- Top of page
- PATIENTS AND METHODS
Radical retropubic prostatectomy (RRP) is the most common method of treating localized prostate cancer, being offered to ≈ 70 000 men every year . While RRP remains the reference standard, concerns about associated discomfort, impact on quality of life, morbidity and loss of work have all resulted in a quest for an alternative to RRP. Laparoscopic prostatectomy , recently developed in Europe, has emerged as a viable alternative but requires considerable experience and training, and is therefore not embraced by the American urological community. We have explored laparoscopic prostatectomy under expert supervision but chose an alternative, using a recently available robotic surgical system . This ergonomic master-slave system offered three-dimensional visualization, wristed instrumentation, intuitive finger-controlled movements and a comfortable seated position for the surgeon. All these features translate into a more anatomical, easily learned and more reproducible prostatectomy than the laparoscopic approach .
Using this robotic system, we developed a technique of anatomical prostatectomy at our institution, the Vattikuti Institute prostatectomy (VIP), and started using VIP in November 2000 [3,4]; we have performed 400 such operations (to December 2002). Our current approach is based on the conventional anatomical ‘open’ prostatectomy, modified by technical nuances derived from robotic technology, as elaborated by the da Vinci surgical systemTM (Intuitive Surgical, Sunnyvale, CA, USA). The present study is an extension of our previously reported prospective comparison between robotic and conventional RRP in which the initial 30 robotic cases were conducted using the classical Montsouris technique [4,5]. The data and technique have since been developed. The aim of the present report was to prospectively compare surgical, oncological and functional outcomes of VIP and RRP.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
The study was a single-institution, prospective, unrandomized comparison of outcomes using VIP and RRP in 300 patients (100 RRP and 200 VIP) with clinically localized prostate cancer who chose surgical treatment between October 1999 and December 2002 at our institution.
Patients were included if they were surgical candidates; the choice between RRP and VIP was offered to all patients who had a 10-year life-expectancy and had prostate cancer of Gleason score ≥ 6. Attempts were initially made to randomize the subjects, but none of the patients agreed. Ironically, all the patients approved of randomization as a concept until they were asked to participate (unpublished data). Therefore, the patients were assigned to one of the two treatment arms based on their personal preferences. No attempt was made to pre-select patients for VIP. However, we conducted a statistical sensitivity analysis to ensure that the two groups were comparable at baseline. The local Institutional Review Board evaluated and approved the study.
RRP was performed by one of the eight urologists at the Vattikuti Institute, who have a combined experience of > 1400 RRPs; VIP was performed by a one team (lead by M.M.), which included a console surgeon (M.M.) and a surgical team at the patients’ side. The present analysis includes data from the patients operated by M.M.
RRP was conducted using the anatomical technique [1,6,7] and VIP using the previously described technique . All patients were followed on a standard clinical pathway and, barring complications, the RRP patients were discharged within 48 h and VIP patients either on the same day or the next morning. All patients were sent home with an indwelling Foley catheter; these were left in place for 14 days after RRP, while a cystogram was taken after 4–7 days in VIP patients, and if there was no extravasation the catheter was removed, or it was left for an additional week.
For baseline variables, the preoperative information was collected on a custom-designed data-sheet, particularly recording age, previous abdominal surgery, comorbidity, body mass index (BMI), serum PSA, biopsy Gleason score and clinical stage. Variables recorded during and after surgery included the details of each surgical procedure, entered on a comprehensive data-collection sheet completed by a third party (recording the times of entry into the operative room, induction of anaesthesia, and for the various steps of the operation). Additional data measured were blood loss (determined by the anaesthesiologist), the postoperative pain score (using a visual analogue score), days of hospitalization, haemoglobin level at discharge and duration of catheterization. Any untoward event within 30 days of surgery was recorded as a perioperative complication.
For histopathological data, the surgical specimen was inked and processed for histopathological analysis . Margins were considered positive if there was tumour present at the inked margin in the RRP series. In the VIP patients, apical margin was considered positive if cancer was seen in the intraoperative distal biopsies. The PSA level after surgery was recorded during 3-monthly clinic visits and entered into the database. Biochemical recurrence was defined as two successive PSA levels of > 0.02 ng/mL.
For functional outcome, functional data were collected by a third-party telephone interviewer who asked the patients about preoperative sexual function, ability to obtain an erection, recent sexual intercourse (within 4 weeks), use of sildenafil and the number of pads or liners used to manage urinary incontinence; continence was defined as using no pads or a liner for security reasons only.
Data were analysed using standard statistical software, with between-group comparisons of nominal variables by Fisher's exact test and the chi-square analysis, while continuous variables were compared using anova. The days to continence, erection and intercourse were recorded and used in the time dependent-survival analysis, using the Kaplan-Meier method .
- Top of page
- PATIENTS AND METHODS
The study cohort included 100 consecutive RRP and 200 VIP patients; their demographic variables before surgery are summarized in Table 1. The groups were comparable in age, BMI, PSA, prostate volume, clinical stage, Gleason score, comorbidity and previous abdominal surgery.
Table 1. Baseline and surgical variables in 300 contemporary patients undergoing prostatectomy
|Variable||RRP (100)||VIP (200)||P|
|Age, years|| 63.1 (42.8–72)|| 59.9 (40–72)||NS|
|Serum PSA, ng/mL|| 7.3 (1.9–35)|| 6.4 (0.6–41)||NS|
|Prostate volume, mL|| 48.4 (24.2–70)|| 58.8 (18–140)||NS|
|Clinical stage, %|
|T1a|| 0|| 0.5||NS|
|T1c|| 59|| 49|| |
|T2a|| 10|| 10|| |
|T2b|| 35|| 39|| |
|T3a|| 4|| 1.5|| |
|Mean|| 6.6|| 6.5||NS|
|2–4|| 0|| 0|| |
|5|| 3|| 0|| |
|6|| 49|| 67|| |
|7|| 35|| 28|| |
|8|| 10|| 4|| |
|9–10|| 3|| 2|| |
|Mean BMI|| 27.6 (17–41)|| 27.7 (19–38)||NS|
|Previous abdominal and hernia surgery, %|| 19|| 20||NS|
|Charlson score|| 2.5|| 2.3||NS|
|Operative time, min||163 (86–395)||160 (71–315)||NS|
|Estimated blood loss, mL||910 (200–5000)||153 (25–750)||< 0.001|
|Intraoperative blood transfusion, %|
|Autologous|| 56|| 0||< 0.001|
|Banked|| 11|| 0|| |
|Total|| 67|| 0|| |
|Postop. pain score|| 7 (4–10)|| 3 (1–7)||< 0.05|
|Discharge Hb, g/L|| 101 (69–146)||130 (73–151)||< 0.05|
|Hospital stay, days|| 3.5 (3–6)|| 1.2 (< 1–5)||< 0.05|
|Discharged < 24 h, %|| 0|| 93||< 0.001|
The mean operative duration (Table 1; incision or dissection to closure) was comparable for RRP (163 min) and VIP (160 min) and not significantly different. The estimated blood loss differed significantly (P < 0.001); the distribution of blood loss in the ranges 0–200, 201-500, 501–1000 and > 1001 mL was 2%, 32%, 43% and 23% during RRP, and 81%, 17%, 2% and none during VIP (P < 0.05). The need for blood transfusion was greater in the RRP group (67% vs none, P < 0.001), with banked blood transfused in 11% during RRP.
The mean pain score on the first day after surgery differed significantly (P = 0.05; Table 1) and the mean discharge haemoglobin level was greater in the VIP group. The mean hospital stay was longer for RRP (P < 0.05), with 93% of patients after VIP discharged in < 24 h, while none were discharged within this time after RRP (P < 0.001). We currently discharge (each day) the first two patients within 4 h of surgery while the third stays overnight.
The percentage cancer, Gleason scores and pathological stages were comparable between the groups (Table 2); 9% of the VIP and 23% (P < 0.05) of the RRP patients had tumour at the inked margin (see methods). There was a significant difference in the proportion of patients with an undetectable PSA at a mean follow-up of 556 and 236 days, respectively (P < 0.05). As shown in Table 2, there were no deaths in either treatment arm. Overall, there were four times as many complications in the RRP arm (20% vs 5%).
Table 2. The outcome after surgery in 300 contemporary patients undergoing prostatectomy
|Variable||RRP (100)||VIP (200)||P|
|catheterization, days|| 15.8 (7–28)|| 7 (1–18)||< 0.05|
|follow-up, days||556||236||< 0.05|
|Undetectable PSA, %|| 85|| 92||NS|
|% cancer in specimen|| 18.3 (5–90)|| 19 (1–80)||NS|
|Mean|| 6.6|| 6.9||NS|
|2–4, %|| 0|| 0|| |
|5|| 1|| 0.5|| |
|6|| 41|| 43|| |
|7|| 38|| 40|| |
|8|| 18|| 8|| |
|9–10|| 2|| 2.5|| |
|Pathological stage, %|
|T2a|| 18|| 15||NS|
|T2b|| 75|| 72|| |
|T3a|| 4|| 7|| |
|T3b|| 3|| 6|| |
|Positive node|| 2|| 1||NS|
|Margin positivity in organ-confined cancers (pT2a-T3a), %|
|Extensive (> 1 mm)|| 15|| 1||< 0.05|
|Focal (≤ 1 mm)|| 8|| 5|| |
|Aborted|| 1|| 2|| |
|Conversion|| – || 0||NS|
|Rectal injuries|| 1|| 0||NS|
|Postoperative ileus|| 3|| 3||NS|
|Wound dehiscence/hernia|| 1|| 2||NS|
|Postop. fever/pneumonia)|| 4|| 0||< 0.05|
|Lymphocele|| 2|| 0||NS|
|Obturator neuropathy|| 2|| 0||NS|
|DVT|| 1|| 1||NS|
|Postoperative MI|| 1|| 0|| |
|Postop. bleeding/re-exploration|| 4|| 1||NS|
|Total|| 20|| 5||< 0.05|
We have initiated a formal quality-of-life study using the Expanded Prostate Cancer Composite (EPIC) quality-of-life instrument, developed and validated by Wei et al.. The EPIC study requires 140 completed responses in each treatment arm to have a 90% power at a significance of 0.0042; the results of the EPIC survey will be presented later. We are also collecting functional data through physician interviews and a third-party telephone survey. For the purpose of this paper we report the data on 120 patients who participated in the telephone interview. The urinary continence results are summarized in Fig. 1a; survival analysis was used to compute the probability of return to continence in the two groups . As shown, patients achieved continence much quicker after VIP than after RRP, and the 50% return of continence occurred in 160 and 44 days, respectively (P < 0.05).
Figure 1. Kaplan-Meier analysis of return of a continence, b erections and c, intercourse in the RRP (green) and VIP (red) groups. The methods were described previously [10,27].
Download figure to PowerPoint
Sexual function was also evaluated using the same interviews and the response compiled as (a) ability to achieve erections and (b) to have an erection strong enough for intercourse. The analysis was limited to patients who classified themselves as having normal preoperative erections and sexual intercourse, and those who had a bilateral nerve-sparing procedure. The preoperative sexual function was cross-checked with a baseline patient-reported sexual health inventory of males ; potent patients had a preoperative inventory score of ≥ 18. In Fig. 1b,c the probability of return of erection and intercourse is calculated for both groups ; patients after VIP had a more rapid return of erection (50% return at a mean follow-up of 180 days, vs 440 days after RRP, P < 0.05). The return of intercourse was also quicker after VIP, with half the patients achieving intercourse at a mean follow-up of 340 days (Fig. 2c). Half the RRP patients have not yet achieved return of intercourse at 700 days (P < 0.05). Of the VIP and RRP patients, 42% and 65%, respectively, are known to be using sildenafil at the time of analysis. We routinely encourage patients to try sildenafil early, in an attempt to prevent possible corporal fibrosis .
- Top of page
- PATIENTS AND METHODS
In this study we compared the operative, oncological and functional results of RRP and VIP at the same institution. While the study was not randomized it was well suited to compare the two procedures because of its single-centre setting, prospective data collection, inclusion of consecutive patients, adequate sample size and use of third-party telephone interviews for assessing the functional results. The methodology was further strengthened because the baseline characteristics were comparable in the two arms. The results of the analysis clearly show that in our hands VIP is a safer and more effective procedure than RRP in treating clinically localized prostate cancer, and produces comparable oncological control. The other major benefits of VIP are its short hospitalization and catheterization times, with less pain, minimal blood loss, higher haemoglobin levels and quicker return of continence and potency after surgery. There were fewer complications and the procedure has now been used in > 400 patients, with reproducible results.
We are particularly pleased with the early functional outcome, which shows that patients recover much faster after VIP than RRP, both in their continence and potency. This early return of functional outcome could be attributed to better visualization, anatomical dissection, reduced blood loss and improved ability to reconstruct the anatomy using robotic assistance . These results from VIP are comparable with most published RRP [10,15–18] and laparoscopic radical prostatectomy data [19–22].
Previously VIP took longer than RRP, but with experience there is no difference in the operative duration at our institution . We are currently operating on three patients per day and two are discharged within 4 h of surgery. There could be theoretical concerns that the present results are unique to our institution and cannot be duplicated at other centres. Indeed, we have an experienced surgeon, a committed team, and 2 years experience of > 400 cases. However, based on our experience, we consider strongly that unlike laparoscopic prostatectomy, which is difficult and time-consuming to learn , VIP can be perfected by any committed team lead by a surgeon skilled in RRP.
One caveat of this study is that while one committed team performed VIP, eight different surgeons in the same institution undertook the RRPs. This is a valid concern but does not seem to affect the results, because every open surgeon had conducted at least 100 RRPs and they had a combined experience of > 1400 cases. Furthermore, the RRP data are no different from those in other published series of RRP from major USA centres [10,15–17]. Therefore, despite these methodological concerns, we consider that the results of this comparative study are valid and provide useful information to be used by other centres embarking on robotic prostatectomy.
A potential disadvantage of this technology is the lack of tactile sensation, but three-dimensional visualization allowed us to compensate for this handicap. A more real disadvantage is cost; while the reduced need for blood transfusion, avoidance of cell-savers, minimal laboratory testing, shorter hospital stay, early return to work and ability to manage three patients a day may offset the cost a little, the costs of the da Vinci Surgical System and the recurring costs of consumables remain substantial, at least at present. We are currently comparing the costs in a formal study and the results will be presented soon.
Another caveat requiring clarification is our use of intraoperative biopsy data to classify margin status. While this approach may appear biased in favour of VIP, importantly, robotic assistance provides exquisite details of the apical anatomy and sphincter complex, and provides the tools to excise precisely any additional suspicious tissue without damaging the surrounding important neuromuscular structures. Therefore, in view of the benefits to the patient, we felt compelled to use the extra help provided by the robot in taking 2–3 mm of extra margin. A similar approach has been used by other centres during RRP , the only difference being that using the robot is simpler and more precise.
Few other studies have reported robot-assisted prostatectomy using the da Vinci system; Abbou et al. published a case report of this procedure in one patient. Binder and Kramer  and Pasticier et al. described preliminary results in two small series of patients. However, neither of these groups compared their results with ‘open’ RRP; the present is the first prospective study comparing RRP and VIP, and provides a contemporary framework for the objective evaluation of results and conclusions.
In conclusion, we have established the feasibility and safety of VIP for managing clinically localized prostate cancer; this study strongly suggests that surgeons experienced in conventional RRP can learn robot-assisted prostatectomy within a reasonable time and with no unduly higher rate of complications.