Gilling 2018.

Methods	Study design: double‐blinded, parallel, randomised clinical study (2:1) Statistical design: non‐inferiority Setting/Country: multicenter (17 centres)/ multicountry (Australia 1, New Zealand 1, UK 3, USA 12) Dates when study was conducted: October 2015‐December 2016
Participants	Ethnicity: Asian 5, Hispanic 5, black 4, white 168, other 1, unknown 1 (participant did not provide race during baseline data gathering) Inclusion criteria LUTS due to BPH Men aged 45‐80 years Prostate size 30 mL‐80 mL (measured with TRUS) IPSS ≥ 12, Qmax < 15 mL/s Normal range of serum creatinine History of inadequate response, contraindication, or refusal to medical therapy Exclusion criteria BMI ≥ 42 History of prostate cancer or current/suspected bladder cancer Prostate cancer should be ruled out before participation to the satisfaction of the investigator if PSA is above acceptable thresholds History of actively treated bladder cancer within the past 2 years Bladder calculus or clinically significant bladder diverticulum (e.g. pouch size > 20% of full bladder size) Active infection, including urinary tract infection Prostatitis treated with antibiotics within 1 year of enrolment Ever been diagnosed with a urethral stricture, meatal stenosis, or bladder neck contracture Damage to external urinary sphincter Diagnosis of stress urinary incontinence that requires treatment or daily pad or device use PVR > 300 mL Urinary retention at time of enrolment or has been catheterised in the 14 days prior to the surgical procedure History of intermittent self‐catheterisation Previous prostate surgery or history of other lower urinary tract surgery such as e.g. urinary diversion, artificial urinary sphincter or penile prosthesis Taking anticoagulants (if medication cannot be stopped before and after procedure) or known coagulopathy (except aspirin < 100 mg/day) Any severe illness that would prevent complete study participation or confound study results Participants using systematic immune‐suppressants including corticosteroids; unable to withhold NSAIDs, including aspirin for 3‐5 days prior to treatment except for low‐dose aspirin (e.g. ≤ 100 mg) Participants using anticholinergics specifically for bladder problems. Use of medications with anticholinergic properties was allowable provided the participant did not have documented adverse urinary adverse events from these medications Contraindication to general or spinal anaesthesia Any other disease or condition(s) that would interfere with completion of the study and follow‐up assessments, would increase risks of the procedure, or in the judgment of the investigator would potentially interfere with compliance to this study or would adversely affect outcomes Unwilling to accept a transfusion should one be required Total number of participants randomly assigned: Screened: 275 Eligible: 184 Aquablation number of all participants randomly assigned: 117 age (years): 66.0 ± 7.3 prostate volume (mL): 54.1 ± 16.2 PSA (ng/mL): 3.7 ± 3.0 IPSS: 22.9 ± 6.0 Qmax (mL/s): 9.4 ± 3.0 TURP: monopolar 36 (55.4%), bipolar 29 (44.6%)) number of all participants randomly assigned: 67 age (years): 65.8 ± 7.2 prostate volume (mL): 51.8 ± 13.8 PSA (ng/mL): 3.3 ± 2.3 IPSS: 22.2 ± 6.1 Qmax (mL/s): 9.1 ± 2.7
Interventions	Aquablation Using the Aquabeam System (PROCEPT BioRobotics, Redwood Shores, California, USA). A 24 F handpiece probe similar to a rigid cystoscope is inserted into the prostatic urethra and locked into place using a bed‐mounted rigid arm. Under real‐time prostate visualisation using transrectal ultrasound, the surgeon uses a console to mark the target resection contour. Under the surgeon’s control, the ablation of tissue is robotically executed using a high‐velocity waterjet to resect adenomatous tissue while avoiding the verumontanum and the ejaculatory ducts. TURP (monopolar 36 (55.4%), bipolar 29 (44.6%)) TURP was performed according to standard practice Follow‐up: 12 months
Outcomes	Primary outcome Change in IPSS from baseline to 12 months How measured: questionnaire Time to measured: not reported Time to reported: baseline, 1 month, 3 months, 6 months, 12 months Secondary outcome Resection time, total operative time, length of hospital stay, reoperation or reintervention rate Worsening of sexual function through 6 months, change in incontinence, pelvic pain, quality of life, duration of bladder catheterization, Work Productivity and Activity Impairment, relationship between prostate size reduction and change in symptoms scores How measured: IIEF (sexual function), MSHQ‐EjD (sexual function), EuroQoL‐5D (QoL), TRUS (prostate size), Incontinence Severity Index and the others were not reported Time to measured: not reported Time to reported: baseline, 1 month, 3 months, 6 months, 12 months Safety outcome Adverse events How measured: Clavien–Dindo classification of adverse events Time points measured: not reported Time points reported: 12 months post‐treatment Subgroup: pre‐planned subgroup analysis Baseline IPSS (< 20 versus ≥ 20) Prostate size (< 50 versus ≥ 50 mL) Age (< 65 versus ≥ 65 years)
Funding Sources	PROCEPT BioRobotics
Declarations of interest	Dr Peter Gilling, Dr Paul Anderson, Dr Mihir Desai, Dr Alexis E. Te and Dr Mark DeGuenther report financial interest and/or other relationship with PROCEPT® Biorobotics
Notes	Manuscript preparation supported by PROCEPT® BioRobotics Protocol: NCT02505919 Language of publication: English
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from publication: "Randomization was obtained through a web‐based system and was stratified by study site" Comment: we considered this method of random sequence generation to have low risk of bias
Allocation concealment (selection bias)	Low risk	Quote from publication: "Randomization was obtained through a web‐based system and was stratified by study site" Comment: random sequence generation seems to have been carried out by the central computer randomisation system. This method may ensure allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote from publication: "Baseline evaluation and study treatment were provided by an unblinded research team" Comment: while participants were blinded (protocol), surgeon was not blinded; therefore we considered risk of performance bias to be high
Blinding of outcome assessment (detection bias) Susceptible: Urologic symptom scores, Qulaity of life, Erectile function, Ejaculatory function, Minor adverse events	Low risk	Quote from publication: "participants were blinded" (protocol), "A separate blinded team (coordinator and physician) conducted the follow‐up visits and will do so out to the completion of the trial" Comment: outcome assessor was blinded; therefore we considered risk of detection bias to be low
Blinding of outcome assessment (detection bias) Not susceptible: Major adverse events, Retreatment, Acute urinary retention, Indwelling urinary catheter, Hospital stay	Low risk	Quote from publication: "All adverse events were adjudicated by an independent clinical events committee blinded to treatment assignment" Comment: outcome assessor was blinded; therefore we considered risk of detection bias to be low
Incomplete outcome data (attrition bias) Urologic symptom scores and quality of life	Low risk	Comment: 3/117 (2.5%) in Aquablation and 7/67 (10.4%) in TURP participants were not included in the analysis. Owing to the small number of participants lost to follow‐up and this not affecting the effect size, we considered risk of attrition bias to be low
Incomplete outcome data (attrition bias) Major and minor adverse events	Low risk	Comment: 1/117 (0.8%) in Aquablation and 2/67 (2.9%) in TURP participants were not included in the analysis but missing outcome data balanced in numbers across intervention group. Owing to the small number of participants lost to follow‐up, we considered risk of attrition bias to be low
Incomplete outcome data (attrition bias) Retreatment	Low risk	Comment: 1/117 (0.8%) in Aquablation and 2/67 (2.9%) in TURP participants were not included in the analysis but missing outcome data balanced in numbers across intervention group. Owing to the small number of participants lost to follow‐up, risk of attrition bias was considered to be low
Incomplete outcome data (attrition bias) Erectile and ejaculatory function	High risk	Quote from publication: "Since IIEF and MSHQ assume that a man is sexually active, those who were not sexually active at baseline or the study visit were excluded from this analysis" Comment: analyses of these outcomes were not consistent during the follow‐up period (i.e. included participants may differ at each measured time point). As a result, analyses for these outcomes up to 12 months were limited to 45/117 (38.4%) participants in the Aquablation arm and 19/67 (28.3%) participants in the TURP arm for erectile function and to 79/117 (67.5%) participants in the Aquablation arm and 42/67 (62.6%) participants in the TURP arm for ejaculatory function; therefore we considered risk of attrition bias to be high
Incomplete outcome data (attrition bias) Acute urinary retention	Low risk	Comment: 1/117 (0.8%) in Aquablation and 2/67 (2.9%) in TURP participants were not included in the analysis but missing outcome data balanced in numbers across intervention group. Owing to the small number of participants lost to follow‐up, we considered risk of attrition bias to be low
Incomplete outcome data (attrition bias) Indwelling urinary catheter	Low risk	Comment: the study did not address this outcome but study investigator provided this outcome. 7/117 (5.9%) in Aquablation and 4/67 (5.9%) in TURP participants were not included in the analysis but missing outcome data balanced in numbers across intervention group. Owing to the small number of participants lost to follow‐up, we considered risk of attrition bias to be low
Incomplete outcome data (attrition bias) Hospital stay	Low risk	Comment: 1/117 (0.8%) in Aquablation and 2/67 (2.9%) in TURP participants were not included in the analysis but missing outcome data balanced in numbers across intervention group. Owing to the small number of participants lost to follow‐up, we considered risk of attrition bias to be low
Selective reporting (reporting bias)	Unclear risk	Comment: protocol (NCT02505919) was provided but all secondary outcomes in the articles were not reported in the protocol. Secondary outcomes were not pre‐specified and several quality‐of‐life measurements were not reported (EuroQoL‐5D); therefore we considered risk of reporting bias to be unclear
Other bias	Low risk	Comment: no other sources of bias could be found; therefore we considered risk of other bias to be low

BMI: body mass index; BPH: benign prostate hyperplasia; IIEF: International index of erectile function; IPSS: International prostate symptom score; LUTS: lower urinary tract symptom; MSHQ‐EjD: Male Sexual Health Questionnaire for Ejaculatory Dysfunction; NSAID: nonsteroidal anti‐inflammatory drugs; PSA: prostate specific antigen; PVR: post voided residuals; Q_max: maximum urine flow; QoL: quality of life; TRUS: transrectal ultrasound; TURP: transurethral resection of the prostate