Benfield 2005.

Methods	Study design: parallel RCT Time frame: 2001 to 2004 Follow‐up period: 3 years Primary endpoint: change in standardised height z score
Participants	Country: Mexico, USA Setting: multicentre (17 centres) Age: 0 to 20 years First cadaveric or living kidney transplant; enrolment at transplantation; randomisation 6 months after transplantation of participants without previous rejection if clinical or histologic evidence of rejection in protocol biopsy absent Number: withdrawal group (73); maintenance group (59) Mean age ±SD (years): withdrawal group (11 ± 5); maintenance group (12 ± 6) Sex (female): withdrawal group (44%); maintenance group (37%) Donor source (living donors): withdrawal group (64%); maintenance group (69%) Exclusion criteria: not reported
Interventions	Treatment group Steroid withdrawal 6 to 12 months after transplantation Control group Steroid maintenance Baseline immunosuppression Basiliximab: day 0 and 4 CsA or TAC CsA trough level: weeks 1 to 2: 175 to 400 ng/mL; week 3 to month 3: 175 to 300 ng/mL; thereafter: 50 to 250 ng/mL; TAC trough level: weeks 1 to 4: 10 to 15 ng/mL; thereafter: 5 to 10 ng/mL SRL: starting on day 1 with 6 mg/m2/d adjusted to trough level: 10 to 20 ng/mL Steroids IV methylprednisone: day 0 and 1: 10 mg/kg Oral prednisone: starting on day 2 with 2 mg/kg/d, tapered to 0.15 mg/kg/d by day 74 Withdrawal group: withdrawal by end of month 12 after transplantation Maintenance group: maintained on 0.15 mg/kg/d
Outcomes	Mortality Graft loss Acute rejection CrCl (mL/min) Malignancy (PTLD)
Notes	The study was terminated on 13 August 2004 due to an unanticipated high incidence of post‐transplant lymphoproliferative disease; 19 patients developed PTLD (before randomisation: 10) Did not report number screened for eligibility 142/274 enrolled participants were not randomised (52% drop out before randomisation), because of rejection (40), graft loss (9), death (2), had not yet reached 6 month protocol biopsy when study was stopped (35), adverse events (16). protocol violation (4), lost to follow‐up/withdrawal of consent (5), other reasons (31) Contact with study authors for additional information: authors contacted 8 July 2013; no response received
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Stated 'centrally randomised' but no further information provided.
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding (performance bias and detection bias) All outcomes	Low risk	Stated 'in a placebo controlled double‐blinded fashion' but no further information provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Stated 'in a placebo controlled double‐blinded fashion' but no further information provided
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Stated 'in a placebo controlled double‐blinded fashion' but no further information provided. Outcomes are objective hard endpoints
Incomplete outcome data (attrition bias) All outcomes	Low risk	Total number of patients by group not reported for outcomes; ITT analysis performed
Selective reporting (reporting bias)	Low risk	Primary outcomes for this review have been reported
Other bias	High risk	High drop‐out rate before randomisation (52%) Choice of calcineurin inhibitor was centre specific (TAC or CsA) Support provided by NIH UO1‐A1‐46135 and Wyeth Pharmaceuticals The study was terminated early due to an unanticipated high incidence of PTLD