Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease affecting the blood vessels and connective tissues, causing damage to the skin, joints, and organs. The cardiovascular involvements of SLE are often present with pericarditis and sometimes with coronary arteritis [1]. However, aortic aneurysm formation and aortic dissection are rarely seen in SLE patients [2]. SLE patients have been found to be at a significantly increased risk of developing aortic aneurysm and dissection in comparison to non-SLE controls, and the incidence of aortic aneurysm in SLE patients is significantly higher than that of the non-SLE controls [3]. Wang et al. [4] reported in a retrospective study of 15,209 SLE patients that 20 (0.13%) patients had aortic aneurysms and 13 (0.09%) patients developed aortic dissection. Choudhary et al. [5] reported an 11.5% incidence of unruptured aneurysms in a setting of 104 SLE patients. The incidence, etiology, risk factors, and outcomes of aortic aneurysm and dissection in SLE patients were largely unknown. The present article aims to highlight the clinical features of this patient setting.

Methods

PubMed was searched for publications reporting aortic aneurysm or dissection due to SLE published between 2000 and 2017. The search terms included “systemic lupus erythematosus”, “aortic aneurysm”, “aortic dissection”, and “aortic dilatation”. Bibliographic references were tracked for ensuring the completion of literature retrieval.

Data were carefully extracted for study population, demographics, diagnostic techniques, clinical features of aortic aneurysm or dissection in SLE patients, associated disorders, treatment, follow-up length, and patients’ outcomes, etc.

Quantitative data were presented as mean ± standard deviation with range and median values, and were compared by using the independent t-test. The frequencies were compared by Fisher’s exact test. p < 0.05 was considered statistically significant.

Results

A total of 35 articles describing a single case or case series involving 40 patients were collected [2, 6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39]. There were 11 males and 29 females, with a male-to-female ratio of 1:2.6. The patients were at the age of 44.6 ± 16.8 (range 9–76; median 45) years (n = 40). No age difference was found between male and female patients (47.7 ± 22.0 years vs. 43.4 ± 14.7 years; p = 0.473). The SLE history was reported in 23 patients: in 3 patients, it was described as “new onset” [8, 10] and “long-standing” [9]. In the remaining 20 patients, the SLE history was 14.6 ± 11.5 (range 0–38; median 14) years (n = 21). It showed a longer SLE history in female patients than in male patients, which lacked significance (17.1 ± 11.9 years vs. 6.6 ± 4.7 years; p = 0.073). The duration of steroid use was 13.3 ± 9.4 (range 0–32; median 14) years (n = 19). In female patients, the duration of steroid use was longer than in the male patients, without significant difference (15.5 ± 9.6 years vs. 8.5 ± 7.6 years; p = 0.134). No difference was found between the SLE history and the duration of steroid use (p = 0.555). As many as 26 (65%) patients had a total of 51 associated disorders, of which hypertension and chronic renal failure were the most common (Table 1).

Table 1 51 associated conditions in 26 patients
Full size table

Aortic aneurysm occurred in 27 (67.5%) patients [6, 9, 11,12,13,14,15, 17, 19, 20, 23,24,25,26,27, 29, 31,32,33,34,35,36, 38], aortic dissection in 11 (27.5%) patients [2, 7, 8, 10, 16, 18, 25, 28, 30, 37, 39], concurrent aortic aneurysm and dissection in 2 (5%) patients [21, 22]. A comparison of patient information of the former two groups was shown in Table 2.

Table 2 A comparison of patient information of patients with aortic aneurysm and aortic dilation
Full size table

Aortic aneurysm often affected SLE patients in the abdominal aorta, whereas aortic dissection did not show any location predilection (Table 3). The type of aortic dissection was described in 12 patients: type A, 7 cases [2, 6, 7, 15, 21, 29, 36] and type B, 5 cases [20, 25, 27, 33, 38]. One patient with type A aortic dissection had a history of type B aortic dissection [15]. In aortic aneurysm patients, 2 (7.4%) were with pseudoaneurysms [13, 31], and aneurysmal rupture occurred in 4 (14.8%) patients [6, 13, 14, 20] (one pseudoaneurysm ruptured [13]). In contrast, in the aortic dissection group, abdominal rupture occurred in 1 (7.7%) patient with descending aortic dissection [35]. A mural thrombus was seen in 6 (22.2%) patients of the aortic aneurysm group [11, 13, 29, 31, 33, 35], while intramural hematoma was observed in 2 (15.4%) patients with aortic dissection [10, 37]. Aortic wall atherosclerosis and/or calcification occurred in 8 (8/27, 29.6%) aortic aneurysm patients [6, 9, 11, 29, 31, 34, 36, 38] and in 1 (1/9, 9.1%) aortic dissection patient [2]. In all, there was an occurrence of aortic atherosclerosis in 3 (3/9, 33.3%) patients [31, 34, 38], aortic wall calcification in 2 (2/9, 22.2%) patients [6, 11], and concurrent atherosclerosis and calcification in 4 (4/9, 44.4%) patients [2, 9, 29, 36]. The dimensions of 39 measurements of 36 aneurysms from 26 patients were shown, with a mean of 57.8 ± 22.4 (range 33–110; median 47) mm (n = 33; Table 4). The descending aorta showed the largest dimension, followed by the abdominal aorta, which extended >60 mm in average.

Table 3 Locations of the aortic aneurysm/dissection by incorporating the 2 patients with concurrent aortic aneurysm and dissection, n (%)
Full size table
Table 4 Dimensions of 39 measurements of 36 aneurysms from 26 patients (20 aortic aneurysm patients, 5 aortic dissection patients, and 1 aortic dissection with chronic dilation patient)
Full size table

One patient suddenly died and lost the chance of treatment [34], and the treatment was not described for another patient [28]. Of the remaining 38 patients, 24 (63.2%) were surgically treated, 7 (18.4%) were interventionally treated, 2 (5.3%) were treated by hybrid procedures, and 5 (13.2%) were conservatively treated. A total of 27 (69.2%) experienced event-free survival, 4 (10.3%) were complicated (2 of them eventually died), and 9 (23.1%) died (Tables 5 and 6).

Table 5 Treatment of choice of aortic aneurysm and aortic dissection
Full size table

In the surgical group, 3 (12%) patients were operated on an urgent basis [13, 30, 37], 6 (24%) patients died, the death causes of 4 patients were given as late aortic dissection/rupture in 2 (50%) patients [31, 35], renal failure in 1 patient [37], and neurological sequelae in 1 patient [13]. In the interventional group, 2 (28.6%) patients underwent urgent interventions [18, 20]. As many as 4 patients had a complication: aortoesophageal fistula in 1 patient [14], and endoleaks in 3 patients with a respective type I, II, and III endoleaks in these patients [18, 21, 29].

Event-free survival of the surgical, interventional, and conservative groups did not differ significantly (75% vs. 42.9% vs. 80%; χ2 = 4.258; p = 0.119), and the mortality rates of the three groups were 25%, 12.5%, and 20%, respectively (Table 6).

Table 6 Patients’ outcomes of 38 patients (27 aortic aneurysm/9 aortic dissection/2 aortic aneurysm and dissection patients), n (%)
Full size table

Patients were at a follow-up of 19.6 ± 23.0 (range 1–78; median 9) months (n = 23). Postoperative use of immunosuppressive agents was described for 6 patients [7, 18, 20, 22, 25, 26], and in 2 of them, a tapering dose was applied during the therapeutic course [7, 25].

No significances were found in the recovery and mortality rates between surgical and interventional treatments (recovery: 73.9% [17/23] vs. 57.1% [4/7]; χ2 = 0.719; p = 0.640; mortality: 26.1% [7/23] vs. 0% [0/7]; χ2 = 2.779; p = 0.154), or between aortic aneurysm and aortic dissection patients (recovery: 75% [12/16] vs. 57.1% [4/7]; χ2 = 0.138; p = 1.000; mortality: 18.8% [3/16] vs. 42.9% [3/7]; χ2 = 0.806; p = 0.633; Table 6). Pathological studies of the surgical specimens of the aorta and/or aortic valve were reported in 11 patients (Table 7).

Table 7 Pathological studies of the surgical specimens of the aorta from 22 patients and of the aortic valve from 3 patients, n (%)
Full size table

A comparison of risk factors between aortic aneurysm and aortic dissection revealed that only cumulative steroid dose was much higher in aortic dissection patients than in the aortic aneurysm group (Table 8).

Table 8 A comparison of risk factors between aortic aneurysm and aortic dissection
Full size table

Discussion

The etiology of aortic aneurysm and dissection in SLE patients can be multifactorial. It was considered a result of pathological changes in the aorta of SLE patients, such as atherosclerosis, vasculitis, and cystic medial necrosis [40]. Esdaile et al. [41] reported that SLE itself represented an independent factor for premature atherosclerosis development, as patients with longer duration of SLE suffered more cardiovascular events. Clinical observations proved that prolonged steroid therapy in SLE patients predisposed to aortic aneurysm formation due to adverse effects in accelerating atherosclerosis, hypertension, and mucoid degeneration, and even cystic medial necrosis of the aortic wall by inhibition of formation of granulation tissue and chondroitin sulfate [42]. In short, the possible pathogenesis of aortic pathology was attributed to atherosclerosis, aortic elastic tissue degeneration, and vasculitis [4]. Men with SLE had a significantly higher incidence of aortic aneurysm, whereas women with SLE carried a significantly higher risk of aortic dissection [4]. Aoyagi et al. [7] described that the chronic inflammatory effect of SLE, a long-term history of smoking, the presence of systemic hypertension, and atherosclerosis contributed to the development of aortic dissection. However, Roman et al. [43] found that long-term treatment with corticosteroids reduced the risk of atherosclerosis.

Wei et al. [39] reported a mean age of 39.3 years of 21 SLE patients with aortic dissection and Kurata et al. [40] reported that the mean age of 35 SLE patients with aortic aneurysm was 44.5 years. Given that the female:male ratio in SLE is 10:1, a 2.6:1 ratio in aortic events still pointed to a higher risk in male SLE patients.

The pathologic study of aorta specimens of SLE patients showed Marfan-like changes, including cystic medial necrosis and elastic fiber disruption with deposition of mucopolysaccharides [42]. Guard et al. [44] found the pathological changes resembling those of Takayasu arteritis, including lymphoplasmacytic infiltrations and obliterative endarteritis of vasa vasorum in the aorta of SLE patients with aortic dissection. On excised aortic valve, although there were no findings of atherosclerosis, infection, cystic medial necrosis, or aortitis, mucinous and hyaline degeneration was observed, which seemed compatible with the findings previously reported in an SLE patient [19]. In one case, hyaline degeneration was present, with a diminution in both the smooth muscle cells and elastic fiber layers on the aneurysmal wall [45]. It was observed that atherosclerosis mostly affected the abdominal aorta. Subsequently, different pathological changes of aortic aneurysm were found according to different locations: a medial degeneration in the thoracic aorta and atherosclerosis in the abdominal aorta. Furthermore, mucoid degeneration might be a remarkable change of the aortic wall and/or aortic valve in SLE patients. Aortic valve insufficiency might be due to mucoid degeneration in addition to associated aortic cusp fenestration caused by infective endocarditis [46]. Cervera et al. [46] reported that valvular abnormalities were found in 44% of patients with SLE, and the incidence of valvular lesions increased with an increasingly detectable rate of the valve disorders by echocardiography. Aortic aneurysmal formation is actually an immune-related process, where the inflammatory cells, such as T and B cells, macrophages, mast cells, dendritic cells, and neutrophils, accumulate in the adventitia of the aortic wall, thereby triggering the inflammatory response [12].

Aoyagi et al. [7] analyzed 15 cases of aortic dissection in SLE patients including 14 cases from the literature and 1 case of their own. They concluded that the clinical features of the patient setting were onset at a young age, association with systemic disorders, prolonged steroid use, and presence of hypertension. The gender disparity of the incidence of aortic disorders secondary to SLE has been noted. The incidence of an aortic aneurysm was significantly higher in men with SLE than those without SLE, but the incidence of aortic dissection was significantly higher in women with SLE than those without SLE [4]. It has been proposed that female patients would have less potentiality to develop aortic aneurysm and dissection, as estrogens have antioxidant, antiplatelet, and vasodilating properties, thereby preventing the aorta from atherosclerosis [40]. In most SLE patients, aortic aneurysms often develop in their third–fifth decade of age, much earlier than the normal population. Moreover, a group of SLE patients for abdominal aortic aneurysm repair showed much younger ages than non-SLE control patients for the same operation [31].

Kurata et al. [40] summarized the patient information of 26 articles reported between 1969 and 2008, and found that the mean aneurysm age was 44.5 ± 15.2 years and the mean steroid duration was 12.6 ± 9.73 years, which were compatible to the results of the present study (Table 9). However, the incidences of hypertension and atherosclerosis shown in this study were much lower than in the previous report.

Table 9 A comparison of the statistical results of aortic aneurysm/dissection in SLE patients between Kurata’s and the present study
Full size table

Although an unruptured aortic aneurysm usually does not cause symptoms, it may incur severe complications if left untreated, e.g., right ventricular outflow tract obstruction, coronary artery occlusion, aortic valve regurgitation, complete heart block, or refractory ventricular arrhythmias [4]. The correlative studies by Kurate et al. [40] revealed that the SLE age was positively related to the aneurysm age and inversely related to long-term survival. These conclusions facilitated the treatment of choice in such patients.

Endovascular thoracic aortic aneurysm repair has emerged as an alternative to conventional open surgery considering its advantages over open surgery in terms of lower surgical morbidity and mortality. Nevertheless, its clinical use is limited due to the concern of spinal ischemia, which often requires visceral and intercostal arterial reconstructions [6]. A recently developed hybrid procedure consisting of retrograde visceral revascularization and endoluminal exclusion has facilitated the treatment of thoracoabdominal aortic aneurysms in high-risk patients eligible for open surgery [6]. It was once recommended to administer a low-dose steroid to prevent aortic aneurysm formation and occurrence of aortic dissection [47], but the outcomes warrant long-term observations. The associated immunosuppressive diseases including antiphospholipid antibody syndrome in SLE patient may exaggerate SLE severity by predisposing to thrombosis and atherosclerosis. Concurrent therapy of the associated immunosuppressive disease is an important measure to control SLE and complications [34].

The incomplete data from the literature has led to some statistical bias in this study. The incidence of atherosclerosis was underestimated rather than a true one. Prospective studies based on large patient populations for accurate evaluations of the aortic complications in SLE patients are warranted.

SLE patients have significant risks of developing aortic aneurysm and dissection. Hypertension, prolonged steroid use, and aortic pathological changes related to SLE can be predominant risk factors of the occurrence of aortic aneurysm and dissection in such patients. The intergroup difference in cumulative steroid doses hint that higher cumulative steroid doses predicting aortic dissection. Aortic aneurysm and dissection occur more commonly in the abdominal aorta than in the other segments of aorta. Patients with open aortic operations tend to show a higher mortality rate than the other groups. Interventional therapy is minimally invasive, but postinterventional endoleaks are a concerning problem, which anticipates plausible resolutions.