Background/Objectives: Pulmonary arterial hypertension (PAH) occurs in >12% of systemic sclerosis (SSc) patients and has a high mortality. Annual screening for all SSc patients is the standard of care and can include spirometry (RFT), serum N-terminal pro-B-type natriuretic peptide (NT-proBNP) and echocardiography (TTE). We sought to evaluate the accuracy of NT-proBNP alone to screen for SSc-PAH. Methods: We defined groups as low, intermediate and high risk of PAH according to RFT, TTE and RHC (right heart catheter) parameters. NT-proBNP thresholds predictive of risk group were defined, and ROC curves were used to evaluate the predictive accuracy of the calculated cut-points to diagnose SSc-PAH within the following 12 months. Results: A total of 820 SSc patients had NT-proBNP recorded 12 months prior to RHC, or if no RHC had been performed, within 12 months of the most recently recorded annual visit. The majority (335, 40.85%) were assigned to low risk of PAH, 189 (23.05%) to intermediate risk and 176 (21.46%) to high risk. NT-proBNP level 186.6 ng/L defined the high risk. NT-proBNP ≥210 ng/L with RFT (the ‘ASIG algorithm’) had the highest predictive accuracy, with sensitivity of 86.75% (77.52–93.19%), specificity of 60.71% (58.52–62.87%), PPV of 8.47% (6.69–10.55%), NPV of 99.09% (98.38–99.55%) and AUC of 0.74 (0.70–0.78). NT-proBNP ≥ 210 ng/L alone had sensitivity of 67.05% (56.21–76.70), specificity of 71.91% (70.03–73.72), PPV of 8.25% (6.34–10.52), NPV of 98.30% (97.57–98.86) and AUC of 0.69 (0.64–0.74). NT-proBNP thresholds associated with the three defined risk groups performed similarly overall. Conclusions: NT-proBNP alone, compared to the ASIG algorithm, had a slight reduction in NPV and AUC to predict PAH within 12 months. When NT-proBNP is ≥210 ng/L, the NPV remains high (98.0%).
Brown et al. (Sun,) studied this question.