MORTALITY AND CAUSES OF DEATH OF 344 DANISH PATIENTS WITH SYSTEMIC-SCLEROSIS (SCLERODERMA)

Objective. To determine survival, mortality and causes of death in Dan ish patients with systemic sclerosis (scleroderma), and to analyse how these parameters are influenced by demographic variables and the exte nt of skin involvement. Methods. A cohort of 344 patients with inciden t systemic sclerosis (SSc) after 1 January 1960 was retrospectively id entified, representing 3716 patient-years of follow-up. The vital stat us at the end of the study ultimo 1996 was established by reviewing th e clinical charts and by contacting the Danish Central Person Register . The causes of death were based on information obtained from the hosp ital charts and the autopsy reports (80%) and death certificates (20%) . The extent of skin sclerosis was described by means of a two-subset model (limited and diffuse involvement) and a three-subset model (digi tal, extremity and truncal involvement) in all patients. Expected numb ers of deaths were calculated by means of age- and sex-specific mortal ity rates of the general Danish population. Results. Crude mortality r ates were 6.5 and 3.9%/yr, respectively, in men and women. The standar dized mortality ratio (SMR) was 2.9 (95% CI 2.5-3.4) and was not signi ficantly influenced by sex or age at disease onset. Mortality rates we re significantly increased in patients with diffuse skin sclerosis (SM R 4.5, 95% CI 3.5-5.7) and patients under the age of 35 yr (SMR 13, 95 % CI 2.7-37). Of the 160 patients who died during the study, 41 deaths (26%) were SSc related, accounting for one-third of the excess mortal ity. Conclusion. We found a significantly increased mortality in patie nts with SSc, particularly in the subset of patients with diffuse skin sclerosis and in young patients. The excess mortality was due to an i ncrease in both the SSc-related mortality and the unrelated mortality. However. all other subsets also had a significantly increased risk of death, mainly due to an increased risk of death due to causes unrelat ed to SSc, cancer among others. The Fact that the unrelated mortality in SSc was found to be 2-fold increased suggests that the excess morta lity that was classified as unrelated should more appropriately be ter med indirectly SSc related.