2266 - Inpatient Mortality Related to Cerebrovascular Disease in Systemic Lupus Erythematosus: The Impact of Delay in Diagnosis of Neuropsychiatric Lupus

Tuesday, November 8, 2011: 9:00 AM-6:00 PM
Hall F2 - Poster Hall (McCormick Place West)
Jamal A. Mikdashi, Univ of Maryland Schl of Med, Baltimore, MD
Presentation Number: 2266

Background/Purpose: Despite improvements in the process of care of stroke patients, mortality related to cerebrovascular disease (CVD) in systemic lupus erythematosus (SLE) patients during the first- ever stroke remains high when compared to patient with other causes for their first-ever stroke. Purpose: To examine the predictors of inpatient mortality related to CVD in SLE and determine whether mortality associated with first-ever stroke is related to the disease effects, comorbidities or quality of care.

Method: Inpatient mortality was examined in 143 SLE patients admitted to a tertiary care stroke center with validated first-ever stroke between January 2000 and January 2011. Mortality was defined as all-cause fatal event from day of first-ever stroke through the 28 day of follow up. Demographics, clinical manifestations and processes of care were compared between those who died (n= 14) and age- and gender –matched controls who survived the stroke event [3 controls for every patient] (N= 42). Risk factors predictive of mortality were determined using univariable analyses between survivals and non survivals. Multivariate analyses and Cox proportional hazards analyses of significant variables were examined to determine their contribution to mortality related to CVD. Severity of SLE disease activity (SLEDAI), damage (SDI) and stroke severity were adjusted for during analyses.

Result: Intractable disease activity and cumulative damage involving pulmonary and renal organs were higher in patients who died compared to those who survived the first-ever stroke. Both groups had comparable disease duration, ethnicity, socioeconomic features, smoking and alcohol use, stroke type (ischemic, hemorrhagic) and Charlson comorbidity score. Death related to cerebral vasculitis (50 %) appeared to be more common than death related to hemorrhagic (29 %) or ischemic stroke (21 %). The median time to diagnosis of neuropsychiatric SLE (NPSLE) was longer for patients who died as compared to patients who survived the stroke event [7days v 3 days; p < 0.026].

Table 1.  Univariable analyses and Cox hazard analyses of factors associated with mortality related to CVD

                                                 Univariable  Model                               Proportional Hazard Model

Factors                           Odds Ratio (95 % Confidence Interval)         Hazard Ratio (95 % Confidence interval)

Seizure                                           3.6  (1.1 - 12.3)                               2.7 (1.0 - 7.6)

Cognitive impairment                      3.7 (0.9 - 19.3)                                 2.1 ( 0.7 - 6.4)

Thrombocytopenia                           4.8   (1.3 – 17.6)                              3.3 (1.2 – 9.3)

Lupus anticoagulant                        3.3 (0.9 - 15.4)                                 2.5 (0.8 - 7.0)

C Reactive Protein                           1.2 ( 0.48 - 3.1)                               0.9 ( 0.3 - 2.9)

Delay in NPSLE diagnosis                 7.0 (1.8 - 27.1)                                 4.2 (1.4 - 11.6)

Conclusion: Despite the admission to a designated stroke center, mortality related to CVD in SLE remains significantly high. Quality improvement strategies targeting early diagnosis and timely intervention of neuropsychiatric SLE may reduce the associated morbidity and mortality among SLE patients.


Keywords: morbidity and mortality, neuropsychiatric disorders and quality of care

Disclosure: J. A. Mikdashi, None.