Wu 2014.

Methods	Study design: randomized, parallel‐group, multi‐centre, safety assessor‐blinded study Sample size calculation: powered to demonstrate that recovery of the TOFR to 0.9 after sugammadex 2 mg/kg is ≥ 2 times faster than after neostigmine 50 μg/kg
Participants	Number of randomized participants: Chinese 247, Caucasian 61, all in all 308 Inclusion criteria: age 18 to 64 years, ASA I to III, scheduled for elective surgery under general anaesthesia, allowing stable neuromuscular monitoring, which requires neuromuscular blockade using rocuronium; compliant with dose/visit schedules, and used an accepted method of contraception (if applicable). Chinese participants had to be born in China, to have never emigrated out of China, and to have a Chinese home address. Similarly, Caucasian participants had to be born in Europe, to have never emigrated out of Europe, and to have a European home address Exclusion criteria: anatomical malformations expected to lead to difficult tracheal intubation; neuromuscular disorders affecting NMB; significant renal/hepatic dysfunction (as determined by the investigator); (family) history of malignant hyperthermia; allergy to general anaesthesia medications; contraindication to study drugs; or clinically significant condition that may interfere with the trial (as determined by the investigator)
Interventions	Anaesthesia: Anaesthesia was induced and maintained with IV propofol according to clinical needs of the participant. Opioids could be administered according to local practice NMBA: single intubating dose: rocuronium 0.6 mg/kg; maintenance dose: rocuronium 0.1 to 0.2 mg/kg Comparison: sugammadex 2 mg/kg (Chinese n = 126, Caucasian n = 29) vs neostigmine 50 μg/kg plus atropine 10 to 20 μg/kg (Chinese n = 121, Caucasian n = 32) Administration time of sugammadex or neostigmine: reappearance of T2
Outcomes	Primary efficacy variable: time from start of administration of sugammadex or neostigmine/atropine to recovery of TOFR to 0.9 Secondary efficacy variable: time to recovery of the TOFR to 0.7 and 0.8 Safety assessments: adverse events, serious adverse events, vital signs, and physical examination findings
Notes	Publication type: peer‐reviewed article Country: 6 sites in China and 4 sites in Europe (2 sites in Denmark and 1 site each in Belgium and Norway) Conversions: Median + Range to Mean + SD following guidelines from Hozo 2005 Handling of adverse events: More detailed information regarding number of adverse events possibly, probably, or definitely related to study drug was provided by the authors through email correspondence, and we used these updated numbers in the review Authors’ conclusions: Both Chinese and Caucasian participants recovered from NMB significantly faster after sugammadex 2 mg/kg than after neostigmine 50 μg/kg, with recovery that was ˜ 5.7 times (P < 0.0001) faster with sugammadex than with neostigmine in Chinese participants. Sugammadex was generally well tolerated Contact: first trial author Xinmin Wu contacted by email: xmwu2784@hotmail.com on 15.04.2016; no reply received; email sent to last author Woo 15.05.2016; replied 21.07.2016
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Eligible participants were randomized via a central randomization system. Sponsor produced a computer‐generated randomization schedule with treatment codes in blocks, using a validated SAS‐based application. The schedule associated each treatment code with a participant number, and participants were randomized in a 1:1 ratio
Allocation concealment (selection bias)	Low risk	Central allocation (secondary to central randomization)
Blinding of participants (performance bias)	Low risk	Participants were blinded *
Blinding of personnel (performance bias)	High risk	Personnel in the OR were not blinded
Blinding of primary outcome assessment (detection bias)	High risk	TOF‐watch assessor was not blinded
Blinding of safety assessment (detection bias)	Low risk	Safety assessments were performed by a safety assessor who was blinded to the treatment administered
Incomplete outcome data (attrition bias) All outcomes	Low risk	All participants are accounted for, and missing outcome data are balanced in numbers across intervention groups, with similar reasons for missing data across groups: Of 247 randomized Chinese participants, 16 discontinued the study, and one who completed the study had missing efficacy data. Hence, 231 Chinese participants received study treatment and were included in the safety analysis (AST group), and 230 Chinese subjects with evaluable data were included in the efficacy analysis (full analysis set; sugammadex n = 119, neostigmine n = 111). In total, 61 Caucasian participants were randomized, 60 of whom received treatment (AST group) and 59 who had evaluable data (full analysis set; sugammadex n = 29, neostigmine n = 30)
Selective reporting (reporting bias)	Low risk	Study protocol is available on clinicaltrials.gov (NCT00825812), and all of the study’s prespecified outcomes of interest to the review have been reported in the prespecified way
Funding bias	High risk	Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Whitehouse Station, NJ, USA,. provided financial support to the study. Medical writing support was provided by Melanie More of Prime Medica Ltd., Knutsford, Cheshire, UK. This assistance was funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Whitehouse Station, NJ, USA. HR is an employee of MSD, Oss, The Netherlands, and TW is an employee of Merck Sharp & Dohme Corp., Whitehouse Station, NJ, USA, both of whom may own stock and/or hold stock options in the Company. EA was formerly an employee of MSD, Oss, The Netherlands. XW, SY, JL, BV, LX, CC, VD, YY, HO, and YH work for institutions that received research funding from Merck Sharp & Dohme Corp., Whitehouse Station, NJ, USA. BV and VD have also received research funding from Merck & Co., Inc. for previous studies
Other bias	Low risk	Appears free of other sources of bias. Study sample size calculation designed to address this review's primary outcome. Baseline characteristics comparable within participant groups