BJUI. Study Type Therapy (RCT) Level of Evidence 1b OBJECTIVE

; 2011 Lower Urinary Tract COMARISON OF B-TUR VS M-TUR IN A MULTICENTRE RCT SETTING MAMOULAKIS ET AL BJUI Results. from an international multicentre double-blind randomized controlled trial on the perioperative efficacy and safety of bipolar vs monopolar transurethral resection of the prostate Charalampos Mamoulakis*,, Andreas Skolarikos, Michael Schulze, Cesare M. Scoffone, Jens J. Rassweiler, Gerasimos Alivizatos, Roberto M. Scarpa and Jean J.M.C.H. de la Rosette* *Department of Urology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands, Department of Urology, University Hospital of Heraklion, University of Crete Medical School, Heraklion, Crete, Second Department of Urology, Sismanoglio Hospital, University of Athens Medical School, Athens, Greece, Department of Urology, SLK Kliniken Heilbronn, University of Heidelberg, Heilbronn, Germany, and Department of Urology, San Luigi Hospital, University of Turin, Orbassano, Turin, Italy Accepted for publication 20 January 2011 Study Type Therapy (RCT) Level of Evidence 1b OBJECTIVE What s known on the subject? and What does the study add? Short-term efficacy is similar but B-TUR is preferable due to a more favourable safety. a) first multicentre RCT, b) adequate quality, c) experience with a new bipolar device, d) morbidity standardize using the modified Clavien classification system. To compare the perioperative efficacy and safety of bipolar (B-) and monopolar transurethral resection of the prostate (M- TUR) in an international multicentre double-blind randomized controlled trial using the bipolar system AUTOCON II 400 ESU for the first time. ATIENTS AND METHODS From July 2006 to June 2009, consecutive transurethral resection of the prostate (TUR) candidates with benign prostatic obstruction were prospectively recruited in four academic urological centres, randomized 1:1 into an M-TUR or B-TUR arm and followed up for 6 weeks after surgery. A total of 295 eligible patients were enrolled. Of these, 279 patients received treatment (M-TUR, n = 138; B-TUR, n = 141) and were analysed for immediate postoperative outcomes and perioperative safety. In all, 268 patients (M-TUR, n = 129; B-TUR, n = 139) were analysed for efficacy, which was quantified using changes in maximum urinary flow rate, postvoid residual urine volume and International rostate Symptom Score at 6 weeks compared with baseline. Safety was estimated using sodium and haemoglobin level changes immediately after surgery and perioperative complication occurrence graded according to the modified Clavien classification system. Secondary outcomes included operationresection time, resection rate, capsular perforation and catheterization time. RESULTS No significant differences were detected between each study arm except that postoperative decreases in sodium levels favoured B-TUR ( 0.8 vs 2.5 mmol/l, for B-TUR and M-TUR, respectively; = 0.003). The lowest values were 131 mmol/l (B-TUR) and 106 mmol/l (M-TUR). Nine patients ranged between 125 and 130 mmol/l and the values for three patients were <125 mmol/l after M-TUR. The greatest decrease was 9 mmol/l after B- TUR (two patients). In nine patients (M- TUR) the decrease was between 9 and 34 mmol/l. These results were not translated into a significant difference in TUR-syndrome rates (1/138: 0.7% vs 0/141: 0.0%, for M-TUR and B-TUR, respectively; = 0.495). CONCLUSIONS In contrast to the previous available evidence, no clinical advantage for B-TUR was shown. erioperative efficacy, safety and secondary outcomes were comparable between study arms. The potentially improved safety of B-TUR that is attributed to the elimination of dilutional hyponatraemia risk, a risk still present with M-TUR, did not translate into a significant clinical benefit in experienced hands. KEYWORDS benign prostatic hyperplasia, electrosurgery, randomized controlled trial, transurethral resection of prostate, treatment outcome 240 2011 109, 240 248 doi:10.1111/j.1464-410x.2011.10222.x

COMARISON OF B-TUR VS M-TUR IN A MULTICENTRE RCT SETTING TABLE 1 referred general settings and conditions in each centre (Netherlands) (Germany) (Greece) (Italy) July 2006 February 2009 October 2006 June 2009 January 2007 September 2008 January 2008 May 2009 Recruitment period M-TUR B-TUR M-TUR B-TUR M-TUR B-TUR M-TUR B-TUR Resectoscope (Karl 24 Fr 24 Fr 24F or 26F* 24F or 26F* 26 F 26 F 24 F 24 F Storz) Intermittent flow 47/47 (100) 46/46 (100) 24/33 (72.7) 18/32 (56.3) 0/35 (0.0) 0/38 (0.0) 0/23 (0.0) 0/25 (0.0) patients (%) Continuous flow 0/47 (0.0) 0/46 (0.0) 9/33 (27.3) 14/32 (43.7) 35/35 (100) 38/38 (100) 23/23 (100) 25/25 (100) patients (%) Cutting power 120 W 5 230 W 5 240 W 5 200 W 5 Coagulating power 80 W 5 80 W 5 120 W 5 60 W 5 Irrigation fluid during sorbitol 5% normal saline sorbitol-manitol normal saline water for injection normal saline glycine 1.5% normal saline surgery Type of anaesthesia General (%) 19/47 (40.4) 11/46 (23.9) 31/33 (93.9) 32/32 (100) 2/35 (5.7) 3/38 (7.9) 1/23 (4.3) 0/25 (0.0) Spinal (%) 28/47 (59.6) 35/46 (76.1) 2/33 (6.1) 0/32 (0.0) 33/35 (94.3) 35/38 (92.1) 22/23 (95.7) 25/25 (100) atients with 0/47 (0.0) 0/46 (0.0) 6/33 (18.2) 2/32 (6.3) 0/35 (0.0) 0/38 (0.0) 10/23 (43.5) 12/25 (48.0) suprapubic trocar Bladder catheter (three-way) 20 Fr 20 Fr 22 Fr 22 Fr 20 Fr 20 Fr 20 Fr 20 Fr *The sizes of the intermittent flow and the continuous flow resectoscope were 24 F and 26 F, respectively. roportions do not differ significantly between arms in the total populaiton. urisole (Fresenius-Kabi AG, Bad Homburg): sorbitol 27%-mannitol 5.4%. INTRODUCTION Monopolar TUR (M-TUR) is considered to be the surgical gold standard for benign prostatic obstruction (BO), mainly because of its well documented long-term efficacy [1]. In the absence of strong, high-quality evidence favouring newer surgical methods [2,3], M- TUR remains clinically and cost-effective [4 6]. Although significant technical improvements during the past decades have reduced negative outcomes of the procedure, there are still concerns regarding complications such as TUR-syndrome, bleeding and urethral strictures [7,8]. In a large-scale, multicentre, observational study it was shown that TUR mortality and morbidity have anadays decreased but morbidity remains high (11.1%) [9]. The incorporation of bipolar technology has been the most significant modification to TUR; B-TUR addresses a fundamental flaw of M-TUR in that it can be performed using normal saline irrigation and has had promising results [10]. The bipolar systems Gyrus (ACMI Southborough, MA, USA), TURis (Olympus, Tokyo, Japan), S(a)line Resectoscope (Richard Wolf GmbH, Knittlingen, Germany), Tekno Tom (Tekno- Medical, Tuttlingen, Germany) and AUTOCON II 400 ESU (Karl Storz Endoscope, Tuttlingen, Germany) adhere to distinct electrophysiological principles regarding current flow [11,12]. Consequently, efficacy and, principally, safety should be cautiously and separately evaluated for each system [13]. We compared the perioperative efficacy and safety of B-TUR with that of M-TUR in an international multicentre randomized controlled trial (RCT) setting using the AUTOCON II 400 ESU, for the first time, with the aim of fulfilling the Cochrane collaboration criteria for RCT quality assessment [14]. ATIENTS AND METHODS STUDY DESIGN From July 2006 to June 2009, consecutive TUR candidates with BO were prospectively recruited in four academic urological centres (: Academic Medical Centre, University of Amsterdam, Amsterdam, the Netherlands; : SLK Kliniken Heilbronn, University of Heidelberg, Heilbronn, Germany; : Sismanoglio Hospital, University of Athens Medical School, Athens, Greece; : San Luigi Hospital, University of Turin, Orbassano, Turin, Italy). The trial was initially intended to accrue patients in Centre 1 (Netherlands Trial Register, Registration No. NTR703). Centres 2 4 subsequently agreed to follow the same protocol in order to increase the number of cases, to obtain insights into differences among surgeons and to guarantee longer-term follow-up data. The trial was conducted according to the Declaration of Helsinki principles in compliance with the approved protocol by an independent local medical ethics committee at each centre, Good Clinical ractice guidelines and applicable regulatory requirements. ATIENTS AND STUDY ROTOCOL atients who were able to void spontaneously (maximum urinary flow rate [Q max ] 15 ml/s, International rostate Symptom Score [ISS] 13, voided volume 125 ml) and patients on indwelling or intermittent catheterization [IC] were eligible. Exclusion criteria included: suspected malignant disease of the lower urinary tract including prostate cancer; impaired detrusor function (neurogenic bladder, postvoid residual urine volume [VR] >400 ml); active urinary tract infection [UTI]; 2011 241

MAMOULAKIS ET AL. 5-α reductase inhibitor or α-blocker consumption within the last 3 months or 2 weeks before surgery, respectively; previous prostate surgery; cardiovascular disease necessitating anticoagulation continuation that might jeopardize outcome; and immunosuppression. The baseline evaluation included medical history, digital rectal examination, selfreported ISS, prostate volume estimation on transrectal ultrasound [TRUS], uroflowmetry, full blood count, prostate specific antigen [SA], creatinine, electrolytes, urine analysis/ urine culture. Efficacy and safety (primary outcomes) were compared between arms. Efficacy was quantified by Q max, VR and ISS improvement at 6 weeks after surgery. Safety was estimated by changes in sodium and haemoglobin levels immediately after surgery and perioperative complications were graded according to the modified Clavien classification system [15] as previously proposed for TUR procedures [16]. Secondary outcomes included operationresection time, resection rate (resected tissue weight divided by resection time), capsular perforation and catheterization time. Sample size determination was based on sodium level changes immediately after surgery (main outcome). Using a one-sided t-test, setting the probability of type-i error (α) at 0.050, type-ii error (β) at 0.100 and based on previous studies reporting on immediate postoperative sodium level decreases [17], the sample size in was calculated to be 47 patients per arm for detecting a difference in means of 1.6 ± 3.3 mmol/l (S: power and sample size calculation 2.1.31). We opted for a one-sided test because B-TUR is intended to reduce the decline in sodium levels by allowing performance in normal saline. A total of 295 patients were enrolled and randomized 1:1 into an M-TUR or B-TUR arm after written informed consent was obtained. Data were entered by one investigator per centre through a trial webbased central electronic system (http:// www.turp.nl). Randomization numbers were assigned using a stratified permuted computer algorithm after completing eligibility screening. Stratification factors included prostate volume and age. Randomization was performed within 24 h prior to treatment application, which was done blindly among centres through the TABLE 2 atient distribution and baseline patient characteristics M-TUR B-TUR atients randomized 149 146 A. atients analysed (B+C): 138 141 Age, years 68.9 (8.7) 69.4 (8.4) 0.655 Body mass index, kg/m 2 26.2 (3.1) 26.0 (3.1) 0.578 rostate volume, ml 63.2 (30.2) 64.0 (29.4) 0.765 SA, ng/ml 5.3 (7.7) 5.7 (7.9) 0.663 Creatinine, μmol/l 112.9 (96.9) 94.7 (31.3) 0.042 Serum sodium, mmol/l 139.9 (3.6) 140.6 (3.3) 0.142 Haemoglobin, mmol/l 8.8 (0.9) 8.9 (0.8) 0.470 B. atients on catheterization 63 (45.7) 50 (35.5) 0.107 preoperatively (%): Age, years 70.4 (9.5) 72.2 (8.6) 0.297 Body mass index, kg/m 2 26.2 (3.3) 25.3 (2.5) 0.088 rostate volume, ml 69.5 (37.2) 72.6 (35.7) 0.635 SA, ng/ml 7.2 (10.5) 9.0 (11.6) 0.411 Creatinine, μmol/l 130.6 (121.6) 96.7 (32.0) 0.067 C. atients not on catheterization 75 (54.3) 91 (64.5) 0.107 preoperatively (%): Age, years 67.7 (7.9) 67.9 (7.9) 0.884 Body mass index, kg/m 2 26.2 (3.0) 26.4 (3.3) 0.659 rostate volume, ml 57.9 (21.6) 59.3 (24.3) 0.748 SA, ng/ml 3.8 (3.9) 3.9 (4.0) 0.882 Creatinine, μmol/l 98.5 (68.5) 93.6 (30.9) 0.534 Q max, ml/s* 8.7 (2.3) 9.2 (3.3) 0.344 VR, ml* 91.6 (70.4) 88.1 (67.9) 0.739 ISS* 23.0 (5.2) 23.2 (5.0) 0.787 QoL score* 4.2 (1.1) 4.3 (1.1) 0.491 *Applicable only in patients not on catheterization before surgery. Data are presented as mean (SD). central electronic system to minimize potential selection bias and guarantee allocation concealment. The nature of the intervention made blinding of treatment providers impossible and thus surgeons were not blinded. However, the outcome assessors were different from the surgeons. Outcome assessors and patients were both blinded for the intervention type (double-blind RCT). ostoperative bladder irrigation with normal saline was applied until efflux was sufficiently clear. One urologist per centre (the outcome assessor) made the decision on catheter removal. Catheters were normally removed within 24 h after irrigation was stopped. SURGICAL ROCEDURES All procedures were performed using the AUTOCON II 400 ESU radiofrequency generator, compatible resectoscopes and disposables (Karl Storz Endoscope, Tuttlingen, Germany). AUTOCON II 400 ESU, introduced in 2007, provides an automode applying only the minimum amount of high frequency current together with voltage regulation. It has a liquid crystal touchscreen offering a multitude of preprogrammed monopolar and bipolar modes. The compatible bipolar resectoscope comes with various sized loops for single or multiple uses. Bipolar functionality is achieved through the special resectoscope mirrored dual-loop distal tip design consisting of an active (resection loop) and passive electrode opposed. The system is considered to be truly bipolar as it meets the 1998 International Electrotechnical Commission definition (http://www.iec.ch) that requires both electrodes to be attached to a single-port system. The resection technique used was at the investigator s discretion in each centre, but there was no difference between study arms within centres. Both procedures were performed using a 24/26-F bipolar or 242 2011

COMARISON OF B-TUR VS M-TUR IN A MULTICENTRE RCT SETTING FIG. 1. Flow diagram of the phases of the present RCT (: Academic Medical Centre, University of Amsterdam, Amsterdam, the Netherlands; : SLK Kliniken Heilbronn, University of Heidelberg, Heilbronn, Germany; : Sismanoglio Hospital, University of Athens Medical School, Athens, Greece; : San Luigi Hospital, University of Turin, Orbassano, Turin, Italy). TUI, transurethral incision of the prostate. n = 48 Allocated to M-TUR (n = 149) Did not receive allocated intervention (n = 10) - Operation not according to protocol (B-TUR, TUI; n = 2) - Not operated (anaesthesiological reasons; n = 7) - Not operated (withdrew consent; n = 1) Received allocated intervention (n = 139) n = 47 n = 35 n = 34 n = 40 n = 35 Discontinued intervention (n = 0) Analysed immediately after surgery (n = 138) (Excluded (n = 1) owing to bladder cancer; centre) Withdrew consent after the operation (n = 3) Death from myocardial infarction (n = 2) (9 days and 5 weeks postoperatively) Discontinued owing to complication (n = 2) - Ischaemic colitis on day 2 after surgery - ulmonary embolism on day 9 after surgery Failure (Re-operation, permanent catheter; n = 2) n = 45 n = 27 n = 34 Analysed for efficacy at 6 weeks (n = 129) monopolar resectoscope of continuous or intermittent flow. For B-TUR, a 24-F bipolar cutting loop (catalogue code: 27040 GV) was used, the preferred preprogrammed cutting and coagulating effect was set to level 5 and the bladder was irrigated with normal saline intraoperatively. For M-TUR, a 24/26-F monopolar cutting loop (diameter 0.8 mm, delivered with a wire diameter 0.35 mm) was used, cutting and coagulating power across centres was 120 240 W and 60 120 W, respectively and the bladder was irrigated with hypotonic solution intraoperatively. All operations were performed under general or spinal anaesthesia by a senior urological surgeon in each centre. atients were treated similarly apart from the surgical technique used. They received antibiotics and discontinued anticoagulants perioperatively Assessed for eligibility (n = 295) n = 26 n = 23 n = 23 Enrolment Randomization Allocation Analysis 1 Follow-Up Analysis 2 n = 46 n = 46 n = 45 Excluded (n = 0) according to the protocol of each centre. A three-way Foley catheter was used for bladder irrigation with normal saline after surgery in all cases. The preferred general settings and conditions used in each centre are summarized in Table 1. STATISTICAL ANALYSIS Not meeting inclusion criteria (n = 0) Refused to participate (n = 0) Other reasons (n = 0) n = 35 n = 40 Allocated to B-TUR (n = 146) Did not receive allocated intervention (n = 5) - Operation not according to protocol (Optical urethrotomy; n = 1) - Not operated (anaesthesiological reasons; n = 3) - Not operated (withdrew consent; n = 1) Received allocated intervention (n = 141) n = 32 n = 38 Discontinued intervention (n = 0) n = 25 n = 25 Analysed immediately after surgery (n = 141) (Excluded (n = 0)) Withdrew consent after the operation (n = 0) Death after the operation (n = 0) Discontinued owing to complication (n = 1) - decompensation of cerebral microangiopathy (dementia) on day 5 after surgery Failure (Re-operation; n = 1) n = 31 n = 38 n = 25 Analysed for efficacy at 6 weeks (n = 139) All variables were treated as continuous (including ISS because of its reasonable amount of categories; 0 35) apart from quality of life (QoL) score (ordinal) and those addressing complications (categorical). Continuous variables were tested for normality and equality of variances using the Shapiro Wilk W-test and the Levene F-test, respectively. Baseline patient characteristics (age, body mass index, prostate volume, SA and creatinine levels) were tested using univariate ANOVA models including a fourlevel nominal variable indicating the centre of data origin to adjust for potential differences among centres. Continuous outcome variables were tested in univariate or repeated measure ANOVA models (for single-measure or repeated measure outcome analysis, respectively) adjusted for centre and baseline characteristics that were found to differ significantly between arms (if any). QoL score was analysed using the Mann Whitney U-test or Wilcoxon test for single-measure or repeated measure outcome analysis, respectively. Categorical variables were analysed using the earson chi-squared test with continuity correction or Fisher s exact test. Descriptive statistics were presented as mean (SD), irrespective of the variable distribution, for uniformity purposes. Data were analysed using the Statistical ackage for Social Sciences (SSS 15.0 for Windows, Chicago, IL, USA) with a two-tailed value 0.050 considered to indicate statistical significance. RESULTS The flow diagram showing the RCT phases is shown in Fig. 1. Assessment of outcomes was based on >80% of randomized patients. All patients receiving the allocated intervention were included in the analysis except one, who was excluded in retrospect owing to a suspicion of bladder cancer invading the prostate on preresection cystoscopy, which was verified on biopsy. ost-treatment withdrawal rates did not differ significantly (9/138: 6.5% vs 2/141: 1.4%, for M-TUR and B-TUR, respectively; = 0.060; Fig. 1). Baseline patient characteristics were comparable between study arms (Table 2). Statistical and/or clinical efficacy was shown for each surgical method with no significant differences between arms (Tables 3,4). No differences were detected in safety or secondary outcomes except that the drop in sodium level was significantly greater after M-TUR ( 2.5 vs 0.8 mmol/l; = 0.003; Table 5). The lowest values were 131 and 106 mmol/l (patient with TUR-syndrome) after B-TUR and M-TUR, respectively. The sodium values in nine patients were 125 130 mmol/l and in three patients the value was <125 mmol/l after M-TUR. Sodium levels decreased by the greatest amount, 9 mmol/l, in two patients after B-TUR. In nine patients in the M-TUR arm the decrease 2011 243

MAMOULAKIS ET AL. was 9 34 mmol/l. These results were not translated into a significant difference in TUR-syndrome rates (1/138: 0.7% vs 0/141: 0.0%, for M-TUR and B-TUR, respectively; = 0.495; Table 6), TUR-syndrome being defined as sodium levels <125 mmol/l and one or more circulatory and/or neurological symptoms. Haemoglobin levels decreased significantly in each study arm but, the decrease was similar between arms. Fifty-five complications were recorded in 44 patients (morbidity: 44/279 patients; 15.8%). Two complications in the M-TUR arm were fatal (myocardial infarction) vs no complication in the B-TUR arm (2/138 vs 0/141 patients, = 0.244; overall mortality was 2/279 patients; 0.7%). The morbidity did not differ significantly between arms (26/138 vs 18/141, for M-TUR and B-TUR, respectively; = 0.220). Specific complication rates and their grading based on the modified Clavien classification system are shown in Tables 6 and 7. No differences were detected in any specific complication rate or Clavien grade between arms. Three patients presented acute urine retention (AUR) after catheter removal and finally failed to void. Two of them were re-operated for residual adenoma and one was placed permanently on IC. These cases were considered treatment failures rather than complications based on the modified Clavien classification system (Table 7). DISCUSSION Although B-TUR was introduced several years ago, its true merits over M-TUR remain unclear [18] and the vivid interest of the urological community. The high level of interest in B- and M-TUR is reflected in the number of RCTs that appeared in the literature in just one year [19 25] after the publication of meta-analysis by Mamoulakis et al. [17]. This RCT-based profound metaanalysis, which compared the efficacy and safety of the two techniques [17], was the largest one compared to those previously published [2 5,26] and provided evidence for the first time that no clinically relevant differences exist between the techniques in short-term efficacy and complication rates (AUR after catheter removal, transfusions or urethral strictures) but that B-TUR was preferable owing to a more favourable safety profile defined by clinically relevant differences regarding TUR-syndrome and clot retention rates. In addition, this study [17] pointed out the lack of well designed TABLE 3 Efficacy at 6 weeks after surgery in patients on catheterization before surgery M-TUR B-TUR Number of patients 54 49 Q max, ml/s 20.0 (9.2) 16.6 (8.2) 0.073 VR, ml 40.4 (39.6) 40.1 (70.8) 0.998 ISS 10.0 (6.1) 9.9 (5.7) 0.930 QoL score 1.5 (1.3) 2.1 (1.6) 0.104 Data are presented as mean (SD). TABLE 4 Efficacy at 6 weeks after surgery in patients not on catheterization before surgery M-TUR B-TUR Number of patients 75 90 Q max, ml/s reoperatively 8.7 (2.3) 9.2 (3.3) 0.526 ostoperatively 19.9 (11.0) 19.3 (7.0) <0.001 <0.001 VR, ml reoperatively 91.6 (70.4) 88.1 (67.9) 0.852 ostoperatively 29.6 (39.6) 23.5 (27.2) <0.001 <0.001 ISS reoperatively 23.0 (5.2) 23.2 (5.0) 0.675 ostoperatively 9.5 (5.1) 9.4 (4.4) <0.001 <0.001 HRQL score reoperatively 4.2 (1.1) 4.3 (1.1) 0.886 ostoperatively 1.5 (1.3) 1.7 (1.3) <0.001 <0.001 Data are presented as mean (SD). international/multicentre RCTs and the low methodological quality of the existing trials as assessed by the use of the Dutch Cochrane Collaboration checklist, which is in accordance with Table 8.5.a of the Cochrane Handbook for Systematic Reviews of Interventions [14]. The present study is the first to be based on data derived from an international, multicentre RCT in the field and its methods fulfil the suggested quality criteria [14]. Futher more it is the first reported clinical experience of using the AUTOCON II 400 ESU. Operators were easily accustomed to the dual-loop tip design of the resectoscope, coagulation was effective and visibility throughout the procedure excellent. Cutting was sharp and effortless, occurring smoothly when the orange glow of activated plasma surrounded the resection loop after a very short initiation delay on tissue contact. The perioperative efficacy, safety and secondary outcomes in the present study did not differ significantly between arms. Comparable results on efficacy parameters are in accordance with those previously published [17]. The present trial is the first to standardize morbidity using the modified Clavien classification system. No significant differences were detected in safety apart from in postoperative decline in sodium levels. Sodium levels decreased significantly in both arms, as previously reported in a large RCT focusing on dilutional hyponatraemia [22]. Nevertheless, the decline was significantly greater after M-TUR, in line with previously published results [17,22]. In accordance with all individual RCTs these results were not translated into significant differences in TURsyndrome rates [17,22,24,25]. Nevertheless, 244 2011

COMARISON OF B-TUR VS M-TUR IN A MULTICENTRE RCT SETTING TABLE 5 Immediate-postoperative serum sodium changes, haemoglobin changes and secondary outcomes M-TUR B-TUR Number of patients 138 141 Serum sodium, mmol/l reoperatively 139.9 (3.6) 140.6 (3.3) ostoperatively 137.4 (5.4) 139.8 (2.6) Change 2.5 (5.3) 0.8 (3.2) 0.003 <0.001 0.002 Haemoglobin, mmol/l reoperatively 8.8 (0.9) 8.9 (0.8) ostoperatively 7.9 (1.0) 8.1 (0.9) Change 0.9 (0.8) 0.8 (0.7) 0.548 <0.001 <0.001 Operation time, min 50.8 (18.8) 52.0 (21.7) 0.702 Resection time, min 38.9 (16.8) 39.7 (19.0) 0.822 Resection rate, ml/min 0.8 (0.4) 0.7 (0.3) 0.081 Capsular perforation, n (%) 32 (23.2) 31 (22.1) 0.948 Minor 28 29 0.672 Severe 4 2 Catheterization time, days 3.0 (1.2) 3.1 (1.2) 0.436 Data are presented as mean (SD). TABLE 6 Complications during the perioperative period M-TUR B-TUR Number of patients 138 141 Febrile UTI (%) 2 (1.4) 1 (0.7) 0.620 Clot retention (%) 7 (5.1) 9 (6.4) 0.831 Blood transfusion (%) 4 (2.9) 9 (6.4) 0.273 TUR-syndrome (%) 1 (0.7) 0 (0.0) 0.495 AUR (%) 8 (5.8) 5 (3.5) 0.543 Urethral strictures (%) 5 (3.6) 2 (1.4) 0.279 Other (%)* 4 (2.9) 1 (0.7) Total number of complications 31 27 *Cerebral microangiopathy decompensation (dementia), ischemic colitis, pulmonary embolism, myocardial infarction (death) (see Table 7). Based on the modified Clavien classification system the total number of complications is M-TUR: n = 29, B-TUR: n = 26. This is because three cases with AUR were considered failures rather than complications (patients failed to void finally; M-TUR: n = 2, B- TUR: n = 1; see Table 7). they show that while B-TUR may not prevent fluid absorption, it eliminates the risk of dilutional hyponatraemia, which is significantly increased with M-TUR [17]. Therefore, bipolar technology fulfils the expectation to provide more time for larger resection, coagulation and training without compromising safety. Although the clinical significance of this protective effect might not be considered such an important issue in experienced hands, B-TUR might be preferable, especially in less experienced settings. A major M-TUR complication is bleeding, which is clinically relevant mainly if it results in clot retention, transfusion or reoperation. The haemostatic capacity of the bipolar current has been reported to be superior in ex vivo studies, possibly because of deeper coagulation depths and the cut-and-seal effect [10]. The present RCT failed to show differences in bleeding tendency between arms. Haemoglobin levels decreased similarly and clot retention or transfusion rates did not differ significantly. These data are in accordance with the general clinical impression [18] and most individual RCT results on haemoglobin decline [17,22], clot retention [17,21] or transfusion rates [17,20,21,24]. Nevertheless, there are still contrary reports on haemoglobin decline [20,23,24], clot retention [27] and transfusion rate [23], favouring B-TUR. In addition, a previously general pooled and subgroup analysis (using the plasmakinetic device) of RCT results on clot retention and transfusion rates showed a significant difference favouring B-TUR [17]. Catheterization duration has been reported to be significantly shorter for B-TUR, in a pooled, sensitivity analysis of double-blind RCTs [17]. This result was not verified in the present RCT. Consistent with all previous RCTs [17,19 21,24], no significant differences were detected in any other complications, such as AUR after catheter removal and urethral strictures. Thus, as previously noted [17], the serious concerns raised about bipolar technology regarding urethral strictures [28,29] are not justified. Nevertheless, accrual of increased numbers of patients and/or longer follow-up, which is still limited [19,20,30] might change these results [17]. The short follow-up period is a potential limitation of the present study, however, our aim was to focus mainly on the perioperative results, for which a longer follow-up period is redundant. In conclusion, the present study provides a valuable addition to the international literature in that it is based on several trial characteristics that were not included in previously published RCTs in the field. These may be summarized as follows: a) it is the first international, multicentre RCT; b) its methodology fulfils the suggested quality criteria; c) it presents data on the first clinical experience with a relatively new bipolar device; d) it attempts to standardize perioperative morbidity using the modified Clavien classification system. In contrast to the previous available evidence, the present trial shared no clinical advantage for B-TUR. erioperative efficacy, safety and secondary outcomes were shown to be comparable in both surgical techniques, while the potentially improved safety of B-TUR 2011 245

MAMOULAKIS ET AL. TABLE 7 Complications during the postoperative period classified according to the modified Clavien system atient ID* Number of complications Grade Centre M-TUR B-TUR M-TUR B-TUR (%) Complication Management Netherlands 41 5, 44 1 2 I Clot retention (n = 8) Bedside clot evacuation via Germany 4, 37 61 2 1 catheter ± catheter Italy 7 28 1 1 change ± bladder irrigation Netherlands 49 11, 40 1 2 AUR (n = 10) Bedside recatheterization Germany 6, 8, 36 22 3 1 (spontaneous voiding Italy 2, 14 28 2 1 after catheter trial) Greece 58 80 1 1 Bulbar urethral stricture (n = 2) Dilation Grade I rate (%) 11/37.9 9/26 (34.6) 0.799 20/55 (36.4) Netherllands 44, 50 0 2 II Haematuria (n = 13) Transfusion Germany 35, 50, 70 13, 18, 42, 48, 62 3 5 Greece 52 75 1 1 Italy 25 0 1 Netherlands 73 0 1 Febrile UTI (n = 3) Antibiotics Germany 34 1 0 Italy 34 1 0 Germany 13 0 1 Cerebral microangiopathy decompensation (dementia) (n = 1) harmacological treatment Grade II rate (%) 6/29 (20.7) 11/26 (42.3) 0.083 17/55 (30.9) Netherlands 27, 63 50, 92 2 2 III Clot retention (n = 8) Endoscopic evacuation ± Germany 70 42, 62 1 2 revision Italy 25 0 1 Greece 2, 28, 32, 75 4 1 Bulbar urethral Optical urethrotomy 73 stricture (n = 5) Germany 25 1 0 Ischaemic colitis (n = 1) Bowel surgery (colostoma) Grade III rate (%) 8/29 (27.6) 6/26 (23.1) 0.702 14/55 (25.5) Germany 50 1 0 IV TUR-syndrome (n = 1) Germany 70 1 0 ulmonary embolism (n = 1) Admission to ICU Admission to ICU Grade IV rate (%) 2/29 (6.9) 0/26 (0.0) 0.492 2/55 (3.6) Netherlands 39 1 0 V Death due to Greece 56 1 0 myocardial infarction (n = 2) Grade V rate (%) 2/30 (6.7) 0/25 (0.0) 0.495 2/55 (3.6) Grand Total 29 26 55 246 2011

COMARISON OF B-TUR VS M-TUR IN A MULTICENTRE RCT SETTING TABLE 7 Continued atients that failed to void Netherlands 35 45 1 1 Treatment AUR (n = 3) Re-operation or permanent Germany 45 1 0 Failure IC (failure to void after cathter trial) Failure rate (%) 2/138 (1.4) 1/141 (0.7) 3/279 (1.1) 0.620 *The distribution of patients with at least one complication (n = 44) across grades (when highest grade counted) between arms does not differ significantly ( = 0589); Discovered at 6 weeks postoperatively; atient withdrawn from the study; ICU = intensive care unit. that is attributable to the elimination of dilutional hyponatraemia risk, which is still present with M-TUR, did not translate into a significant clinical benefit in experienced hands. ACKNOWLEDGEMENTS We acknowledge the support of Karl Storz Endoscope, Tuttlingen, Germany in providing all the equipment used in this trial. We would like to thank Mrs S.A. Lagerveld-Zaaijer and Mrs A.D. Bakker, trial nurses at the Trial Centre of Urology, Academic Medical Centre, University of Amsterdam, Amsterdam, the Netherlands for administrative and technical support. CONFLICT OF INTEREST None declared. REFERENCES 1 Reich O, Gratzke C, Stief CG. Techniques and long-term results of surgical procedures for BH. Eur Urol 2006; 49: 970 8 2 Burke N, Whelan J, Goeree L et al. Systematic review and meta-analysis of transurethral resection of the prostate versus minimally invasive procedures for the treatment of benign prostatic obstruction. Urology 2010; 75: 1015 22 3 Ahyai SA, Gilling Kaplan SA et al. Meta-analysis of functional outcomes and complications following transurethral procedures for lower urinary tract symptoms resulting from benign prostatic enlargement. Eur Urol 2010; 58: 384 97 4 Lourenco T, ickard R, Vale L et al. Alternative approaches to endoscopic ablation for benign enlargement of the prostate: systematic review of randomised controlled trials. BMJ 2008; 337: a449 5 Lourenco T, Armstrong N, N Dow J et al. Systematic review and economic modelling of effectiveness and cost utility of surgical treatments for men with benign prostatic enlargement. Health Technol Assess 2008; 12: 1 146, 169 515 6 Lourenco T, ickard R, Vale L et al. Minimally invasive treatments for benign prostatic enlargement: systematic review of randomised controlled trials. BMJ 2008; 337: a1662 7 Rassweiler J, Teber D, Kuntz R, Hofmann R. Complications of transurethral resection of the prostate (TUR)-incidence, management, and prevention. Eur Urol 2006; 50: 969 79 8 Hawary A, Mukhtar K, Sinclair A, earce I. Transurethral resection of the prostate syndrome: almost gone but not forgotten. J Endourol 2009; 23: 2013 20 9 Reich O, Gratzke C, Bachmann A et al. Morbidity, mortality and early outcome of transurethral resection of the prostate: a prospective multicenter evaluation of 10,654 patients. J Urol 2008; 180: 246 9 10 Mamoulakis C, Trompetter M, de la Rosette J. Bipolar transurethral resection of the prostate: the golden standard reclaims its leading position. Curr Opin Urol 2009; 19: 26 32 11 Rassweiler J, Schulze M, Stock C, Teber D, De La Rosette J. Bipolar transurethral resection of the prostate-technical modifications and early clinical experience. Minim Invasive Ther Allied Technol 2007; 16: 11 21 12 Issa MM. Technological advances in transurethral resection of the prostate: bipolar versus monopolar TUR. J Endourol 2008; 22: 1587 95 13 Faul, Schlenker B, Gratzke C, Stief CG, Reich O, Hahn RG. Clinical and technical aspects of bipolar transurethral prostate resection. Scand J Urol Nephrol 2008; 42: 318 23 14 Higgins JT, Altman DG. Steme JAC on behalf of the Cochrane Statistical Methods Group and the Cochrane Bias Methods Group. In Higgins JT, Green S eds, Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.D. Available at: http://www.cochranehandbook.org/. Updated March 2011. The Cochrane Collaboration 2011. 15 Dindo D, Demartines N, Clavien A. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205 13 16 Mamoulakis C, Efthimiou I, Kazoulis S, Christoulakis I, Sofras F. The modified Clavien classification system: a standardized platform for reporting complications in transurethral resection of the prostate. World J Urol 2011; 29: 205 10 17 Mamoulakis C, Ubbink DT, de la Rosette JJMCH. Bipolar versus monopolar transurethral resection of the prostate: a systematic review and meta-analysis of randomized controlled trials. Eur Urol 2009; 56: 798 809 18 Reich O. Bipolar transurethral resection of the prostate: what did we learn, and where do we go from here? Eur Urol 2009; 56: 796 7 19 Michielsen D, Coomans D. Urethral strictures and bipolar transurethral resection in saline of the prostate: fact or fiction? J Endourol 2010; 24: 1333 7 2011 247

MAMOULAKIS ET AL. 20 Chen Q, Zhang L, Fan QL, Zhou J, eng YB, Wang Z. Bipolar transurethral resection in saline vs traditional monopolar resection of the prostate: results of a randomized trial with a 2-year follow-up. BJU Int 2010; 106: 1339 43 21 Singhania, Nandini D, Sarita F, Hemant, Hemalata I. Transurethral resection of prostate: a comparison of standard monopolar versus bipolar saline resection. Int Braz J Urol 2010; 36: 183 9 22 Michielsen D, Coomans D, Braeckman JG, Umbrain V. Bipolar transurethral resection in saline: the solution to avoid hyponatraemia and transurethral resection syndrome. Scand J Urol Nephrol 2010; 44: 228 35 23 Fagerström T, Nyman CR, Hahn RG. Bipolar transurethral resection of the prostate causes less bleeding than the monopolar technique: a single-centre randomized trial of 202 patients. BJU Int 2010; 105: 1560 4 24 Kong CH, Ibrahim MF, Zainuddin ZM. A prospective, randomized clinical trial comparing bipolar plasma kinetic resection of the prostate versus conventional monopolar transurethral resection of the prostate in the treatment of benign prostatic hyperplasia. Ann Saudi Med 2009; 29: 429 32 25 Chen Q, Zhang L, Liu YJ, Lu JD, Wang GM. Bipolar transurethral resection in saline system versus traditional monopolar resection system in treating large-volume benign prostatic hyperplasia. Urol Int 2009; 83: 55 9 26 Mamoulakis C, de la Rosette JJ, Ubbink DT. Re: Burke et al.: systematic review and meta-analysis of transurethral resection of the prostate versus minimally invasive procedures for the treatment of benign prostatic obstruction. Urology 2010; 75: 1235 6 27 Erturhan S, Erbagci A, Seckiner I, Yagci F, Ustun A. lasmakinetic resection of the prostate versus standard transurethral resection of the prostate: a prospective randomized trial with 1-year follow-up. rostate Cancer rostatic Dis 2007; 10: 97 100 28 Tefekli A, Muslumanoglu AY, Baykal M, Binbay M, Tas A, Altunrende F. A hybrid technique using bipolar energy in transurethral prostate surgery: a prospective, randomized comparison. J Urol 2005; 174: 1339 43 29 Reich O. Editorial comment on: a prospective randomized study comparing monopolar and bipolar transurethral resection of prostate using transurethral resection in saline (TURIS) system. Eur Urol 2007; 52: 523 4 30 Autorino R, Damiano R, Di Lorenzo G et al. Four-year outcome of a prospective randomised trial comparing bipolar plasmakinetic and monopolar transurethral resection of the prostate. Eur Urol 2009; 55: 922 9 Correspondence: Charalampos Mamoulakis, Department of Urology, University Hospital of Heraklion, O Box 1031, 71001, Heraklion, Crete, Greece. e-mail: c.mamoulakis@amc.uva.nl Abbreviations: B-TUR, bipolar TUR; M- TUR, monopolar TUR; RCT, randomized controlled trial; Q max, maximum urinary flow rate; VR, postvoid residual urine volume; HRQL, health-related quality of life; AUR, acute urinary retention. 248 2011