Background Several studies have investigated the predictors of morbidity and mortality in Systemic sclerosis (SSc). However long-term follow-up data from inception cohorts of early SSc patients are limited.

Objectives To identify predictors of morbidity and mortality in a single centre inception cohort of early SSc patients at long-term follow-up.

Methods Our inception cohort comprised SSc patients who fulfilled the American College of Rheumatology criteria, were recruited within 12 months of disease onset and followed prospectively for at least 3 years. Clinical manifestations, laboratory and lung function tests were recorded for each patient at baseline and at 3rd and 6th years of follow up. Multivariate regression analysis and Cox proportional hazard models were used to identify predictors (clinical manifestations, laboratory and lung function tests at baseline) of morbidity and mortality in SSc, respectively.

Results A total of 114 patients (96 female, mean age at diagnosis 48.1±13.5 years, 53 diffuse SSc subtype) were included in the study, from January 1997 to December 2012. All patients were followed for at least 3 years and 84 patients for at least 6 years. Twenty (17.5%) of 114 patients died during a mean follow up of 101.8±48.5 months. In multivariate regression analysis predictors for major SSc outcomes at 6 years were: diffuse subtype (OR: 6.2, p=0.015), anti-scl-70 positivity (OR: 3.9, p=0.05), esophageal involvement (OR: 6.5, p=0.017) and digital ulcers (OR: 8.6, p=0.003) at baseline for the development of pulmonary fibrosis (PF). The presence of digital ulcers at baseline was a predictor for the development of arrhythmias (OR: 3.7, p=0.05) and the presence of arrhythmias at baseline for the development of pulmonary hypertension (PH) (OR: 5.4, p=0.039).

Cox proportional hazard models multivariate analysis revealed that independent predictors of mortality were: male gender (HR: 3.3, p=0.023), diffuse type (HR: 8.7, p=0.004), PF at baseline (HR: 2.7, p=0.05), PH based on echocardiography at baseline (HR: 12.7, p=0.001) FVC <80% (HR: 2.74, p=0.042) and DLCO <60% of predicted value at baseline (HR: 2.97, p=0.019).

Conclusions Results from long-term follow-up data from a single centre inception cohort indicate that diffuse SSc subtype, anti-Scl-70 positivity, esophageal involvement and digital ulcers at baseline are independent predictors for the development of PF. Male gender, diffuse subtype, PF, PH and decreased FVC and DLCO at baseline are prognostic factors of mortality.

Disclosure of Interest None declared