Background/Purpose:
In Systemic
Sclerosis (SSc), pulmonary arterial hypertension (PAH) is often diagnosed at an
advanced stage. Serum levels of N-terminal pro-brain natriuretic peptide (NT-proBNP) are typically elevated in
established SSc-PAH, but it is not known if the serum levels rise prior to PAH
diagnosis, and it is not known if NT-proBNP levels differ between PAH and
pulmonary hypertension secondary to interstitial lung disease (PH-ILD); which
is resistant to treatment. Here, we utilized serial serum samples from a large,
prospective and unselected SSc cohort, and approached these two questions by
testing NT-proBNP levels in sera
obtained before and after PAH or PH-ILD diagnoses.

<>Methods: The
study cohort included patients from the prospective, observational Oslo
University Hospital SSc cohort (n=298) who met the 2013 ACR/EULAR
classification criteria
and had (A) baseline and follow-up
serum samples available for NT-proBNP analyses, (B) baseline and follow-up data on systolic pulmonary artery pressure (sPAP) by echocardiography (ECHO) and diffusion
capacity for carbon monoxide (DLCO), and (C) follow-up data on PAH and PH-ILD
diagnoses by right heart catheterization (RHC).

<>Results: The study cohort included 298 consecutive SSc patients
(243 female, 78 diffuse cutaneous SSc) aged
48 ± 15.4 years at disease onset. During an observation period of 11.5  ± 8.0 years 52/298 patients developed PH,
of those 32/52 were diagnosed with PAH, and 20/52 had PH-ILD. Serum levels of
NT-pro-BNP obtained at baseline (mean 1.4 years before PH diagnoses) did not
differ between cases with and without PH (Table 1). There was, however, a
difference in NT-proBNP levels between PH and non-PH cases in serum samples
obtained at follow-up, mean 2.6 years after the PH
diagnoses. Development of PH was positively associated with sPAP
values obtained at baseline, mean 1.6 years before PH diagnosis, and at
follow-up: Likewise, PH was also associated with DLCO at baseline and at
follow-up (Table 1). None of these parameters could significantly segregate PAH
from PH-ILD.

<>Conclusion: In this prospective SSc cohort we did not find any
difference in baseline NT-proBNP levels between PH and non PH cases; indicating
that NT-proBNP analyses are not useful for predicting PH development. In
contrast, we found that the values of sPAP and DLCO differed
between PH and non PH cases both at baseline and follow-up. None of the markers
tested could differentiate between PAH and PH-ILD. Hence, we are still missing
predictive biomarkers for the development of PAH.

Table 1: Associations
between PH and baseline and follow up characteristics