Search

Background Interstitial lung disease (ILD) is a leading cause of mortality in systemic sclerosis (SSc). We have shown that total extent of lung disease on high-resolution CT (HRCT) lung, reported using a simple grading system,[1] can determine prognosis in SSc-ILD. Objectives We assessed factors that predicted deterioration in ILD using a time varying covariate model. Methods SSc patients with a baseline HRCT around the time of diagnosis of ILD were identified through the Australian Scleroderma Cohort Study (ASCS). All HRCTs and pulmonary function tests (PFTs) performed during follow-up were retrieved. Demographic and disease-related data were prospectively collected in the ASCS database. HRCTs were reported by three blinded raters (two respiratory physicians, one radiologist) following a training session. Scans were graded according to the percentage of lung disease seen; >20% - extensive, <20% - limited, unclear - indeterminate. Indeterminate results were converted to limited or extensive using an FVC threshold of 70%. The composite outcome variable was deterioration (defined as need for home oxygen or lung transplantation) or death. Demographic and disease related data including HRCT grade was compared to outcome using a Weibull hazards regression model with time-varying covariates. Results Among 172 patients followed for mean+/-SD of 3.47+/-2.93 years, there were 33 outcome events. In time-varying covariate models (based on 1309 serial PFTs and 353 serial HRCTs in the 172 patients), serial DLCO/VA (ml/min/mmHg/L) (adjusted hazards ration (aHR) =0.4, 95% CI: 0.3-0.7, p=0.001), FVC (dL) (aHR =0.9, 95% CI: 0.8-0.97, p=0.008) and extensive disease score (aHR =3.3, 95% CI: 1.1-9.9, p=0.03), were strongly predictive of outcome. Conclusions In follow-up of patients with SSc-ILD findings of extensive disease according to this simple staging system or lower DLCO/VA or FVC are indicative of increased risk of poor disease outcome. Considering the cost and radiation risk of

BACKGROUND: Interstitial lung disease encompasses a diverse group of chronic lung conditions characterised by distressing dyspnoea, fatigue, reduced exercise tolerance and poor health-related quality of life. Exercise training is one of the few treatments to induce positive changes in exercise tolerance and symptoms, however there is marked variability in response. The aetiology and severity of interstitial lung disease may influence the response to treatment. The aims of this project are to establish the impact of exercise training across the range of disease severity and to identify whether there is an optimal time for patients with interstitial lung disease to receive exercise training. METHODS/DESIGN: One hundred and sixteen participants with interstitial lung disease recruited from three tertiary institutions will be randomised to either an exercise training group (supervised exercise training twice weekly for eight weeks) or a usual care group (weekly telephone support). The 6-minute walk distance, peripheral muscle strength, health-related quality of life, dyspnoea, anxiety and depression will be measured by a blinded assessor at baseline, immediately following the intervention and at six months following the intervention. The primary outcome will be change in 6-minute walk distance following the intervention, with planned subgroup analyses for participants with idiopathic pulmonary fibrosis, dust-related interstitial lung disease and connective-tissue related interstitial lung disease. The effects of disease severity on outcomes will be evaluated using important markers of disease severity and survival, such as forced vital capacity, carbon monoxide transfer factor and pulmonary hypertension. DISCUSSION: This trial will provide certainty regarding the role of exercise training in interstitial lung disease and will identify at what time point within the disease process this treatment is most effective. The results from this study will inform and optimise the

Aim: We sought to determine whether the extent of disease on highresolution CT lung (HRCT), measured using a simple grading system, is predictive of decline and mortality in systemic sclerosis-related interstitial lung disease (SSc-ILD), independently of pulmonary function tests (PFTs) and other prognostic variables. Method(s): SSc patients with a baseline HRCT performed at the time of ILD diagnosis were identified. All HRCTs and PFTs performed during follow-up were retrieved. Demographical and disease-related data were prospectively collected. HRCTs were graded according to the percentage of lung disease seen; >20% - extensive, <20% - limited, unclear - indeterminate. Indeterminate HRCTs were converted to limited or extensive using an FVC threshold of 70%. The composite outcome variable was deterioration (defined as need for home oxygen or lung transplantation), or death. Result(s): Among 172 patients followed for mean +/- SD of 3.47 +/- 2.93 years, there were 33 outcome events. In Weibull multivariable hazards regression modelling, baseline HRCT grade was independently predictive of outcome, with an adjusted hazard ratio (aHR) = 3.0, 95% CI: 1.2-7.5, p = 0.02. In time-varying covariate models (based on 1309 serial PFTs and 353 serial HRCTs in 172 patients), serial DLCO/VA (ml/min/mmHg/L) (aHR = 0.4, 95% CI: 0.3-0.7, p = 0.001) and FVC (dL) (aHR = 0.9, 95% CI: 0.8-0.97, p = 0.008), were also strongly predictive of outcome. Conclusion(s): Extensive changes (>20%) on HRCT at baseline, reported using a semi-quantitative grading system, are associated with a three-fold increased risk of deterioration or death in SSc-ILD, compared with limited changes. Serial PFTs are informative in follow-up of patients.

Introduction. Interstitial lung disease (ILD) is a leading cause of mortality in SSc. We have shown that total extent of disease on high-resolution CT (HRCT) lung at baseline, reported using a simple grading system, predicts prognosis in SSc-ILD (5-fold increased hazard of decline/death with extensive compared with limited changes, P=0.002). Here we sought to determine factors that predict deterioration and mortality in SSc-ILD using a time-varying covariate survival model. Method. SSc patients with a baseline HRCT around the time of diagnosis of ILD were identified through the Australian Scleroderma Cohort Study (ASCS), wherein demographic and disease-related data are prospectively collected. All serial HRCTs and pulmonary function tests (PFTs) performed during follow-up were retrieved. HRCTs were reported in a blinded manner and graded according to the percentage of lung disease seen; >20%: extensive; <20%: limited; unclear: indeterminate. Indeterminate results were converted to limited or extensive using an FVC threshold of 70%. The composite outcome variable was deterioration defined as need for home oxygen, or death. A priori predictors of outcome, including HRCT grade, PFTs, demographic and disease-related variables were identified and quantified using a time-varying Cox proportional hazards regression model that included all serial HRCTs and PFTs. Results. Patient characteristics are summarized in Table 1. During follow-up, 28 (17.4%) patients changed HRCT grade, mostly from limited to extensive. After exclusion of indeterminate HRCT scores, the correlation between HRCT grade (limited vs extensive) and FVC (>=70 vs <70%) at each time-point was small (p=0.2, P=0.003). In multivariable, time-dependent Cox regression analysis using 1178 time-points in 161 patients followed for mean (S.D.) of 4.9 (3.9) years, disease subtype, smoking, age and disease duration at each timepoint were not predictive of outcome. However, as shown in Table 2, FVC (L) (HR=0.08, 95% CI

Introduction. Interstitial lung disease (ILD) is a leading cause of mortality in SSc. It has been proposed that extent of total disease on high-resolution CT (HRCT) lung, reported using a simple grading system, be used to determine prognosis in SSc-ILD. Method. SSc patients with a baseline HRCT around the time of ILD diagnosis were identified through the Australian Scleroderma Cohort Study (ASCS). All HRCTs and pulmonary function tests (PFTs) performed during follow-up were retrieved. Demographic and disease- related data were prospectively collected. HRCTs were reported by three blinded raters (two respiratory physicians, one radiologist) following a training session. Scans were graded according to the percentage of lung disease seen: >20%: extensive; <20%: limited; unclear: indeterminate. Indeterminate results were converted to limited or extensive using an FVC threshold of 70%. The composite outcome variable was deterioration defined as need for home oxygen, or death. Kaplan-Meier (K-M) survival curves and Cox proportional hazards regression models were used to determine the association between independent variables (including HRCT grade) and outcome. Results. Among 137 patients, 107 (78%) were females. Of the patients, 58 patients had dcSSc, 71 lcSSc, 7 overlap syndrome and 1 sine. The mean (S.D.) age and disease duration at baseline HRCT were 54.5 (12.6) and 9.8 (9.6) years, respectively. Ninety-eight patients were classified as having limited and 37 as having extensive changes on baseline HRCT. Inter-rater agreement was good (Cronbach's alpha=0.60). During mean (S.D.) follow-up of 3.5 (3.1) years, there were 16 'first' outcome events (5 home O2, 11 deaths). K-M event-free survival estimates according to HRCT grade are depicted in Fig. 1. In Cox regression modelling, HRCT grade was significantly predictive of deterioration or death, with an unadjusted hazard ratio (HR)=5.0, 95% CI 1.8, 13.7, P=0.002. However, in models that also included time-varying covariates

disease on high-resolution CT (HRCT) lung, reported using a simple grading system, is predictive of decline and mortality in SSc-related interstitial lung disease (SSc-ILD), independently of pulmonary function tests (PFTs) and other prognostic variables.Methods. SSc patients with a baseline HRCT performed at the time of ILD diagnosis were identified. All HRCTs and PFTs performed during follow-up were retrieved. Demographic and disease-related data were prospectively collected. HRCTs were graded according to the percentage of lung disease: >20%: extensive; <20%: limited; unclear: indeterminate. Indeterminate HRCTs were converted to limited or extensive using a forced vital capacity threshold of 70%. The composite outcome variable was deterioration (need for home oxygen or lung transplantation), or death.Results. Among 172 patients followed for mean (s.d.) of 3.5 (2.9) years, there were 30 outcome events. In Weibull multivariable hazards regression modelling, baseline HRCT grade was independently predictive of outcome, with an adjusted hazard ratio (aHR) = 3.0, 95% CI 1.2, 7.5 and P = 0.02. In time-varying covariate models (based on 1309 serial PFTs and 353 serial HRCTs in 172 patients), serial diffusing capacity of the lung for carbon monoxide by alveolar volume ratio (ml/min/mmHg/l) (aHR = 0.4; 95% CI 0.3, 0.7; P = 0.001) and forced vital capacity (dl) (aHR = 0.9; 95% CI 0.8, 0.97; P = 0.008), were also strongly predictive of outcome.Conclusion. Extensive disease (>20%) on HRCT at baseline, reported using a semi-quantitative grading system, is associated with a three-fold increased risk of deterioration or death in SSc-ILD, compared with limited disease. Serial PFTs are informative in follow-up of patients. © The Author 2012. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved.