Neutropenia at diagnosis of systemic lupus erythematosus: prevalence and clinical and serological correlations

Our cohort

Abstract

Introduction: hematological manifestations are frequent in systemic erythematosus lupus (SLE). Leukopenia is seen in between 50 to 60% of cases, but only 17% has a leukocyte count <1,000/mm3. Neutropenia in patients with leukopenia occurs between 20-40% of cases, depending on the cut-off value used. Possible described mechanisms for neutropenia are: an increase in destruction of granulocytes by anti-neutrophil antibodies, opsonization and destruction by monocytes; change in the splenic and marginal neutrophil pool; a diminished production in the bone marrow. The formation of NETs contributes to the production of INF-1 from plasmocytes and dendritic cells, causing endothelial damage and pro-thrombotic changes. NETosis and apoptotic abnormal clearence promote the formation of antigens and subsequent autoantibodies. Infectious consequences of neutropenia in SLE are still unknown. The objectives of this article were to know the prevalence of neutropenia at diagnosis of SLE in our hospital, and secondly to determine its correlation with other variables of the disease and to investigate whether it’s related with a greater probability of infections. Materials and methods: descriptive, retrospective study. Patients with diagnosis of SLE (SLICC 2012) from our cohort were included. Demographic and related to disease variables were stated. Activity scale: SLEDAI-2k. Patients were divided into two groups according to the presence or absence of neutropenia (<1.500/mm3) and multivariate analysis was performed to clinical and analytical variables. A subgroup with severe neutropenia (<500/mm3) was evaluated. Multivariate analysis was performed to detect correlations between a diminished neutrophil count and clinical manifestations, disease severity, autoantibodies pro le, infections, and associated diseases. In neutropenic patients, the presence of viral or bacterial infection and the use of GM-GSF were evaluated. Statistical analysis was performed as mean +/-SD for continuous variables and percentage for categorical variables. T-Test or Mann-Whitney were used to compare independent variables according to distribution. Student’s T and Chi-Square for qualitative variables. Statistical signi cance: p<0.05. Results: 70 patients were included. Female 59 (84%), mean age 38.6 years (1872). Leukopenia 24 (34%), lymphopenia 30 (42.8%), neutropenia 12 (17%), severe 2 (2.8%), thrombocytopenia 7 (10%). Neutropenic group: Sicca 12 (100%), neutrophil/lymphocyte index (NLI) 1.33 (DS 0.69), infections: EBV IgM+1/12, parvovirus and CMV negative 2/12. BMA 1/12, without pathologic findings. GM-GSF 2/12. Infections: 2/12 (urinary). Conclusions: we observed a correlation between Sicca symptoms, leuko and lymphopenia, and a lower NLI. The clinical signi cance of these findings was uncertain. The presence of infection was low (16%). Two required GM-GSF, having not presented adverse events.