Suburethral slingplasty evaluation study in North Queensland, Australia: The SUSPEND trial

Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59 Blackwell Publishing, Ltd. Original Article The SUSPEND trial Suburethral slingplasty evaluation study in North Queensland, Australia: The SUSPEND trial Yik Nyok LIM, 1 Reinhold MULLER, 2 Audrey CORSTIAANS, 3 Hans P. DIETZ, 4 Christopher BARRY 1 and Ajay RANE 1 1 James Cook University, 2 School of Public Health, James Cook University, 3 Townsville Hospital, Queensland and 4 Royal Prince Alfred Hospital, Sydney, Australia Abstract Objective: To compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence. Methods: Following ethics approval, 195 (3 65) patients with urodynamic stress incontinence were randomly assigned to undergo suburethral slingplasty with the Tension-free Vaginal Tape (TVT; Gynecare, Ethicon, Somerville, NJ, USA), Intravaginal Sling (IVS; Tyco Healthcare, Mansfield, MA, USA), or Suprapubic Arc Sling (SPARC; American Medical Systems, Minnetonka, MN, USA). The patients were blinded to the type of sling implanted. Main outcome measures were: (1) operative and short-term complications; (2) pre- and postoperative symptomatology; and (3) pre- and postoperative urodynamic findings. Results: There was a statistically significant increased rate of sling protrusion (13.1% vs. 3.3% and 1.7%; P = 0.04) in the SPARC group when compared to TVT and IVS. Otherwise, there were no significant differences between the groups with respect to the incidences of other operative complications, patients main subjective outcomes, satisfaction rates, or postoperative urodynamic findings. The overall objective stress incontinence cure rates were 87.9%, 81.5% and 72.4% for the TVT, IVS, SPARC groups respectively (P = 0.11). Conclusions: All three slings appear quite successful for the treatment of stress incontinence. The SPARC tapes showed more sling protrusion complications and a trend towards lower objective cure rates; probably as a result of the insertion method used in this study which favoured a loose SPARC sling placement. The authors recommend that the SPARC slings be left tighter than TVT, or for the cough test to be carried out. Key words: complications, polypropylene, stress incontinence, suburethral sling, surgery. Introduction Ever since the advent of the Tension-free Vaginal Tape (TVT; Gynecare, Ethicon, Somerville, NJ, USA) suburethral slings (SUS) have probably become the most commonly used technique for the surgical treatment of urinary stress incontinence. 1,2 The popularity of these slings originates from several well-established observations: they are relatively simple to implant and have been shown to shorten operating time, hospital stay and recovery time while providing similar efficacy to the traditional approach of retropubic colposuspension. 3 5 Cost-utility analyses have also suggested that the TVT appears more cost effective than colposuspension in the short term. 6 8 Initial reports have suggested that the subjective and objective stress incontinence cure rates of TVT exceed 80% at 5 7 years follow-up. 9,10 Furthermore, the use of synthetic slings negates the need for harvesting autologous fascia and the use of cadaveric fascia or xenographs, thus avoiding the potential risk of viral and prion transmission. Although the use of synthetic materials does carry the potential problem of sling infection, rejection and local tissue erosion, the use of macroporous (pore sizes > 75 µm) and monofilament polypropylene mesh for the construction of TVT seems to have reduced these complications. For instance, in a series of 1455 TVT cases in Finland, there was only a 0.7% incidence of defective healing of the vaginal mucosa and 0.8% incidence of wound infection. 11 In addition, the Austrian registry with data of 2795 TVT cases only identified Correspondence: Dr Yik N Lim, Women s and Children s Institute, Townsville Hospital, PO Box 670, Townsville, Queensland 4810, Australia. Email: yik_lim@health.qld.gov.au Received 6 July 2004; accepted 2 November 2004. 52

The SUSPEND trial a 0.1% incidence of cases where the TVT had to be removed for various reasons. 12 Currently, several types of polypropylene slings are being marketed worldwide. Each of these slings has been heavily advertised with various claims of improvement of design and biomechanical properties. However, with the exception of TVT, there is little published data on the safety and efficacy of these other devices. The three most commonly used synthetic retropubic slings in Australia are: (1) TVT; (2) Intravaginal Sling (IVS); and (3) Suprapubic Arc Sling (SPARC). Although they are all made of polypropylene, there are significant differences between the slings. The TVT and SPARC tapes are made of macroporous monofilament threads, whilst the IVS tapes are made of multifilament threads with much smaller pore sizes. There are also significant in-vitro mechanical differences between these tapes. 13 The SPARC tapes come with an absorbable tensioning suture that is designed to aid tape tension adjustment (Fig. 1). In addition, these insertion devices (trocars) differ with regards to their diameter, shapes and ergonomics; and their insertion methods are different. In contrast to the TVT and IVS tapes, the SPARC tapes are inserted with a top to bottom (craniocaudal) approach, rather than the more traditional bottom to top (caudocranial) approach. As there is little comparative evidence on the safety and efficacy of these slings in published work, a single blinded, randomised clinical trial was conducted to compare these slings. Patients and methods Following approval from the Townsville Health District Ethics Committee, 195 consenting patients with urodynamic stress incontinence (USI) were randomly allocated in a balanced way (three groups of 65 patients each) to undergo suburethral slingplasty with either TVT, IVS or SPARC. All patients recruited for this study had either failed conservative management for symptomatic stress incontinence, or required prophylactic incontinence surgery during prolapse repair for occult stress incontinence (OSI, no preoperative subjective complaint of urinary stress leakage but found to have USI). Exclusion criteria were a past history of urogenital malignancy, fistula or pelvic radiotherapy. The sample size was chosen empirically and the patients were blinded to the type of slings being implanted. Preoperative evaluation included clinical history and examination, urinalysis and culture, and urodynamic study (UDS). Where UDS had already been carried out elsewhere, an attempt to obtain the full report from the referring doctor was made instead. All proven urinary tract infections were treated prior to surgical intervention. Intra- and peri-operative complications were recorded through chart reviews. Postoperatively, the patients were reviewed at 6 12 weeks, with further clinical history, examination and, where possible, UDS. For the purpose of this study, the following definitions were used: (1) USI is defined as positive cough leak with Figure 1 Photographs of four different tapes magnified 5 6 fold. 300 ml of water either in supine or erect position during UDS; (2) intrinsic sphincter deficiency (ISD) is defined as maximum urethral closure pressure < 20 cm water; (3) improvement of a particular symptom denotes at least 50% reduction in frequency of occurrence when compared to preoperative state; and (4) dysfunctional voiding is defined Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59 53

Y. N. Lim et al. as postvoid residual volume > 100 ml and maximum urine flow rate < 10th centile according to Liverpool nomogram. 14 Surgical technique All procedures were carried out by or under the supervision of the senior author (Ajay Rane), who has significant experience in the implantation of SUS. In all cases where bladder or urethral perforation occurred, an indwelling catheter was left in situ for 24 h and then removed for trial of void. Tension-free Vaginal Tape insertion The TVT s were implanted as described by Ulmsten et al. previously with the following exceptions: (1) most cases were carried out under general anaesthesia; and (2) cough test was not used. 15 As we do not believe that there should be any tension under the urethra, we always make sure that there is a gap between the tape and the urethra. Intravaginal Sling insertion The dissection and trocar insertion were carried out as described for TVT. However, as the IVS tape is smoother than the TVT, it does not come with a plastic sheath and its tension can be adjusted easily without one. In addition, the IVS tape is also much less elastic than TVT, meaning that allowance for the pretensioning effect seen in TVT is unnecessary. The rest of the procedure was completed as for TVT. Suprapubic Arc Sling insertion The procedure started off with the dissections as in cases of TVT, but then instead of passing a trocar from the vagina up to the suprapubic region, two narrow needles were passed with a top-down fashion (from suprapubic region) into the suburethral region. The SPARC tape was then attached to the two needles which acted as a guide to bring the tape from the vaginal area to suprapubic region. Subsequently the tape tension was adjusted as in the cases of TVT. Statistical analysis Of the 195 patients recruited, four, five and four patients from the TVT, IVS and SPARC groups respectively, had to be excluded from statistical analysis because of incomplete or missing hospital charts. It was known that all patients had undergone preoperative UDS, however some missing information were encountered in specific variables as indicated in the respective tables. Differences between groups were tested using exact versions of chi-square tests. Age was found to be normally distributed and consequently parametric statistics were used. Urine flowmetry proved skewly distributed and non-parametric statistics were used accordingly. The association between the sling type and cure of USI was adjusted for ISD and urethrotomy using logistic regression analysis. Results Table 1 compares patient demographics, past history, preoperative symptomatology and also preoperative urodynamic findings between the three randomised groups. Randomisation proved successful in that no significant differences between the groups were found. However, a trend (P = 0.06) for the TVT group to contain fewer patients with the preoperative diagnosis of ISD has been noted. Table 2 lists details of the mode of anaesthesia, concomitant surgery, intra- and peri-operative, and short-term complications. It has to be pointed out that the relatively high incidences of concomitant urethrotomies were a result of the high incidence of preoperative voiding dysfunction in the study groups. Otherwise, no significant differences between the three groups were observed with the single exception that the incidence of sling protrusion was significantly associated with the type of sling implanted, with observed incidences of 3.3%, 1.7% and 13.1% for TVT, IVS and SPARC, respectively. All cases with sling protrusion were initially treated with topical vaginal oestrogen. However, nine of these 11 (81.8%) patients still required trimming under general anaesthesia. The changes in pre- and postoperative subjective outcome measures between the three sling groups are displayed in Table 3. When patients with OSI were excluded from statistical analysis, the subjective stress incontinence cure rates were 48/58 (82.8%), 50/56 (83.9%) and 45/57 (78.9%) for the TVT, IVS and SPARC groups respectively (P = 0.64). If the patients with missing charts who could not be analysed were considered treatment failures, then the subjective stress incontinence cure rates of this sensitivity analysis were 48/65 (73.8%), 47/65 (72.3%) and 45/65 (69.2%) for TVT, IVS and SPARC groups respectively (P = 0.84). None of the changes in any of the other investigated symptoms was significantly associated with the type of sling implanted. Table 4 details changes in urodynamic findings and compares these across the groups. The numbers of patients who declined postoperative UDS and had to be excluded from this analysis were three, six, and three for the TVT, IVS and SPARC groups respectively. If these cases were considered treatment failures (sensitivity analysis), the objective cure rates would be 51/65 (78.5%), 44/65 (67.7%) and 42/65 (64.6%) for the TVT, IVS and SPARC groups respectively (P = 0.19). Otherwise, no significant association between the changes of the investigated parameters and the type of sling was found. Table 5 lists the findings of urinary free flowmetry. Neither the changes (in percentiles) of the maximum flow rates nor the average flow rates differed significantly between the sling groups. Discussion To our knowledge, this is the first reported randomised controlled trial comparing the three most popular retropubic suburethral slings in Australia. The overall subjective stress incontinence cure and patient satisfaction rates were excellent, 54 Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59

The SUSPEND trial Table 1 Control of randomisation: patient demographics, past history, and preoperative symptomatology and urodynamic findings from the study to compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence Variable TVT (n = 61) IVS (n = 60) SPARC (n = 61) P-value Patient demographics: Age 56.4 (SD11.9) 58.4 (SD11.8) 58.2 (SD11.6) 0.61 Postmenopausal 44 (72.1%) 47 (78.3%) 39 (63.9%) 0.22 Vaginal parity 2.6 (SD1.3), m = 4 2.7 (SD1.3), m = 1 2.9 (SD1.6), m = 6 0.64 Past history of instrumental delivery 38 (66.7%), m = 4 38 (64.4%), m = 1 32 (58.2%), m = 6 0.64 Previous vaginal repair 17 (27.9%) 19 (31.7%) 10 (16.4%) 0.14 Previous hysterectomy 31 (50.8%) 32 (53.3%) 22 (36.1%) 0.13 Previous incontinence surgery 10 (16.4%) 11 (18.3%) 8 (13.1%) 0.72 Pre-operative symptomatology: Urgency 42 (68.9%) 33 (55.0%) 41 (67.2%) 0.22 Urge incontinence 29 (47.5%) 27 (45.0%) 32 (52.5%) 0.70 Stress incontinence 58 (95.1%) 56 (93.3%) 58 (95.1%) 0.85 Poor stream 9 (14.8%) 7 (11.7%) 8 (13.1%) 0.96 Incomplete emptying 11 (18.0%) 18 (30.0%) 16 (26.2%) 0.28 Bladder Pain 8 (13.1%) 6 (10.0%) 6 (9.8%) 0.87 Pre-operative urodynamic findings: USI (Inclusion criteria) 61 (100%) 60 (100%) 61 (100%) ISD 4 (6.8%), m = 2 13 (22.0%), m = 1 10 (16.9%), m = 2 0.06 Detrusor overactivity 3 (5.0%), m = 1 6 (10.2%), m = 1 3 (5.1%), m = 2 0.56 Increased bladder sensation 14 (23.3%), m = 1 6 (10.2%), m = 1 7 (11.9%), m = 2 0.11 Dysfunctional voiding 15 (25.0%), m = 1 12 (20.3%), m = 1 9 (15.3%), m = 2 0.41 MUCP (cm H 2 O) 41.1 (SD21.4), m = 21 38.8 (SD22.5), m = 10 43.3 (SD25.0), m = 14 0.62 Qmax percentile 19 (4 38), m = 7 13 (6 44), m = 4 27 (4 57), m = 5 0.60 Qave percentile 6 (1 34), m = 7 9 (4 34), m = 5 14 (3 67), m = 5 0.17 SD = standard deviation; m = no. of cases with missing data; USI = urodynamic stress incontinence; ISD = intrinsic sphincter deficiency; MUCP = maximum urethral closure pressure; Qmax percentile = maximum urine flow rate percentile on Liverpool nomogram; Qave percentile = average urine flow rate on Liverpool nomogram. Table 2 Mode of anaesthesia, concomitant surgery, intra- and peri-operative complications, short-term complications and follow-up duration from the study to compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence Variable TVT (n = 61) IVS (n = 60) SPARC (n = 61) P = value Operative details: GA 60 (98.4%) 56 (93.3%) 59 (96.7%) 0.31 Urethrotomy 11 (18.0%) 10 (16.7%) 4 (6.6%) 0.14 AVR 19 (31.1%) 13 (21.7%) 22 (36.1%) 0.23 Hysterectomy 2 (3.3%) 2 (3.3%) 4 (6.6%) 0.73 Enterocele repair 5 (8.2%) 2 (3.3%) 2 (3.3%) 0.51 PIVS 0 (0.0%) 1 (1.7%) 2 (3.3%) 0.55 TVSSF 2 (3.3%) 1 (1.7%) 1 (1.6%) 0.98 PVR 35 (57.4%) 30 (50.0%) 27 (44.3%) 0.36 Intra-operative complications: Urethral puncture 0 (0.0%) 0 (0.0%) 3 (4.9%) 0.11 Bladder puncture 1 (1.6%) 2 (3.3%) 4 (6.6%) 0.41 Peri-operative complications: IDC > 3 days 0 (0.0%) 0 (0.0%) 0 (0.0%) Short-term complications: Sling protrusion 2 (3.3%) 1 (1.7%) 8 (13.1%) 0.04* Sling infection 0 (0.0%) 0 (0.0%) 0 (0.0%) 0.33 GA = general anaesthesia; AVR = interior vaginal repair; PIVS = posterior intravaginal slingplasty; TVSSF = transvaginal sacrospinous fixation; PVR = posterior vaginal repair. *Statistically significant. Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59 55

Y. N. Lim et al. Table 3 Subjective outcome measures from the study to compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence Symptom TVT (n = 61) IVS (n = 60) SPARC (n = 61) P-value Stress incontinence 0.83 Pre and post symptom free 3 (4.9%) 4 (6.7%) 3 (5.0%) Cured 48 (78.7%) 47 (78.3%) 45 (75.0%) Improved but not cured 10 (16.4%) 7 (11.7%) 10 (16.7%) No improvement or deterioration 0 (0.0%) 2 (3.3%) 2 (3.3%) Acquired (de novo) 0 (0.0%) 0 (0.0%) 0 (0.0%) Urinary Frequency 0.90 Pre and Post symptom free 33 (54.1%) 22 (36.7%) 17 (28.3%) Cured 21 (34.4%) 20 (33.3%) 20 (33.3%) Improved but not cured 4 (6.6%) 9 (15.0%) 12 (20.0%) No improvement or deterioration 1 (1.6%) 4 (6.7%) 8 (13.3%) Acquired (de novo) 2 (3.3%) 5 (8.3%) 3 (5.0%) Urinary urgency 0.06 Pre and post symptom free 15 (24.6%) 22 (36.7%) 17 (28.3%) Cured 20 (32.8%) 20 (33.3%) 20 (33.3%) Improved but not cured 22 (36.1%) 9 (15.0%) 12 (20.0%) No improvement or deterioration 0 (0.0%) 4 (6.7%) 8 (13.3%) Acquired (de novo) 4 (6.6%) 5 (8.3%) 3 (5.0%) Urge incontinence 0.06 Pre and post symptom free 28 (45.9%) 32 (53.3%) 23 (38.3%) Cured 19 (31.1%) 18 (30.0%) 24 (40.0%) Improved but not cured 10 (16.4%) 8 (13.3%) 23 (6.7%) No improvement or deterioration 0 (0.0%) 1 (1.7%) 4 (6.7%) Acquired (de novo) 4 (6.6%) 1 (1.7%) 6 (10.0%) Incomplete bladder emptying 0.48 Pre and post symptom free 48 (78.7%) 40 (66.7%) 43 (71.7%) Cured 9 (14.8%) 17 (28.3%) 15 (25.0%) Improved but not cured 2 (3.3%) 1 (1.7%) 0 (0.0%) No improvement or deterioration 0 (0.0%) 0 (0.0%) 0 (0.0%) Acquired (de novo) 2 (3.3%) 2 (3.3%) 2 (3.3%) Satisfied with surgery 51 (83.9) 50 (83.3) 52 (85.0) 0.99 All P-values refer to exact versions of chi-square tests between the three groups. comparable to those reported in published material and similar for all three slings. However, there was a trend (P = 0.10) for more patients in the SPARC group to have persistent USI postoperatively. This trend for differences in objective cure rates reduced but persisted even after logistic regression analysis was carried out to adjust for the (non-significant) differences in the incidence of preoperative ISD and concomitant urethrotomy, since it has been previously shown to reduce the effectiveness of TVT. 16,17 In this study, where there were preoperative incidence of ISD or concomitant urethrotomy, the overall objective stress incontinence cure rates became 74.1% and 68.0% respectively. Otherwise, there was only a trend for slight differences in postoperative overactive bladder symptoms between the three groups. As the numbers of patients affected were small and the preoperative bladder complaints were not controlled; these observations are probably not clinically significant. In addition, we have observed a 4 8 fold increase in the risk of sling protrusion when the SPARC slings were compared to TVT and IVS in this study. This high sling protrusion rate with SPARC tapes has never been reported with the limited available publisihed material. In view of the observations that the TVT and SPARC slings are very similar mechanically in vitro, we postulate that the increased incidence of sling protrusion in the SPARC group to be related to our SPARC implantation method. 13 Due to the elasticity of the TVT, when the plastic sheaths are being removed after sling placement, there is a pretensioning effect that causes the tape to spring back up superiorly. The tensioning suture of the SPARC sling limits this effect, so when placed with similar tension as the TVT, tended to become looser. With the much less elastic IVS tapes, the surgeon would naturally leave it in a more snug-fit fashion under the urethra, and thus would not become as loose as a SPARC sling. The results of this study also concur with the findings of a case-control study comparing SPARC slings with the TVT s, whereby the SPARC slings were found to be looser and more elastic on Valsalva manoeuvre on 3-dimensional ultrasound scans. 18 This might also be the reason why there was a trend for slightly lower objective stress incontinence cure rates with SPARC slings in this study. A follow up study to evaluate these patients with 3-dimentional ultrasound is 56 Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59

The SUSPEND trial Table 4 Objective urodynamic outcome measures from the study to compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence Variable TVT (n = 58) IVS (n = 54) SPARC (n = 58) P-value USI (inclusion criteria): 0.11 Cured 51 (87.9%) 44 (81.5%) 42 (72.4%) Still present 7 (12.1%) 10 (18.5%) 16 (27.6%) ISD (MUCP < 20 cm H 2 O): m = 1 m = 0 m = 0 0.10 Absent pre and post 43 (75.4%) 37 (68.5%) 45 (77.6%) Cured 3 (21.4%) 9 (16.7%) 6 (10.3%) Present Pre and Post 1 (1.8%) 4 (7.4%) 4 (6.9%) Acquired (de novo) 10 (17.5%) 4 (7.4%) 3 (5.2%) Detrusor overactivity 0.26 Absent pre and post 53 (91.4%) 47 (87.0%) 55 (94.8%) Cured 3 (5.2%) 4 (7.4%) 0 (0.0%) Present pre and post 0 (0.0%) 1 (1.9%) 2 (3.4%) Acquired (de novo) 2 (3.4%) 2 (3.7%) 1 (1.7%) Increased bladder sensation 0.17 Absent pre and post 41 (70.7%) 47 (87.0%) 49 (84.5%) Cured 9 (15.5%) 4 (7.4%) 7 (12.1%) Present pre and post 5 (8.6%) 2 (3.7%) 0 (0.0%) Acquired (de novo) 3 (5.2%) 1 (1.9%) 2 (3.4%) Dysfunctional voiding 0.76 Absent Pre and Post 40 (69.0%) 40 (74.1%) 45 (77.6%) Cured 8 (13.8%) 8 (14.8%) 4 (6.9%) Present pre and post 7 (12.1%) 4 (7.4%) 5 (8.6%) Acquired (de novo) 3 (5.2%) 2 (3.7%) 4 (6.9%) m denotes missing data. Table 5 Objective changes in urine free flowmetry from the study to compare the safety and efficacy of three types of suburethral slings for the treatment of urodynamic stress incontinence Variable TVT (n = 61) IVS (n = 60) SPARC (n = 61) P-value Pre-op Qmax percentile 19 (4 38), m = 7 13 (6 44), m = 4 27 (4 57), m = 5 Post-op Qmax percentile 4 (1 17), m = 7 9 (3 36), m = 6 15 (1 48), m = 4 Qmax percentile change 6 ( 23 ± 2), m = 11 7 ( 18 ± 6), m = 9 3 ( 23 ± 9), m = 8 0.12 Pre-op Qave percentile 6 (1 34), m = 7 9 (4 34), m = 5 14 (3 67), m = 5 Post-op Qave percentile 2 (0 12), m = 7 8 (2 31), m = 7 12 (1 30), m = 8 Qave percentile change 3 ( 24 0), m = 11 3 ( 11 ± 4), m = 11 4 ( 19 ± 9), m = 8 0.51 Displayed are medians as measures of central tendency and quartiles (25 75%); P-values refer to exact versions of Kruskall-Wallis tests between the 3 groups. currently underway to investigate this further. We would therefore suggest that perhaps the SPARC tape should be left tighter when compared to TVT, or for the cough test to be carried out intraoperatively. More recently, the differences between monofilament (TVT, SPARC) and multifilament (IVS) tapes have come under the spotlight. In a Danish case-control study involving 127 cases of TVT and 42 cases of IVS, Glavind and Sander described only 1 (0.8%) case of sling erosion in the TVT group, but six (14.3%) cases of defective healing, erosion or extrusion in the IVS group at follow up visits ranging from 2 weeks to 12 months. 19 The authors suggested that the increased risk of tape complications could be a result of the microporous component of IVS tape harbouring bacteria while precluding the entrance of macrophages. However, in a randomised comparison study between 50 cases of TVT and 50 cases of IVS with a median follow up duration of 13.5 months, Rechberger and colleagues did not find any statistically significant differences in complication rates, or stress incontinence improvement and cure rates between the two groups. 20 They only reported a statistically significant increase in the incidence of prolonged postoperative urinary retention (> 1 day postoperatively) in the TVT group (20%) when compared to those in the IVS group (4%). Apart from these two studies, other publications were limited to the old IVS tapes that were made with either Nylon or polytetrafluoroethylene meshes. In this study, we have not observed any sling infection cases at 6 12 weeks follow-up visit. However, due to the conflicting nature of the limited available data, such a theoretical risk is difficult to prove or quantify Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59 57

Y. N. Lim et al. at this stage. We will, however, continue to follow these patients up in order to assess for any differences in the infection risk with longer term follow up. We have also observed a non-significant trend (P = 0.11) for urethral injury to occur more frequently with the use of SPARC slings. However, all three cases involved occurred during the early stages (first 6 months) of the trial and it is likely that these might have been influenced by the learning curve effect. The incidences of intraoperative bladder injury were generally low (1.6 6.6%) in all three groups and did not differ significantly between the groups. This is consistent with the findings of the randomised trial of Rechberger et al. where this complication occurred in 4% and 8% of the TVT and IVS groups respectively (P = 0.34). 20 The 6.6% bladder injury rate of the SPARC group in this study is also comparable to the 10.5% incidence reported in the French multicentre prospective series with 104 SPARC slings. 21 However, in a retrospective study of 140 cases of SPARC slings, Kobashi et al. only reported five cases of retropubic haematoma (of which one required percutaneous drainage) and one case of small bowel injury. 22 Again, as a result of the conflicting nature of the limited available evidence and also the potential for a lack of power to detect significant differences in this study; it is difficult to conclude whether the craniocaudal approach of SPARC devices and the blunt tip of the IVS devices might actually reduce the incidences of intraoperative bladder injury when compared to the utility of the TVT s sharp trocars. Although all patients had undergone preoperative UDS, we have unfortunately not been able to gather data on all the parameters for analysis purpose because some referral centres: (1) did not measure urethral pressure profile or uroflowmetry; or (2) could not be contacted to provide the full UDS report. However, when the available data were analysed, there was a trend for all patients to show slight reduction in urine flow rates following a suburethral sling procedure. This is consistent with the findings of the majority of published work and would suggest that perhaps these procedures are not tensionless after all. 23 25 Conclusions This study has reconfirmed that all three tested polypropylene suburethral slings are highly effective for the treatment of USI in the short-term. The incidences of major complications are also rare. However, compared to TVT and IVS devices, the SPARC slings were associated with significantly higher incidence of sling protrusion at 6 12 weeks follow-up in this study. This might be related to the relative lack of pretensioning effect of the SPARC slings when inserted with similar technique to TVT. We would recommend that the SPARC slings be left tighter than the TVT s, or for the cough test to be carried out intraoperatively. As there is currently a paucity of comparison data on the suburethral slings, there is now an urgent need for more randomised controlled studies like this to establish if the non-tvt devices operate as well as the TVT. We will continue to follow up these patients and report on the medium to long-term results in due course. References 1 Ulmsten U, Petros P. Intravaginal slingplasty (IVS). An ambulatory surgical procedure for the treatment of female urinary incontinence. Scand J Urol Nephrol. 1995; 29: 346 348. 2 Bemelmans BL, Chapple CR. Are slings now the gold standard treatment for the management of female urinary stress incontinence and if so what technique? Curr Opin Urol. 2003; 13: 301 307. 3 Ward KL, Hilton P: United Kingdom and Ireland Tension-free Vaginal Tape Trial Group. Prospective multicentre randomised trial of tension-free vaginal tape and colposuspension as primary treatment for stress incontinence. BMJ. 2002; 325: 67. 4 Valpas A, Kivela A, Penttinen J et al. Tension-free vaginal tape and laparoscopic mesh colposuspension in the treatment of stress urinary incontinence: immediate outcome and complications a randomized clinical trial. Acta Obstet Gynecol Scand. 2003; 82: 665 671. 5 Ustun Y, Engin-Ustun Y, Gungor M, Tezcan S. Tension-free vaginal tape compared with laparoscopic Burch urethropexy. J Am Assoc Gynecol. Laparoscop. 2003; 10: 386 389. Erratum in: J Am Assoc Gynecol Laparosc. 2003; 10: 581. 6 Quievy A, Couturier F, Prudhon C, Abram F, al Salti R., Ansieau JP. [Economic comparison of 2 surgical techniques for the treatment of stress urinary incontinence in women: Burch s technique versus the TVT technique.] Prog Urol. 2001; 11: 347 353 (in French). 7 Cody J, Wyness L, Wallace S et al. Systematic review of the clinical effectiveness and cost-effectiveness of tension-free vaginal tape for treatment of urinary stress incontinence. Health Technol Assess. 2003: 7: 1 189. 8 Manca A, Sculpher MJ, Ward K, Hilton P. A cost-utility analysis of tension-free vaginal tape versus colposuspension for primary urodynamic stress incontinence. BJOG. 2003; 110: 255 262. 9 Nilsson CG, Kuuva N, Falconer C, Rezapour M, Ulmsten U. Long-term results of the tension-free vaginal tape (TVT) procedure for surgical treatment of female stress urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct. 2001; 12: S5 S8. 10 Nilsson CG, Rezapour M, Falconer C. Seven-year Follow-up of the Tension-Free Vaginal Tape Procedure for Treatment of Urinary Incontinence. Obstet Gynecol. 2004; 104: 1259 1262. 11 Kuuva N, Nilsson CG. A nationwide analysis of complications associated with the tension-free vaginal tape (TVT) procedure. Acta Obstet Gynecol Scand. 2002; 81: 72 77. 12 Tamussino KF, Hanzal E, Kolle D, Ralph G, Riss PA: Austrian Urogynecology Working Group. Tension-free vaginal tape operation: results of the Austrian Registry. Obstet Gynecol. 2001; 98: 732 736. 13 Dietz. HP, Vancaillie P, Svehla M, Walsh W, Steensma AB, Vancaillie TG. Mechanical properties of urogynecologic implant materials. Int Urogynecol J Pelvic Floor Dysfunct. 2003; 14: 239 243. 14 Haylen BT Ashby D, Sutherst JR, Frazer MI, West CR. Maximum and average urine flow rates in normal male and female populations the Liverpool nomograms. Br J Urol. 1989; 64: 30 38. 58 Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59

The SUSPEND trial 15 Ulmsten U, Henriksson L, Johnson P, Varhos G. An ambulatory surgical procedure under local anaesthesia for treatment of female urinary incontinence. J Urogyneol J Pelvic Floor Dysfunct. 1996; 7: 81 85. 16 Rezapour M, Falconer C, Ulmsten U. Tension-free vaginal tape (TVT) in stress incontinent women with intrinsic sphincter deficiency (ISD) a long-term follow-up. Int Urogynecol J Pelvic Floor Dysfunct. 2001; 12: S12 S14. 17 Djelouat T, Avances C, Dubon O, Viale S, Boukaram M, Costa P. [Efficacy of suburethral TVT in cases of stress urinary incontinence associated with severe sphincter insufficiency. Report of 22 cases] Prog. Urol. 2002; 12: 1251 1255 (in French). 18 Dietz. HP, Foote AJ, Mak HL, Wilson PD. TVT and Sparc suburethral slings: a case-control series. Int Urogynecol J Pelvic Floor Dysfunct. 2004; 15: 129 131. 19 Glavind K, Sander P. Erosion, defective healing and extrusion after tension-free urethropexy for the treatment of stress urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct. 2004; 15: 179 182. 20 Rechberger T, Rzezniczuk K, Skorupski P et al. A randomized comparison between monofilament and multifilament tapes for stress incontinence surgery. Int Urogynecol J Pelvic Floor Dysfunct. 2003; 14: 432 436. 21 Deval B, Levardon M, Samain E et al. A French multicentre clinical trial of SPARC for stress urinary incontinence. Eur Urol. 2003; 44: 254 258. 22 Kobashi KC, Govier FE. Perioperative complications: The first 140 polypropylene pubovaginal slings. J Urol. 2003; 170: 1918 1921. 23 Sander P, Moller LM, Rudnicki PM, Lose G. Does the tension-free vaginal tape procedure affect the voiding phase? Pressure-flow studies before and 1 year after surgery. BJU Int. 2002; 89: 694 698. 24 Al-Badr A, Ross S, Soroka D, Minassian VA, Karahalios A, Drutz HP. Voiding patterns and urodynamics after a tensionfree vaginal tape procedure. J Obstet Gynecol Can. 2003; 25: 725 730. 25 Lukacz. ES, Luber KM, Nager CW. The effects of the tensionfree vaginal tape on voiding function: a prospective evaluation. Int Urogynecol J Pelvic Floor Dysfunct. 2004; 15: 32 38. Australian and New Zealand Journal of Obstetrics and Gynaecology 2005; 45: 52 59 59