Tornetta 2013.

Methods	Randomised controlled trial
Participants	Setting: multicentre study (22 listed in trial registration document) in USA Size: 156 participants, 80 allocated to locking plate and 76 to intramedullary nail group Recruitment period: February 2007 to (estimated) December 2012, expected date of completion December 2013) Baseline characteristics: mean age 51 years (range 16‐90), 71 men, 55 women (of the 126 completing follow‐up), mean Injury Severity Score 12.6 (range 9‐43), 34% fractures were open Inclusion criteria: skeletally mature fracture of the metaphyseal distal femur with or without intra‐articular extension and with or without a total knee arthroplasty fracture requiring operative treatment amenable to either intramedullary nail or plate informed consent obtained participant is English speaking Exclusion criteria: fracture of the metaphyseal distal femur with intra‐articular comminution fracture with vascular injury (Gustillo Type IIIC injury) requiring repair pathological fracture known metabolic bone disease contralateral distal femur fractures (bilateral injury) or ipsilateral lower extremity injury that would compromise function of the knee retained hardware or existing deformity in the affected limb that would complicate intramedullary nailing or plating symptomatic knee arthritis soft tissue injuries compromising either treatment method with nail or plate surgical delay > 3 weeks for closed fractures or 24 hours for open fractures immunocompromised unable to comply with postoperative rehabilitation protocols or instructions (i.e. head injured or mentally impaired) current or impending incarceration unlikely to follow‐up in surgeon's estimation
Interventions	Participants were allocated to either: reamed, interlocking retrograde intramedullary nail open reduction and internal fixation with locking plate
Outcomes	Schedule: 3, 6 and 12 months; 12‐month follow‐up reported so far; planned to be up to 24 months Outcomes: patient‐reported functional and quality of life scores (Short‐Form 12 questions version 2 (SF‐12v2), EuroQol‐5D (EQ‐5D), Short Musculoskeletal Function Assessment (SMFA), Knee Society Score), clinical assessment, re‐operation, non‐union, malunion, infection, compartment syndrome, knee range of motion
Notes	Communication with authors was unsuccessful. This is likely to be an interim report of available 1‐year data prior to study completion, which is estimated to be December 2015
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "randomization scheme was with permutated blocks for open and closed fractures using a HIPAA [Health Insurance Portability and Accountability Act]‐compliant computer‐based system" Comment: appropriate sequence generation
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "masking: open label" (trial registration) Comment: the nature of the interventions means that the operating surgeons were not blinded. This also confirms that there was also no participant blinding
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "masking: open label" (trial registration) This confirms that there was no blinding of outcome assessors Additionally, outcome assessment is poorly reported
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Quote: "156 patients were randomized ... 126 patients were followed" Comment: 81% follow‐up but allocations of those participants who were lost were not reported
Selective reporting (reporting bias)	Unclear risk	The majority of outcomes described in the trial registry entry have been reported in the conference abstract. However, no data were reported beyond the 1‐year stage despite follow‐up planned out to 2 years
Imbalance in baseline characteristics (selection bias)	Unclear risk	Quote: "There were no differences in demographic information or injury pattern" Comment: however, there are no supporting data
Care provider expertise (performance bias)	Unclear risk	Providers of interventions were not stated