Objectives To examine the standardized mortality ratios (SMRs) and predictive factors for malignancy in a cohort of southern Chinese patients with inflammatory myopathies (IM).
Methods All patients who were diagnosed to have polymyositis (PM), dermatomyositis (DM) or amyotrophic dermatomyositis (ADM) between 2000 and 2010 in the three regional hospitals in Hong Kong were studied. Diagnosis was made according to the Peter and Bohan criteria. Demographic data, clinical presentation, clinical status at last clinic visit and causes of death were retrieved from the hospital database. The sex- and age-adjusted standardized mortality ratios (SMRs) in comparison to the general population for cancer and non-cancer associated DM/ADM and PM were calculated. Cumulative survival of the patients over time was studied by the Kaplan-Meier’s method. Demographic data, presentation at diagnosis (DM vs ADM vs PM), concomitant rheumatic diseases, extra-muscular manifestations were analyzed as predictors for mortality by the Cox regression model.
Results 215 patients (64.6% women) with IM were studied (125 DM, 75 PM, 15 ADM). IM was idiopathic in 124 (57.6%) patients, malignancy-related in 52 (24.2%) patients and in 39 (18.1%) patients with concomitant rheumatic diseases. The mean age of disease onset was 51.5±16.3 years. 81 patients (37.6%) died after a mean follow-up period of 4.7±4.6 years. The age- and sex-adjusted SMRs for DM/ADM and PM were 6.4[5.3-7.7], 7.5[5.8-9.6]; (both p<0.001) respectively and the corresponding figures for non-cancer and cancer associated DM/ADM and PM were 2.5[2.1-3.2], 16.8[10-28.2] and 7.8[6.1-10.3], 55.5[13-222]; (all p<0.001) respectively. The cumulative survival rates at 1, 3 and 5 years after diagnosis of DM/ADM and PM patients were 91.7%, 80.1%, 67.4% and 90.2%, 79.4%, 61.8% respectively (p=0.82). The most frequent causes of death were, in descending order, infection 31 (38.2%), malignancies 29 (35.8%), cardiovascular diseases 10 (12.3%), interstitial lung disease 5 (6%). On univariate analysis, the age at IM diagnosis (HR 1.04[1.03-1.06]; p<0.001), male gender (HR 2.36[1.49-3.71]; p<0.001), history of smoking (HR 2.53[1.58-4.01]; p<0.001), presence of associated malignancies (HR 4.18[2.65-6.57]; p<0.001), the diagnosis of DM (HR 1.94[1.39-8.14]; p=0.008), oropharyngeal muscle involvement (dysphagia) (HR 2.28[1.41-3.69]; p<0.001) were predictors for mortality whereas the presence of concomitant rheumatic diseases (HR 0.15[0.04-0.49]; p=0.002), use of azathioprine (HR 0.46[0.29-0.74]; p=0.001), use of hydroxychloroquine (HR 0.34[0.18-0.65]; p=0.03) were protective factors for mortality. Cox regression analysis revealed that the age at IM diagnosis (HR 1.04[1.02-1.05]; p<0.001), presence of malignancy (HR 2.29[1.33-3.93]; p=0.003), oropharyngeal muscle involvement (dysphagia) (HR 2.61[1.52-4.48]; p<0.001) were independent predictors for mortality in IM
Conclusions Idiopathic IM has a poor survival than IM associated with rheumatic diseases. DM/ADM and PM both have significantly increased mortality compared to the general population and to the similar extent. The presence of associated malignancies is a major determinant of survival in dermatomyositis patients but less important in predicting mortality in PM.
Disclosure of Interest None Declared