An evidence-based strategy to screen for pulmonary arterial hypertension in systemic sclerosis

Background Clinical practice guidelines recommend screening all systemic sclerosis (SSc) patients for pulmonary arterial hypertension (PAH) with yearly echocardiograms. There is a paucity of evidence to support these guidelines. Research question Can a prediction model identify SSc patients with a very low probability of PAH and therefore not requiring annual screening echocardiogram? Study design and methods We performed a case-control study of 925 unselected SSc subjects nested in a multi-centered, longitudinal cohort. The probability of PAH for each subject was calculated using the results of multivariate logistic regression models. A cut-off was identified for the estimated probability of PAH below which no subject developed PAH (100% sensitivity). Results Study subjects were predominantly female (87.5%), with mean (SD) age 58.6 (11.7) years and disease duration of 18.2 (12.2) years. Thirty-seven subjects developed PAH during 5407.97 person-years of observation (incidence rate 0.68 per 100 person-years). Shortness of breath (SOB), diffusing capacity for carbon monoxide (DLCO) and NT-proBNP were independent predictors of PAH. All SSc-PAH cases had a probability of PAH of >1.1%. Subjects below this cut-off, none of whom had PAH, accounted for 46.2% of the study population. Interpretation A simple prediction model identified subjects at very low probability of PAH who could potentially forego annual screening echocardiogram. This represents almost half of SSc subjects in a general SSc population. This study, which is the first evidence-based study for the rational use of follow-up echocardiograms in an unselected SSc cohort, requires validation. The scoring system is freely available online at http://pahtool.ladydavis.ca.