Background/Purpose: Mortality is 3 to 5 times greater in SLE patients then it is in the general population – especially among younger patients where it can be over 10 times greater.

Aim: The aim of this study is to determine the age-specific predictors of mortality in Lupus patients.

Methods: All patients followed in a longitudinal cohort of lupus patients since 1970 were studied. Potential predictors for mortality were sex, coronary artery disease(CAD) ever (MI or Angina), Adjusted Mean SLEDAI-2K (AMS), SLICC/ACR Damage Index (SDI), current use of steroids, antimalarials, immunosuppressives, and infection. Each of the variables was evaluated at every clinic visit and used to predict mortality. Prediction models were evaluated using all of the deaths and for all visits with age < 40, for ages between 40 & 60 and for ages ≥ 60 separately. In the models with separate ages, AMS was evaluated only for the visits included in the period of analysis.

Statistical Analysis: Risk factors for death were evaluated using time-dependent covariate survival analysis.

Results: 1439 patients are included in the analysis. Depending on duration of followup, patients could be included in one or more age interval.  967 were seen at ages < 40, 730 were seen at ages between 40 & 60 and 262 were seen at ages ≥ 60.  In total, there are 1264 (87.8%) female, 958 (67%) caucasian, 178 (12%) black, 140 (10%) Asian and 163 (11%) other. Age at SLE diagnosis was 30.3 ± 13.6.

A total of 211 patients died. 51 (24.2%) deaths occurred in patients aged < 40, 80 (37.9%) deaths in patients between the ages of 40 & 60 and 80 (37.9%) deaths occurred in patients over 60 years old.

Table 1 shows the hazard ratio for each of the potential predictors for mortality for age-specific intervals.  The proportion of patients who died at each age interval increases from 5.3% for ages < 40, 11.0% for ages between 40 & 60, 30.5% for ages ≥ 60.

Disease duration is a predictor when all age groups are combined but not in age-specific analysis. Sex differences are only seen in older age with male at an increased hazard. AMS in the age specific intervals, SDI and steroids increases the hazard for death in all age intervals (with borderline significance in younger patients for steroids). Presence of infection increases the risk of death in patients aged < 60 while previous CAD ever increases the chance of mortality only in patients between the ages of 40 & 60. Antimalarials have a protective effect for patients under the age of 60. Immunosuppressives have a negligible effect in all age groups.

Conclusion:    Predictors for mortality change with age intervals. Disease activity, damage and steroids are predictors of mortality in all age intervals. In older patients is male gender is a predictor and in younger patients infection and CAD ever are predictors for mortality. The use of antimalarials is protective for individuals younger than 60.