S.4.1 N-terminal pro-brain natriuretic peptide levels predict incident pulmonary arterial hypertension in SSc
Thakkar, V. ; Stevens, W. ; Prior, D. ; Byron, J. ; Patterson, K. ; Hissaria, P. ; Moore, O. ; Roddy, J. ; Zochling, J. ; Sahhar, J. ; Nash, P. ; Tymms, K. ; Youssef, P. ; Proudman, S. ; Nikpour, M. ; Launay, D. ; Sitbon, O. ; Cordier, J. F. ; Hachulla, E. ; Mouthon, L. ; Gressin, V. ; Rottat, L. ; Clerson, P. ; Simonneau, G. ; Humbert, M. ; Carreira, P. ; Carmona, L. ; Joven, B. E. ; Denton, C. P. ; Allanore, Y. ; Walker, U. A. ; Matucci-Cerinic, M. ; Muller-Ladner, U. ; Hsu, V. ; Cheng, Q. ; Steen, V.
In: Rheumatology, 2012, vol. 51, p. ii8-ii10
Ajouter à la liste personnelle- Summary
- Introduction. Pulmonary arterial hypertension (PAH) is a major cause of mortality in SSc. NT-proBNP may be a useful biomarker of prevalent PAH but its role in screening for incident PAH has not been evaluated. Methods. Patients recruited into the Australian Scleroderma Cohort Study undergo annual echocardiography, pulmonary function tests (PFTs), 6-min walk test (6MWT) and have serum NT-proBNP measured (ElecsysproBNP II). The diagnosis of PAH is based on Dana point criteria at right heart catheterization (RHC). Patients with LV dysfunction or eGFR <30 ml/min were excluded from this study. Four clinical groups were selected. Group 1 (n = 20) had definite PAH with pre-treatment sera assayed. Group 2 (n = 30) were considered ‘at risk' for PAH based on (i) sPAP on echo >36 mmHg, (ii) FVC/DLCO% >1.6 and no significant ILD, (iii) DLCO <50% or (iv) resting mPAP of 20-25 mmHg at RHC. Group 3 (n = 19) had interstitial lung disease (ILD) but no evidence of PAH on echo or RHC. Group 4 (n = 31) were SSc patients without cardiopulmonary disease. Analysis of variance with two-group comparisons were used to determine differences in clinical characteristics and NT-proBNP level among groups. Simple and multiple linear regression were used to determine correlates of NT-proBNP. ROC curve analysis was performed to determine the optimal NT-proBNP threshold for detection of PAH. Results. Patient characteristics are summarized in Table 1. As seen in Fig. 1, patients in Group 1 (PAH) had significantly higher mean NT-proBNP levels than patients in Group 4 (SSc controls; P < 0.0001). In addition, patients in Group 2 (‘at risk') had significantly higher NT-proBNP levels than those in Group 4 (SSc controls; P = 0.008). In simple linear regression, NT-proBNP was positively correlated with echo parameters (P < 0.0001). Among patients with PAH, NT-proBNP was positively correlated with mPAP on RHC (adjusted estimate = 0.048, 95% CI 0.01, 0.09, P = 0.019), independently of corrected DLCO, FVC/DLCO% ratio and 6MWD. An NT-proBNP cut-point of> 189.2 pg/ml had a likelihood ratio of 26.4 for presence of PAH (c-statistic = 0.9; sensitivity 85%; specificity 97%). An NT-proBNP level <82.9 pg/ml had a likelihood ratio of 6.8 for exclusion of PAH (sensitivity 67.7%, specificity 90%). Fig. 1NT-proBNP level according to group. Table 1. Group 1, PAH Group 2, at risk Group 3, ILD Group 4, SSc controls P-value Number 20 90 19 31 NA Sex, n (%) Female 16 (80) 25 (83) 14 (74) 31 (100) 0.041 Male 4 (22) 5 (17) 5 (26) 0 Disease subtype, n Limited 16 28 6 13 22 <0.0001 Diffuse 3 2 0 8 MCTD 1 0 1 Age at SSc onset, mean (s.d.), years 42 (114) 52 (15) 40 (16) 41 (13) 0.011 Age at time of study, mean (s.d.), years 62 (10.3) 66 (11.8) 51.1 (12.7) 48.4 (10) <0.0001 Disease duration at study, mean (s.d.), years 20.4 (12.8) 14.5 (10.4) 10.8 (7.9) 7.6 (7.2) 0.0001 Echo parameters, mean (s.d.) TR velocity, ms/s 3.81 (0.69) 2.95 (0.28) 2.58 (0.38) 2.20 (0.18) <0.0001 sPAP, mmHg 65.25 (27.82) 43.8 (7.16) 33.54 (7.73) 26.2 (2.57) <0.0001 RHC parameters, mean (s.d.) mPAP, mmHg 39.50 (12.36) NA NA NA NA mftAP, mmHg 9.94 (2.96) PVR 6.16 (3.10) 6MWD, m 335 (96) 401 (114) 458 (84) 502 (74) 0.0001 WHO-FC 1 0 5 2 22 <0.0001 II 3 10 10 4 III 12 13 6 2 IV 3 0 1 0 PFT FVC, per cent predicted 71.6 (24.7) 100.6 (19.5) 71 (14.6) 100.9 (13.12) <0.0001 Corr DLCO% 43.5 (12.2) 59.8 (14.7) 47.4 (13.8) 1.48 87.2 (14.3) <0.0001 fvC:0lCO ratio 1.68 (0.47) 1.75 (0.46) (0.35) 1.17 (0.19) <0.0001 NT-ProBNP, pg/ml 1791 (2509) 278 (243) 133 (87) 75 (43) <0.0001 Conclusions. Our findings suggest that in absence of LV dysfunction, NT-proBNP is a useful screening biomarker for PAH in SSc, with levels >189.2 pg/ml and <82.9 pg/ml defining patients with a high and low likelihood of PAH, respectively. Further prospective studies are required in unselected patients in order to confirm these findings