)
3 (50)
2 (22)
8 (16)
1 (14)
.356a
Required SMA intervention
0
0
0
2 (4)
0
1.000a
Bowel resection
0
0
2 (4)
0
.761a
P
Clinical course
1 (11)
Early (<30 days)
d
d
0
1
d
Late
d
d
1
1
d
Morphologic changes on follow-up CT images Available
7 (88)
6 (100)
9 (100)
49 (94)
5 (63)
Remodeling
2 (29)
0
4 (44)
28 (57)
5 (100)
.006a
NA
Complete
1 (14)
0
2 (22)
15 (31)
3 (60)
.242a
Partial
1 (14)
0
2 (22)
13 (27)
2 (40)
.596a
No change
5 (71)
6 (100)
5 (56)
20 (41)
0
.004a
Progression
0
0
0
1 (2)
0
1.000a
Aneurysmal change
0
0
0
2 (4)
0
1.000a
CT, Computed tomography; NA, not applicable; SMA, superior mesenteric artery. Complete remodeling was defined as no residual arterial dissection or arterial luminal narrowing. Partial remodeling was defined as improved luminal patency with residual luminal narrowing on the reconstructed CT angiography image. a Fisher exact test.
with conservative treatment. In contrast, interventional treatment was required for 4.5% of all patients with SISMAD.
During the follow-up period, eight (7%) confirmed deaths occurred. The causes of death were pneumonia (n ¼ 2), advanced cancer (n ¼ 1), hemorrhagic stroke
Journal of Vascular Surgery Volume
-,
Number
Heo et al
7
-
Table IV. Comparison of treatment results of symptomatic spontaneous isolated superior mesenteric artery dissection (SISMAD) patients (N ¼ 83) by use of antithrombotic therapy Antithrombotic therapy (n ¼ 27),a No. (%)
No antithrombotic therapy (n ¼ 56), No. (%)
Pb
Initial pain relief
8 (30)
11 (58)
.404
Recurrent pain
6 (22)
10 (18)
.726
Required SMA intervention
2 (7)
0
.103
Bowel resection
.246
Clinical course (n ¼ 83; 100%)
2 (7)
1 (2)
Early (<30 days)
1
0
Late
1
1
Morphologic changes on follow-up CT images (n ¼ 76 [92%]) Available
25 (93)
51 (91)
.323
Complete remodelingc
7 (28)
14 (28)
1.000
Partial remodelingd
7 (28)
11 (22)
.574
No change
11 (44)
25 (49)
.808
0
1 (2)
.671
1 (4)
1 (2)
.547
Distal progression Aneurysmal change
CT, Computed tomography; SMA, superior mesenteric artery. a The antithrombotic therapy group included 14 patients who received anticoagulant therapy with unfractionated heparin or low-molecular-weight heparin followed by oral antiplatelet agent (n ¼ 6) or not and 13 patients who received antiplatelet therapy only. b Fisher exact test. c Complete remodeling was defined as no residual arterial dissection or luminal narrowing. d Partial remodeling was defined as improved luminal patency but showing residual luminal narrowing on the reconstructed CT angiography image.
(n ¼ 1), fall-related injury (n ¼ 1), and unidentified cause (n ¼ 3). No deaths were thought to be due to SMA rupture or bowel ischemia.
DISCUSSION The main purpose of this study was to determine the long-term results of conservative treatment and to define the natural angiographic as well as clinical course of the patients with SISMAD. In addition, we aimed to identify angiographic features predictive of the late clinical outcome of disease. In our series, 92% of the patients with SISMAD were male, and the mean age of the patients was mid-50s (mean, 54.7 6 10.8 years; range, 33-85 years). Overall, 76% of the patients presented with sudden-onset severe abdominal or back pain. In the past, open surgery was performed to treat patients with SISMAD. However, the surgical mortality rate was high.17 Recently, conservative treatment5,7,11,13,15,19,20 or endovascular treatment with SMA stenting16,17,21-23 has been recommended for SISMAD. Each treatment strategy has specific advantages and disadvantages. For example, conservative treatment is noninvasive and does not leave foreign material in the artery. However, conservative treatment carries potential risks of progression of arterial dissection, SMA occlusion resulting in bowel gangrene, aneurysmal change, and SMA rupture. Moreover, conservative treatment can take longer to achieve symptom relief compared with SMA stent placement. SMA stenting can promptly improve SMA blood
flow and also reduce the risk of dissection progression. However, SMA stenting is associated with risk of procedure-related complications, can occlude collateral branches of the SMA, and can result in late development of stent-related complications such as stent thrombosis or stent restenosis. However, in practice, SMA rupture is rarely reported in patients with SISMAD.24 Gobble et al17 thoroughly reviewed the literature related to vascular surgery and radiology from 1998 to 2008 and summarized 106 documented cases of SISMAD (96 symptomatic and 10 asymptomatic). These patients were treated with expectant management (n ¼ 56), anticoagulation (n ¼ 23), open surgery (n ¼ 22), or endovascular stent placement (n ¼ 5). Among the 56 patients who underwent expectant management, only 55% were successfully managed. For patients in whom expectant management provided inadequate pain relief, open surgery or endovascular stent placement in the SMA was performed as a secondary treatment. Gobble et al also reported a higher mortality rate after surgical treatment in patients who failed to respond to expectant treatment compared with the other treatment groups. We believe that the higher mortality is related to the surgical treatment of SISMAD and not to the expectant management itself. A recent survey of 622 Chinese patients with SISMAD demonstrated that most of these patients were symptomatic and typically resided in large cities.14 They were managed with conservative (63.2%), surgical (3.2%), or endovascular (33.6%) treatment. Higher rates of
8
Heo et al
Journal of Vascular Surgery ---
2016
Fig 3. Remodeling of spontaneous isolated superior mesenteric artery dissection (SISMAD) lesions on follow-up computed tomography (CT) angiography. A, CT angiogram at initial presentation showing type III SISMAD in a 45-year-old man (left), follow-up CT angiogram at 2 months (middle), and follow-up CT angiogram at 6 months showing complete remodeling (right). B, CT angiogram at initial presentation showing type III SISMAD in a 57year-old man (left), catheter angiography showing developed mesenteric collateral circulation on the second day (middle), and follow-up CT angiogram at 7 months showing partial superior mesenteric artery (SMA) remodeling (right).
Fig 4. Operative findings in the six spontaneous isolated superior mesenteric artery dissection (SISMAD) patients who underwent laparotomy. SMA, Superior mesenteric artery.
symptom relief after surgical or endovascular treatment were observed compared with those after conservative treatment. It appears that many asymptomatic, undetected SISMAD patients were not included in their
cohort and long-term results were not described in their report. It is also likely that interventional treatment was performed earlier with lower threshold criteria than in our practice.
Journal of Vascular Surgery Volume
-,
Number
Heo et al
9
-
Various treatment algorithms have been proposed for patients with SISMAD.10,11,13,15,25 Li et al10 suggested conservative treatment in stable patients with patent true lumen flow, endovascular treatment in cases of true lumen stenosis and occlusion, and surgical treatment when endovascular therapy fails or there are signs of bowel infarction or arterial rupture. Cho et al11 recommended anticoagulant therapy for patients with symptomatic SISMAD with luminal narrowing of the SMA. Kim et al13 also recommended conservative treatment unless persistent abdominal pain or peritonitis signs are present. Most proposed treatment algorithms recommend conservative treatment for uncomplicated SISMAD; interventional treatment is generally recommended in cases of persistent abdominal pain, dissection progression, aneurysmal change of the arterial lesion, rupture, or bowel gangrene. These recommendations have usually been based on early treatment results from a small number of patients. No randomized prospective studies have yet compared the long-term results of endovascular treatment with those of conservative treatment for patients with SISMAD to determine the optimal treatment of SISMAD. Moreover, angiographic features that can adequately predict late complications or the natural course of SISMAD have not yet been identified. We observed that prolonged unrelieved abdominal symptoms were relatively common and lowered the quality of life and interfered with social life. However, these symptoms were rarely associated with bowel gangrene, even in cases of SMA occlusion (type III). Progression of arterial dissection and aneurysmal changes were also rarely detected. Moreover, SMA rupture was not found on our long-term follow-up examinations. Furthermore, we did not observe any difference in long-term clinical outcomes (SMA narrowing or occlusion [type II or III]) among the patients with SISMAD. About half of the symptomatic SISMAD lesions showed complete or partial remodeling on follow-up CT angiography. Interestingly, SMA lesion remodeling was more frequent in type III lesions, whereas “no change” was more frequent in type I lesions. We hypothesize that the early development of mesenteric collateral channels may play a role in the early resolution of pain in patients with occluded SMA (type III SISMAD). We previously reported clinical features of patients with SISMAD by angiographic type.8 In this study, we modified our SISMAD classification scheme by adding the status of the outflow branch from the false lumen of the type I lesion, in addition to the patency of the false or true lumen of the SMA and the entry or re-entry of the SMA dissection. We hypothesized that the presence of the outflow branch from the false lumen might reduce the risk of progression of the arterial dissection. Among the patients who underwent conservative therapy, 32% were treated with antithrombotic therapy with
anticoagulant or antiplatelet agents. In contrast, 68% were not treated with antithrombotic therapy. No significant differences in clinical or morphologic changes were observed between the two groups. Currently, we do not routinely use antithrombotic agents. We used therapeutic doses of low-molecular-weight heparin only for patients with occluded or severely compromised SMA flow. The primary purpose of anticoagulation therapy in patients with SISMAD is to reduce the risk of bowel ischemia secondary to the SMA thrombotic occlusion at the diseased arterial segment. We propose that the concept of anticoagulation therapy for patients with arterial dissection originated from experiences with patients with spontaneous dissection of the carotid artery.26 In current practice, anticoagulation therapy is well accepted as the primary treatment of cervical artery dissection, whereas no consensus has been reached with respect to anticoagulation therapy in the context of SISMAD management. In addition, we were concerned about the potential risk of anticoagulation therapy to lead to further subintimal hemorrhage at the site of SMA dissection. Our long-term observations of patients with SISMAD revealed that the majority of patients with SISMAD showed pain resolution with conservative treatment. Moreover, the majority of patients exhibited morphologic improvements or showed no change on CT images, regardless of angiographic type. Recurrent abdominal pain developed in about one in five of all patients after conservative treatment. However, most of these cases were successfully managed conservatively. Bowel gangrene was rare; moreover, interventional treatment was required in only 4.5% of all patients with SISMAD and 6% of all symptomatic patients with SISMAD. Despite that we extended duration of follow-up and increased the number of SISMAD patients to our previous observation,7 this study still has the limitation of a small number of patients, precluding a multivariate analysis for risk factors of pain recurrence or requirement of interventional treatment.
CONCLUSIONS Our long-term observation of SISMAD patients demonstrated that the majority of SISMAD patients have a benign natural clinical course. Therefore, we recommend conservative management for all patients with SISMAD as the first-line treatment strategy regardless of the angiographic features. Primary stenting of the SMA can be considered for patients with severe prolonged abdominal pain while they are being conservatively managed, for patients with clinical signs of bowel gangrene, or in symptomatic patients with short life expectancy. After conservative treatment, a careful long-term follow-up is recommended, considering the potential risk of late occurrence of bowel stricture, especially in symptomatic patients with SISMAD.
10
Heo et al
Journal of Vascular Surgery ---
AUTHOR CONTRIBUTIONS Conception and design: SH, YW Analysis and interpretation: SH, YW Data collection: SH, YW, SY, YJ, KB, DK Writing the article: SH, YW Critical revision of the article: SH, YW Final approval of the article: SH, YW, SY, YJ, KB, DK Statistical analysis: SH, SY Obtained funding: Not applicable Overall responsibility: YW
REFERENCES 1. Foord AG, Lewis RD. Primary dissecting aneurysms of peripheral and pulmonary arteries: dissecting hemorrhage of media. Arch Pathol 1959;68:553-77. 2. Bauersfeld SR. Dissecting aneurysm of the aorta; a presentation of 15 cases and a review of the recent literature. Ann Intern Med 1947;26:873-89. 3. Goueffic Y, Costargent A, Dupas B, Heymann MF, Chaillou P, Patra P. Superior mesenteric artery dissection: case report. J Vasc Surg 2002;35:1003-5. 4. Javerliat I, Becquemin JP, d’Audiffret A. Spontaneous isolated dissection of the superior mesenteric artery. Eur J Vasc Endovasc Surg 2003;25:180-4. 5. Cho YP, Ko GY, Kim HK, Moon KM, Kwon TW. Conservative management of symptomatic spontaneous isolated dissection of the superior mesenteric artery. Br J Surg 2009;96: 720-3. 6. Li Z, Ding H, Shan Z, Du J, Yao C, Chang G, et al. Initial and middle-term outcome of treatment for spontaneous isolated dissection of superior mesenteric artery. Medicine (Baltimore) 2015;94:e2058. 7. Park YJ, Park KB, Kim DI, Do YS, Kim DK, Kim YW. Natural history of spontaneous isolated superior mesenteric artery dissection derived from follow-up after conservative treatment. J Vasc Surg 2011;54:1727-33. 8. Yun WS, Kim YW, Park KB, Cho SK, Do YS, Lee KB, et al. Clinical and angiographic follow-up of spontaneous isolated superior mesenteric artery dissection. Eur J Vasc Endovasc Surg 2009;37:572-7. 9. Tomita K, Obara H, Sekimoto Y, Matsubara K, Watada S, Fujimura N, et al. Evolution of computed tomographic characteristics of spontaneous isolated superior mesenteric artery dissection during conservative management. Circ J 2016;80:1452-9. 10. Li DL, He YY, Alkalei AM, Chen XD, Jin W, Li M, et al. Management strategy for spontaneous isolated dissection of the superior mesenteric artery based on morphologic classification. J Vasc Surg 2014;59:165-72. 11. Cho BS, Lee MS, Lee MK, Choi YJ, Kim CN, Kang YJ, et al. Treatment guidelines for isolated dissection of the superior mesenteric artery based on follow-up CT findings. Eur J Vasc Endovasc Surg 2011;41:780-5. 12. Jia ZZ, Zhao JW, Tian F, Li SQ, Wang K, Wang Y, et al. Initial and middle-term results of treatment for symptomatic
2016
spontaneous isolated dissection of superior mesenteric artery. Eur J Vasc Endovasc Surg 2013;45:502-8. 13. Kim HK, Jung HK, Cho J, Lee JM, Huh S. Clinical and radiologic course of symptomatic spontaneous isolated dissection of the superior mesenteric artery treated with conservative management. J Vasc Surg 2014;59:465-72. 14. Luan JY, Guan X, Li X, Wang CM, Li TR, Zhang L, et al. Isolated superior mesenteric artery dissection in China. J Vasc Surg 2016;63:530-6. 15. Min SI, Yoon KC, Min SK, Ahn SH, Jae HJ, Chung JW, et al. Current strategy for the treatment of symptomatic spontaneous isolated dissection of superior mesenteric artery. J Vasc Surg 2011;54:461-6. 16. Li N, Lu QS, Zhou J, Bao JM, Zhao ZQ, Jing ZP. Endovascular stent placement for treatment of spontaneous isolated dissection of the superior mesenteric artery. Ann Vasc Surg 2014;28:445-51. 17. Gobble RM, Brill ER, Rockman CB, Hecht EM, Lamparello PJ, Jacobowitz GR, et al. Endovascular treatment of spontaneous dissections of the superior mesenteric artery. J Vasc Surg 2009;50:1326-32. 18. Dreier JP, Lurtzing F, Kappmeier M, Bohner G, Klingebiel R, Leistner S, et al. Delayed occlusion after internal carotid artery dissection under heparin. Cerebrovasc Dis 2004;18: 296-303. 19. Zhang X, Sun Y, Chen Z, Li X. Therapeutic regimen options for isolated superior mesenteric artery dissection. Vasc Endovascular Surg 2012;46:277-82. 20. Ko SH, Hye R, Frankel DA. Management of spontaneous isolated visceral artery dissection. Ann Vasc Surg 2015;29: 470-4. 21. AbuRahma AF, Stone PA, Bates MC, Welch CA. Angioplasty/ stenting of the superior mesenteric artery and celiac trunk: early and late outcomes. J Endovasc Ther 2003;10:1046-53. 22. Ozaki T, Kimura M, Yoshimura N, Hori Y, Takano T, Kamura T, et al. Endovascular treatment of spontaneous isolated dissecting aneurysm of the superior mesenteric artery using stent-assisted coil embolization. Cardiovasc Intervent Radiol 2006;29:435-7. 23. Casella IB, Bosch MA, Sousa WO Jr. Isolated spontaneous dissection of the superior mesenteric artery treated by percutaneous stent placement: case report. J Vasc Surg 2008;47:197-200. 24. Tameo MN, Dougherty MJ, Calligaro KD. Spontaneous dissection with rupture of the superior mesenteric artery from segmental arterial mediolysis. J Vasc Surg 2011;53: 1107-12. 25. Roussel A, Pellenc Q, Corcos O, Tresson P, Cerceau P, Francis F, et al. Spontaneous and isolated dissection of the superior mesenteric artery: proposal of a management algorithm. Ann Vasc Surg 2015;29:475-81. 26. Rao AS, Makaroun MS, Marone LK, Cho JS, Rhee R, Chaer RA. Long-term outcomes of internal carotid artery dissection. J Vasc Surg 2011;54:370-4; discussion: 375.
Submitted Aug 5, 2016; accepted Oct 24, 2016.
Treatment strategy based on the natural course for patients with spontaneous isolated superior mesenteric artery dissection Seon-Hee Heo, MD,a Young-Wook Kim, MD, PhD,a Shin-Young Woo, RN, MPH,a Yang-Jin Park, MD, PhD,a Kwang-Bo Park, MD, PhD,b and Duk-Kyung Kim, MD, PhD,c Seoul, Republic of Korea
ABSTRACT Objective: The optimal treatment of spontaneous isolated superior mesenteric artery dissection (SISMAD) is still not known, and we sought to determine an optimal treatment strategy for patients with SISMAD based on its natural clinical course. Methods: We retrospectively reviewed consecutive patients with SISMAD treated from 2001 through 2016. Diagnosis and angiographic type of SISMAD were determined with contrast-enhanced computed tomography (CT) scan, and the clinical features were obtained using a fixed form questionnaire. All patients were treated conservatively, except for five who unselectively underwent primary interventional treatment. For the follow-up examinations, clinical features and morphologic changes of superior mesenteric artery dissection were examined with CT angiography every 6 to 12 months. Results: During the past 15 years, 116 patients with SISMAD (male, 92%; mean age, 54.7 6 10.8 years; symptomatic, 76%) were encountered. Clinical features and morphologic changes on CT examinations were available in 100% and 88% of the patients, respectively, during the mean follow-up of 53 6 39 months (range, 1-173 months). Of 83 symptomatic patients managed conservatively, 96% achieved pain resolution; 4% experienced prolonged pain, including one patient with bowel gangrene. After pain resolution, 20% of patients developed late recurrence of abdominal pain, which was relieved with conservative management, whereas two patients (12%) required surgery to treat bowel stricture. Follow-up examinations (n ¼ 102) by CT angiography revealed no change in 34%, partial or complete remodeling in 63%, aneurysmal change in 2%, and dissection progression in 1% of the patients. Antithrombotic therapy offered no beneficial effects on either clinical or morphologic outcomes. Conclusions: With conservative treatment, the majority of patients with SISMAD showed clinical improvement and no morphologic changes during long-term follow-up. We thus recommend a conservative management strategy as the first-line treatment for patients with SISMAD, regardless of angiographic type. (J Vasc Surg 2016;-:1-10.)
Spontaneous isolated superior mesenteric artery dissection (SISMAD) is a rare vascular disorder. A classic autopsy series1 reported an incidence of superior mesenteric artery (SMA) dissection of 0.08%. However, this autopsy series included many stillborn and neonatal
From the Division of Vascular Surgery,a Department of Radiology,b and Department of Cardiology,c Samsung Medical Center, Sungkyunkwan University School of Medicine. Author conflict of interest: none. Presented at the International Forum at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Correspondence: Young-Wook Kim, MD, PhD, Division of Vascular Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Irwon-ro 81, Gangnam-gu, Seoul 06351, Korea (e-mail: young52.kim@ samsung.com; [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.2016.10.109
deaths, and if stillbirths and neonates were excluded, its prevalence was 0.09%. Bauersfeld2 reported the first case series of SISMAD in 1947. Since then, an increasing number of case reports and case series have described this rare vascular disorder.3-5 Current advanced imaging techniques, such as abdominal computed tomography (CT), have facilitated the detection of SISMAD.6 Previous reports7,8 have indicated that SISMAD is more prevalent in men in their 50s. Moreover, SISMAD can be detected in asymptomatic patients8 on contrastenhanced CT, with arterial dissection usually starting at the proximal SMA.9 Although the precise mechanism underlying this observation is unknown, SISMAD is more frequently reported in Asian countries.7-14 Currently, open surgical treatment is rarely attempted for SISMAD, but whether primary SMA stenting or conservative treatment is the optimal initial management strategy of SISMAD remains controversial. Some authors7,13,15 have proposed conservative treatment, whereas others16,17 have recommended primary SMA 1
2
Journal of Vascular Surgery
Heo et al
---
stenting. No prospective study has yet been performed to examine whether conservative or endovascular treatment is more efficacious or longer lasting free of a poor outcome. Moreover, previous reports regarding an optimal treatment for SISMAD have been based on short-term results with relatively small numbers of patients. In this study, we investigated the natural course of patients with SISMAD by reviewing the long-term results of conservative management. In addition, we aimed to identify angiographic features predictive of late disease outcomes.
METHODS This study was approved by the Institutional Review Board of Samsung Medical Center. An informed consent was waived for this retrospective review.
ARTICLE HIGHLIGHTS d
d
d
d
d
d
Study subjects. After approval by the Institutional Review Board, we reviewed a prospectively collected database of the patients with SISMAD. Information was collected from 2001 through 2016. Diagnosis and collection of patient information. SISMAD was diagnosed according to the typical duallumen signs of SMA on axial views of contrastenhanced CT or dissected segment of the SMA on CT angiography. Catheter-based angiography was performed selectively for the purposes of SMA stenting or confirmation of mesenteric collateral circulation. When the true and false SMA lumens were completely occluded, diagnosis of SISMAD was made when the SMA dissection was confirmed on another slice or previous CT images. Patients with concomitant aortic and SMA dissection or iatrogenic SMA dissection were excluded from this study. For angiographic classification of SISMAD, we modified our previous classification system8 by adding type Ib (presence of the outflow branch from the false lumen in patients with type I SISMAD). Angiographic type is based on the patency of true and false lumens, presence of entry and re-entry sites, and presence of the outflow branch from the false lumen of the dissected SMA segment (Fig 1). To determine the SISMAD angiographic type, various CT images were reviewed, including axial, coronal, and sagittal views. Reconstructed images were also reviewed. At the initial presentation of each patient with SISMAD, we obtained clinical features and coexisting diseases with a fixed form questionnaire (Table I). We measured the pain severity score on a 10-point visual analog scale and investigated the duration of abdominal pain before pain resolution. Pain resolution was defined as pain relief requiring no more analgesics and oral intake without postprandial pain. Treatment of SISMAD. For treatment of SISMAD, all patients were treated conservatively, with the exception of
2016
Significance: This study examines outcomes of conservative treatment for patients with spontaneous isolated superior mesenteric artery dissection (SISMAD). Type of Research: Retrospective cohort study Take Home Message: Conservative treatment of 83 symptomatic patients with SISMAD resulted in pain resolution in 96%. Late recurrence occurred in 20% of patients and was successfully managed conservatively in 88%. Recommendation: The authors suggest conservative treatment as initial therapy for patients with SISMAD, regardless of angiographic type. Strength of Recommendation: 2. Weak Level of Evidence: B. Moderate
five patients who unselectively underwent primary interventional treatment (open surgery, three; primary SMA stenting, two) early in the study period. Conservative treatment included close observation of patients for development of bowel ischemia, fasting, intravenous fluid therapy or total parenteral nutrition therapy until oral intake was available, and pain control with intravenous nonsteroidal anti-inflammatory drugs or narcotic analgesics. After pain relief, patients were allowed to progress from a sip of water to regular food step by step. Regarding antithrombotic therapy, we routinely administered anticoagulation therapy with intravenous heparin or low-molecular-weight heparin or antiplatelet agent until 2009, after which we selectively used antithrombotic therapy only for patients showing an occlusion (type III) or severe compromised blood flow (type IIb) to the SMA. In selecting the anticoagulation agent or antiplatelet agent, the presence of atherosclerotic change and the use of an antiplatelet agent before the onset of SISMAD were considered. Antithrombotic therapy was discontinued after the acute phase of the SMA dissection. This change was motivated by the concern that the risk of intramural hemorrhage at the dissected segment of the SMA would be increased by antithrombotic therapy.18 SMA intervention (either surgical or endovascular) was considered only when abdominal pain persisted for a week or longer with conservative treatment, when the patient experienced pain suggestive of bowel gangrene, or in symptomatic patients with short life expectancy, such as patients with coexisting advanced cancer. Follow-up examinations. Once abdominal pain subsided with conservative treatment and the patients were able to tolerate oral intake, they were periodically followed up at an outpatient clinic to assess the recurrence of abdominal symptoms and morphologic
Journal of Vascular Surgery Volume
-,
Number
Heo et al
3
-
Fig 1. Angiographic classification and subtype frequencies of spontaneous isolated superior mesenteric artery dissection (SISMAD) according to lumen patency, superior mesenteric artery (SMA) entry or re-entry sites, and presence of distal outflow from the false lumen.
changes of the SMA lesion at 6-month intervals. If no symptom recurrence or interval changes on CT images occurred after 1 year, CT angiography was performed annually. During the follow-up period, the frequency of persistent or recurrent abdominal symptoms (pain or discomfort), requirement of surgical or endovascular intervention, and morphologic changes on follow-up CT images were reviewed. Follow-up results were compared by angiographic type. For follow-up of morphologic changes in the SMA lesion, multidetector CT angiography was performed to
examine the extent of distal or proximal progression of the arterial dissection, occlusion or remodeling of the false or true lumen, aneurysmal changes of the false lumen, or development of another visceral artery dissection. Complete remodeling was defined as the absence of residual arterial dissection and the absence of arterial narrowing on the follow-up CT image. Partial remodeling was defined as improved luminal patency of the SMA but with luminal narrowing or arterial dissection on the follow-up CT image. Aneurysmal dilation was defined as focal SMA dilation >150% of the normal adjacent SMA.13
4
Journal of Vascular Surgery
Heo et al
---
Table I. History-taking sheet at initial presentation for patients with spontaneous isolated superior mesenteric artery dissection (SISMAD) Name Patient ID No. Date of birth
(yy/mm/dd)
Sex
Male/female
Abdominal symptoms
Absent Abdominal/back pain Nausea/vomiting Bloody stool
Symptom onset
Sudden/gradual
Pain location
Epigastric or upper abdominal Whole abdomen Back
Pain nature
Steady/intermittent
Pain aggravation by eating
Yes/no
Pain severity score
( ) on a 10-point visual analog scale
Duration of initial pain
( ) days ( ) hours
Abdominal tenderness
Absent/present
Coexisting disease or risk
Hypertension Ex or current smoker Connective tissue disease Intra-abdominal cancer
ID, Identification.
Statistical analysis. Continuous data are reported as mean 6 standard deviation (range), whereas categorical data are reported as number (%). Data were compared with the Kruskal-Wallis test or Fisher exact test. All statistical tests were conducted at the .05 level using SPSS software version 23 (IBM Corp, Armonk, NY).
RESULTS During the past 15 years, 116 patients with SISMAD (mean age, 54.7 6 10.8 years; male, 92%; symptomatic, n ¼ 88 [76%]) were encountered at a single institution. All the patients received contrast-enhanced CT scans or CT angiography. Among them, six patients (5.3%) underwent selective catheter-based angiography for the purpose of SMA stenting (n ¼ 3), confirmation of mesenteric collateral circulation due to recurrent abdominal pain (n ¼ 1), and preoperative evaluation before rectal cancer surgery (n ¼ 1). During the followup period (mean, 53 6 39 months; range, 1-173 months), clinical features and characteristics were obtained from all patients. These included 78 patients (67%) who visited outpatient clinics and 38 patients (33%) who completed telephone surveys. Follow-up CT angiographic examinations were conducted in 102 patients (88%). In the 88 symptomatic patients, the mean abdominal pain severity score was 8 on a 10-point visual analog scale. The mode of pain onset was sudden in 88% of
2016
the patients and aggravated by food intake in 20% of the patients. After five patients who unselectively underwent interventional treatment were excluded, 83 symptomatic patients (94%) were treated with conservative management. Eighty patients (96%) showed pain resolution within a mean period of 3.7 days (range, 1-12 days) after pain onset, whereas abdominal pain was incompletely relieved in three patients (4%) during a mean period of 40 days (range, 19-61 days) with conservative treatment. Of the three patients with prolonged, unrelieved pain, one was treated with SMA stenting, one was treated with open surgery, and one refused further treatment (Fig 2). Of the 80 patients who showed pain resolution with conservative treatment, 16 (20%) developed recurrent abdominal pain during the follow-up period. The time at onset of the recurrent abdominal pain varied from 4 days to 943 days (mean, 190 days; median, 103 days) after the initial resolution of abdominal pain. The mean number of pain recurrences was 1.7 6 0.9 (range, 1-3) in 16 patients with recurrent abdominal symptoms; among them, 5 patients were hospitalized for pain control and nutrition support, whereas the other 11 patients could be treated as outpatients. The recurrent abdominal pain was relieved with conservative management in 14 patients (88%); 2 patients (12%) required bowel resection for treatment of small bowel stricture. These patients showed a very short segmental narrowing of the small bowel. Symptomatic patients were more common in type IIa (90%), IIb (85%), and III (100%) angiographic types than in type Ia (42%) or Ib (50%; P < .001). However, no significant differences were observed between the angiographic types regarding pain severity, characteristics, or duration at initial presentation (Table II). Comparison of clinical outcomes in symptomatic patients during the follow-up period between angiographic types did not identify any significant differences in recurrent abdominal symptoms, requirement for SMA intervention, or requirement for bowel resection between groups (Table III). Moreover, antithrombotic therapy did not yield any beneficial effects with respect to either clinical or morphologic outcomes of SISMAD (Table IV). During the follow-up period, 102 (88%) patients underwent follow-up CT angiography. On the follow-up CT angiography scans, 64 (63%) remodeled dissected lesions (partially remodeled, 37%; completely remodeled, 26%), 35 (34%) lesions categorized as no change, 2 (2%) lesions with aneurysmal changes, and 1 (1%) lesion with SMA dissection progression were observed. Mean time of duration of complete remodeling was 16 6 16 months (range, 3-63 months) after initial pain relief. It was shown that 61% of complete remodeling was found within 12 months and 81% of complete remodeling was found within 24 months after initial pain relief. No SMA ruptures occurred during the follow-up period. Comparison of the
Journal of Vascular Surgery Volume
-,
Number
Heo et al
5
-
Fig 2. Treatments of spontaneous isolated superior mesenteric artery dissection (SISMAD) and their long-term results. SMA, Superior mesenteric artery.
morphologic changes between angiographic types in symptomatic patients with SISMAD revealed that SMA lesion remodeling was more frequent in type III lesions, whereas “no change” was more frequent in type I lesions (Table III). Fig 3 illustrates typical SMA lesion morphologic changes as observed by follow-up CT angiography. Among all patients with SISMAD, six (5%) underwent laparotomy. Four patients underwent laparotomy within 1 month, whereas two patients underwent laparotomy at 1 month after symptom onset. The six laparotomies included three retrograde aorto-SMA bypasses for symptomatic patients, which were performed early in the study on an unselective basis (Fig 4). Among these three patients, one (0.8%) showed thrombotic SMA occlusion and small bowel gangrene, whereas the other two showed no evidence of bowel gangrene. For the patient with bowel gangrene, we performed segmental resection of the gangrenous small bowel and SMA revascularization. Two (1.7%) delayed laparotomies were required because of recurrent abdominal pain at 1.5 months and
3.8 months after initial pain resolution, respectively. At the laparotomies, we found short segmental stricture of the small bowel but no evidence of bowel ischemia. We performed segmental resection of the small bowel without SMA intervention for those patients. There were 28 asymptomatic patients in whom SISMAD was incidentally detected during the workup for abdominal cancer (n ¼ 8), health care screening (n ¼ 2), and workup for abdominal aortic aneurysm (n ¼ 3) or other benign diseases (n ¼ 15). When we reviewed angiographic features in 28 asymptomatic patients, we found that they consisted of 17 (61%) type I and 11 (39%) type II. There was no patient with type III. We found that type I SISMAD was more common in asymptomatic patients. During the follow-up period, none of the asymptomatic patients developed recurrent abdominal symptoms or required late interventional treatment. When the treatment results of symptomatic and asymptomatic patients were assessed together, 95% of all patients with SISMAD showed pain resolution
6
Journal of Vascular Surgery
Heo et al
---
2016
Table II. Comparison of demographic and clinical features of spontaneous isolated superior mesenteric artery dissection (SISMAD) patients by angiographic type (N ¼ 116) Type I Type
Type Ia
No. of patients
19 (16)
Age, years
62 (33-80)
Male
Type II Type Ib
Type IIa
Type IIb
Type III
12 (10)
10 (9)
67 (58)
8 (7) 57.5 (40-85)
.054a
61 (91)
8 (100)
.554b
9 (90)
57 (85)
8 (100)
<.001b
6 (50)
1 (10)
10 (15)
0
d
7 (5-10)
8 (3-10)
8 (6-10)
7 (88)
6 (100)
9 (100)
0
1 (17)
2 (22)
1 (1-14)
1.5 (1-8)
2 (1-4)
0
0
0
50.5 (39-78)
49.5 (45-80)
16 (84)
12 (100)
10 (100)
Symptomatic patients
8 (42)
6 (50)
Asymptomatic patients
11 (58)
Pain severity score Sudden onset of pain Pain after eating Duration of pain, days Prolonged pain >1 week
P NA
50 (36-82)
9 (2-10)
.804a
50 (88)
5 (63)
.179b
10/49 (20)
3/7 (43)
.364b
2.5 (1-11)
.681a
2 (3)
1 (12.5)
.453b
8 (1-10)
1 (1-19)
Coexisting disease Hypertension
7 (37)
3 (25)
6 (60)
24 (36)
0
.094b
Ex or current smoker
6 (32)
5 (42)
5 (50)
27 (40)
4 (50)
.846b
Connective tissue disease
1 (5)
0
0
0
0
.422b
Intra-abdominal cancer
1 (5)
4 (33)
0
7 (10)
0
.099b
NA, Not applicable. Pain severity score was measured by patients on a 10-point visual analog scale. Hypertension was defined as systolic blood pressure >140 mm Hg, diastolic blood pressure >90 mm Hg, or use of antihypertensive medication. Continuous data are shown as median (range); categorical data are expressed as number (%). a Kruskal-Wallis test. b Fisher exact test.
Table III. Long-term follow-up results after conservative treatment for symptomatic spontaneous isolated superior mesenteric artery dissection (SISMAD) patients (n ¼ 83) by angiographic type Angiographic type, No. (%) Ia (n ¼ 8)
Ib (n ¼ 6)
IIa (n ¼ 9)
IIb (n ¼ 52)
III (n ¼ 8)
Initial pain relief
6 (75)
3 (50)
7 (78)
40 (80)
5 (71)
.570a
Recurrent pain
2 (25)
3 (50)
2 (22)
8 (16)
1 (14)
.356a
Required SMA intervention
0
0
0
2 (4)
0
1.000a
Bowel resection
0
0
2 (4)
0
.761a
P
Clinical course
1 (11)
Early (<30 days)
d
d
0
1
d
Late
d
d
1
1
d
Morphologic changes on follow-up CT images Available
7 (88)
6 (100)
9 (100)
49 (94)
5 (63)
Remodeling
2 (29)
0
4 (44)
28 (57)
5 (100)
.006a
NA
Complete
1 (14)
0
2 (22)
15 (31)
3 (60)
.242a
Partial
1 (14)
0
2 (22)
13 (27)
2 (40)
.596a
No change
5 (71)
6 (100)
5 (56)
20 (41)
0
.004a
Progression
0
0
0
1 (2)
0
1.000a
Aneurysmal change
0
0
0
2 (4)
0
1.000a
CT, Computed tomography; NA, not applicable; SMA, superior mesenteric artery. Complete remodeling was defined as no residual arterial dissection or arterial luminal narrowing. Partial remodeling was defined as improved luminal patency with residual luminal narrowing on the reconstructed CT angiography image. a Fisher exact test.
with conservative treatment. In contrast, interventional treatment was required for 4.5% of all patients with SISMAD.
During the follow-up period, eight (7%) confirmed deaths occurred. The causes of death were pneumonia (n ¼ 2), advanced cancer (n ¼ 1), hemorrhagic stroke
Journal of Vascular Surgery Volume
-,
Number
Heo et al
7
-
Table IV. Comparison of treatment results of symptomatic spontaneous isolated superior mesenteric artery dissection (SISMAD) patients (N ¼ 83) by use of antithrombotic therapy Antithrombotic therapy (n ¼ 27),a No. (%)
No antithrombotic therapy (n ¼ 56), No. (%)
Pb
Initial pain relief
8 (30)
11 (58)
.404
Recurrent pain
6 (22)
10 (18)
.726
Required SMA intervention
2 (7)
0
.103
Bowel resection
.246
Clinical course (n ¼ 83; 100%)
2 (7)
1 (2)
Early (<30 days)
1
0
Late
1
1
Morphologic changes on follow-up CT images (n ¼ 76 [92%]) Available
25 (93)
51 (91)
.323
Complete remodelingc
7 (28)
14 (28)
1.000
Partial remodelingd
7 (28)
11 (22)
.574
No change
11 (44)
25 (49)
.808
0
1 (2)
.671
1 (4)
1 (2)
.547
Distal progression Aneurysmal change
CT, Computed tomography; SMA, superior mesenteric artery. a The antithrombotic therapy group included 14 patients who received anticoagulant therapy with unfractionated heparin or low-molecular-weight heparin followed by oral antiplatelet agent (n ¼ 6) or not and 13 patients who received antiplatelet therapy only. b Fisher exact test. c Complete remodeling was defined as no residual arterial dissection or luminal narrowing. d Partial remodeling was defined as improved luminal patency but showing residual luminal narrowing on the reconstructed CT angiography image.
(n ¼ 1), fall-related injury (n ¼ 1), and unidentified cause (n ¼ 3). No deaths were thought to be due to SMA rupture or bowel ischemia.
DISCUSSION The main purpose of this study was to determine the long-term results of conservative treatment and to define the natural angiographic as well as clinical course of the patients with SISMAD. In addition, we aimed to identify angiographic features predictive of the late clinical outcome of disease. In our series, 92% of the patients with SISMAD were male, and the mean age of the patients was mid-50s (mean, 54.7 6 10.8 years; range, 33-85 years). Overall, 76% of the patients presented with sudden-onset severe abdominal or back pain. In the past, open surgery was performed to treat patients with SISMAD. However, the surgical mortality rate was high.17 Recently, conservative treatment5,7,11,13,15,19,20 or endovascular treatment with SMA stenting16,17,21-23 has been recommended for SISMAD. Each treatment strategy has specific advantages and disadvantages. For example, conservative treatment is noninvasive and does not leave foreign material in the artery. However, conservative treatment carries potential risks of progression of arterial dissection, SMA occlusion resulting in bowel gangrene, aneurysmal change, and SMA rupture. Moreover, conservative treatment can take longer to achieve symptom relief compared with SMA stent placement. SMA stenting can promptly improve SMA blood
flow and also reduce the risk of dissection progression. However, SMA stenting is associated with risk of procedure-related complications, can occlude collateral branches of the SMA, and can result in late development of stent-related complications such as stent thrombosis or stent restenosis. However, in practice, SMA rupture is rarely reported in patients with SISMAD.24 Gobble et al17 thoroughly reviewed the literature related to vascular surgery and radiology from 1998 to 2008 and summarized 106 documented cases of SISMAD (96 symptomatic and 10 asymptomatic). These patients were treated with expectant management (n ¼ 56), anticoagulation (n ¼ 23), open surgery (n ¼ 22), or endovascular stent placement (n ¼ 5). Among the 56 patients who underwent expectant management, only 55% were successfully managed. For patients in whom expectant management provided inadequate pain relief, open surgery or endovascular stent placement in the SMA was performed as a secondary treatment. Gobble et al also reported a higher mortality rate after surgical treatment in patients who failed to respond to expectant treatment compared with the other treatment groups. We believe that the higher mortality is related to the surgical treatment of SISMAD and not to the expectant management itself. A recent survey of 622 Chinese patients with SISMAD demonstrated that most of these patients were symptomatic and typically resided in large cities.14 They were managed with conservative (63.2%), surgical (3.2%), or endovascular (33.6%) treatment. Higher rates of
8
Heo et al
Journal of Vascular Surgery ---
2016
Fig 3. Remodeling of spontaneous isolated superior mesenteric artery dissection (SISMAD) lesions on follow-up computed tomography (CT) angiography. A, CT angiogram at initial presentation showing type III SISMAD in a 45-year-old man (left), follow-up CT angiogram at 2 months (middle), and follow-up CT angiogram at 6 months showing complete remodeling (right). B, CT angiogram at initial presentation showing type III SISMAD in a 57year-old man (left), catheter angiography showing developed mesenteric collateral circulation on the second day (middle), and follow-up CT angiogram at 7 months showing partial superior mesenteric artery (SMA) remodeling (right).
Fig 4. Operative findings in the six spontaneous isolated superior mesenteric artery dissection (SISMAD) patients who underwent laparotomy. SMA, Superior mesenteric artery.
symptom relief after surgical or endovascular treatment were observed compared with those after conservative treatment. It appears that many asymptomatic, undetected SISMAD patients were not included in their
cohort and long-term results were not described in their report. It is also likely that interventional treatment was performed earlier with lower threshold criteria than in our practice.
Journal of Vascular Surgery Volume
-,
Number
Heo et al
9
-
Various treatment algorithms have been proposed for patients with SISMAD.10,11,13,15,25 Li et al10 suggested conservative treatment in stable patients with patent true lumen flow, endovascular treatment in cases of true lumen stenosis and occlusion, and surgical treatment when endovascular therapy fails or there are signs of bowel infarction or arterial rupture. Cho et al11 recommended anticoagulant therapy for patients with symptomatic SISMAD with luminal narrowing of the SMA. Kim et al13 also recommended conservative treatment unless persistent abdominal pain or peritonitis signs are present. Most proposed treatment algorithms recommend conservative treatment for uncomplicated SISMAD; interventional treatment is generally recommended in cases of persistent abdominal pain, dissection progression, aneurysmal change of the arterial lesion, rupture, or bowel gangrene. These recommendations have usually been based on early treatment results from a small number of patients. No randomized prospective studies have yet compared the long-term results of endovascular treatment with those of conservative treatment for patients with SISMAD to determine the optimal treatment of SISMAD. Moreover, angiographic features that can adequately predict late complications or the natural course of SISMAD have not yet been identified. We observed that prolonged unrelieved abdominal symptoms were relatively common and lowered the quality of life and interfered with social life. However, these symptoms were rarely associated with bowel gangrene, even in cases of SMA occlusion (type III). Progression of arterial dissection and aneurysmal changes were also rarely detected. Moreover, SMA rupture was not found on our long-term follow-up examinations. Furthermore, we did not observe any difference in long-term clinical outcomes (SMA narrowing or occlusion [type II or III]) among the patients with SISMAD. About half of the symptomatic SISMAD lesions showed complete or partial remodeling on follow-up CT angiography. Interestingly, SMA lesion remodeling was more frequent in type III lesions, whereas “no change” was more frequent in type I lesions. We hypothesize that the early development of mesenteric collateral channels may play a role in the early resolution of pain in patients with occluded SMA (type III SISMAD). We previously reported clinical features of patients with SISMAD by angiographic type.8 In this study, we modified our SISMAD classification scheme by adding the status of the outflow branch from the false lumen of the type I lesion, in addition to the patency of the false or true lumen of the SMA and the entry or re-entry of the SMA dissection. We hypothesized that the presence of the outflow branch from the false lumen might reduce the risk of progression of the arterial dissection. Among the patients who underwent conservative therapy, 32% were treated with antithrombotic therapy with
anticoagulant or antiplatelet agents. In contrast, 68% were not treated with antithrombotic therapy. No significant differences in clinical or morphologic changes were observed between the two groups. Currently, we do not routinely use antithrombotic agents. We used therapeutic doses of low-molecular-weight heparin only for patients with occluded or severely compromised SMA flow. The primary purpose of anticoagulation therapy in patients with SISMAD is to reduce the risk of bowel ischemia secondary to the SMA thrombotic occlusion at the diseased arterial segment. We propose that the concept of anticoagulation therapy for patients with arterial dissection originated from experiences with patients with spontaneous dissection of the carotid artery.26 In current practice, anticoagulation therapy is well accepted as the primary treatment of cervical artery dissection, whereas no consensus has been reached with respect to anticoagulation therapy in the context of SISMAD management. In addition, we were concerned about the potential risk of anticoagulation therapy to lead to further subintimal hemorrhage at the site of SMA dissection. Our long-term observations of patients with SISMAD revealed that the majority of patients with SISMAD showed pain resolution with conservative treatment. Moreover, the majority of patients exhibited morphologic improvements or showed no change on CT images, regardless of angiographic type. Recurrent abdominal pain developed in about one in five of all patients after conservative treatment. However, most of these cases were successfully managed conservatively. Bowel gangrene was rare; moreover, interventional treatment was required in only 4.5% of all patients with SISMAD and 6% of all symptomatic patients with SISMAD. Despite that we extended duration of follow-up and increased the number of SISMAD patients to our previous observation,7 this study still has the limitation of a small number of patients, precluding a multivariate analysis for risk factors of pain recurrence or requirement of interventional treatment.
CONCLUSIONS Our long-term observation of SISMAD patients demonstrated that the majority of SISMAD patients have a benign natural clinical course. Therefore, we recommend conservative management for all patients with SISMAD as the first-line treatment strategy regardless of the angiographic features. Primary stenting of the SMA can be considered for patients with severe prolonged abdominal pain while they are being conservatively managed, for patients with clinical signs of bowel gangrene, or in symptomatic patients with short life expectancy. After conservative treatment, a careful long-term follow-up is recommended, considering the potential risk of late occurrence of bowel stricture, especially in symptomatic patients with SISMAD.
10
Heo et al
Journal of Vascular Surgery ---
AUTHOR CONTRIBUTIONS Conception and design: SH, YW Analysis and interpretation: SH, YW Data collection: SH, YW, SY, YJ, KB, DK Writing the article: SH, YW Critical revision of the article: SH, YW Final approval of the article: SH, YW, SY, YJ, KB, DK Statistical analysis: SH, SY Obtained funding: Not applicable Overall responsibility: YW
REFERENCES 1. Foord AG, Lewis RD. Primary dissecting aneurysms of peripheral and pulmonary arteries: dissecting hemorrhage of media. Arch Pathol 1959;68:553-77. 2. Bauersfeld SR. Dissecting aneurysm of the aorta; a presentation of 15 cases and a review of the recent literature. Ann Intern Med 1947;26:873-89. 3. Goueffic Y, Costargent A, Dupas B, Heymann MF, Chaillou P, Patra P. Superior mesenteric artery dissection: case report. J Vasc Surg 2002;35:1003-5. 4. Javerliat I, Becquemin JP, d’Audiffret A. Spontaneous isolated dissection of the superior mesenteric artery. Eur J Vasc Endovasc Surg 2003;25:180-4. 5. Cho YP, Ko GY, Kim HK, Moon KM, Kwon TW. Conservative management of symptomatic spontaneous isolated dissection of the superior mesenteric artery. Br J Surg 2009;96: 720-3. 6. Li Z, Ding H, Shan Z, Du J, Yao C, Chang G, et al. Initial and middle-term outcome of treatment for spontaneous isolated dissection of superior mesenteric artery. Medicine (Baltimore) 2015;94:e2058. 7. Park YJ, Park KB, Kim DI, Do YS, Kim DK, Kim YW. Natural history of spontaneous isolated superior mesenteric artery dissection derived from follow-up after conservative treatment. J Vasc Surg 2011;54:1727-33. 8. Yun WS, Kim YW, Park KB, Cho SK, Do YS, Lee KB, et al. Clinical and angiographic follow-up of spontaneous isolated superior mesenteric artery dissection. Eur J Vasc Endovasc Surg 2009;37:572-7. 9. Tomita K, Obara H, Sekimoto Y, Matsubara K, Watada S, Fujimura N, et al. Evolution of computed tomographic characteristics of spontaneous isolated superior mesenteric artery dissection during conservative management. Circ J 2016;80:1452-9. 10. Li DL, He YY, Alkalei AM, Chen XD, Jin W, Li M, et al. Management strategy for spontaneous isolated dissection of the superior mesenteric artery based on morphologic classification. J Vasc Surg 2014;59:165-72. 11. Cho BS, Lee MS, Lee MK, Choi YJ, Kim CN, Kang YJ, et al. Treatment guidelines for isolated dissection of the superior mesenteric artery based on follow-up CT findings. Eur J Vasc Endovasc Surg 2011;41:780-5. 12. Jia ZZ, Zhao JW, Tian F, Li SQ, Wang K, Wang Y, et al. Initial and middle-term results of treatment for symptomatic
2016
spontaneous isolated dissection of superior mesenteric artery. Eur J Vasc Endovasc Surg 2013;45:502-8. 13. Kim HK, Jung HK, Cho J, Lee JM, Huh S. Clinical and radiologic course of symptomatic spontaneous isolated dissection of the superior mesenteric artery treated with conservative management. J Vasc Surg 2014;59:465-72. 14. Luan JY, Guan X, Li X, Wang CM, Li TR, Zhang L, et al. Isolated superior mesenteric artery dissection in China. J Vasc Surg 2016;63:530-6. 15. Min SI, Yoon KC, Min SK, Ahn SH, Jae HJ, Chung JW, et al. Current strategy for the treatment of symptomatic spontaneous isolated dissection of superior mesenteric artery. J Vasc Surg 2011;54:461-6. 16. Li N, Lu QS, Zhou J, Bao JM, Zhao ZQ, Jing ZP. Endovascular stent placement for treatment of spontaneous isolated dissection of the superior mesenteric artery. Ann Vasc Surg 2014;28:445-51. 17. Gobble RM, Brill ER, Rockman CB, Hecht EM, Lamparello PJ, Jacobowitz GR, et al. Endovascular treatment of spontaneous dissections of the superior mesenteric artery. J Vasc Surg 2009;50:1326-32. 18. Dreier JP, Lurtzing F, Kappmeier M, Bohner G, Klingebiel R, Leistner S, et al. Delayed occlusion after internal carotid artery dissection under heparin. Cerebrovasc Dis 2004;18: 296-303. 19. Zhang X, Sun Y, Chen Z, Li X. Therapeutic regimen options for isolated superior mesenteric artery dissection. Vasc Endovascular Surg 2012;46:277-82. 20. Ko SH, Hye R, Frankel DA. Management of spontaneous isolated visceral artery dissection. Ann Vasc Surg 2015;29: 470-4. 21. AbuRahma AF, Stone PA, Bates MC, Welch CA. Angioplasty/ stenting of the superior mesenteric artery and celiac trunk: early and late outcomes. J Endovasc Ther 2003;10:1046-53. 22. Ozaki T, Kimura M, Yoshimura N, Hori Y, Takano T, Kamura T, et al. Endovascular treatment of spontaneous isolated dissecting aneurysm of the superior mesenteric artery using stent-assisted coil embolization. Cardiovasc Intervent Radiol 2006;29:435-7. 23. Casella IB, Bosch MA, Sousa WO Jr. Isolated spontaneous dissection of the superior mesenteric artery treated by percutaneous stent placement: case report. J Vasc Surg 2008;47:197-200. 24. Tameo MN, Dougherty MJ, Calligaro KD. Spontaneous dissection with rupture of the superior mesenteric artery from segmental arterial mediolysis. J Vasc Surg 2011;53: 1107-12. 25. Roussel A, Pellenc Q, Corcos O, Tresson P, Cerceau P, Francis F, et al. Spontaneous and isolated dissection of the superior mesenteric artery: proposal of a management algorithm. Ann Vasc Surg 2015;29:475-81. 26. Rao AS, Makaroun MS, Marone LK, Cho JS, Rhee R, Chaer RA. Long-term outcomes of internal carotid artery dissection. J Vasc Surg 2011;54:370-4; discussion: 375.
Submitted Aug 5, 2016; accepted Oct 24, 2016.