Risk factors of pulmonary arterial hypertension in patients with systemic lupus erythematosus

: Pulmonary arterial symptoms in systemic lupus erythematosus patients are non-specic and early diagnosis and intervention are challenging. It remains essential to explore risk factors for PAH in SLE patients to identify high risk patients and allow intensive monitoring. Methods From to 84 with and and 160 with but without PAH were enrolled. Clinical manifestations and laboratory test results were compared between the two groups to identify predictors of PAH. Candidate PAH risk factors were further compared among SLE-PAH patients with different characteristics. C-reactive (CRP) complement 3 (C3), serum creatinine, urine protein, anti-double stranded DNA (anti-dsDNA) antibodies, antinuclear antibody (anti-ANA), anticardiolipin antibodies immunoglobulin G (ACA-IgG), ACA immunoglobulin M (ACA-IgM), anti-u1 ribonucleoprotein (anti-U1RNP) antibodies, anti-Sjögren's-syndrome-related antigen A (anti-SSA) antibodies, and anti-Sjögren's-syndrome-related antigen B (anti-SSB) antibodies. Indicators were compared between the case group and the control group using bivariate logistic regression. Indicators that were statistically signicant were labeled as candidate SLE-PAH risk factors. Among the 84 cases with SLE-PAH, patients were grouped according to PAH severity (mild to moderate versus severe), SLE disease activity (SLEDAI < 8 points versus SLEDAI ≥ 8 points), stable, and deterioration), and duration of PAH ((< 1 year and ≥ 1 year).


Introduction
Systemic lupus erythematosus (SLE) is a complex chronic autoimmune disease involving multi-system, a wide spectrum of manifestations, and overabundance of immunological and laboratory abnormalities [1].
Among all the complications, pulmonary arterial hypertension (PAH) is a major contributor of SLE related death [2,3]. PAH is characterized by vascular proliferation and remodeling [4,5], which results in progressively increase pulmonary vascular resistance and right ventricular failure and death [6]. The prevalence of PAH in SLE is estimated to be 0.5-17.5% [7]. The prevalence of SLE-related PAH in Chinese population is around 3.8%, which is the third leading cause of death in patients with SLE after mental lupus and lupus nephritis [8].
Most of the early symptoms of PAH, such as dyspnea and fatigue, can be mild and non-speci c.
Symptoms of PAH in lupus can also be caused by other conditions such as interstitial lung disease, all make early diagnosis challenging. Recent novel therapies, especially therapies that targets abnormalities in prostacyclin, endothelin, and nitric oxides pathways have been shown to improve patients' functional status, pulmonary hemodynamics, and a potential to slow disease progression [9]. However, prevention PAH progression is extremely challenged and rarely achieved, especially for patients with connective tissue disease-related PAH [10]. For SLE-associated PAH, immunosuppressive treatment may bene t patients with relatively mild disease at baseline, but limited bene t for patients with severe PAH [11][12][13].
Early diagnosis and intervention are essential to improve patient outcomes.
Some clinical manifestations and immunologic test have been shown to predict PAH in SLE patients [14,15]. However, PAH is a complex condition, risk factors may have different association for certain disease characteristics. In the current study, we aimed to identify early clinical and laboratory predictor for SLE associated PAH and other SLE characteristics to help early diagnosis and identify higher risk population and allow for more intensive monitoring and intervention if necessary.

Participants
Patients diagnosed with SLE in the Department of Rheumatology and Immunology of Guangdong General Hospital and Guangdong Academy of Medical Sciences from January 2010 to December 2018 were eligible for the study. A total of 84 patients who were diagnosed with SLE and PAH were included as the case group, and 160 patients who were diagnosed with SLE but did not have PAH were included as the control group. The study was approved by the Guangdong General Hospital and Guangdong Academy of Medical Sciences. Informed consent was obtained.
Remission was de ned as remaining in low risk status in accordance to the 2015 European pulmonary hypertension guidelines [19]. Stable was de ned as the same risk status for 6 months (excluding high risk status). Deterioration was de ned as moving from lower risk status to higher risk status (low risk status to intermediate/high risk status, intermediate risk status to high risk status, or high risk status with worsening indicators).

Statistical analysis
Data were analyzed using SPSS version 19.0 statistical software (SPSS Inc, Chicago, Illinois). Continuous variables were expressed as mean (standard deviation, SD) if following normal distribution, and median (interquartile range, IQR) if not. Categorical variables were expressed as count (percentage). Between groups comparison was performed using student t-test for continuous variables and chi-square test for categorical variables. Logistic regression was used for risk factor analysis. Associations between candidate SLE-PAH risk factors and different classi cation group were compared using logistic regression to identify risk factors for different PAH prognostic outcomes. P < 0.05 indicated that the difference is statistically signi cant.

Results
A total of 84 patients with SLE and PAH and 160 patients with SLE but without PAH were included. On average, cases were 36.2 (10.7) years old (range, 22-53) and 89.3% were female. In the control group, the mean (SD) age was 34.8 (11.5) years old (range, 24-55) and 88.1% were female. No statistically signi cant difference in age and sex were observed in the two groups (p > 0.05 for both).  Table 2 were observed between the two groups.

Associations between candidate risk factors and PAH in SLE patients
Among collected patient characteristics, Raynaud's phenomenon, digital vasculitis, pericardial effusion, pulmonary interstitial lesions, positive anti-U1RNP antibodies, and positive ACA-IgG were associated with signi cantly higher risk of PAH in SLE patients (Table 3). Among them, digital vasculitis had the stronger association with PAH in SLE patients (OR 4.12, 95% CI: 1.48-11.49), followed by pulmonary interstitial lesions (OR 3.10, 95% CI: 1.55-6.21).    signi cant differences were observed in other characteristics.

Discussion
We found that clinical manifestations including Raynaud's phenomenon, digital vasculitis, pericardial effusion, and pulmonary interstitial lesions, and immunologic abnormalities including anti-U1RNP positive and ACA-IgG positive were predictors for SLE associated PAH. Among the risk factors, positive anti-U1RNP was independently associated with severe PAH and more active disease. Digital vasculitis was independently associated with SLE alleviation, while pericardial effusion was associated with deterioration. Pericardial effusion was associated with longer PAH duration.
Our ndings were consistent with previous studies. In a study involving 41 SLE patients with PAH and 106 SLE patients without PAH, serositis, Raynaud's phenomenon, anticardiolipin antibodies, and anti-U1RNP were associated with higher risk of SLE-PAH [15]. In another cross-sectional study, Raynaud's phenomenon was associated with elevated pulmonary artery systolic pressure [20]. Luo et al found that pleural effusions frequently accumulate in patients with PAH associated with CTD [21]. Wang et al observed in a large SLE-PAH cohort that, pericarditis, pleuritis and anti-RNP positivity are associated with higher SLE activity and vasculopathy, suggesting that higher disease activity and vasculopathy may contribute to PAH development in SLE [8]. Study from the same cohort con rmed that long SLE duration and interstitial lung disease et al were associated with higher risk of PAH in SLE patients [2]. In addition, anti-U1RNP and antiphospholipid antibodies were associated with higher risk of PAH [2]. Consistent with our results, anti-U1RNP is associated higher risk of PAH [2,8,22]. Antiphospholipid antibodies, including ACA in our work, also is a strong predictor of PAH in SLE patients [2,23,24].
Consistently observed in various population and studies, anti-U1RNP antibody was associated with higher disease activity and more severe PAH. Anti-U1RNP antibody is a speci c antibody for mixed connective tissue disease but also present in SLE patients [25]. Anti-U1RNP antibody can up-regulate the expression of adhesion molecules (CAM) and major histocompatibility complex II (MHC II) molecules on pulmonary artery endothelial cells, which play important roles in the development of PAH [26]. In our work, positive anti-U1RNP antibody was associated with more severe PAH, suggesting that vascular lesion involving anti-U1RNP antibody may be involved in PAH development and positive U1RNP antibody indicates worse prognosis among PAH patients. Together with previous studies, our results suggest that among patients with SLE and patients with mild to moderate SLE-PAH, those with anti-U1RNP antibody are at higher risk of developing PAH, especially severe PAH. These patients should be more closely monitored. At the same time, positive anti-U1RNP antibody also is associated with higher disease activity in our study. Annual PAH screening is recommended by some researchers, especially for SLE patients with other risk factors [8].
Thrombosis has been suggested as one mechanism of PAH in patients with connective tissue disease [7]. Studies have found that anti-phospholipid antibodies are associated with PAH in SLE and suggested that thrombosis involves in PAH pathogenesis [8, 14,23]. Though it is worth noted that some studies found negative results between antiphospholipid antibodies and PAH [27]. In our work, positive ACA-IgG was associated with PAH, but not associated severity of PAH. Thromboembolic disease, pulmonary vasculitis, and hypoxia and brosis from interstitial lung disease all have been suggested to involve in PAH in SLE patients [28]. It is possible that antiphospholipid antibodies is only involved in a certain pathways and that explains the heterogeneous associations observed.
Our study has several strengths. First, we have relatively large sample size. Second, we assessed the association between risk factors and different SLE, and SLE-PAH characteristics. As observed, different factors may predict different characteristics, which may be a result of various underlying mechanisms involved. Our study also has several limitations. First of all, our study is a cross-sectional study. It is di cult to interpret some of the observed associations with cross-sectional data. For example, pericardial effusion was more commonly observed in patients with PAH longer than 1 year. However, it is not clear whether patients with pericardial effusion shared certain risk factors with earlier onset of PAH, or the observed association is solely a result of longer SLE duration. Longitudinal data that is able to differentiate the temporal relationship between risk factors and PAH development and progression are needed to elucidate the associations. Second, our results are based on single center data. As a tertiary hospital, it is possible we are seeing patients with more severe SLE and more advanced PAH. Multicenter data involving patients at different stage are needed to test the generalizability of these results.

Conclusions
In summary, we con rmed several risk factors that are associated with PAH in SLE patients, and identi ed different risk factors for PAH and SLE characteristics. Speci cally, Raynaud's phenomenon, digital vasculitis, pericardial effusion, and pulmonary interstitial lesions, and immunologic abnormalities including anti-U1RNP positive and ACA-IgG positive were predictors for SLE associated PAH. Positive anti-U1RNP was independently associated with severe PAH and more active disease. Our study suggested that for patients with above risk factors, more intensive monitoring and interventions if necessary should be given to improve these patients' prognosis.