Background Hematological complications are frequent in systemic lupus erythematosus (SLE) and are considered as good prognosis compared to other organs involvement. However, little is known about the impact of these manifestations in hospital mortality in SLE patients.

Objectives To determine the prognostic impact of hospitalized patients with SLE with and without hematological manifestations

Methods We performed a retrospective case-control study nested in a cohort of patients hospitalized for active SLE (ACR 1997) from January 2009 to July 2012. Case definition: patients with hematologic manifestation attributable to SLE. Control definition: SLE patient without hematologic manifestation. Severe hematologic manifestation was defined as follows: thrombocytopenia <30,000 K/μL, neutropenia <500 K/μL and hemolytic anemia with <7g/dL. We excluded cases associated with drug toxicity. Information was obtained from medical records of patients during first admissions and included clinical, laboratory and immunological variables. Statistical analysis included descriptive statistics, Student's t-test, standard chi-square according to the behavior of the variables; we identified predictors of hospital death by multivariate analysis; odds ratios (OR) with confidence intervals (CI) 95% were also calculated.

Results We studied 83 cases (mean age 33.3±14.2 years) and 123 controls (mean age 30.5±11.6 years). There were no differences in immune variables. Thrombocytopenia (severe in 62%) was the most common hematological feature followed by hemolytic anemia (severe in 37%) and neutropenia (severe in 89%). Renal (65% vs 45%) and neurologic (26% vs 10%) manifestations were more prevalent in control group. The mean SLEDAI in cases on admission was 9.7±6.8 and in controls 11.0±7.1 (p=0.2). The number of organs involved in the cases was 2±1 and 3±1 control group (p<0.01). There were 24 hospital deaths, 19 (23%) in cases and 5 (4.1%) in controls (p<0.01). The main causes of death in the cases were: disease activity, gastrointestinal hemorrhage, stroke, pneumonia and sepsis and in the controls, they were diffuse alveolar hemorrhage, infections and kidney failure. In multivariable analysis the main variables associated with mortality in cases were hematologic SLE (OR 9.5, 95% CI 3.3-27.6, p<0.01) and low C3 (OR 3.5, 95% CI 1.08-11.8, p=0.04).

Conclusions Our study suggests that hospitalized SLE patients with hematologic complications have a higher mortality in comparison with non-hematologic SLE. The number of organs involved and SLEDAI score on admission had no impact on mortality. The presence of hematologic activity and low C3 are factors associated with mortality. The hematologic abnormalities in SLE should have higher scores on indexes of disease activity.

References

1. Hepburn AL, Rheumatology 2010;49:2243-2254.

2. Zhao H, Platelets 2010;21:380-385.

3. Jallouli M, Lupus 2012;21:682-687

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.5170