Risk factors for development of pulmonary arterial hypertension in Australian systemic sclerosis patients: results from a large multicenter cohort study

Abstract

Background: Pulmonary arterial hypertension (PAH) is the leading cause of mortality in patients with systemic sclerosis (SSc). We sought to determine the incidence, prevalence and risk factors for PAH development in a large Australian SSc cohort. Methods: PAH was diagnosed on right heart catheterisation (mPAP >25 and PAWP <15 mmHg at rest). Patients with PH secondary to interstitial lung disease (ILD; defined as abnormal HRCT scan and FVC < 60 %) were excluded. Summary statistics, chi-square tests, univariate and multivariable logistic regression along with post-estimation diagnostics were used to determine the associations of different combinations of risk factors with PAH. Results: Among 1579 SSc patients, 8.4 % (132 patients) were diagnosed with PAH over a mean (±SD) follow-up of 3.2 (±2.5) years. The incidence of PAH in this cohort was 0.7 % per annum. Of these, 68.9 % had limited disease subtype (lcSSc). In multivariable regression analysis, the presence of anti-centromere antibody (ACA) (OR 1.6, 95 % CI 1.1-2.5, p = 0.03), oesphageal stricture (OR 2.0, 95 % CI 1.2-3.3, p = 0.006), calcinosis (OR 1.9, 95 % CI 1.2-2.9, p = 0.003), sicca symptoms (OR 1.6, 95 % CI 1.1-2.5, p = 0.03), mild ILD (OR 2.3, 95 % CI 1.5-3.7, p < 0.001) and digital ulcers (OR 1.6, 95 % CI 1.0-2.4, p = 0.03) were predictive of PAH. This model had an area under the curve of 0.7 and concordance of 91.8 %. When analysed by disease subtype, the presence of calcinosis (OR 2.2, 95 % CI 1.4-3.7, p = 0.01), sicca symptoms (OR 2.6, 95 % CI 1.5-4.6, p = 0.001), mild ILD (OR 2.3, 95 % CI 1.4-3.8, p = 0.001) and digital ulcers (OR 1.9, 95 % CI 1.2-3.7, p = 0.01) were predictive of PAH in lcSSc; and oesophageal stricture (OR 4.4, 95 % CI 1.9-10.5, p = 0.001), mild ILD (OR 2.8, 95 % CI 1.2-6.8, p = 0.02) and ACA (OR 5.2, 95 % CI 1.8-14.8, p = 0.002) were predictive of PAH in dcSSc. Conclusions: The incidence and prevalence of PAH in this cohort are 0.7 % per annum and 8.4 %, respectively. The clinical-serologic risk factors for PAH differ based on disease subtype. In both subtypes, mild ILD is associated with PAH, suggesting the possibility of common pathogenic mechanisms underlying both of these disease manifestations. This model identifies a subset of patients at an appreciably higher risk of developing PAH, who should be screened and would in future, benefit from preventative therapies.