An evidence-based approach to the management of steroid sensitive nephrotic syndrome in children

Elisabeth Hodson

Centre for Kidney Research, The Children's Hospital at Westmead, Westmead, NSW 2145, Australia

The incidence of childhood nephrotic syndrome is 1-2 per 100,000 children aged below 15 years. The incidence appears higher in children from the Indian subcontinent and South-East Asia. Before antibiotics and corticosteroids, the mortality was around 70%. With the introduction of steroids in 1951, mortality fell to 9% since 70-80% had steroid sensitive nephrotic syndrome (SSNS). Because of this impressive before-after effect, steroids were never tested in randomised placebo-controlled trials (RCTs). Many children relapsed and developed steroid side effects prompting the use of non-steroid immunosuppressive agents. In 1966 the International Study of Kidney Disease in Childhood (ISKDC) was set up to study the natural history of nephrotic syndrome and to trial non-steroid agents. The ISKDC proposed a standard steroid regime of daily prednisone (60mg/m2/day) for 4 weeks and prednisone (40mg/m2/day) given on 3 consecutive days of 7 against which to test other agents. In 1979 the Arbeitsgemeinschaft für Pœdiatrische Nephrologie (APN) showed that prednisone given every 48 hours was more effective than intermittent dosage. However the relapse rate on the ISKDC or APN standard regimen remained around 70% so that different durations of steroid therapy in the first episode of SSNS were investigated. To integrate the results of RCTs, we have carried out a systematic review and meta-analysis. A meta-analysis of 5 RCTs showed that the risk for relapse was significantly reduced when steroids were given for 3 months or more compared with the standard regimen in the first episode of SSNS. The risk for relapse fell by 13% for each increase of 1 month over the standard regimen up to 7 months. Cyclophosphamide, chlorambucil, cyclosporin and levamisole are effective in increasing the relapse-free period in SSNS. However all have adverse effects so we carried out a systematic review to determine the relative efficacy of these agents. This confirmed the efficacy of all agents but did not provide data on which agent should be preferred. Thus systematic reviews can provide a rational basis for the treatment of SSNS and show where unanswered questions require further adequately powered randomised trials.

Invited lecture at the 6th annual convention of Pediatric Nephrology of the Philippines, Manila, July 2001

Correspondence
Elisabeth Hodson
ElisaH@chw.edu.au