Background: Neuropsychiatric involvement in systemic lupus erythematosus is complex and several clinical presentations are related to this disease such as: convulsions, chronic headache, transverse myelitis, vascular brain disease, psychosis and neural cognitive dysfunction.
Objectives: To assess the efficacy and safety of cyclophosphamide and methylprednisolone in the treatment of neuropsychiatric manifestations of systemic lupus erythematosus on mortality and side effects.
Search strategy: We searched EMBASE, LILACS, Cochrane Controlled Trials Register and MEDLINE up to and including December 1999, additional articles were sought through handsearching in relevant journals, using the search strategy described in the Cochrane Handbook [Dickersin 1994]. There were no language restrictions.
Selection criteria: All randomized controlled trials which compared cyclophosphamide to methylprednisolone were to be included. Patients of any age and gender were included if they fulfilled the criterion of the American Rheumatology Association for the diagnosis of systemic lupus erythematosus and presented with any one of the following neuropsychiatric events; convulsions, organic brain syndrome; cranial neuropathy. Outcome measures included the following: a) Overall mortality (primary event); b) Motor and psychiatric deficit (primary event); c) Clinical improvement (secondary event).
Data collection and analysis: The analysis planned was to do the following: Data would be independently extracted by the two reviewers and cross-checked. The methodological quality of each trial would be assessed by the same two reviewers. Details of the randomisation (generation and concealment), blinding, and the number of patients lost on follow-up would be recorded. The results of each RCT would be summarised on an intention-to-treat basis in 2 x 2 tables for each outcome. External validity would be defined by characteristics of the participants, the interventions and the outcomes. If appropriate, RCTs would be stratified based on control group and category of disease in accordance to the clinical homogeneity (external validity). The results obtained from these different methods are very similar, and therefore, only the results from the Risk Difference method, with the corresponding 95% confidence interval would be presented in this review. The fixed effects model would be used if there was no significant statistical heterogeneity.
Main results: We found no randomised controlled trials comparing cyclophosphamide versus methylprednisolone for the treatment of neuropsychiatric involvement in the systemic lupus erythematosus.
Reviewer's conclusions: Cyclophosphamide regimen treatment is a form of care in neuropsychiatric involvement in systemic lupus erythematosus with no evidence to prove better effectiveness and safety when compared with methylprednisolone. This systematic review found no randomised controlled trials and its findings must be interpreted as 'no evidence of effect' and not as 'evidence of no effect'.