Significance of Pulmonary Arterial Pressure as a Prognostic Indicator in Lung-Dominant Connective Tissue Disease

Background Lung-dominant connective tissue disease (LD-CTD) is a new concept for classifying the subset of patients with interstitial pneumonia who have clinical features suggesting an associated CTD, but whose features fall short of a clear diagnosis of CTD under the current rheumatologic classification systems. The impact of mean pulmonary arterial pressure (MPAP) in LD-CTD has not been sufficiently elucidated. Objectives To evaluate the survival impact of MPAP measured during the initial evaluation in patients with LD-CTD. Methods We retrospectively analyzed the initial evaluation data of 100 LD-CTD patients undergoing pulmonary function test, 6-min walk test (6MWT), and right heart catheterization (RHC). Results The mean MPAP was 16.2±4.4 mm Hg, and 18 patients had MPAP≥20 mm Hg. A univariate Cox proportional hazard model showed that MPAP and several variables have a statistically significant impact on survival. With stepwise, multivariate Cox proportional analysis, MPAP (HR  = 1.293; 95% CI 1.130–1.480; p<0.001) and mean forced vital capacity (FVC) % predicted (HR = 0.958; 95% CI 0.930–0.986; p = 0.004) were shown to be independent determinants of survival. Conclusions Higher MPAP and lower %FVC at the initial evaluation were significant independent prognostic factors of LD-CTD. MPAP evaluation provides additional information of disease status and will help physicians to predict mortality in LD-CTD.

Recently, Fischer et al. [11] proposed ''lung-dominant CTD'' (LD-CTD) as a new concept for classifying the subset of patients with interstitial pneumonia who have clinical features suggesting an associated CTD, but whose features fall short of a clear diagnosis of CTD under the current rheumatologic classification systems. Little information is available regarding prognostic factors in LD-CTD [12]. Considering the importance of PH in ILD and CTD, we assumed that PH might occur in the clinical course of LD-CTD patients and contribute to poor prognosis. The aim of this study was to evaluate whether MPAP predicts survival in LD-CTD patients in whom background, pulmonary function test, 6min walk test (6MWT), and right heart catheterization (RHC) could be evaluated at the initial evaluation.

Study subjects
Six hundred nine patients with ILD were identified from among patients who underwent systematic evaluations between May 2007 and August 2012 at Tosei General Hospital. Systematic evaluations included pulmonary function test, 6MWT, and RHC. LD-CTD was diagnosed based on the criteria proposed by Fischer et al. when specific autoantibodies were present in the absence of extrathoracic features of a definite CTD [11]. Serological tests, including those for antinuclear antibody (ANA), rheumatoid factor (RF), anti-citrullinated peptide (CCP), anti-Scl 70 (Scl-70), anti-Ro/SSA, anti-La/SSB, anti-double-stranded DNA (dsDNA), anti-Smith, anti-ribonucleoprotein (RNP), anti-tRNA synthetase (Jo-1), and anti-centromere antibodies, were performed routinely when ILD was first suspected. Anti-PM-Scl antibodies were unmeasured. Patients were considered to have LD-CTD if they had one of the specific autoantibodies within one year of initial evaluation (Table 1). CTD was diagnosed when patients fulfilled the established criteria for rheumatoid arthritis, systemic lupus erythematosus (SLE), SSc, polymyositis/dermatomyositis, Sjogren's syndrome (SjS), or MCTD [13][14][15][16][17][18]. Clinically amyopathic dermatomyositis (CADM) was diagnosed when a patient had a skin rash characteristic of DM without clinical evidence of muscle disease and with little or no increase in the serum creatine kinase (CK) level [16,19]. We excluded any patient with a pre-existing diagnosis of CTD. The following cases were also excluded: (1) patients had been treated with corticosteroids, immunosuppressive agents, PH targeted therapy and oxygen therapy before initial evaluation, (2) RHC was not performed within one month of initial evaluation, or pulmonary artery wedge pressure (PAWP) on RHC was over 15 mm Hg. Finally, 100 patients with LD-CTD were taken as subjects ( Figure 1).
In this study, PH is defined by MPAP$20 mm Hg (including borderline PH). A previous study indicated that MPAP 20 mm Hg is optimal for predicting the prognosis in IPF patients [2]. This study was approved by the Tosei General Hospital Institutional Review Board (IRB No.398). Patient approval and/ or informed consent were waived because the study involved a retrospective review of patient records. Our institutional review board determined that ethical approval was not necessary and did not require the patient's approval or informed consent.

Measurements
We studied patients' characteristics, pulmonary function tests, PaO2, 6MWT, and hemodynamics. All patients underwent spirometry (CHESTAC-55V; Chest, Tokyo, Japan), according to the method described in the American Thoracic Society (ATS) 1994 update [20]. Single-breath diffusing capacity of the lung for carbon monoxide (DLCO) was also measured (CHESTAC-55V). The values for forced vital capacity (FVC) and DLCO were related to % predicted values [21]. 6MWT was conducted according to the ATS statement [22]. Briefly, all patients were tested under standardized conditions by trained technicians. Patients were instructed to walk as far as possible in 6 min, and the distance that patients could walk was recorded [23]. Oxygen saturation was also measured by pulse oximetry at rest for 5 min prior to and immediately after the test. Dyspnea was assessed with the modified Medical Research Council score (MMRC) scale, which includes 5 grades (0-4) of various physiological activities that provoke dyspnea [24]. After the patients had read the descriptive phrases, they selected the number that best corresponded to their level of dyspnea in daily living. Patients underwent Doppler echocardiography. Estimated right ventricular systolic pressure (ERVSP) was measured by the peak tricuspid regurgitant flow velocity using the modified Bernoulli equation [25][26]. RHC was performed using a Swan-Ganz catheter percutaneously via either the cubital vein or the femoral vein. At the time of catheterization, the pressure transducers were adjusted to reflect the midthorax in each patient before beginning the procedure. Cardiac output was measured with a thermodilution method. PAWP was measured at the end-expiratory phase.

Statistical analysis
Survival status was analyzed in February 2014. Continuous variables were expressed as means 6 SD. Categorical variables were summarized by frequency. The MMRC score was analyzed as a continuous variable. Distribution of continuous variables was evaluated using the Shapiro-Wilk test. If both variables had a normal distribution, correlations were calculated using Pearson's correlation test. If either variable had a nonnormal distribution, correlations were calculated using Spearman's correlation test. When two continuous variables were compared, the t test was used for normal distributions and the Mann-Whitney test was used for nonnormal distributions.Univariate Cox's proportional hazard models were used to examine the association of selected variables with survival. Variables that were significant (p,0.05) in the univariate analysis were included in the multivariate model. To avoid multicolinearity, only one of the highly correlated variables (coefficient of correlation $0.9) was entered in the multivariate model, if present. A stepwise multivariate Cox's proportional hazards model was then used for variables that were revealed to be significant with the univariate model, in order to select more significant variables. Using the methods of Kaplan-Meier and the log-rank test, we studied the impact on survival of variables. All tests were performed at a significance level of p,0.05. Analysis was completed using IBM SPSS statistics version 19.

Results
The baseline characteristics of 100 patients are summarized in Table 2, and the autoantibody profile of the study cohort is shown in Table 3. ANA was the most frequently positive autoantibody. Seventy-two patients were positive for one antibody, 23 were positive for two antibodies, and 5 had three positive serological tests. Fifty patients underwent surgical lung biopsy. RHC data are summarized in Table 2 In the follow-up period, six patients showed a definite CTD. Rheumatoid arthritis was diagnosed in two cases, SSc in two, polymyositis in one, and SjS in one using the established criteria [13,[15][16][17].
Fifty-four patients received pharmacologic therapy during the follow-up period. Thirty-nine patients were treated with corticosteroid and immunosuppressive agent, thirty-seven patients were treated with pirfenidone, and five patients were treated with PH   The mean observation period was 2.7461.63 years. Twentyfour patients died during the observation period, 11 due to advanced respiratory failure, 9 due to acute exacerbation, 1 due to lung cancer, and 3 due to unknown causes. There were no deaths associated with PH. The univariate Cox regression model ( MPAP and %FVC (r = 20.09; p = 0.447). In addition, there was no association between any autoantibody and clinical outcomes (data not shown). Table 6 shows the baseline characteristics and physiology of patients using a cutoff point of 20 mm Hg of MPAP. %DLCO, PaO 2 , 6MWD, and lowest SpO 2 were significantly lower in those with MPAP$20 mm Hg. MMRC, PAWP, and PVR were significantly higher in those with MPAP$20 mm Hg. Figure 3 shows a Kaplan-Meier curve that reveals significantly worse survival among patients whose MPAP was $20 mm Hg than among those whose MPAP was ,20 mm Hg (log-rank test p = 0.005). The 5-year survival rates from the initial diagnosis were 50.0 and 82.0%, respectively.  Table 3. Autoantibody profile of the study cohort.

Serological tests N = 100
High-titer ANA (. Discussion This is the first study, to our knowledge, to report the impact of MPAP on survival in LD-CTD patients. This study included patients with milder pulmonary function impairment (mean FVC 81.1%, mean DLCO 56.4%) and milder MPAP (mean MPAP 16.2 mm Hg) than subjects of previous studies, however, a higher MPAP was an independent prognostic predictor [2,12,27]. This result shows the importance of evaluating MPAP at an early time in patients with LD-CTD.
Using the methods of Kaplan-Meier and the log-rank test, patients with MPAP$20 mm Hg had a significantly lower survival rate than those with MPAP,20 mm Hg in LD-CTD. Kimura et al. previously demonstrated the importance of PH in IPF patients at their initial workup using a cutoff point (MPAP 20 mm Hg) in RHC [2]. Our studies also demonstrated that borderline PH was found to be closely associated with prognosis in LD-CTD, the same as in IPF. The strength of our study was that RHC was performed on all patients within one month of initial evaluation.
The pathogenesis of PH is likely complex and driven by multiple mechanisms [28][29]. As shown in Table 6, patients with PH (MPAP$20 mm Hg) were found to have a lower PaO 2 and lowest SpO 2 during the 6MWT. Previous studies showed that hypoxia induces vasoconstriction and vascular remodeling through various factors, such as vascular endothelial growth factor and hypoxia-inducible factor 1 alpha [30]. Our results indicate that hypoxia may play a crucial role in the development of PH in LD-CTD. Previous studies also suggest a role for autoimmunity in the development of the pulmonary vascular changes in patients with CTD [28]. Therefore, immune and inflammatory mechanisms may also play a significant role in the pathogenesis of PH in patients with LD-CTD. Further investigation will be required to determine whether this is the case.
Another important point of this study is that LD-CTD has several prognostic factors. The univariate Cox regression model demonstrated that %FVC, %DLCO, PaO 2 , MMRC, 6MWD, lowest SpO2, MPAP, and PVR had statistically significant impacts on survival. With stepwise, multivariate Cox proportional analysis, MPAP and %FVC were shown to be independent determinants of survival. Alhamad et al. [12] showed that a lower resting SpO2 level and lower serum albumin were associated with an increased risk of death in LD-CTD, using a univariate Cox model. However, their study did not identify independent predictors of mortality due to the small sample size and low number of deaths. Therefore, this study is the first to identify independent prognostic factors in LD-CTD.
CTD-ILD patients tend to be treated with immunosuppressive therapies, although there have been few systematic and prospec-  tive studies [31][32]. In addition, a previous study suggested that a minority of CTD-PH patients could also benefit from immunosuppressive therapies [33]. LD-CTD is a new term to describe a subset of ILD patients with features of CTD. However, it is unknown whether the therapeutic strategies and responses of immunosuppressive therapies are the same in LD-CTD as they are in CTD-ILD. In this study, thirty-nine patients were treated with corticosteroid and an immunosuppressive agent. Further trials of therapy are needed for LD-CTD, and may lead to characterization of the subgroup of patients who can benefit from specific therapies. Our study has several limitations. First, this is a retrospective observational study. Collection of additional prospective data is warranted to confirm our findings. Second, LD-CTD was diagnosed on the basis of the proposed criteria without relying on the suggested histopathological features of CTD [11]. However, the central aim of this study was to investigate the prognostic ability of MPAP and other variables in patients with LD-CTD, and the clinical and physiologic evidence. Since not all patients underwent surgical lung biopsy, exclusion of these patients from the study would introduce significant selection bias. Therefore, we included all patients with LD-CTD who were diagnosed on the basis of serological criteria. Finally, our results are subject to treatment bias. Indications for therapy and choice of drug after the initial evaluation were not uniform among patients, limiting the evaluation of the prognosis.
In conclusion, we demonstrated that %FVC and MPAP have a statistically significant impact on survival in LD-CTD patients. MPAP evaluation provides additional information of disease status and will help physicians to predict mortality in LD-CTD.