Segmental arterial mediolysis (SAM) is a rare vasculopathy of unknown etiology characterized by disruption of the arterial medial layer, with resultant susceptibility to vessel dissection, hemorrhage, and ischemia. Since the first case of SAM described by Slavin and Gonzalez-Vitale in 1976 (1), approximately 50 cases have been reported in the literature (2).
Although the abdominal visceral arteries are most frequently affected in SAM (3), any vessel may be involved, including the retroperitoneal (4), intracranial (2, 5, 6), and coronary arteries (7–9). The histopathologic changes begin with vacuolar degeneration of smooth muscle cells in the arterial media, followed by fibrin deposition at the medial–adventitial junction (7). This in turn predisposes to dissecting aneurysms (3, 10). The angiographic appearance of SAM is variable, ranging from arterial dilation to aneurysm formation (single or multiple) to stenoses or occlusion, frequently with dissection (3, 11). Correspondingly, symptoms arise both from stenoses and occlusions (e.g., postprandial pain from intestinal ischemia) and from dissections and aneurysms (e.g., sudden and catastrophic intraperitoneal bleeding). In contrast with true vasculitis, inflammatory cells in SAM are present inconsistently and, when present, are thought to be secondary rather than primary to the pathogenesis of the disease (1).
The differential diagnosis of SAM includes atherosclerosis, fibromuscular dysplasia (FMD), infection (e.g., mycotic aneurysm and endocarditis), connective tissue diseases (e.g., Behçet's disease and polyarteritis nodosa [PAN]), neurofibromatosis, and inherited defects in vessel wall structural proteins (e.g., type IV Ehlers-Danlos syndrome and Marfan's syndrome) (Table 1). Herein, we describe two cases of SAM seen at our institution over the past 20 years and review the salient clinical presentation and treatment of SAM. We report characteristics that may be helpful in distinguishing cases of SAM from other entities in the differential diagnosis.
Table 1. Clinical and laboratory features distinguishing SAM from its mimics*
Type IV Ehlers-Danlos syndrome
Other rare mimics include relapsing polychondritis, Cogan's syndrome, aortitis of tertiary syphilis, and the aortitis associated with seronegative spondylarthritides. SAM = segmental arterial mediolysis; FMD = fibromuscular dysplasia; PAN = polyarteritis nodosa; AAV = ANCA-associated vasculitis; GCA = giant cell arteritis; TA = Takayasu arteritis; BD = Behçet's disease; KD = Kawasaki disease; NF = neurofibromatosis; PXE = pseudoxanthoma elasticum; occas = occasionally; GI = gastrointestinal; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; ANCA = antineutrophil cytoplasmic antibody; HBsAg = hepatitis B surface antigen.
Upper extremity claudication and thoracic back pain are relatively common.
Charcot joints can develop as a result of peripheral neuropathy. Also, bone lesions such as pseudarthrosis and bone dysplasia are common.
The patient was a 25-year-old woman who was admitted to the hospital with an 11-month history of intermittent episodes of anorexia, abdominal pain, and diarrhea. Symptoms had persisted despite discontinuation of oral contraception and initiation of low-dose aspirin therapy.
Her past medical history was unremarkable. Family history was unremarkable except for benign hypermobility syndrome in the patient's mother. On physical examination, the patient was normotensive and had normal height and arm span. She had no carotid, subclavian, abdominal, or femoral bruits. Skin, chest, abdominal, and neurologic examinations were normal. Joint examination was remarkable only for hyperextensibility of the knees, reducible flexion contractures of the fingers, and hammer toe deformities of the feet.
Complete blood cell count revealed anemia, with a hemoglobin level of 11 gm/dl. Serum creatinine, liver enzymes, amylase, and lipase were normal, and a urine pregnancy test was negative. Antinuclear antibody assay was positive at a low titer of 1:80. The erythrocyte sedimentation rate was 20 mm/hour. The remainder of the serologic, metabolic, immunologic, and hematologic evaluations was within normal limits, including negative hepatitis serologies; negative double-stranded DNA, anti-Sm, and anti-RNP antibodies; and normal complement C3 and C4 levels.
Computed tomography (CT) of the abdomen showed thickening of the colonic wall with mucosal enhancement and fat stranding surrounding the splenic flexure. Colonoscopy revealed ischemic colitis of the splenic flexure. Biopsy samples of the ischemic areas were not obtained due to risk of possible perforation. Biopsy samples of the nonischemic areas were normal, as was magnetic resonance angiography (MRA) of the abdomen. Conventional mesenteric angiography revealed focal stenoses of the right and left hepatic arteries, occlusion of the left colic artery near the splenic flexure with collateral vessel formation, and hyperemia of multiple branches of the splenic artery (Figure 1).
Because of persistent ischemic colitis, the patient underwent a partial colectomy of the splenic flexure. Vascular pathology of the colonic arteries showed patchy, isolated destruction of the arterial media involving both the internal and external elastic laminae (Figure 1). In a few sections the media was absent, with direct juxtaposition of the intima and the adventitia. In areas of medial destruction, there was intimal proliferation with marked luminal narrowing. All of the lesions were of a similar age. There was no evidence of inflammation, and giant cells, neutrophils, and cholesterol deposits were absent. After 2 years of followup, the patient remains asymptomatic.
The patient was a 51-year-old man who was admitted to the hospital with acute burning and tearing epigastric pain that began while eating. A full cardiac evaluation and CT scan of the abdomen were unrevealing. He was discharged the next day with a tentative diagnosis of biliary colic. However, because of unremitting abdominal discomfort he was readmitted 2 days later.
Prior to the initial painful episode, he had no history of epigastric pain. He denied a change in bowel habits, loss of appetite, or weight loss. There was no history of fever, chills, sweats, rashes, oral ulcers, genital lesions, hepatitis, or exposures to sexually transmitted diseases. He did not smoke and only drank alcohol occasionally. He admitted to minor chronic headaches but denied visual changes or jaw pain. Other medical history was noncontributory, and he took no long-term medications. Physical examination revealed normal vital signs with full and symmetric pulses throughout. There were no peripheral thrills or bruits. Abdominal examination was abnormal only for mild guarding, without rebound, and there were no masses, organomegaly, or bruits.
Complete blood cell count, urinalysis, basic chemistry panel, and liver function tests were all normal except for an albumin level of 3.2 gm/dl. The C-reactive protein level and erythrocyte sedimentation rate were normal at 3.9 mg/liter and 30 mm/hour, respectively. The remainder of the serologic and immunologic evaluation was within normal limits.
An abdominal ultrasound was within normal limits, but initial CT angiography showed a celiac artery aneurysm with possible dissection. Repeat CT angiography a few days later documented extensive aneurysmal dilatation (up to 1.8 cm) of the celiac axis from the proximal celiac artery to the bifurcation, involving the hepatic and splenic arteries with mural thrombus in the splenic artery. MRA of the abdomen showed similar findings.
The patient was managed conservatively. A followup CT angiogram in 2 weeks revealed enhanced wall thickness of the celiac axis and decreased aneurysmal dilatation of the splenic artery, compatible with evolution and partial improvement of SAM. Despite the encouraging imaging studies, the patient continued to experience epigastric pain, and it was decided to proceed with resection of the celiac and common hepatic artery aneurysms with aortoceliac artery bypass. Histologic examination of the surgical specimen revealed extensive deterioration of the media with marked intimal hyperplasia. There were no giant cells, granulomas, or signs of vascular inflammation, and there were no significant cholesterol plaques. Two years have passed since the surgical repair and the patient remains asymptomatic.
SAM, also known as segmental medial arteriolysis, is a rare vasculopathy characterized by noninflammatory degeneration of the medial layer of muscular arteries (1) and, occasionally, adjacent veins (12). Originally described by Slavin and Gonzalez-Vitale as “segmental mediolytic arteritis” in 1976 (1), Slavin and colleagues later proposed a change in name to “segmental arterial mediolysis” due to a lack of consistent evidence of true inflammation in both the clinical presentation and the histologic features of the disease (10).
SAM is a pathologic diagnosis that is defined by characteristic histologic features on surgical specimens obtained from affected anatomic sites. Although it can be difficult to diagnose, clinical clues are usually present that point to the diagnosis of SAM on the basis of history, physical examination, and initial laboratory evaluation. The combination of clinical features and surgical pathology findings usually allows the discrimination of SAM from its mimics (Tables 1 and 2). For example, although atherosclerosis, a common vasculopathy, is usually widespread throughout many regions of the vascular tree, SAM typically is limited to vessels in only one anatomic site (3). In addition, atherosclerosis typically occurs at the branch points of vessels in patients with traditional cardiovascular risk factors, whereas these features are absent in SAM (13). Furthermore, although atherosclerosis typically occurs in middle-aged and elderly adults, SAM may present at any age.
Table 2. Key vascular, histopathologic, and radiographic features of SAM and its mimics*
M = medium-sized arteries (0.5–1.0 mm in diameter); L = large arteries (>1.0 mm in diameter); S = small arteries (<0.5 mm in diameter); DVTs = deep venous thromboses. See Table 1 for additional abbreviations.
Most commonly affected vessel sizes are reported here, but it should be noted that extension to other vascular sizes or other areas of the vascular tree is relatively common.
M, L, veins
Visceral mesenteric arteries
Outer layer of media predominantly affected; alternating stenoses and aneurysms with disruption of elastin
“String of beads” appearance; stenoses, aneurysms, dissections, thrombosis
Any vessel layer may be affected; in the most common type (medial fibrodysplasia), thick, collagenized regions alternate with thinned areas of media
Classic “string of beads” appearance; stenoses and narrowing; late angiographic appearance is indistinguishable from SAM
Yes; often strikingly segmental
Focal areas of necrotizing panarteritis
Characteristic microaneurysms predominantly at vessel branch points
Characteristic fibrous plaques with calcification and foamy macrophages
Shaggy, irregular appearance; predisposition to branch points; widespread distribution
It is a particular challenge to distinguish SAM from FMD, especially since SAM is often considered an early lesion of FMD (3, 10). Classically, FMD presents in young females and has a predisposition for the renal arteries, causing premature hypertension. SAM, on the other hand, may present at any age, has no sex predisposition, and most commonly affects the celiac artery and its branches. Arterial dissection and hemorrhage are also much more common in SAM than in classic FMD. In addition to FMD, another disease entity classified on the disease spectrum of SAM is cystic medial necrosis (CMN) (1). However, typically, CMN occurs in the aorta and great vessels of patients with Marfan's syndrome (14), whereas these vessels are not typically affected by SAM.
We reviewed all cases of biopsy-confirmed SAM occurring at our institution over the past 20 years. The two cases described here represent the results of our medical record review. Although many more cases of SAM were suspected on the basis of clinical and radiologic features, these were the only two cases with histopathologic confirmation. Of note, the cases identified in our institution lacked physical signs and symptoms and laboratory indicators of systemic inflammation, helping to distinguish SAM from the inflammatory vasculitides.
The most dramatic presentation of SAM is sudden, life-threatening hemorrhage of the abdomen, retroperitoneum, or brain (3, 15). Hemorrhage results from either aneurysm rupture or dissections occurring as a result of weakening along the plane separating the outer media from the adventitia (3). Interestingly, each of the cases of SAM identified at our institution presented in a relatively benign fashion, with abdominal pain as the chief symptom. Ischemic colitis, as in case 1, has been reported as an example of a relatively less acute presentation of SAM (16). More benign presentations of SAM could easily escape clinical diagnosis, and therefore, SAM may be substantially more common than is suggested by the literature (3, 17).
Angiography can reveal several patterns that are consistent with SAM, including single or multiple aneurysms, dissections, stenoses, and occlusions (3). As seen in our cases and in another report by Michael and colleagues (18), lesions of SAM may evolve rapidly over the course of weeks on serial angiography. Despite patterns on angiography that are suggestive of SAM, histopathology remains the gold standard for definitive diagnosis (Table 2). This is especially important in the case of PAN, which can have an angiographic appearance identical to that of SAM (3). Lack of inflammation on arterial biopsy samples in SAM allows these two entities to be readily distinguished. Patients with PAN also generally have clinical evidence of systemic inflammation, where as patients with SAM do not.
The discrimination of SAM from systemic inflammatory vasculitides is particularly important, since corticosteroids and immunosuppressive agents, which are crucial in the treatment of the inflammatory vasculitides, have no proven benefit in SAM (19). Without any evidence of an inflammatory etiology, the use of immunosuppressive regimens in SAM exposes the patient to undue risks, including infection and poor wound healing, and could possibly worsen the prognosis (19). Treatment of SAM involves embolization, surgical bypass, or resection of the injured arteries (20). The long-term prognosis of SAM is somewhat uncertain, since its natural history has not been thoroughly characterized despite its initial description more than 30 years ago. It is known, however, that although cases of SAM complicated by intraabdominal hemorrhage have a mortality approaching 50% (21), the most common scenario is of long-term disease-free survival following embolization, bypass, or resection of the affected areas. There have even been reports of complete spontaneous resolution of the vascular lesions of SAM (18). Our cases were both followed for 2 years with no recurrence of disease.
In summary, SAM is a rare but important cause of unexplained vascular lesions in patients in whom other inflammatory, infectious, or heritable diseases have been ruled out. The diagnosis should be considered when a patient presents with unexplained acute-onset abdominal pain with or without intraabdominal bleeding. SAM should also be kept in mind when aneurysms, stenoses, and occlusions are identified in medium-sized and large vessels, especially when these lesions are limited to a single anatomic location. Conventional angiography is more sensitive than CT or MRA, and should be used after more conventional methods of imaging are unrevealing. If a diagnosis of SAM is suspected, a multidisciplinary approach involving consultation with interventional radiology and vascular or general surgery should be promptly pursued.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Baker-LePain had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Stone, Fye.
Acquisition of data. Baker-LePain, Stone, Mattis.
Analysis and interpretation of data. Baker-LePain, Stone, Mattis, Nakamura, Fye.
We dedicate this article to Dr. Kenneth H. Fye, a beloved master clinician, teacher, and friend, who had a special passion for raising awareness about SAM.