Abstract
Introduction: Whether sarcoidosis patients are at a higher risk for death compared to the rest of the population is questionable due to conflicting results from a handful of longitudinal studies.
Aims and objectives: To compare the mortality of individuals with sarcoidosis vs the general population.
Methods: The Swedish National Patient Register (2003-2013) was used to identify patients with incident sarcoidosis (ICD codes: D86/135) defined as at least two visits, >14 days apart, with at least one where sarcoidosis was the primary diagnosis. They were required to be 18-85 yrs old with no lymphoma diagnosis ±3 months from the first sarcoidosis visit. Controls identified from the Total Population Register were matched to cases 10:1 on age, sex and county. Death was identified from the Cause of Death Register and premature death was defined as death age <70 yrs. We estimated age- and sex-adjusted mortality rates using Poisson models, and hazard ratios (HR) for death using Cox models, adjusted for age, sex, region, birth country, education, and comorbidity (Charlson index).
Results: We identified 8229 sarcoidosis patients and 81402 controls (mean age 49 yrs, 56% male). During follow-up, 456 deaths occurred in the sarcoidosis group and 2659 in the controls (mortality rate 11.3 vs 6.6/1000 person-yrs, respectively). For all-cause death, the HR was 1.65 (95% CI 1.50-1.83) comparing sarcoidosis to controls, and 1.68 (95% CI 1.45-1.96) for premature death.
Conclusions: Compared to the general population, individuals with sarcoidosis have a greater risk for death and are more likely to die prematurely. This shows sarcoidosis is not always a self-resolving disease and may cause severe long-term consequences.
- Copyright ©the authors 2017