Frequencies of borderline pulmonary hypertension before and after the DETECT algorithm: results from a prospective systemic sclerosis cohort

Frequencies of borderline pulmonary hypertension before and after the DETECT algorithm: results... Abstract Objective The DETECT algorithm was developed for screening patients with SSc at high risk of pulmonary arterial hypertension (PAH). We evaluated the impact of this algorithm in a SSc population. Methods Patients from the unselected, prospective Oslo University Hospital SSc study were divided into the Early and DETECT cohorts, respectively, depending on whether an incident right heart catheterization (RHC) was performed before (2009–13) or after (2014–17) the DETECT algorithm was instituted. A PAH diagnosis and patient risk stratification (low, intermediate and high risk) were performed according to 2015 European Society of Cardiology guidelines. Results At the time of the incident RHC, PAH frequency was similar between the DETECT (15/84 with PAH; 18%) and Early (16/77; 21%) cohorts, but more patients had borderline pulmonary hypertension (PH) in the DETECT (31%) than in the Early (17%) cohort. PAH risk levels were distributed differently. In the DETECT cohort, 27% and 27% were at low and high risk, respectively, at the time of PAH diagnosis. In the Early cohort, 19 and 44% were at low and high risk, respectively. A follow-up RHC, performed after [mean (SD)] 2.4 (1.8) years, showed that 39% of patients with borderline PH in the Early cohort had developed PAH. Conclusion The DETECT algorithm did not alter PAH incidence in this unselected SSc population. However, it appeared to affect the risk distribution at the time of PAH diagnosis and increased the frequency of borderline PH cases. These findings may translate into novel opportunities for earlier PAH treatment and, possibly, prevention. scleroderma and related disorders, respiratory, cardiovascular, epidemiology, autoinflammatory conditions Rheumatology key messages The number of new pulmonary arterial hypertension cases per year remained constant before and after DETECT. After DETECT, risk profiles and functional classes at pulmonary arterial hypertension diagnosis tended to improve. The number of borderline pulmonary hypertension cases increased after DETECT. Introduction SSc is a progressive autoimmune disorder characterized by fibrosis of the skin and internal organs, obliterative vascular pathology and distinct serum autoantibodies [1, 2]. SSc is associated with high mortality; the major cause of death is pre-capillary pulmonary hypertension (PH) [3–6]. The onset of PH in SSc is insidious, and it is often diagnosed late, with advanced stage vessel pathology [7]. The dominant forms of pre-capillary PH in SSc are pulmonary arterial hypertension (PAH), which is amenable to therapy (see below), and PH secondary to interstitial lung disease (ILD), which is less responsive to treatment [8, 9]. Recent studies have indicated that, in SSc-associated PAH, the effects of targeted PAH therapies on outcome appear to be most pronounced in patients with early stage PAH [10]. Therefore, it is crucial to diagnose PAH as early as possible. The European Society of Cardiology and the European Respiratory Society (ESC/ERS) guidelines from 2009 suggested that patients with symptomatic SSc should be screened with echocardiography (ECHO) to determine the risk of PAH. These guidelines were modified in 2015, to recommend annual screening with ECHO for all patients with SSc [11, 12]. In 2014, an evidence-based PAH screening algorithm was developed (DETECT) to identify patients with PAH at asymptomatic stages [13]. The authors demonstrated that DETECT could detect PAH with higher sensitivity than the ESC/ERS 2009 guidelines [13]. The gold standard criterion for PH diagnosis is a right heart catheterization (RHC) that demonstrates a mean pulmonary arterial pressure (mPAP) ⩾25 mmHg [12]. However, the mPAP threshold of 25 mmHg was extrapolated primarily from early data and expert opinion, originally published four decades ago, and currently, this threshold is mostly considered arbitrary [14]. Measurements in healthy individuals have shown that the normal resting mPAP is within the range of 14–17 mmHg, with very few outliers [15]. The clinical relevance of mPAP values in the upper normal range remains unknown, but mPAP >17 mmHg was shown to be associated with adverse events and reduced survival [15]. An analysis of RHC data obtained from the National Clinical Quality Program for All Veterans in the United States (the VA-CART national haemodynamic database) demonstrated that borderline PH, defined as mPAP = 19 − 24 mmHg, was a common independent risk factor for adverse clinical outcomes and mortality [16]. In SSc, borderline elevations of mPAP were shown to be associated with decreased exercise capacity and a high risk of developing PAH within a few years [17]. However, there are no guidelines for following up or treating patients with mPAP values that indicate borderline PH. The present study aimed to: evaluate the incidence of PAH and borderline PH in an unselected SSc cohort, before and after implementation of the DETECT algorithm; assess the WHO functional class of patients with borderline PH and PAH; and stratify patients with borderline PH and PAH in the risk groups defined by the 2015 ESC/ERS guidelines. Methods Study cohort and clinical parameters At the Oslo University Hospital (OUH), all patients with SSc are included in an ongoing, prospective, observational SSc cohort. All patients are followed by rheumatologists at OUH, and clinical, laboratory and imaging parameters are systematically recorded in the Norwegian Systemic Connective Tissue Disease and Vasculitis Registry [18]. The current study included all patients from the OUH SSc cohort that met the 2013 EULAR/ACR classification criteria for SSc and had undergone a primary, diagnostic RHC in the period of January 2009 to March 2017 [19]. The study cohort was divided into two groups: patients that underwent a first RHC in January 2009 to December 2013 (Early cohort), and patients that underwent a primary RHC in January 2014 to March 2017 (DETECT cohort). The DETECT cohort comprised patients that were additionally screened for PH with the DETECT algorithm [13]. Data on patient demographics, clinical parameters and SSc subsets were retrieved from the Norwegian Systemic Connective Tissue Disease and Vasculitis Registry [18]. SSc subsets were defined as lcSSc and dcSSc [20]. The observation period was defined as the time from the incident RHC until the study end, in June 2017, or death. The time to follow-up RHC was defined as the time from the incident RHC to the follow-up RHC. This study complied with the Declaration of Helsinki. The Regional Committee of Health and Medical Research Ethics in South-East Norway approved the research protocol (No.2006/119). Informed consent was obtained from all included subjects. PH surveillance In the time period from 2009 to 2013, the OUH protocol for PH surveillance in SSc included annual visits with a complete clinical examination, an ECHO, which provided an estimated systolic pulmonary arterial pressure, pulmonary function tests, a 6-min walking distance test and a measurement of N-terminal pro-brain natriuretic peptide (NT-proBNP). A referral for RHC was indicated when the clinical assessment pointed to a suspicion of PH, typically due to increasing or unexplained dyspnoea, a significant decline in the percentage of diffusing lung capacity for carbon monoxide (DLCO%), increasing NT-proBNP levels and a systolic pressure >40 mmHg on the ECHO. In January 2014, the DETECT algorithm was added to the OUH-SSc protocol, and all patients were referred to RHC when the DETECT score was >35. RHC procedures and diagnostic procedures A primary, diagnostic RHC was performed in all included patients from January 2009 to March 2017, with a Swan–Ganz pulmonary artery thermodilution catheter (Baxter Health Care Corp, Santa Ana, CA, USA). A follow-up RHC was conducted when clinically indicated. The mPAP and mean pulmonary capillary wedge (PCW) pressure were recorded (mmHg), cardiac output was measured with thermodilution, and the cardiac index was calculated by dividing cardiac output by the body surface area [21]. PCW was documented according to existing guidelines with breath holding close to end-expiration. PH was diagnosed, according to the updated European Society of Cardiology guidelines, as a mPAP ⩾25 mmHg, measured with RHC. The diagnosis was further divided into pre- and postcapillary PH, based on a threshold PCW of 15 mmHg [12, 22]. Borderline PH was defined as a mPAP of 20–24 mmHg [16, 22]. The PH diagnosis was performed by an experienced cardiologist. The criteria for a PAH (WHO group 1) diagnosis were: (i) the presence of precapillary PH; (ii) the absence of significant ILD, based on high resolution CT (HRCT) and/or a pulmonary function tests; absence of ILD was demonstrated by < 10% lung fibrosis on the HRCT at baseline and follow-up investigations and/or by a predicted percentage forced vital capacity (FVC%) >70% at baseline and follow-up; and (iii) other causes of pre-capillary PH could be excluded [23, 24]. The PH-ILD (WHO group 3) diagnosis was defined as precapillary PH combined with findings of lung fibrosis >10% with HRCT and/or FVC <70%. Angina pectoris and myocardial infarction were recorded when present. Risk stratification at the time of PAH diagnosis was estimated with the low, intermediate and high risk grouping system defined in the 2015 European Society of Cardiology recommendations for PAH [12]. Patients with ⩾1 parameter of poor prognosis were considered at high risk; those with ⩾1 parameter of intermediate prognosis were considered at intermediate risk. WHO functional classes (FC 1–4), established by the New York Heart Association, were assessed in all patients at the time of RHC. Statistical analyses Analyses were performed with SPSS Statistics version 22 (IBM Corp., Armonk, NY, USA) and STATA version 14 (StataCorp, College Station, TX, USA) software. Comparisons between groups were evaluated with the Pearson χ2 test, Fisher’s exact test and the Kruskal–Wallis t test, as appropriate. For analysing correlations, Pearson or Kendall’s tau-b coefficients were applied, as appropriate. Parameters with more than two variables were analysed with one-way analysis of variance (ANOVA), and post hoc tests were applied. Results Study cohort The total study cohort included 161 patients with SSc that had undergone a primary, diagnostic (incident) RHC at OUH during the study period, from January 2009 to March 2017. The mean age at inclusion was 61 (11.8) years, and the mean observation period was 2.9 (2.4) years. The cohort comprised 126 (78.8%) females, and 78 patients (48.8%) were ACA positive (Table 1). Of the 161 patients, 77 (48.0%) were in the Early cohort (incident RHC performed 2009–13) and 84 (52.0%) were in the DETECT cohort (incident RHC performed 2014–17). Demographic and clinical characteristics did not differ significantly between the Early and DETECT cohorts, except that the Early cohort had a longer observation period [4.2 (2.2) years vs 1.2 (0.8) years, P < 0.001; Table 1]. Table 1 Demographics and clinical characteristics of the SSc cohorts Characteristics  Total cohort  Early cohort  DETECT cohort  P-valuea  (n = 161)  (n = 77)  (n = 84)  Age at RHC, mean (s.d.), years  61 (11.8)  59.6 (12.2)  62.4 (11.3)  0.137  Time from SSc onset to RHC, mean (s.d.), years  6.6 (8.2)  6.3 (8.3)  6.9 (8.1)  0.634  Observation period, mean (s.d.), years  2.9 (2.4)  4.6 (2.2)  1.2 (0.8)  <0.001  Females, n (%)  126 (78.8)  60 (77.9)  66 (78.6)  0.582  Smoker (past or present), n (%)  67 (41.9)  28 (36.4)  39 (46.4)  0.064  lcSSc, n (%)  125 (78.1)  63 (81.8)  63 (75)  0.502  Anti-centromere Ab, n (%)  78 (48.8)  37 (48.1)  34 (40.5)  0.848  Digital ulcers, n (%)  61 (38.1)  31 (40.3)  30 (35.7)  0.911  Scleroderma renal crisis, n (%)  7 (4.4)  4 (5.2)  3 (3.6)  0.652  Baseline modified Rodnan skin score, mean (s.d.)  9.9 (8.8)  9.8 (10.5)  9.9 (9.8)  0.934  RHC results        <0.001      No PH, n (%)  57 (35.4)  29 (37.7)  28 (33.3)        Borderline PH, n (%)  39 (24.2)  13 (16.9)  26 (31.0)        Pulmonary hypertension, n (%)  65 (40.4)  35 (45.5)  30 (35.7)            PAH, n (%)  31 (19.3)  16 (20.8)  15 (17.9)            PH-ILD, n (%)  20 (12.4)  14 (18.2)  6 (7.1)            Post capillary PH, n (%)  14 (8.7)  5 (6.5)  9 (10.7)    Myocardial infarction, n (%)  14 (8.8)  8 (10.4)  6 (7.1)  0.557  Angina pectoris, n (%)  14 (10.1)  9 (11.7)  5 (6)  0.414  Characteristics  Total cohort  Early cohort  DETECT cohort  P-valuea  (n = 161)  (n = 77)  (n = 84)  Age at RHC, mean (s.d.), years  61 (11.8)  59.6 (12.2)  62.4 (11.3)  0.137  Time from SSc onset to RHC, mean (s.d.), years  6.6 (8.2)  6.3 (8.3)  6.9 (8.1)  0.634  Observation period, mean (s.d.), years  2.9 (2.4)  4.6 (2.2)  1.2 (0.8)  <0.001  Females, n (%)  126 (78.8)  60 (77.9)  66 (78.6)  0.582  Smoker (past or present), n (%)  67 (41.9)  28 (36.4)  39 (46.4)  0.064  lcSSc, n (%)  125 (78.1)  63 (81.8)  63 (75)  0.502  Anti-centromere Ab, n (%)  78 (48.8)  37 (48.1)  34 (40.5)  0.848  Digital ulcers, n (%)  61 (38.1)  31 (40.3)  30 (35.7)  0.911  Scleroderma renal crisis, n (%)  7 (4.4)  4 (5.2)  3 (3.6)  0.652  Baseline modified Rodnan skin score, mean (s.d.)  9.9 (8.8)  9.8 (10.5)  9.9 (9.8)  0.934  RHC results        <0.001      No PH, n (%)  57 (35.4)  29 (37.7)  28 (33.3)        Borderline PH, n (%)  39 (24.2)  13 (16.9)  26 (31.0)        Pulmonary hypertension, n (%)  65 (40.4)  35 (45.5)  30 (35.7)            PAH, n (%)  31 (19.3)  16 (20.8)  15 (17.9)            PH-ILD, n (%)  20 (12.4)  14 (18.2)  6 (7.1)            Post capillary PH, n (%)  14 (8.7)  5 (6.5)  9 (10.7)    Myocardial infarction, n (%)  14 (8.8)  8 (10.4)  6 (7.1)  0.557  Angina pectoris, n (%)  14 (10.1)  9 (11.7)  5 (6)  0.414  Demographics and clinical characteristics of all patients at OUH diagnosed with SSc that underwent an incident right heart catheterization (RHC) in January 2009 to March 2017 (total cohort); the Early and DETECT cohorts underwent incident RHCs, respectively, prior to (2009–13) and after (2014–17) establishment of the DETECT algorithm. aP-values indicate differences between the Early and DETECT cohorts. Ab: antibody; lcSSc: limited cutaneous SSc; PAH: pulmonary arterial hypertension; PH: pulmonary hypertension; PH-ILD: interstitial lung disease-associated pulmonary hypertension; RHC: right heart catheterization. Trends in RHC frequency and diagnostic outcome over the study period In addition to the incident RHC, at least one follow-up RHC procedure was performed in 69 (42.9%) patients from the study cohort, including 49 (63.6%) in the Early cohort and 20 (23.8%) in the DETECT cohort. The number of incident and follow-up RHCs conducted per year throughout the study period increased after 2013 (supplementary Fig. S1, available at Rheumatology Online). The outcomes of these RHC investigations (no PH, borderline PH or PAH) are shown in Fig. 1. The distribution of mPAP values (<17, 17–20, 20–24 or >25 mmHg) in the two cohorts were: n = 16 (20.8%), n = 13 (16.9%), n = 13 (16.9%) and n = 35 (45.5%), respectively, in the Early cohort, compared with n = 15 (19.3%), n = 10 (12%), n = 28 (33.7%) and n = 30 (36.1%), respectively, in the DETECT cohort (P = 0.146). Fig. 1 View largeDownload slide Frequencies of PAH, borderline PH and no PH Numbers of patients, from 2009 until 2016, diagnosed with PAH (red line), borderline PH (blue line) or no PH (normal mean arterial pressure; green line) at the time of right heart catheterization. PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Fig. 1 View largeDownload slide Frequencies of PAH, borderline PH and no PH Numbers of patients, from 2009 until 2016, diagnosed with PAH (red line), borderline PH (blue line) or no PH (normal mean arterial pressure; green line) at the time of right heart catheterization. PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Results from the incident RHC in the Early and DETECT cohorts The incident RHC was diagnostic for PAH in 16 (20.8%) patients in the Early cohort and 15 (17.9%) patients in the DETECT cohort (Table 1). There were no significant differences between the cohorts in the haemodynamic parameters at the incident RHC (supplementary Table S1, available at Rheumatology Online). In the Early cohort, 13 (16.9%) patients were diagnosed with borderline PH, compared with 26 (31.0%) patients in the DETECT cohort (Table 1). The corresponding mPAP values were 22.9 (1.2) mmHg and 21.8 (1.5) mmHg, respectively; P = 0.019 (Table 2). Table 2 Demographics and clinical characteristics of borderline PH Characteristics  Early cohort  DETECT cohort  P-value  (n = 13)  (n = 26)  Age at RHC, mean (s.d.), years  56.6 (13.4)  61.9 (11.0)  0.199  Time from SSc onset to RHC, mean (s.d.), years  9.9 (8.8)  9.1 (10.8)  0.769  Females, n (%)  8 (61.5)  23 (80.5)  0.090  lcSSc, n (%)  5 (38.5)  16 (61.5)  0.247  Anti-centromere Ab, n (%)  3 (23.1)  11 (47.8)  0.143  FVC/DLCO ratio, mean (s.d.)  1.9 (0.7)  1.8 (0.7)  0.557  Extent of fibrosis, mean (s.d.), %  7.9 (10.4)  6.1 (13.4)  0.684  At diagnosis, mean (s.d.)            mPAP, mmHg  22.9 (1.2)  21.8 (1.5)  0.019      CO, l/min  5.7 (1.3)  5.7 (1.5)  0.965      CI, l/min/m2  3.2 (0.6)  3.2 (0.7)  0.913  Follow-up RHC, n (%)  10 (76.9)  5 (19.2)    Time between incident and follow-up RHCs, mean (s.d.), years  2.4 (1.8)  0.7 (0.7)  <0.001  Developed PH, n (%)  5 (50)  2 (40)  0.828  Characteristics  Early cohort  DETECT cohort  P-value  (n = 13)  (n = 26)  Age at RHC, mean (s.d.), years  56.6 (13.4)  61.9 (11.0)  0.199  Time from SSc onset to RHC, mean (s.d.), years  9.9 (8.8)  9.1 (10.8)  0.769  Females, n (%)  8 (61.5)  23 (80.5)  0.090  lcSSc, n (%)  5 (38.5)  16 (61.5)  0.247  Anti-centromere Ab, n (%)  3 (23.1)  11 (47.8)  0.143  FVC/DLCO ratio, mean (s.d.)  1.9 (0.7)  1.8 (0.7)  0.557  Extent of fibrosis, mean (s.d.), %  7.9 (10.4)  6.1 (13.4)  0.684  At diagnosis, mean (s.d.)            mPAP, mmHg  22.9 (1.2)  21.8 (1.5)  0.019      CO, l/min  5.7 (1.3)  5.7 (1.5)  0.965      CI, l/min/m2  3.2 (0.6)  3.2 (0.7)  0.913  Follow-up RHC, n (%)  10 (76.9)  5 (19.2)    Time between incident and follow-up RHCs, mean (s.d.), years  2.4 (1.8)  0.7 (0.7)  <0.001  Developed PH, n (%)  5 (50)  2 (40)  0.828  Demographics, clinical characteristics and right heart catheterization (RHC) data for patients diagnosed with borderline pulmonary hypertension in the Early cohort (2009–13) and the DETECT cohort (2014–17). Ab: antibody; CI: cardiac input; CO: cardiac output; DLCO: diffusing lung capacity for carbon monoxide; FVC: forced vital capacity; lcSSc: limited cutaneous SSc; mPAP: mean pulmonary arterial pressure; n: number; RHC: right heart catheterization; PH: pulmonary hypertension. In the DETECT cohort, 28 patients (37.7%) had mPAPs <20 mmHg (no PH). These patients had significantly lower FVC/DLCO ratios, lower frequencies of systolic pulmonary arterial pressure >40 mmHg, higher frequencies of DLCO >60%, lower NT-proBNP levels and lower FCs than patients diagnosed with PH at the time of a referral for RHC (Table 3). Moreover, the DETECT cohort had 46 (54.8%) patients with a DLCO <60% and 53 (64.3%) patients with a disease duration >3 years (measured from disease onset at the time of referral for RHC; Table 3). We also found that the DETECT scores calculated before referral to RHC differed significantly between the mPAP subgroups measured at the incident RHC (P < 0.001) (supplementary Fig. S2, available at Rheumatology Online). Table 3 Clinical characteristics of DETECT cohort, segregated by PH subtype Characteristic  No PH (n = 28)  Borderline PH (n = 26)  PAH (n = 15)  PH-ILD (n = 6)  Post-capPH (n = 9)  P-valuea  Age at RHC, mean (s.d.), years  60.3 (13.6)  59.9 (10.9)  66.8 (9.6)  65 (9.5)  65.7 (8.9)  0.497  Time from SSc onset to RHC, mean (s.d.), years  6.9 (6.7)  9.7 (11.3)  6.3 (6.2)  6.7 (7.6)  3.9 (3.9)  0.330  Onset to RHC >3 years, n (%)  17 (60.7)  16 (61.5)  9 (60)  5 (83.3)  6 (66.7)  0.401  DLCO <60%, n (%)  6a (21.4)  14a (53.8)  14a (93.3)  6a (100)  6a (66.7)  <0.001  FVC/DLCO ratio, mean (s.d.)  1.6a (0.4)  1.8 (0.8)  2.9a (1.2)  2.7a (1.0)  2.1 (0.5)  <0.001*  NT-proBNP, mean (s.d.), µmol/l  32 (19.7)  39 (66.9)  189 (221.9)  38 (31.1)  341 (1024.9)  0.007  6MWD, mean (s.d.), m  514 (62.8)  502 (109.7)  356 (207.9)  456 (167.4)  278 (222.1)  0.747  sPAP, mean (s.d.), mmHg  31a (10.2)  31 (11.8)  52a (26)  53a (9.8)  49a (15.7)  0.006*  sPAP >30 mmHg, n (s.d.)  11 (39.3)  10 (38.5)  13 (86.7)  6 (100)  7 (77.8)  0.003  sPAP >40 mmHg, n (s.d.)  4a (14.3)  7 (26.9)  11a (73.3)  5a (83.3)  3 (33.3)  0.001*  FC, n (%)  19 (67.9)  26 (100)  15 (100)  5 (83.3)  8 (88.9)        FC 1 and 2, n (%)  17 (60.7)  22 (84.6)  7 (46.7)  2 (33.3)  5 (55.6)  0.008      FC 3 and 4, n (%)  2 (7.1)  4 (15.4)  8 (53.3)  3 (50)  3 (33.3)    Characteristic  No PH (n = 28)  Borderline PH (n = 26)  PAH (n = 15)  PH-ILD (n = 6)  Post-capPH (n = 9)  P-valuea  Age at RHC, mean (s.d.), years  60.3 (13.6)  59.9 (10.9)  66.8 (9.6)  65 (9.5)  65.7 (8.9)  0.497  Time from SSc onset to RHC, mean (s.d.), years  6.9 (6.7)  9.7 (11.3)  6.3 (6.2)  6.7 (7.6)  3.9 (3.9)  0.330  Onset to RHC >3 years, n (%)  17 (60.7)  16 (61.5)  9 (60)  5 (83.3)  6 (66.7)  0.401  DLCO <60%, n (%)  6a (21.4)  14a (53.8)  14a (93.3)  6a (100)  6a (66.7)  <0.001  FVC/DLCO ratio, mean (s.d.)  1.6a (0.4)  1.8 (0.8)  2.9a (1.2)  2.7a (1.0)  2.1 (0.5)  <0.001*  NT-proBNP, mean (s.d.), µmol/l  32 (19.7)  39 (66.9)  189 (221.9)  38 (31.1)  341 (1024.9)  0.007  6MWD, mean (s.d.), m  514 (62.8)  502 (109.7)  356 (207.9)  456 (167.4)  278 (222.1)  0.747  sPAP, mean (s.d.), mmHg  31a (10.2)  31 (11.8)  52a (26)  53a (9.8)  49a (15.7)  0.006*  sPAP >30 mmHg, n (s.d.)  11 (39.3)  10 (38.5)  13 (86.7)  6 (100)  7 (77.8)  0.003  sPAP >40 mmHg, n (s.d.)  4a (14.3)  7 (26.9)  11a (73.3)  5a (83.3)  3 (33.3)  0.001*  FC, n (%)  19 (67.9)  26 (100)  15 (100)  5 (83.3)  8 (88.9)        FC 1 and 2, n (%)  17 (60.7)  22 (84.6)  7 (46.7)  2 (33.3)  5 (55.6)  0.008      FC 3 and 4, n (%)  2 (7.1)  4 (15.4)  8 (53.3)  3 (50)  3 (33.3)    Clinical characteristics of 84 patients with SSc in the DETECT cohort, prior to a referral for right heart catheterization (RHC), segregated according to the outcome of the incident RHC. aSignificant findings in an ad hoc test compared with no PH. P-values were tested with one-way ANOVA. *significant findings in an ad hoc test compared to no PH. 6MWD: 6-min walking distance; DLCO: diffusing lung capacity for carbon monoxide; FC: functional class; FVC: forced vital capacity; NT-proBNP: N-terminal pro-brain natriuretic peptide; PAH: pulmonary arterial hypertension; PH: pulmonary hypertension; PH-ILD: interstitial lung disease associated pulmonary hypertension; post-capPH: post-capillary PH; RHC: right heart catheterization; sPAP: systolic pulmonary arterial pressure measured on ECHO. Functional status and risk assessment in patients with PAH or borderline PH At the time of the PAH diagnosis, the Early and DETECT cohorts were similar in New York Heart Association functional classes and risk stratifications, according to the 2015 ESC/ERS guidelines. In the DETECT cohort, four patients (26.7%) were at low risk, seven (46.7%) were at intermediate risk and four (26.7%) were at high risk of PAH poor prognosis. Corresponding numbers in the Early cohort were three (18.8%), six (37.5%) and seven (43.8%), respectively (Fig. 2A and B). Distribution of the clinical features for risk stratification is shown in supplementary Table S2, available at Rheumatology Online. Fig. 2 View largeDownload slide Functional class and risk stratification of patients with PAH or borderline PH diagnosis Results are shown for different diagnostic subsets of the DETECT and Early cohorts. (A) Functional class (NYHA 1–4) and (B) risk stratification (low-high), measured at the time of PAH diagnosis. (C) Functional class and (D) risk stratification, measured at the time of the borderline PH diagnosis. Differences between groups are indicated with P-values (at the top of each panel). NYHA: New York Heart Association; PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Fig. 2 View largeDownload slide Functional class and risk stratification of patients with PAH or borderline PH diagnosis Results are shown for different diagnostic subsets of the DETECT and Early cohorts. (A) Functional class (NYHA 1–4) and (B) risk stratification (low-high), measured at the time of PAH diagnosis. (C) Functional class and (D) risk stratification, measured at the time of the borderline PH diagnosis. Differences between groups are indicated with P-values (at the top of each panel). NYHA: New York Heart Association; PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. At the time of the borderline PH diagnosis, a trend of improvement in functional class and significantly lower risk levels were observed in the DETECT cohort compared with the Early cohort. Specifically, in the Early cohort, two patients (22.2%) were at low risk, six (66.7%) were at intermediate risk and one patient (11.1%) was at high risk of PAH poor prognosis. Corresponding numbers in the DETECT cohort for patients at low, intermediate or high risk were 13 (59.1%), nine (40.9%) and zero (0%), respectively; P = 0.030 (Fig. 2C and D). Results from the follow-up RHC in the study cohort In the Early cohort, 26 patients (63%) had borderline or no PH on the incident RHC. These patients were subjected to a follow-up RHC after a mean of 2.4 (1.8) years (Fig. 3, left panel). The follow up RHC results showed that, among the 10 patients with borderline PH on the incident RHC, five (38.5%) developed PH after a mean of 2.2 (1.6) years. Among the 16 patients with no PH, eight (28.6%) developed borderline PH after a mean of 2.6 (2.3) years, and four (14.3%) developed PH after a mean of 3 (1.2) years (Fig. 3). Of the remaining 15 patients with SSc in the Early cohort that did not undergo a follow-up RHC, four patients died, five were followed at their local department of rheumatology, five were clinically stable and one is currently awaiting a follow-up RHC, due to suspicion of PH development. Of the 77 patients with SSc in the Early cohort, 27 patients (35.1%) died during the observation period; the proportions of deaths were not significantly different among the PH subgroups. Fig. 3 View largeDownload slide Flowchart of follow-up right heart catheterization Flowchart of follow-up right heart catheterization (RHC) in patients with SSc without pulmonary hypertension (no PH) and with borderline PH at baseline in the Early and DETECT cohorts. PH: pulmonary hypertension. Fig. 3 View largeDownload slide Flowchart of follow-up right heart catheterization Flowchart of follow-up right heart catheterization (RHC) in patients with SSc without pulmonary hypertension (no PH) and with borderline PH at baseline in the Early and DETECT cohorts. PH: pulmonary hypertension. In the DETECT cohort, seven patients (13%) underwent a follow-up RHC after a mean of 0.7 (0.7) years after the incident RHC (Fig. 3). Due to the short follow-up time and few patients, no further analyses or comparisons were conducted within the DETECT cohort, regarding new onset PH or survival. Discussion There is a need to improve the early diagnosis of PAH in SSc to optimize treatment effects and improve survival. The DETECT algorithm was established in 2014, with the intention of identifying patients with PAH at asymptomatic stages. Our results indicated that, after the introduction of DETECT, the incidence of PAH was stable, but the number of cases diagnosed with borderline PH increased significantly. Interestingly, we also found that, after DETECT was established, patients with borderline PH had significantly lower risk scores at diagnosis than patients with borderline PH identified in the Early, pre-DETECT cohort. To our knowledge, this study was the first to compare the relative frequencies of PAH and borderline PH in a population-based SSc cohort before and after the introduction of the DETECT algorithm. Other studies focused on the sensitivity and specificity of DETECT. Those studies demonstrated the benefits of DETECT, compared with the ESC/ERS screening guidelines for PAH [25–27] and borderline PH [25, 28]. In our clinical practice, we found no change in the PAH incidence after introducing the DETECT algorithm for PH surveillance. However, we did find that the number of new borderline PH cases increased significantly after including the DETECT algorithm; this increase was accompanied by a concomitant shift in the 2015 ESC/ERS risk score distribution [12]. We are aware that these risk scores were primarily developed for PAH, rather than borderline PH; however, because the mPAP threshold of 25 mmHg was mostly an arbitrary value, and considering the fact that mortality increased in patients with borderline PH, it is clinically important to identify patients with mPAP values of 20–25 mmHg and consider their risk scores [14, 17]. However, the predictive value of risk scores in patients with SSc-related borderline PH remains to be confirmed with larger, prospective studies, preferably in population-based cohorts. Another interesting finding is the lower percentage of cases with PH-ILD in the DETECT cohort compared with the Early cohort. We do not know the exact reason, but it appears likely that the reduction reflects a change in the RHC referral pattern, with relatively fewer referrals of patients with severe lung fibrosis. Of note, we added the DETECT algorithm to annual PH screening in our clinical practice. Therefore, the DETECT algorithm was also performed in patients with DLCO >60%. We are aware that DETECT has not been validated for those patients, but it was unavoidable, because we included all patients with SSc that were considered at PH risk, based on clinical assessments at the annual PH screening. The current study reinforced the notion that patients with SSc and borderline PH are at high risk of developing PAH. In fact, we observed a frequency of 38.5% PAH development after 2.4 years; that frequency was within the range observed in previous studies. In a large UK cohort, 30% of patients with SSc and borderline PH developed PAH. Additionally, in the PHAROS cohort, 55% of patients with borderline PH developed PAH [29, 30]. Given this high risk and the high mortality rate of established PAH, it is a sobering thought that there are no guidelines for following up borderline PH, and there have been no randomized trials on testing strategies for PAH prevention [10, 22, 31, 32]. In animal models, PAH was shown to be driven by inflammatory responses. Local immune activation led to the dysfunction of vascular endothelial cells and smooth muscle cells, which resulted in neo-intimal hyperplasia in small pulmonary arteries [33]. In these models, it was shown that immune-modulation with MMF (a purine inhibitor), which was approved for organ transplantation, could attenuate the development of PAH [34, 35]. It is unknown whether MMF could prevent PAH development in SSc more efficiently than the approved PAH therapies, which target the vascular component of PAH. The latter therapies aim to maximize the properties of vessels. Nevertheless, it is interesting to note that preliminary data from the OUH SSc cohort indicated that the frequency of PAH in patients with SSc was reduced in a subset of patients that had received MMF treatment for skin/lung disease. Randomized clinical trials are warranted to determine whether PAH is preventable in borderline PH and, eventually, to determine optimal prevention strategies. The current study reinforced the notion that the DETECT algorithm has high sensitivity. However, in this study, when we added the algorithm to our clinical screening programme, it had rather low specificity. Nevertheless, our findings suggested that an annual PAH screening programme that included DETECT would be likely to increase the total number of RHC investigations and increase the frequency of patients with SSc that have normal or borderline mPAP. Borderline mPAP justifies an early RHC follow-up, but prospective, randomized clinical trials are necessary to determine whether patients with borderline PH would benefit from this strategy. To avoid unnecessary invasive RHC procedures and to identify patients most likely to develop PH, additional, more specific PAH markers are needed that are suitable for integrating into clinical follow-up procedures, including the DETECT algorithm. Our study had some limitations. First, a follow-up RHC was not performed in the entire SSc cohort, and second, the follow-up period was rather short for the DETECT cohort. The frequency of borderline PH and PAH was rather low, because the OUH cohort was population-based, which covered the whole spectrum of disease. Larger, prospective studies, preferably in population-based cohorts, are needed. The strengths of our study were that our study prospectively enrolled patients with SSc over a 9-year period, and all included patients had undergone an incident RHC. Additionally, we compared the frequencies of PAH and borderline PH in two cohorts, before and after including the DETECT algorithm in the annual PH screening programme. In conclusion, we showed that the number of new PAH cases per year has been stable since 2009, with no apparent change after introducing the DETECT algorithm into PH screening. Starting in 2014, we observed a non-significant increase in the number of newly diagnosed PAH cases at low risk and better functional class status. The total number of RHCs increased, starting in 2014, with a concomitant increase in the number of borderline PH cases. Funding: This work was supported by grants from the Norwegian Women’s Public Health Association. Disclosure statement: A.A. has received honoraria and research grants from Actelion, GSK, Bayer, MSD and United Therapeutics. All other authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Allanore Y, Denton C, Matucci-Cerinic M. Systemic sclerosis. In: Bijlsma JWJ, ed. EULAR Textbook on Rheumatic Diseases.  London: BMJ, 2012; 646– 567. 2 Pattanaik D, Brown M, Postlethwaite B, Postlethwaite A. Pathogenesis of systemic sclerosis. Front Immunol  2015; 6: 272. Google Scholar CrossRef Search ADS PubMed  3 Tyndall AJ, Bannert B, Vonk M et al.   Causes and risk factors for death in systemic sclerosis: a study from the EULAR Scleroderma Trials and Research (EUSTAR) database. Ann Rheum Dis  2010; 69: 1809– 15. http://dx.doi.org/10.1136/ard.2009.114264 Google Scholar CrossRef Search ADS PubMed  4 Rubio-Rivas M, Royo C, Simeón C, Corbella X, Fonollosa V. Mortality and survival in systemic sclerosis: systematic review and meta-analysis. Semin Arthritis Rheum  2014; 44: 208– 19. Google Scholar CrossRef Search ADS PubMed  5 Hoffmann-Vold AM, Molberg O, Midtvedt O, Garen T, Gran JT. Survival and causes of death in an unselected and complete cohort of Norwegian patients with systemic sclerosis. J Rheumatol  2013; 40: 1127– 33. Google Scholar CrossRef Search ADS PubMed  6 Hesselstrand R, Wildt M, Ekmehag B, Wuttge DM, Scheja A. Survival in patients with pulmonary arterial hypertension associated with systemic sclerosis from a Swedish single centre: prognosis still poor and prediction difficult. Scand J Rheumatol  2011; 40: 127– 32. Google Scholar CrossRef Search ADS PubMed  7 Chaisson NF, Hassoun PM. Systemic sclerosis-associated pulmonary arterial hypertension. Chest  2013; 144: 1346– 56. http://dx.doi.org/10.1378/chest.12-2396 Google Scholar CrossRef Search ADS PubMed  8 Hervier B, Meyer A, Dieval C et al.   Pulmonary hypertension in antisynthetase syndrome: prevalence, aetiology and survival. Eur Respir J  2013; 42: 1271– 82. http://dx.doi.org/10.1183/09031936.00156312 Google Scholar CrossRef Search ADS PubMed  9 Le Pavec J, Girgis RE, Lechtzin N et al.   Systemic sclerosis-related pulmonary hypertension associated with interstitial lung disease: impact of pulmonary arterial hypertension therapies. Arthritis Rheum  2011; 63: 2456– 64. http://dx.doi.org/10.1002/art.30423 Google Scholar CrossRef Search ADS PubMed  10 Morrisroe K, Stevens W, Huq M et al.   Survival and quality of life in incident systemic sclerosis-related pulmonary arterial hypertension. Arthritis Res Ther  2017; 19: 122. http://dx.doi.org/10.1186/s13075-017-1341-x Google Scholar CrossRef Search ADS PubMed  11 Galie N, Hoeper MM, Humbert M et al.   Guidelines for the diagnosis and treatment of pulmonary hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J  2009; 30: 2493– 537. http://dx.doi.org/10.1093/eurheartj/ehp297 Google Scholar CrossRef Search ADS PubMed  12 Galie N, Humbert M, Vachiery JL et al.   2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Heart J  2016; 37: 67– 119. http://dx.doi.org/10.1093/eurheartj/ehv317 Google Scholar CrossRef Search ADS PubMed  13 Coghlan JG, Denton CP, Grunig E et al.   Evidence-based detection of pulmonary arterial hypertension in systemic sclerosis: the DETECT study. Ann Rheum Dis  2014; 73: 1340– 9. http://dx.doi.org/10.1136/annrheumdis-2013-203301 Google Scholar CrossRef Search ADS PubMed  14 Hatano S, Strasser T. Primary Pulmonary Hypertension . Geneva, Switzerland: World Health Organization, 1975. 15 Kovacs G, Berghold A, Scheidl S, Olschewski H. Pulmonary arterial pressure during rest and exercise in healthy subjects: a systematic review. Eur Respir J  2009; 34: 888– 94. http://dx.doi.org/10.1183/09031936.00145608 Google Scholar CrossRef Search ADS PubMed  16 Maron BA, Hess E, Maddox TM et al.   Association of borderline pulmonary hypertension with mortality and hospitalization in a large patient cohort: Insights from the Veterans Affairs Clinical Assessment, Reporting, and Tracking Program. Circulation  2016; 133: 1240– 8. http://dx.doi.org/10.1161/CIRCULATIONAHA.115.020207 Google Scholar CrossRef Search ADS PubMed  17 Kovacs G, Avian A, Tscherner M et al.   Characterization of patients with borderline pulmonary arterial pressure. Chest  2014; 146: 1486– 93. http://dx.doi.org/10.1378/chest.14-0194 Google Scholar CrossRef Search ADS PubMed  18 Hoffmann-Vold AM, Gunnarsson R, Garen T, Midtvedt O, Molberg O. Performance of the 2013 American College of Rheumatology/European League Against Rheumatism Classification Criteria for Systemic Sclerosis (SSc) in large, well-defined cohorts of SSc and mixed connective tissue disease. J Rheumatol  2015; 42: 60– 3. Google Scholar CrossRef Search ADS PubMed  19 van den Hoogen F, Khanna D, Fransen J et al.   2013 classification criteria for systemic sclerosis: an American College of Rheumatology/European League against Rheumatism collaborative initiative. Arthritis Rheum  2013; 65: 2737– 47. http://dx.doi.org/10.1002/art.38098 Google Scholar CrossRef Search ADS PubMed  20 LeRoy E, Black C, Fleischmajer R et al.   Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol  1988; 15: 202– 5. Google Scholar PubMed  21 Gude E, Simonsen S, Geiran OR et al.   Pulmonary hypertension in heart transplantation: discrepant prognostic impact of pre-operative compared with 1-year post-operative right heart hemodynamics. J Heart Lung Transplant  2010; 29: 216– 23. http://dx.doi.org/10.1016/j.healun.2009.08.021 Google Scholar CrossRef Search ADS PubMed  22 Hoeper MM, Bogaard HJ, Condliffe R et al.   Definitions and diagnosis of pulmonary hypertension. J Am Coll Cardiol  2013; 62(25 Suppl): D42– 50. Google Scholar CrossRef Search ADS   23 Hoffmann-Vold AM, Aalokken TM, Lund MB et al.   Predictive value of serial high-resolution computed tomography analyses and concurrent lung function tests in systemic sclerosis. Arthritis Rheumatol  2015; 67: 2205– 12. http://dx.doi.org/10.1002/art.39166 Google Scholar CrossRef Search ADS PubMed  24 Hoffmann-Vold A, Tennøe AH, Garen T et al.   High level of chemokine CCL18 is associated with pulmonary function deterioration, lung fibrosis progression and reduced survival in systemic sclerosis. Chest  2016; 150: 299– 306. Google Scholar CrossRef Search ADS PubMed  25 Guillen-Del Castillo A, Callejas-Moraga EL, Garcia G et al.   High sensitivity and negative predictive value of the DETECT algorithm for an early diagnosis of pulmonary arterial hypertension in systemic sclerosis: application in a single center. Arthritis Res Ther  2017; 19: 135. Google Scholar CrossRef Search ADS PubMed  26 Soukup T, Pudil R, Kubinova K et al.   Application of the DETECT algorithm for detection of risk of pulmonary arterial hypertension in systemic sclerosis: data from a Czech tertiary centre. Rheumatology  2016; 55: 109– 14. http://dx.doi.org/10.1093/rheumatology/kev327 Google Scholar CrossRef Search ADS PubMed  27 Hao Y, Thakkar V, Stevens W et al.   A comparison of the predictive accuracy of three screening models for pulmonary arterial hypertension in systemic sclerosis. Arthritis Res Ther  2015; 17: 7. http://dx.doi.org/10.1186/s13075-015-0517-5 Google Scholar CrossRef Search ADS PubMed  28 Visovatti SH, Distler O, Coghlan JG et al.   Borderline pulmonary arterial pressure in systemic sclerosis patients: a post-hoc analysis of the DETECT study. Arthritis Res Ther  2014; 16: 493. http://dx.doi.org/10.1186/s13075-014-0493-1 Google Scholar CrossRef Search ADS PubMed  29 Valerio CJ, Schreiber BE, Handler CE, Denton CP, Coghlan JG. Borderline mean pulmonary artery pressure in patients with systemic sclerosis: transpulmonary gradient predicts risk of developing pulmonary hypertension. Arthritis Rheum  2013; 65: 1074– 84. Google Scholar CrossRef Search ADS PubMed  30 Bae S, Saggar R, Bolster MB et al.   Baseline characteristics and follow-up in patients with normal haemodynamics versus borderline mean pulmonary arterial pressure in systemic sclerosis: results from the PHAROS registry. Ann Rheum Dis  2012; 71: 1335– 42. http://dx.doi.org/10.1136/annrheumdis-2011-200546 Google Scholar CrossRef Search ADS PubMed  31 Kovacs G, Olschewski H. Borderline pulmonary pressures in scleroderma – a ‘pre-pulmonary arterial hypertension’ condition? Arthritis Res Ther  2015; 17: 123. Google Scholar CrossRef Search ADS PubMed  32 Nihtyanova S, Schreiber B, Ong V et al.   Prediction of pulmonary complications and long-term survival in systemic sclerosis. Arthritis Rheumatol  2014; 66: 1625– 35. http://dx.doi.org/10.1002/art.38390 Google Scholar CrossRef Search ADS PubMed  33 Rabinovitch M, Guignabert C, Humbert M, Nicolls MR. Inflammation and immunity in the pathogenesis of pulmonary arterial hypertension. Circ Res  2014; 115: 165– 75. Google Scholar CrossRef Search ADS PubMed  34 Suzuki C, Takahashi M, Morimoto H et al.   Mycophenolate mofetil attenuates pulmonary arterial hypertension in rats. Biochem Biophys Res Commun  2006; 349: 781– 8. http://dx.doi.org/10.1016/j.bbrc.2006.08.109 Google Scholar CrossRef Search ADS PubMed  35 Zhang YF, Zheng Y. The effects of mycophenolate mofetil on cytokines and their receptors in pulmonary arterial hypertension in rats. Scand J Rheumatol  2015; 44: 412– 5. http://dx.doi.org/10.3109/03009742.2015.1023829 Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Frequencies of borderline pulmonary hypertension before and after the DETECT algorithm: results from a prospective systemic sclerosis cohort

Loading next page...
 
/lp/ou_press/frequencies-of-borderline-pulmonary-hypertension-before-and-after-the-muXBLQoU6H
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/kex435
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective The DETECT algorithm was developed for screening patients with SSc at high risk of pulmonary arterial hypertension (PAH). We evaluated the impact of this algorithm in a SSc population. Methods Patients from the unselected, prospective Oslo University Hospital SSc study were divided into the Early and DETECT cohorts, respectively, depending on whether an incident right heart catheterization (RHC) was performed before (2009–13) or after (2014–17) the DETECT algorithm was instituted. A PAH diagnosis and patient risk stratification (low, intermediate and high risk) were performed according to 2015 European Society of Cardiology guidelines. Results At the time of the incident RHC, PAH frequency was similar between the DETECT (15/84 with PAH; 18%) and Early (16/77; 21%) cohorts, but more patients had borderline pulmonary hypertension (PH) in the DETECT (31%) than in the Early (17%) cohort. PAH risk levels were distributed differently. In the DETECT cohort, 27% and 27% were at low and high risk, respectively, at the time of PAH diagnosis. In the Early cohort, 19 and 44% were at low and high risk, respectively. A follow-up RHC, performed after [mean (SD)] 2.4 (1.8) years, showed that 39% of patients with borderline PH in the Early cohort had developed PAH. Conclusion The DETECT algorithm did not alter PAH incidence in this unselected SSc population. However, it appeared to affect the risk distribution at the time of PAH diagnosis and increased the frequency of borderline PH cases. These findings may translate into novel opportunities for earlier PAH treatment and, possibly, prevention. scleroderma and related disorders, respiratory, cardiovascular, epidemiology, autoinflammatory conditions Rheumatology key messages The number of new pulmonary arterial hypertension cases per year remained constant before and after DETECT. After DETECT, risk profiles and functional classes at pulmonary arterial hypertension diagnosis tended to improve. The number of borderline pulmonary hypertension cases increased after DETECT. Introduction SSc is a progressive autoimmune disorder characterized by fibrosis of the skin and internal organs, obliterative vascular pathology and distinct serum autoantibodies [1, 2]. SSc is associated with high mortality; the major cause of death is pre-capillary pulmonary hypertension (PH) [3–6]. The onset of PH in SSc is insidious, and it is often diagnosed late, with advanced stage vessel pathology [7]. The dominant forms of pre-capillary PH in SSc are pulmonary arterial hypertension (PAH), which is amenable to therapy (see below), and PH secondary to interstitial lung disease (ILD), which is less responsive to treatment [8, 9]. Recent studies have indicated that, in SSc-associated PAH, the effects of targeted PAH therapies on outcome appear to be most pronounced in patients with early stage PAH [10]. Therefore, it is crucial to diagnose PAH as early as possible. The European Society of Cardiology and the European Respiratory Society (ESC/ERS) guidelines from 2009 suggested that patients with symptomatic SSc should be screened with echocardiography (ECHO) to determine the risk of PAH. These guidelines were modified in 2015, to recommend annual screening with ECHO for all patients with SSc [11, 12]. In 2014, an evidence-based PAH screening algorithm was developed (DETECT) to identify patients with PAH at asymptomatic stages [13]. The authors demonstrated that DETECT could detect PAH with higher sensitivity than the ESC/ERS 2009 guidelines [13]. The gold standard criterion for PH diagnosis is a right heart catheterization (RHC) that demonstrates a mean pulmonary arterial pressure (mPAP) ⩾25 mmHg [12]. However, the mPAP threshold of 25 mmHg was extrapolated primarily from early data and expert opinion, originally published four decades ago, and currently, this threshold is mostly considered arbitrary [14]. Measurements in healthy individuals have shown that the normal resting mPAP is within the range of 14–17 mmHg, with very few outliers [15]. The clinical relevance of mPAP values in the upper normal range remains unknown, but mPAP >17 mmHg was shown to be associated with adverse events and reduced survival [15]. An analysis of RHC data obtained from the National Clinical Quality Program for All Veterans in the United States (the VA-CART national haemodynamic database) demonstrated that borderline PH, defined as mPAP = 19 − 24 mmHg, was a common independent risk factor for adverse clinical outcomes and mortality [16]. In SSc, borderline elevations of mPAP were shown to be associated with decreased exercise capacity and a high risk of developing PAH within a few years [17]. However, there are no guidelines for following up or treating patients with mPAP values that indicate borderline PH. The present study aimed to: evaluate the incidence of PAH and borderline PH in an unselected SSc cohort, before and after implementation of the DETECT algorithm; assess the WHO functional class of patients with borderline PH and PAH; and stratify patients with borderline PH and PAH in the risk groups defined by the 2015 ESC/ERS guidelines. Methods Study cohort and clinical parameters At the Oslo University Hospital (OUH), all patients with SSc are included in an ongoing, prospective, observational SSc cohort. All patients are followed by rheumatologists at OUH, and clinical, laboratory and imaging parameters are systematically recorded in the Norwegian Systemic Connective Tissue Disease and Vasculitis Registry [18]. The current study included all patients from the OUH SSc cohort that met the 2013 EULAR/ACR classification criteria for SSc and had undergone a primary, diagnostic RHC in the period of January 2009 to March 2017 [19]. The study cohort was divided into two groups: patients that underwent a first RHC in January 2009 to December 2013 (Early cohort), and patients that underwent a primary RHC in January 2014 to March 2017 (DETECT cohort). The DETECT cohort comprised patients that were additionally screened for PH with the DETECT algorithm [13]. Data on patient demographics, clinical parameters and SSc subsets were retrieved from the Norwegian Systemic Connective Tissue Disease and Vasculitis Registry [18]. SSc subsets were defined as lcSSc and dcSSc [20]. The observation period was defined as the time from the incident RHC until the study end, in June 2017, or death. The time to follow-up RHC was defined as the time from the incident RHC to the follow-up RHC. This study complied with the Declaration of Helsinki. The Regional Committee of Health and Medical Research Ethics in South-East Norway approved the research protocol (No.2006/119). Informed consent was obtained from all included subjects. PH surveillance In the time period from 2009 to 2013, the OUH protocol for PH surveillance in SSc included annual visits with a complete clinical examination, an ECHO, which provided an estimated systolic pulmonary arterial pressure, pulmonary function tests, a 6-min walking distance test and a measurement of N-terminal pro-brain natriuretic peptide (NT-proBNP). A referral for RHC was indicated when the clinical assessment pointed to a suspicion of PH, typically due to increasing or unexplained dyspnoea, a significant decline in the percentage of diffusing lung capacity for carbon monoxide (DLCO%), increasing NT-proBNP levels and a systolic pressure >40 mmHg on the ECHO. In January 2014, the DETECT algorithm was added to the OUH-SSc protocol, and all patients were referred to RHC when the DETECT score was >35. RHC procedures and diagnostic procedures A primary, diagnostic RHC was performed in all included patients from January 2009 to March 2017, with a Swan–Ganz pulmonary artery thermodilution catheter (Baxter Health Care Corp, Santa Ana, CA, USA). A follow-up RHC was conducted when clinically indicated. The mPAP and mean pulmonary capillary wedge (PCW) pressure were recorded (mmHg), cardiac output was measured with thermodilution, and the cardiac index was calculated by dividing cardiac output by the body surface area [21]. PCW was documented according to existing guidelines with breath holding close to end-expiration. PH was diagnosed, according to the updated European Society of Cardiology guidelines, as a mPAP ⩾25 mmHg, measured with RHC. The diagnosis was further divided into pre- and postcapillary PH, based on a threshold PCW of 15 mmHg [12, 22]. Borderline PH was defined as a mPAP of 20–24 mmHg [16, 22]. The PH diagnosis was performed by an experienced cardiologist. The criteria for a PAH (WHO group 1) diagnosis were: (i) the presence of precapillary PH; (ii) the absence of significant ILD, based on high resolution CT (HRCT) and/or a pulmonary function tests; absence of ILD was demonstrated by < 10% lung fibrosis on the HRCT at baseline and follow-up investigations and/or by a predicted percentage forced vital capacity (FVC%) >70% at baseline and follow-up; and (iii) other causes of pre-capillary PH could be excluded [23, 24]. The PH-ILD (WHO group 3) diagnosis was defined as precapillary PH combined with findings of lung fibrosis >10% with HRCT and/or FVC <70%. Angina pectoris and myocardial infarction were recorded when present. Risk stratification at the time of PAH diagnosis was estimated with the low, intermediate and high risk grouping system defined in the 2015 European Society of Cardiology recommendations for PAH [12]. Patients with ⩾1 parameter of poor prognosis were considered at high risk; those with ⩾1 parameter of intermediate prognosis were considered at intermediate risk. WHO functional classes (FC 1–4), established by the New York Heart Association, were assessed in all patients at the time of RHC. Statistical analyses Analyses were performed with SPSS Statistics version 22 (IBM Corp., Armonk, NY, USA) and STATA version 14 (StataCorp, College Station, TX, USA) software. Comparisons between groups were evaluated with the Pearson χ2 test, Fisher’s exact test and the Kruskal–Wallis t test, as appropriate. For analysing correlations, Pearson or Kendall’s tau-b coefficients were applied, as appropriate. Parameters with more than two variables were analysed with one-way analysis of variance (ANOVA), and post hoc tests were applied. Results Study cohort The total study cohort included 161 patients with SSc that had undergone a primary, diagnostic (incident) RHC at OUH during the study period, from January 2009 to March 2017. The mean age at inclusion was 61 (11.8) years, and the mean observation period was 2.9 (2.4) years. The cohort comprised 126 (78.8%) females, and 78 patients (48.8%) were ACA positive (Table 1). Of the 161 patients, 77 (48.0%) were in the Early cohort (incident RHC performed 2009–13) and 84 (52.0%) were in the DETECT cohort (incident RHC performed 2014–17). Demographic and clinical characteristics did not differ significantly between the Early and DETECT cohorts, except that the Early cohort had a longer observation period [4.2 (2.2) years vs 1.2 (0.8) years, P < 0.001; Table 1]. Table 1 Demographics and clinical characteristics of the SSc cohorts Characteristics  Total cohort  Early cohort  DETECT cohort  P-valuea  (n = 161)  (n = 77)  (n = 84)  Age at RHC, mean (s.d.), years  61 (11.8)  59.6 (12.2)  62.4 (11.3)  0.137  Time from SSc onset to RHC, mean (s.d.), years  6.6 (8.2)  6.3 (8.3)  6.9 (8.1)  0.634  Observation period, mean (s.d.), years  2.9 (2.4)  4.6 (2.2)  1.2 (0.8)  <0.001  Females, n (%)  126 (78.8)  60 (77.9)  66 (78.6)  0.582  Smoker (past or present), n (%)  67 (41.9)  28 (36.4)  39 (46.4)  0.064  lcSSc, n (%)  125 (78.1)  63 (81.8)  63 (75)  0.502  Anti-centromere Ab, n (%)  78 (48.8)  37 (48.1)  34 (40.5)  0.848  Digital ulcers, n (%)  61 (38.1)  31 (40.3)  30 (35.7)  0.911  Scleroderma renal crisis, n (%)  7 (4.4)  4 (5.2)  3 (3.6)  0.652  Baseline modified Rodnan skin score, mean (s.d.)  9.9 (8.8)  9.8 (10.5)  9.9 (9.8)  0.934  RHC results        <0.001      No PH, n (%)  57 (35.4)  29 (37.7)  28 (33.3)        Borderline PH, n (%)  39 (24.2)  13 (16.9)  26 (31.0)        Pulmonary hypertension, n (%)  65 (40.4)  35 (45.5)  30 (35.7)            PAH, n (%)  31 (19.3)  16 (20.8)  15 (17.9)            PH-ILD, n (%)  20 (12.4)  14 (18.2)  6 (7.1)            Post capillary PH, n (%)  14 (8.7)  5 (6.5)  9 (10.7)    Myocardial infarction, n (%)  14 (8.8)  8 (10.4)  6 (7.1)  0.557  Angina pectoris, n (%)  14 (10.1)  9 (11.7)  5 (6)  0.414  Characteristics  Total cohort  Early cohort  DETECT cohort  P-valuea  (n = 161)  (n = 77)  (n = 84)  Age at RHC, mean (s.d.), years  61 (11.8)  59.6 (12.2)  62.4 (11.3)  0.137  Time from SSc onset to RHC, mean (s.d.), years  6.6 (8.2)  6.3 (8.3)  6.9 (8.1)  0.634  Observation period, mean (s.d.), years  2.9 (2.4)  4.6 (2.2)  1.2 (0.8)  <0.001  Females, n (%)  126 (78.8)  60 (77.9)  66 (78.6)  0.582  Smoker (past or present), n (%)  67 (41.9)  28 (36.4)  39 (46.4)  0.064  lcSSc, n (%)  125 (78.1)  63 (81.8)  63 (75)  0.502  Anti-centromere Ab, n (%)  78 (48.8)  37 (48.1)  34 (40.5)  0.848  Digital ulcers, n (%)  61 (38.1)  31 (40.3)  30 (35.7)  0.911  Scleroderma renal crisis, n (%)  7 (4.4)  4 (5.2)  3 (3.6)  0.652  Baseline modified Rodnan skin score, mean (s.d.)  9.9 (8.8)  9.8 (10.5)  9.9 (9.8)  0.934  RHC results        <0.001      No PH, n (%)  57 (35.4)  29 (37.7)  28 (33.3)        Borderline PH, n (%)  39 (24.2)  13 (16.9)  26 (31.0)        Pulmonary hypertension, n (%)  65 (40.4)  35 (45.5)  30 (35.7)            PAH, n (%)  31 (19.3)  16 (20.8)  15 (17.9)            PH-ILD, n (%)  20 (12.4)  14 (18.2)  6 (7.1)            Post capillary PH, n (%)  14 (8.7)  5 (6.5)  9 (10.7)    Myocardial infarction, n (%)  14 (8.8)  8 (10.4)  6 (7.1)  0.557  Angina pectoris, n (%)  14 (10.1)  9 (11.7)  5 (6)  0.414  Demographics and clinical characteristics of all patients at OUH diagnosed with SSc that underwent an incident right heart catheterization (RHC) in January 2009 to March 2017 (total cohort); the Early and DETECT cohorts underwent incident RHCs, respectively, prior to (2009–13) and after (2014–17) establishment of the DETECT algorithm. aP-values indicate differences between the Early and DETECT cohorts. Ab: antibody; lcSSc: limited cutaneous SSc; PAH: pulmonary arterial hypertension; PH: pulmonary hypertension; PH-ILD: interstitial lung disease-associated pulmonary hypertension; RHC: right heart catheterization. Trends in RHC frequency and diagnostic outcome over the study period In addition to the incident RHC, at least one follow-up RHC procedure was performed in 69 (42.9%) patients from the study cohort, including 49 (63.6%) in the Early cohort and 20 (23.8%) in the DETECT cohort. The number of incident and follow-up RHCs conducted per year throughout the study period increased after 2013 (supplementary Fig. S1, available at Rheumatology Online). The outcomes of these RHC investigations (no PH, borderline PH or PAH) are shown in Fig. 1. The distribution of mPAP values (<17, 17–20, 20–24 or >25 mmHg) in the two cohorts were: n = 16 (20.8%), n = 13 (16.9%), n = 13 (16.9%) and n = 35 (45.5%), respectively, in the Early cohort, compared with n = 15 (19.3%), n = 10 (12%), n = 28 (33.7%) and n = 30 (36.1%), respectively, in the DETECT cohort (P = 0.146). Fig. 1 View largeDownload slide Frequencies of PAH, borderline PH and no PH Numbers of patients, from 2009 until 2016, diagnosed with PAH (red line), borderline PH (blue line) or no PH (normal mean arterial pressure; green line) at the time of right heart catheterization. PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Fig. 1 View largeDownload slide Frequencies of PAH, borderline PH and no PH Numbers of patients, from 2009 until 2016, diagnosed with PAH (red line), borderline PH (blue line) or no PH (normal mean arterial pressure; green line) at the time of right heart catheterization. PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Results from the incident RHC in the Early and DETECT cohorts The incident RHC was diagnostic for PAH in 16 (20.8%) patients in the Early cohort and 15 (17.9%) patients in the DETECT cohort (Table 1). There were no significant differences between the cohorts in the haemodynamic parameters at the incident RHC (supplementary Table S1, available at Rheumatology Online). In the Early cohort, 13 (16.9%) patients were diagnosed with borderline PH, compared with 26 (31.0%) patients in the DETECT cohort (Table 1). The corresponding mPAP values were 22.9 (1.2) mmHg and 21.8 (1.5) mmHg, respectively; P = 0.019 (Table 2). Table 2 Demographics and clinical characteristics of borderline PH Characteristics  Early cohort  DETECT cohort  P-value  (n = 13)  (n = 26)  Age at RHC, mean (s.d.), years  56.6 (13.4)  61.9 (11.0)  0.199  Time from SSc onset to RHC, mean (s.d.), years  9.9 (8.8)  9.1 (10.8)  0.769  Females, n (%)  8 (61.5)  23 (80.5)  0.090  lcSSc, n (%)  5 (38.5)  16 (61.5)  0.247  Anti-centromere Ab, n (%)  3 (23.1)  11 (47.8)  0.143  FVC/DLCO ratio, mean (s.d.)  1.9 (0.7)  1.8 (0.7)  0.557  Extent of fibrosis, mean (s.d.), %  7.9 (10.4)  6.1 (13.4)  0.684  At diagnosis, mean (s.d.)            mPAP, mmHg  22.9 (1.2)  21.8 (1.5)  0.019      CO, l/min  5.7 (1.3)  5.7 (1.5)  0.965      CI, l/min/m2  3.2 (0.6)  3.2 (0.7)  0.913  Follow-up RHC, n (%)  10 (76.9)  5 (19.2)    Time between incident and follow-up RHCs, mean (s.d.), years  2.4 (1.8)  0.7 (0.7)  <0.001  Developed PH, n (%)  5 (50)  2 (40)  0.828  Characteristics  Early cohort  DETECT cohort  P-value  (n = 13)  (n = 26)  Age at RHC, mean (s.d.), years  56.6 (13.4)  61.9 (11.0)  0.199  Time from SSc onset to RHC, mean (s.d.), years  9.9 (8.8)  9.1 (10.8)  0.769  Females, n (%)  8 (61.5)  23 (80.5)  0.090  lcSSc, n (%)  5 (38.5)  16 (61.5)  0.247  Anti-centromere Ab, n (%)  3 (23.1)  11 (47.8)  0.143  FVC/DLCO ratio, mean (s.d.)  1.9 (0.7)  1.8 (0.7)  0.557  Extent of fibrosis, mean (s.d.), %  7.9 (10.4)  6.1 (13.4)  0.684  At diagnosis, mean (s.d.)            mPAP, mmHg  22.9 (1.2)  21.8 (1.5)  0.019      CO, l/min  5.7 (1.3)  5.7 (1.5)  0.965      CI, l/min/m2  3.2 (0.6)  3.2 (0.7)  0.913  Follow-up RHC, n (%)  10 (76.9)  5 (19.2)    Time between incident and follow-up RHCs, mean (s.d.), years  2.4 (1.8)  0.7 (0.7)  <0.001  Developed PH, n (%)  5 (50)  2 (40)  0.828  Demographics, clinical characteristics and right heart catheterization (RHC) data for patients diagnosed with borderline pulmonary hypertension in the Early cohort (2009–13) and the DETECT cohort (2014–17). Ab: antibody; CI: cardiac input; CO: cardiac output; DLCO: diffusing lung capacity for carbon monoxide; FVC: forced vital capacity; lcSSc: limited cutaneous SSc; mPAP: mean pulmonary arterial pressure; n: number; RHC: right heart catheterization; PH: pulmonary hypertension. In the DETECT cohort, 28 patients (37.7%) had mPAPs <20 mmHg (no PH). These patients had significantly lower FVC/DLCO ratios, lower frequencies of systolic pulmonary arterial pressure >40 mmHg, higher frequencies of DLCO >60%, lower NT-proBNP levels and lower FCs than patients diagnosed with PH at the time of a referral for RHC (Table 3). Moreover, the DETECT cohort had 46 (54.8%) patients with a DLCO <60% and 53 (64.3%) patients with a disease duration >3 years (measured from disease onset at the time of referral for RHC; Table 3). We also found that the DETECT scores calculated before referral to RHC differed significantly between the mPAP subgroups measured at the incident RHC (P < 0.001) (supplementary Fig. S2, available at Rheumatology Online). Table 3 Clinical characteristics of DETECT cohort, segregated by PH subtype Characteristic  No PH (n = 28)  Borderline PH (n = 26)  PAH (n = 15)  PH-ILD (n = 6)  Post-capPH (n = 9)  P-valuea  Age at RHC, mean (s.d.), years  60.3 (13.6)  59.9 (10.9)  66.8 (9.6)  65 (9.5)  65.7 (8.9)  0.497  Time from SSc onset to RHC, mean (s.d.), years  6.9 (6.7)  9.7 (11.3)  6.3 (6.2)  6.7 (7.6)  3.9 (3.9)  0.330  Onset to RHC >3 years, n (%)  17 (60.7)  16 (61.5)  9 (60)  5 (83.3)  6 (66.7)  0.401  DLCO <60%, n (%)  6a (21.4)  14a (53.8)  14a (93.3)  6a (100)  6a (66.7)  <0.001  FVC/DLCO ratio, mean (s.d.)  1.6a (0.4)  1.8 (0.8)  2.9a (1.2)  2.7a (1.0)  2.1 (0.5)  <0.001*  NT-proBNP, mean (s.d.), µmol/l  32 (19.7)  39 (66.9)  189 (221.9)  38 (31.1)  341 (1024.9)  0.007  6MWD, mean (s.d.), m  514 (62.8)  502 (109.7)  356 (207.9)  456 (167.4)  278 (222.1)  0.747  sPAP, mean (s.d.), mmHg  31a (10.2)  31 (11.8)  52a (26)  53a (9.8)  49a (15.7)  0.006*  sPAP >30 mmHg, n (s.d.)  11 (39.3)  10 (38.5)  13 (86.7)  6 (100)  7 (77.8)  0.003  sPAP >40 mmHg, n (s.d.)  4a (14.3)  7 (26.9)  11a (73.3)  5a (83.3)  3 (33.3)  0.001*  FC, n (%)  19 (67.9)  26 (100)  15 (100)  5 (83.3)  8 (88.9)        FC 1 and 2, n (%)  17 (60.7)  22 (84.6)  7 (46.7)  2 (33.3)  5 (55.6)  0.008      FC 3 and 4, n (%)  2 (7.1)  4 (15.4)  8 (53.3)  3 (50)  3 (33.3)    Characteristic  No PH (n = 28)  Borderline PH (n = 26)  PAH (n = 15)  PH-ILD (n = 6)  Post-capPH (n = 9)  P-valuea  Age at RHC, mean (s.d.), years  60.3 (13.6)  59.9 (10.9)  66.8 (9.6)  65 (9.5)  65.7 (8.9)  0.497  Time from SSc onset to RHC, mean (s.d.), years  6.9 (6.7)  9.7 (11.3)  6.3 (6.2)  6.7 (7.6)  3.9 (3.9)  0.330  Onset to RHC >3 years, n (%)  17 (60.7)  16 (61.5)  9 (60)  5 (83.3)  6 (66.7)  0.401  DLCO <60%, n (%)  6a (21.4)  14a (53.8)  14a (93.3)  6a (100)  6a (66.7)  <0.001  FVC/DLCO ratio, mean (s.d.)  1.6a (0.4)  1.8 (0.8)  2.9a (1.2)  2.7a (1.0)  2.1 (0.5)  <0.001*  NT-proBNP, mean (s.d.), µmol/l  32 (19.7)  39 (66.9)  189 (221.9)  38 (31.1)  341 (1024.9)  0.007  6MWD, mean (s.d.), m  514 (62.8)  502 (109.7)  356 (207.9)  456 (167.4)  278 (222.1)  0.747  sPAP, mean (s.d.), mmHg  31a (10.2)  31 (11.8)  52a (26)  53a (9.8)  49a (15.7)  0.006*  sPAP >30 mmHg, n (s.d.)  11 (39.3)  10 (38.5)  13 (86.7)  6 (100)  7 (77.8)  0.003  sPAP >40 mmHg, n (s.d.)  4a (14.3)  7 (26.9)  11a (73.3)  5a (83.3)  3 (33.3)  0.001*  FC, n (%)  19 (67.9)  26 (100)  15 (100)  5 (83.3)  8 (88.9)        FC 1 and 2, n (%)  17 (60.7)  22 (84.6)  7 (46.7)  2 (33.3)  5 (55.6)  0.008      FC 3 and 4, n (%)  2 (7.1)  4 (15.4)  8 (53.3)  3 (50)  3 (33.3)    Clinical characteristics of 84 patients with SSc in the DETECT cohort, prior to a referral for right heart catheterization (RHC), segregated according to the outcome of the incident RHC. aSignificant findings in an ad hoc test compared with no PH. P-values were tested with one-way ANOVA. *significant findings in an ad hoc test compared to no PH. 6MWD: 6-min walking distance; DLCO: diffusing lung capacity for carbon monoxide; FC: functional class; FVC: forced vital capacity; NT-proBNP: N-terminal pro-brain natriuretic peptide; PAH: pulmonary arterial hypertension; PH: pulmonary hypertension; PH-ILD: interstitial lung disease associated pulmonary hypertension; post-capPH: post-capillary PH; RHC: right heart catheterization; sPAP: systolic pulmonary arterial pressure measured on ECHO. Functional status and risk assessment in patients with PAH or borderline PH At the time of the PAH diagnosis, the Early and DETECT cohorts were similar in New York Heart Association functional classes and risk stratifications, according to the 2015 ESC/ERS guidelines. In the DETECT cohort, four patients (26.7%) were at low risk, seven (46.7%) were at intermediate risk and four (26.7%) were at high risk of PAH poor prognosis. Corresponding numbers in the Early cohort were three (18.8%), six (37.5%) and seven (43.8%), respectively (Fig. 2A and B). Distribution of the clinical features for risk stratification is shown in supplementary Table S2, available at Rheumatology Online. Fig. 2 View largeDownload slide Functional class and risk stratification of patients with PAH or borderline PH diagnosis Results are shown for different diagnostic subsets of the DETECT and Early cohorts. (A) Functional class (NYHA 1–4) and (B) risk stratification (low-high), measured at the time of PAH diagnosis. (C) Functional class and (D) risk stratification, measured at the time of the borderline PH diagnosis. Differences between groups are indicated with P-values (at the top of each panel). NYHA: New York Heart Association; PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. Fig. 2 View largeDownload slide Functional class and risk stratification of patients with PAH or borderline PH diagnosis Results are shown for different diagnostic subsets of the DETECT and Early cohorts. (A) Functional class (NYHA 1–4) and (B) risk stratification (low-high), measured at the time of PAH diagnosis. (C) Functional class and (D) risk stratification, measured at the time of the borderline PH diagnosis. Differences between groups are indicated with P-values (at the top of each panel). NYHA: New York Heart Association; PAH: pulmonary arterial hypertension, PH: pulmonary hypertension. At the time of the borderline PH diagnosis, a trend of improvement in functional class and significantly lower risk levels were observed in the DETECT cohort compared with the Early cohort. Specifically, in the Early cohort, two patients (22.2%) were at low risk, six (66.7%) were at intermediate risk and one patient (11.1%) was at high risk of PAH poor prognosis. Corresponding numbers in the DETECT cohort for patients at low, intermediate or high risk were 13 (59.1%), nine (40.9%) and zero (0%), respectively; P = 0.030 (Fig. 2C and D). Results from the follow-up RHC in the study cohort In the Early cohort, 26 patients (63%) had borderline or no PH on the incident RHC. These patients were subjected to a follow-up RHC after a mean of 2.4 (1.8) years (Fig. 3, left panel). The follow up RHC results showed that, among the 10 patients with borderline PH on the incident RHC, five (38.5%) developed PH after a mean of 2.2 (1.6) years. Among the 16 patients with no PH, eight (28.6%) developed borderline PH after a mean of 2.6 (2.3) years, and four (14.3%) developed PH after a mean of 3 (1.2) years (Fig. 3). Of the remaining 15 patients with SSc in the Early cohort that did not undergo a follow-up RHC, four patients died, five were followed at their local department of rheumatology, five were clinically stable and one is currently awaiting a follow-up RHC, due to suspicion of PH development. Of the 77 patients with SSc in the Early cohort, 27 patients (35.1%) died during the observation period; the proportions of deaths were not significantly different among the PH subgroups. Fig. 3 View largeDownload slide Flowchart of follow-up right heart catheterization Flowchart of follow-up right heart catheterization (RHC) in patients with SSc without pulmonary hypertension (no PH) and with borderline PH at baseline in the Early and DETECT cohorts. PH: pulmonary hypertension. Fig. 3 View largeDownload slide Flowchart of follow-up right heart catheterization Flowchart of follow-up right heart catheterization (RHC) in patients with SSc without pulmonary hypertension (no PH) and with borderline PH at baseline in the Early and DETECT cohorts. PH: pulmonary hypertension. In the DETECT cohort, seven patients (13%) underwent a follow-up RHC after a mean of 0.7 (0.7) years after the incident RHC (Fig. 3). Due to the short follow-up time and few patients, no further analyses or comparisons were conducted within the DETECT cohort, regarding new onset PH or survival. Discussion There is a need to improve the early diagnosis of PAH in SSc to optimize treatment effects and improve survival. The DETECT algorithm was established in 2014, with the intention of identifying patients with PAH at asymptomatic stages. Our results indicated that, after the introduction of DETECT, the incidence of PAH was stable, but the number of cases diagnosed with borderline PH increased significantly. Interestingly, we also found that, after DETECT was established, patients with borderline PH had significantly lower risk scores at diagnosis than patients with borderline PH identified in the Early, pre-DETECT cohort. To our knowledge, this study was the first to compare the relative frequencies of PAH and borderline PH in a population-based SSc cohort before and after the introduction of the DETECT algorithm. Other studies focused on the sensitivity and specificity of DETECT. Those studies demonstrated the benefits of DETECT, compared with the ESC/ERS screening guidelines for PAH [25–27] and borderline PH [25, 28]. In our clinical practice, we found no change in the PAH incidence after introducing the DETECT algorithm for PH surveillance. However, we did find that the number of new borderline PH cases increased significantly after including the DETECT algorithm; this increase was accompanied by a concomitant shift in the 2015 ESC/ERS risk score distribution [12]. We are aware that these risk scores were primarily developed for PAH, rather than borderline PH; however, because the mPAP threshold of 25 mmHg was mostly an arbitrary value, and considering the fact that mortality increased in patients with borderline PH, it is clinically important to identify patients with mPAP values of 20–25 mmHg and consider their risk scores [14, 17]. However, the predictive value of risk scores in patients with SSc-related borderline PH remains to be confirmed with larger, prospective studies, preferably in population-based cohorts. Another interesting finding is the lower percentage of cases with PH-ILD in the DETECT cohort compared with the Early cohort. We do not know the exact reason, but it appears likely that the reduction reflects a change in the RHC referral pattern, with relatively fewer referrals of patients with severe lung fibrosis. Of note, we added the DETECT algorithm to annual PH screening in our clinical practice. Therefore, the DETECT algorithm was also performed in patients with DLCO >60%. We are aware that DETECT has not been validated for those patients, but it was unavoidable, because we included all patients with SSc that were considered at PH risk, based on clinical assessments at the annual PH screening. The current study reinforced the notion that patients with SSc and borderline PH are at high risk of developing PAH. In fact, we observed a frequency of 38.5% PAH development after 2.4 years; that frequency was within the range observed in previous studies. In a large UK cohort, 30% of patients with SSc and borderline PH developed PAH. Additionally, in the PHAROS cohort, 55% of patients with borderline PH developed PAH [29, 30]. Given this high risk and the high mortality rate of established PAH, it is a sobering thought that there are no guidelines for following up borderline PH, and there have been no randomized trials on testing strategies for PAH prevention [10, 22, 31, 32]. In animal models, PAH was shown to be driven by inflammatory responses. Local immune activation led to the dysfunction of vascular endothelial cells and smooth muscle cells, which resulted in neo-intimal hyperplasia in small pulmonary arteries [33]. In these models, it was shown that immune-modulation with MMF (a purine inhibitor), which was approved for organ transplantation, could attenuate the development of PAH [34, 35]. It is unknown whether MMF could prevent PAH development in SSc more efficiently than the approved PAH therapies, which target the vascular component of PAH. The latter therapies aim to maximize the properties of vessels. Nevertheless, it is interesting to note that preliminary data from the OUH SSc cohort indicated that the frequency of PAH in patients with SSc was reduced in a subset of patients that had received MMF treatment for skin/lung disease. Randomized clinical trials are warranted to determine whether PAH is preventable in borderline PH and, eventually, to determine optimal prevention strategies. The current study reinforced the notion that the DETECT algorithm has high sensitivity. However, in this study, when we added the algorithm to our clinical screening programme, it had rather low specificity. Nevertheless, our findings suggested that an annual PAH screening programme that included DETECT would be likely to increase the total number of RHC investigations and increase the frequency of patients with SSc that have normal or borderline mPAP. Borderline mPAP justifies an early RHC follow-up, but prospective, randomized clinical trials are necessary to determine whether patients with borderline PH would benefit from this strategy. To avoid unnecessary invasive RHC procedures and to identify patients most likely to develop PH, additional, more specific PAH markers are needed that are suitable for integrating into clinical follow-up procedures, including the DETECT algorithm. Our study had some limitations. First, a follow-up RHC was not performed in the entire SSc cohort, and second, the follow-up period was rather short for the DETECT cohort. The frequency of borderline PH and PAH was rather low, because the OUH cohort was population-based, which covered the whole spectrum of disease. Larger, prospective studies, preferably in population-based cohorts, are needed. The strengths of our study were that our study prospectively enrolled patients with SSc over a 9-year period, and all included patients had undergone an incident RHC. Additionally, we compared the frequencies of PAH and borderline PH in two cohorts, before and after including the DETECT algorithm in the annual PH screening programme. In conclusion, we showed that the number of new PAH cases per year has been stable since 2009, with no apparent change after introducing the DETECT algorithm into PH screening. Starting in 2014, we observed a non-significant increase in the number of newly diagnosed PAH cases at low risk and better functional class status. The total number of RHCs increased, starting in 2014, with a concomitant increase in the number of borderline PH cases. Funding: This work was supported by grants from the Norwegian Women’s Public Health Association. Disclosure statement: A.A. has received honoraria and research grants from Actelion, GSK, Bayer, MSD and United Therapeutics. All other authors have declared no conflicts of interest. Supplementary data Supplementary data are available at Rheumatology online. References 1 Allanore Y, Denton C, Matucci-Cerinic M. Systemic sclerosis. In: Bijlsma JWJ, ed. EULAR Textbook on Rheumatic Diseases.  London: BMJ, 2012; 646– 567. 2 Pattanaik D, Brown M, Postlethwaite B, Postlethwaite A. Pathogenesis of systemic sclerosis. Front Immunol  2015; 6: 272. Google Scholar CrossRef Search ADS PubMed  3 Tyndall AJ, Bannert B, Vonk M et al.   Causes and risk factors for death in systemic sclerosis: a study from the EULAR Scleroderma Trials and Research (EUSTAR) database. Ann Rheum Dis  2010; 69: 1809– 15. http://dx.doi.org/10.1136/ard.2009.114264 Google Scholar CrossRef Search ADS PubMed  4 Rubio-Rivas M, Royo C, Simeón C, Corbella X, Fonollosa V. Mortality and survival in systemic sclerosis: systematic review and meta-analysis. Semin Arthritis Rheum  2014; 44: 208– 19. Google Scholar CrossRef Search ADS PubMed  5 Hoffmann-Vold AM, Molberg O, Midtvedt O, Garen T, Gran JT. Survival and causes of death in an unselected and complete cohort of Norwegian patients with systemic sclerosis. J Rheumatol  2013; 40: 1127– 33. Google Scholar CrossRef Search ADS PubMed  6 Hesselstrand R, Wildt M, Ekmehag B, Wuttge DM, Scheja A. Survival in patients with pulmonary arterial hypertension associated with systemic sclerosis from a Swedish single centre: prognosis still poor and prediction difficult. Scand J Rheumatol  2011; 40: 127– 32. Google Scholar CrossRef Search ADS PubMed  7 Chaisson NF, Hassoun PM. Systemic sclerosis-associated pulmonary arterial hypertension. Chest  2013; 144: 1346– 56. http://dx.doi.org/10.1378/chest.12-2396 Google Scholar CrossRef Search ADS PubMed  8 Hervier B, Meyer A, Dieval C et al.   Pulmonary hypertension in antisynthetase syndrome: prevalence, aetiology and survival. Eur Respir J  2013; 42: 1271– 82. http://dx.doi.org/10.1183/09031936.00156312 Google Scholar CrossRef Search ADS PubMed  9 Le Pavec J, Girgis RE, Lechtzin N et al.   Systemic sclerosis-related pulmonary hypertension associated with interstitial lung disease: impact of pulmonary arterial hypertension therapies. Arthritis Rheum  2011; 63: 2456– 64. http://dx.doi.org/10.1002/art.30423 Google Scholar CrossRef Search ADS PubMed  10 Morrisroe K, Stevens W, Huq M et al.   Survival and quality of life in incident systemic sclerosis-related pulmonary arterial hypertension. Arthritis Res Ther  2017; 19: 122. http://dx.doi.org/10.1186/s13075-017-1341-x Google Scholar CrossRef Search ADS PubMed  11 Galie N, Hoeper MM, Humbert M et al.   Guidelines for the diagnosis and treatment of pulmonary hypertension: the Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS), endorsed by the International Society of Heart and Lung Transplantation (ISHLT). Eur Heart J  2009; 30: 2493– 537. http://dx.doi.org/10.1093/eurheartj/ehp297 Google Scholar CrossRef Search ADS PubMed  12 Galie N, Humbert M, Vachiery JL et al.   2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Heart J  2016; 37: 67– 119. http://dx.doi.org/10.1093/eurheartj/ehv317 Google Scholar CrossRef Search ADS PubMed  13 Coghlan JG, Denton CP, Grunig E et al.   Evidence-based detection of pulmonary arterial hypertension in systemic sclerosis: the DETECT study. Ann Rheum Dis  2014; 73: 1340– 9. http://dx.doi.org/10.1136/annrheumdis-2013-203301 Google Scholar CrossRef Search ADS PubMed  14 Hatano S, Strasser T. Primary Pulmonary Hypertension . Geneva, Switzerland: World Health Organization, 1975. 15 Kovacs G, Berghold A, Scheidl S, Olschewski H. Pulmonary arterial pressure during rest and exercise in healthy subjects: a systematic review. Eur Respir J  2009; 34: 888– 94. http://dx.doi.org/10.1183/09031936.00145608 Google Scholar CrossRef Search ADS PubMed  16 Maron BA, Hess E, Maddox TM et al.   Association of borderline pulmonary hypertension with mortality and hospitalization in a large patient cohort: Insights from the Veterans Affairs Clinical Assessment, Reporting, and Tracking Program. Circulation  2016; 133: 1240– 8. http://dx.doi.org/10.1161/CIRCULATIONAHA.115.020207 Google Scholar CrossRef Search ADS PubMed  17 Kovacs G, Avian A, Tscherner M et al.   Characterization of patients with borderline pulmonary arterial pressure. Chest  2014; 146: 1486– 93. http://dx.doi.org/10.1378/chest.14-0194 Google Scholar CrossRef Search ADS PubMed  18 Hoffmann-Vold AM, Gunnarsson R, Garen T, Midtvedt O, Molberg O. Performance of the 2013 American College of Rheumatology/European League Against Rheumatism Classification Criteria for Systemic Sclerosis (SSc) in large, well-defined cohorts of SSc and mixed connective tissue disease. J Rheumatol  2015; 42: 60– 3. Google Scholar CrossRef Search ADS PubMed  19 van den Hoogen F, Khanna D, Fransen J et al.   2013 classification criteria for systemic sclerosis: an American College of Rheumatology/European League against Rheumatism collaborative initiative. Arthritis Rheum  2013; 65: 2737– 47. http://dx.doi.org/10.1002/art.38098 Google Scholar CrossRef Search ADS PubMed  20 LeRoy E, Black C, Fleischmajer R et al.   Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol  1988; 15: 202– 5. Google Scholar PubMed  21 Gude E, Simonsen S, Geiran OR et al.   Pulmonary hypertension in heart transplantation: discrepant prognostic impact of pre-operative compared with 1-year post-operative right heart hemodynamics. J Heart Lung Transplant  2010; 29: 216– 23. http://dx.doi.org/10.1016/j.healun.2009.08.021 Google Scholar CrossRef Search ADS PubMed  22 Hoeper MM, Bogaard HJ, Condliffe R et al.   Definitions and diagnosis of pulmonary hypertension. J Am Coll Cardiol  2013; 62(25 Suppl): D42– 50. Google Scholar CrossRef Search ADS   23 Hoffmann-Vold AM, Aalokken TM, Lund MB et al.   Predictive value of serial high-resolution computed tomography analyses and concurrent lung function tests in systemic sclerosis. Arthritis Rheumatol  2015; 67: 2205– 12. http://dx.doi.org/10.1002/art.39166 Google Scholar CrossRef Search ADS PubMed  24 Hoffmann-Vold A, Tennøe AH, Garen T et al.   High level of chemokine CCL18 is associated with pulmonary function deterioration, lung fibrosis progression and reduced survival in systemic sclerosis. Chest  2016; 150: 299– 306. Google Scholar CrossRef Search ADS PubMed  25 Guillen-Del Castillo A, Callejas-Moraga EL, Garcia G et al.   High sensitivity and negative predictive value of the DETECT algorithm for an early diagnosis of pulmonary arterial hypertension in systemic sclerosis: application in a single center. Arthritis Res Ther  2017; 19: 135. Google Scholar CrossRef Search ADS PubMed  26 Soukup T, Pudil R, Kubinova K et al.   Application of the DETECT algorithm for detection of risk of pulmonary arterial hypertension in systemic sclerosis: data from a Czech tertiary centre. Rheumatology  2016; 55: 109– 14. http://dx.doi.org/10.1093/rheumatology/kev327 Google Scholar CrossRef Search ADS PubMed  27 Hao Y, Thakkar V, Stevens W et al.   A comparison of the predictive accuracy of three screening models for pulmonary arterial hypertension in systemic sclerosis. Arthritis Res Ther  2015; 17: 7. http://dx.doi.org/10.1186/s13075-015-0517-5 Google Scholar CrossRef Search ADS PubMed  28 Visovatti SH, Distler O, Coghlan JG et al.   Borderline pulmonary arterial pressure in systemic sclerosis patients: a post-hoc analysis of the DETECT study. Arthritis Res Ther  2014; 16: 493. http://dx.doi.org/10.1186/s13075-014-0493-1 Google Scholar CrossRef Search ADS PubMed  29 Valerio CJ, Schreiber BE, Handler CE, Denton CP, Coghlan JG. Borderline mean pulmonary artery pressure in patients with systemic sclerosis: transpulmonary gradient predicts risk of developing pulmonary hypertension. Arthritis Rheum  2013; 65: 1074– 84. Google Scholar CrossRef Search ADS PubMed  30 Bae S, Saggar R, Bolster MB et al.   Baseline characteristics and follow-up in patients with normal haemodynamics versus borderline mean pulmonary arterial pressure in systemic sclerosis: results from the PHAROS registry. Ann Rheum Dis  2012; 71: 1335– 42. http://dx.doi.org/10.1136/annrheumdis-2011-200546 Google Scholar CrossRef Search ADS PubMed  31 Kovacs G, Olschewski H. Borderline pulmonary pressures in scleroderma – a ‘pre-pulmonary arterial hypertension’ condition? Arthritis Res Ther  2015; 17: 123. Google Scholar CrossRef Search ADS PubMed  32 Nihtyanova S, Schreiber B, Ong V et al.   Prediction of pulmonary complications and long-term survival in systemic sclerosis. Arthritis Rheumatol  2014; 66: 1625– 35. http://dx.doi.org/10.1002/art.38390 Google Scholar CrossRef Search ADS PubMed  33 Rabinovitch M, Guignabert C, Humbert M, Nicolls MR. Inflammation and immunity in the pathogenesis of pulmonary arterial hypertension. Circ Res  2014; 115: 165– 75. Google Scholar CrossRef Search ADS PubMed  34 Suzuki C, Takahashi M, Morimoto H et al.   Mycophenolate mofetil attenuates pulmonary arterial hypertension in rats. Biochem Biophys Res Commun  2006; 349: 781– 8. http://dx.doi.org/10.1016/j.bbrc.2006.08.109 Google Scholar CrossRef Search ADS PubMed  35 Zhang YF, Zheng Y. The effects of mycophenolate mofetil on cytokines and their receptors in pulmonary arterial hypertension in rats. Scand J Rheumatol  2015; 44: 412– 5. http://dx.doi.org/10.3109/03009742.2015.1023829 Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

RheumatologyOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off