WATER: Aquablation versus Transurethral Resection of the Prostate for BPH
WATER was a double-blind, multicenter noninferiority randomized trial of Aquablation (image-guided robotic waterjet resection) versus transurethral resection of the prostate (TURP) in 181 treated men with moderate-to-severe lower urinary tract symptoms from BPH and prostates of 30 to 80 mL, with primary endpoints of IPSS reduction at 6 months and a 3-month safety composite of procedure-related Clavien-Dindo grade 2 or higher events or grade 1 events causing persistent disability. Aquablation met the noninferiority hypothesis for symptom relief and had fewer persistent grade 1 or grade 2 or higher complications (26% versus 42%), with anejaculation in 10% versus 36% of sexually active men without anejaculation at baseline.[1]
- Design and population
- Multicenter, double-blind noninferiority randomized trial. Men aged 45 to 80 years with moderate-to-severe BPH-related LUTS (IPSS 12 or higher, Qmax below 15 mL/s) and a 30 to 80 mL prostate on transrectal ultrasound; 181 were treated.
- Arms
- Aquablation (AquaBeam robotic waterjet resection) versus transurethral resection of the prostate (TURP)
- Report and follow-up
- 2018 · 6 months for the IPSS endpoint; 3 months for the primary safety endpoint
| Endpoint | Results | Effect | Finding |
|---|---|---|---|
| Mean IPSS reduction from baseline at 6 months (Aquablation minus TURP) | 16.9 points vs 15.1 points | difference 1.8; CI -0.4–4; P <0.0001 (noninferiority) | Met |
| Clavien-Dindo persistent grade 1 or grade 2 or higher procedure-related complications at 3 months | 26% vs 42% | difference -16; P 0.0149 (superiority) | Met |
| Anejaculation among sexually active men | 10% vs 36% | P 0.0003 | Descriptive |
Harms: Persistent grade 1 events at 3 months: 7% vs 25%; Grade 2 or higher events at 3 months: 20% vs 23%; Blood transfusion: 1/116 vs 0/65
Design
Seventeen sites (12 in the United States, 3 in the United Kingdom and 2 in Australia and New Zealand) enrolled men from October 2015 to December 2016. Men were 45 to 80 years old with an IPSS of 12 or higher, a maximum flow rate below 15 mL/s and a transrectal ultrasound volume of 30 to 80 mL. Exclusions included a history of prostate or bladder cancer, neurogenic bladder, urethral stricture, meatal stenosis or bladder neck contracture, stress urinary incontinence, postvoid residual above 300 mL or retention, self-catheterization, prior prostate surgery, anticoagulant or anticholinergic use and severe cardiovascular disease.[2] Of 275 men evaluated, 72 were screen failures and 19 were roll-in cases; 184 were randomized 2:1 (117 to Aquablation and 67 to TURP) and 181 were treated (116 and 65).[2][3] Randomization was stratified by site and baseline IPSS category. Fourteen of the 17 sites had no previous Aquablation experience.[2]
Both primary endpoints were tested for noninferiority first and then for superiority. For the IPSS endpoint, noninferiority required the lower 95% confidence limit of the between-group difference in IPSS reduction to exceed -4.7 points. The safety endpoint counted adverse events judged possibly, probably or definitely procedure-related that were Clavien-Dindo grade 2 or higher, plus any grade 1 event causing persistent disability (ejaculatory or erectile dysfunction or incontinence) through 3 months, adjudicated by an independent clinical events committee; its noninferiority margin was 10 percentage points. The sample of 177 gave 80% power to show superiority on the safety endpoint, assuming event rates of 40% with Aquablation and 65% with TURP.[2]
Patients and follow-up assessors were blinded to treatment for 3 years, with a separate blinded team seeing patients after discharge; all were unblinded after the 3-year visit.[2] At 2 years, 35% of Aquablation patients and 11% of TURP patients guessed that they had received Aquablation (p = 0.0021).[4]
Results
Mean IPSS fell by 16.9 points with Aquablation and 15.1 points with TURP at 6 months, a difference of 1.8 points. The noninferiority hypothesis was met (p < 0.0001); superiority was not reached (p = 0.1346).[1][2] The FDA decision summary for the device gave a 95% confidence interval for the difference of -0.4 to 4.0 points.[5]
The primary safety endpoint occurred in 26% of Aquablation patients and 42% of TURP patients at 3 months; noninferiority was met and superiority followed (p = 0.0149).[1] Persistent grade 1 events occurred in 7% versus 25% (p = 0.0004), the same rates the report gives for procedure-related anejaculation at 3 months, and grade 2 or higher events were similar (20% versus 23%, p = 0.3038).[2] Among sexually active men without anejaculation at baseline, anejaculation was 10% with Aquablation and 36% with TURP (p = 0.0003).[1] Across visits through 5 years, ejaculatory function scores (MSHQ-EjD) averaged 2.7 points worse after TURP (p = 0.0015) and stayed close to baseline after Aquablation; erectile function (IIEF-5) did not change significantly in either group.[2]
At 1 year, IPSS reduction was 15.1 points in both groups, Qmax improved by 10.3 versus 10.6 cc/s (p = 0.8632), and surgical retreatment was 2.6% versus 1.5% (not significant).[6] At 2 years the IPSS reductions were 14.7 and 14.9 points (p = 0.8304) and cumulative surgical retreatment was 4.3% versus 1.5% (p = 0.4219).[4] At 3 years they were 14.4 and 13.9 points (p = 0.6848); 5 of 116 Aquablation patients and 1 of 65 TURP patients had been retreated surgically, none after 20 months.[3]
At 5 years, 58 of 116 Aquablation patients (50%) and 32 of 65 TURP patients (49%) had month-60 follow-up. IPSS reduction was 15.1 versus 13.2 points (p = 0.2764), Qmax improved by 8.7 mL/s (125%) versus 6.3 mL/s (89%), and postvoid residual fell by 62 versus 82 mL (p = 0.3960). Additional BPH therapy, defined as medication or surgery for recurrent symptoms, was needed by 7 of 116 Aquablation patients (6.0%) and 8 of 65 TURP patients (12.3%); by the report's table, 6 Aquablation patients and 1 TURP patient had a surgical procedure.[2] In men with prostates of 50 mL or more, IPSS reduction across all visits was 3.5 points greater after Aquablation (p = 0.0123).[2]
In a subgroup analysis of men with 50 to 80 g prostates, the 6-month IPSS reduction was 4 points greater after Aquablation than after TURP (p = 0.001), the primary safety endpoint occurred in 20% versus 46% (p = 0.008) and anejaculation in 2% versus 41% (p < 0.001). Exploratory analyses showed larger IPSS changes after Aquablation in men with enlarged middle lobes, severe middle lobe obstruction, a low baseline maximum flow rate and a postvoid residual above 100 mL.[7] A separate 5-year report of 62 Aquablation and 34 TURP patients with 50 to 80 mL prostates found IPSS reductions of 14.1 versus 10.8 points across visits (p = 0.02) and additional BPH therapy in 3.2% versus 17.6%.[8]
Surgery and treatment delivery
Aquablation used the AquaBeam Robotic System. A 24 Fr handpiece was placed in the prostatic urethra and locked to a bed-mounted arm; the surgeon marked the resection contour on real-time transrectal ultrasound, and the robot executed the waterjet ablation while avoiding the verumontanum and ejaculatory ducts. Hemostasis used focal non-resective electrocautery or low-pressure inflation of a Foley balloon in the prostatic fossa.[2] In the 50 to 80 mL subgroup, electrocautery was used in 24 patients and balloon inflation in 38.[8] TURP used a monopolar or bipolar loop at the surgeon's preference, followed by continuous bladder irrigation; catheter choice and irrigation duration followed local practice.[2]
Total operative time was 33 minutes with Aquablation and 36 minutes with TURP (p = 0.2752); resection time was 4 versus 27 minutes (p < 0.0001).[1] Hospital stay averaged 1.4 days in both groups and the catheter came out a median of 1 day after surgery in both.[1] One Aquablation patient and no TURP patient needed a blood transfusion, and no late bleeding event required transfusion.[2]
Adverse events
No patient in either arm had de novo erectile dysfunction or incontinence requiring a pad at 3 months.[2] The 1-year report recorded no procedure-related adverse events after month 6.[6] Between years 1 and 2, rates of individual events were low and similar; urinary retention occurred in 1 Aquablation patient (0.9%) and 1 TURP patient (1.5%), and one TURP patient had a stricture or adhesions.[4]
Through 3 years, anejaculation was 11% with Aquablation and 29% with TURP (p = 0.0039), urethral stricture occurred in 1 patient (0.9%) versus 4 (6.2%) (p = 0.0567), and meatal or submeatal stenosis in 3 (2.5%) versus 0 (p = 0.5539).[3] Between years 2 and 3 there were 4 serious adverse events in the Aquablation group (3.4% of patients) and 1 in the TURP group (1.5%); none was attributed to the procedure.[3] Urinary complications from 3 to 5 years did not differ between groups.[2] In the 50 to 80 mL subgroup, grade 3b events occurred in 1 of 62 Aquablation patients (urinary retention) and 2 of 34 TURP patients (bleeding and stricture or adhesions).[8]
Limitations
The sponsor, PROCEPT BioRobotics, funded the trial, and six authors report consulting agreements with it.[2] Prostates above 80 mL were excluded, and the only comparator was monopolar or bipolar TURP.[2] The primary endpoints were measured at 6 and 3 months; later outcomes come from follow-up that was unblinded after year 3, and the 4- and 5-year visits fell during the COVID-19 pandemic, leaving about half of treated patients with month-60 data.[2] Adverse events after month 12 were reported by investigators and not adjudicated by the clinical events committee.[3]
In conversation
- Parallel to GOLIATH (2013). Both are noninferiority randomized trials of a prostate ablation technique against TURP; WATER tested robotic waterjet resection in 30 to 80 mL prostates.
- Parallel to UNBLOCS (2020). Another randomized comparison of an alternative BPH operation with TURP and a blinded patient-reported symptom endpoint; WATER tested robotic waterjet resection.
Same setting St Gallen PAE trial (2018)
The other trials on the BPH procedures timeline compared minimally invasive devices with sham treatment, whereas WATER compared an ablative operation with TURP, so their IPSS changes answer different comparisons.
Guidelines and approvals
- AUA (2026). Clinicians should offer robotic waterjet treatment as an option for patients with prostates of 30 to 80 cc for the treatment of LUTS/BPH . The guideline text describes WATER and its follow-up reports as the randomized evidence.[9]
- EAU (March 2026). Offer Aquablation to patients with moderate-to-severe LUTS/benign prostatic obstruction and a prostate volume of 30 to 80 mL as an alternative to TURP, particularly in patients interested in preserving ejaculatory function . The summary of evidence rates the statement that Aquablation appears as effective as TURP at level 1b.[10]
- FDA (December 21, 2017). The AQUABEAM System was granted De Novo classification (class II, fluid jet system for prostate tissue removal, 21 CFR 876.4350) for the resection and removal of prostate tissue in males with LUTS due to benign prostatic hyperplasia; the decision summary presents WATER as the clinical study supporting safety and effectiveness.[5]
Report history
| Published | Report | Cutoff | Headline results | Citation |
|---|---|---|---|---|
| 2018-01-31 | Primary report (6-month efficacy, 3-month safety) · primary | — | Mean IPSS reduction from baseline at 6 months (Aquablation minus TURP): 16.9 points vs 15.1 points; difference 1.8; CI -0.4–4; P <0.0001 (noninferiority) Met Clavien-Dindo persistent grade 1 or grade 2 or higher procedure-related complications at 3 months: 26% vs 42%; difference -16; P 0.0149 (superiority) Met Anejaculation among sexually active men: 10% vs 36%; P 0.0003 Descriptive | Gilling P, Barber N, Bidair M, et al. "WATER: A Double-Blind, Randomized, Controlled Trial of Aquablation® vs Transurethral Resection of the Prostate in Benign Prostatic Hyperplasia." *J Urol.* 2018;199(5):1252-1261. doi:10.1016/j.juro.2017.12.065 DOI |
| 2018-07-26 | Subgroup analysis (prostate size, middle lobe, severity) · secondary | — | IPSS reduction at 6 months, men with prostates 50 to 80 g (difference, Aquablation minus TURP): Four points greater after Aquablation than after TURP, versus 1.8 points overall (p=0.135). Prespecified and post hoc exploratory subgroup analyses.; difference 4; P 0.001 Persistent grade 1 or grade 2 or higher complications, men with prostates 50 to 80 g: 20% vs 46%; difference -26; P 0.008 Anejaculation among sexually active men with prostates 50 to 80 g: 2% vs 41%; P <0.001 IPSS change by baseline anatomy: Exploratory analysis showed larger IPSS changes after Aquablation in men with enlarged middle lobes, severe middle lobe obstruction, a low baseline maximum flow rate and a postvoid residual above 100 mL.; — | Plante M, Gilling P, Barber N, et al. "Symptom relief and anejaculation after aquablation or transurethral resection of the prostate: subgroup analysis from a blinded randomized trial." *BJU Int.* 2019;123(4):651-660. doi:10.1111/bju.14426 DOI |
| 2018-12-12 | One-year follow-up · updated | — | Mean IPSS reduction at 12 months: 15.1 points vs 15.1 points; — Mean improvement in maximum urinary flow rate at 12 months: 10.3 cc/s vs 10.6 cc/s; P 0.8632 Surgical retreatment for BPH within 1 year: 2.6% vs 1.5%; — | Gilling PJ, Barber N, Bidair M, et al. "Randomized Controlled Trial of Aquablation versus Transurethral Resection of the Prostate in Benign Prostatic Hyperplasia: One-year Outcomes." *Urology.* 2019;125:169-173. doi:10.1016/j.urology.2018.12.002 DOI |
| 2019-04-26 | Two-year follow-up · updated | — | Mean IPSS reduction at 2 years: 14.7 points vs 14.9 points; P 0.8304 Mean improvement in maximum urinary flow rate at 2 years: 11.2 cc/s vs 8.6 cc/s; P 0.1880 Cumulative surgical retreatment for BPH over 2 years: 4.3% vs 1.5%; P 0.4219 | Gilling P, Barber N, Bidair M, et al. "Two-Year Outcomes After Aquablation Compared to TURP: Efficacy and Ejaculatory Improvements Sustained." *Adv Ther.* 2019;36(6):1326-1336. doi:10.1007/s12325-019-00952-3 DOI |
| 2020 | Three-year follow-up · updated | — | Mean IPSS reduction at 3 years: 14.4 points vs 13.9 points; P 0.6848 Mean improvement in maximum urinary flow rate at 3 years: 11.6 cc/s vs 8.2 cc/s; P 0.0848 Cumulative surgical retreatment for BPH through 3 years: 5/116 (4.3%) vs 1/65 (1.5%); P 0.4219 | Gilling P, Barber N, Bidair M, et al. "Three-year outcomes after Aquablation therapy compared to TURP: results from a blinded randomized trial." *Can J Urol.* 2020;27(1):10072-10079. |
| 2022 | Five-year follow-up · updated | — | Mean IPSS reduction at 5 years: 15.1 points vs 13.2 points; P 0.2764 IPSS reduction across all follow-up visits, men with prostates of 50 mL or more: Reduction was 3.5 points greater after Aquablation (repeated measures analysis); no difference in the age or baseline IPSS subgroups.; difference 3.5; P 0.0123 Mean improvement in maximum urinary flow rate at 5 years: 8.7 mL/s (125%) vs 6.3 mL/s (89%); — Mean reduction in postvoid residual at 5 years: 62 mL vs 82 mL; P 0.3960 Additional BPH medication or surgery for recurrent LUTS through 5 years: 7/116 (6.0%) vs 8/65 (12.3%); — | Gilling PJ, Barber N, Bidair M, et al. "Five-year outcomes for Aquablation therapy compared to TURP: results from a double-blind, randomized trial in men with LUTS due to BPH." *Can J Urol.* 2022;29(1):10960-10968. |
| 2023 | Five-year outcomes in 50 to 80 mL prostates · secondary | — | Mean IPSS reduction across 5 years of follow-up, 50 to 80 mL subgroup: 14.1 points vs 10.8 points; P 0.02 Additional BPH medication or surgery through 5 years, 50 to 80 mL subgroup: 3.2% vs 17.6%; — Persistent grade 1 ejaculatory dysfunction, 50 to 80 mL subgroup: 1/62 (1.6%) vs 8/34 (23.5%); — | Oumedjbeur K, Corsi NJ, Bouhadana D, et al. "Aquablation versus TURP: 5-year outcomes of the WATER randomized clinical trial for prostate volumes 50-80 mL." *Can J Urol.* 2023;30(5):11650-11658. |
References
1. Gilling P, Barber N, Bidair M, et al. "WATER: A Double-Blind, Randomized, Controlled Trial of Aquablation® vs Transurethral Resection of the Prostate in Benign Prostatic Hyperplasia." J Urol. 2018;199(5):1252-1261. doi:10.1016/j.juro.2017.12.065
2. Gilling PJ, Barber N, Bidair M, et al. "Five-year outcomes for Aquablation therapy compared to TURP: results from a double-blind, randomized trial in men with LUTS due to BPH." Can J Urol. 2022;29(1):10960-10968. PMID: 35150215
3. Gilling P, Barber N, Bidair M, et al. "Three-year outcomes after Aquablation therapy compared to TURP: results from a blinded randomized trial." Can J Urol. 2020;27(1):10072-10079. PMID: 32065861
4. Gilling P, Barber N, Bidair M, et al. "Two-Year Outcomes After Aquablation Compared to TURP: Efficacy and Ejaculatory Improvements Sustained." Adv Ther. 2019;36(6):1326-1336. doi:10.1007/s12325-019-00952-3
5. U.S. Food and Drug Administration. "De Novo Classification Request for AQUABEAM System: Decision Summary (DEN170024)." Decision date December 21, 2017. accessdata.fda.gov
6. Gilling PJ, Barber N, Bidair M, et al. "Randomized Controlled Trial of Aquablation versus Transurethral Resection of the Prostate in Benign Prostatic Hyperplasia: One-year Outcomes." Urology. 2019;125:169-173. doi:10.1016/j.urology.2018.12.002
7. Plante M, Gilling P, Barber N, et al. "Symptom relief and anejaculation after aquablation or transurethral resection of the prostate: subgroup analysis from a blinded randomized trial." BJU Int. 2019;123(4):651-660. doi:10.1111/bju.14426
8. Oumedjbeur K, Corsi NJ, Bouhadana D, et al. "Aquablation versus TURP: 5-year outcomes of the WATER randomized clinical trial for prostate volumes 50-80 mL." Can J Urol. 2023;30(5):11650-11658. PMID: 37838991
9. Goueli R, Badlani GH, Welliver C, et al. "Management of Lower Urinary Tract Symptoms Attributed to Benign Prostatic Hyperplasia: AUA Guideline (2026) Part III: Procedural/Surgical Management." J Urol. 2026;216(2):161-170. doi:10.1097/JU.0000000000005099. Statement 47.
10. Cornu JN, Elterman D, Hashim H, et al. EAU Guidelines on Non-Neurogenic Male Lower Urinary Tract Symptoms (LUTS). Arnhem: EAU Guidelines Office; update March 2026. Section 5.3.4.a.