J Robotic Surg (2015) 9:291–297 DOI 10.1007/s11701-015-0529-9
ORIGINAL ARTICLE
Does previous transurethral prostate surgery affect oncologic and continence outcomes after RARP? Yu-Kai Su1 • Benjamin F. Katz1 • Shailen S. Sehgal1 • Sue-Jean S. Yu1 Yu-Chen Su2 • Andrew Lightfoot1 • Ziho Lee1 • Elton Llukani1 • Kelly Monahan1 • David I. Lee1
•
Received: 19 February 2015 / Accepted: 17 July 2015 / Published online: 8 August 2015 Ó Springer-Verlag London 2015
Abstract We examined the effect of previous transurethral resection of the prostate (TURP) on multiple oncologic and continence outcomes after robotic-assisted radical prostatectomy (RARP). We performed a retrospective cohort study of a total of 2693 patients from 2007 to 2014 who underwent RARP. Patients were stratified into 49 patients who had previous TURP prior to RARP (group 1) and 2644 patients who had no TURP prior to RARP (group 2). We collected operative variables including
& David I. Lee [email protected]; [email protected] Yu-Kai Su [email protected] Benjamin F. Katz [email protected] Shailen S. Sehgal [email protected] Yu-Chen Su [email protected] Andrew Lightfoot [email protected] Ziho Lee [email protected] Elton Llukani [email protected] Kelly Monahan [email protected] 1
2
Department of Urology, University of Pennsylvania Health System, Penn Presbyterian Medical Center, University of Pennsylvania, 300 Medical Office Building, 51 North 39th Street, Philadelphia, PA 19104, USA Division of Biostatistics, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA
estimated blood loss, operative time, and positive surgical margin (PSM) rates. Urinary continence, defined as 0 pads per day (PPD), and social continence, defined as 1-PPD, were also assessed. American Urological Association Symptoms Score (AUASS), overall ability to function sexually, and Expanded Prostate Cancer Index Composite (EPIC) questionnaire were evaluated at 3 and 12 months after RARP. Weakness of urinary stream (EPIC #4d) at 12 months imposed a greater problem for group 1 patients with prior TURP compared to group 2 patients without prior TURP (p = 0.012). PSM was not statistically significant between the two groups (p = 0.110). Group 1 patients had a greater PSM rate (30.61 %) as compared to group 2 (20.95 %). PSM locations in group 1 patients showed the most common locations at the posterior and apex. The difference between the two groups for AUASS, overall sexual function, estimated blood loss, operative time, urinary continence, and social continence was not statistically significant. We examined the effect of previous TURP on postoperative RARP continence and oncologic outcomes. This data can be used to counsel those with prior TURP before RARP. Keywords Robotic prostatectomy Prostate cancer Transurethral prostate surgery TURP Continence outcomes
Introduction With advancing age being a risk factor for both benign prostatic hyperplasia (BPH) and prostate cancer, it is not uncommon for aging men once treated for BPH to later undergo prostatectomy for prostate cancer [1]. Robot-assisted radical prostatectomy (RARP) is now the dominant
123
292
approach to radical prostatectomy (RP) in the USA [2]. With the increasing prevalence and availability of RARP [2], surgeons are expanding this procedure to patients with prior transurethral resection of the prostate (TURP) [3]. However, concerns have been raised regarding outcomes of these cases due to their potential increased complexity [4]. In the open RP literature, a number of studies have described an increase in surgical difficulty of patients with prior prostate surgery as compared to the surgery naı¨ve. However, the perioperative and postoperative morbidity, complication rates, and oncological outcomes in these patients were not statistically different [5–9]. Conversely, other groups have shown that in those with previous prostate surgery, the outcomes are compromised in several ways in patients undergoing either laparoscopic radical prostatectomy (LRP) or open RP [4, 10, 11]. For example, Jaffe and colleagues recently illustrated that men with prior history of TURP undergoing LRP not only have worse outcomes with regard to positive surgical margin (PSM) rates (21.8 vs. 12.6 %), but also have prolonged operative time (OT), and poorer overall complication rates [10]. Moreover, Colombo et al. have suggested that previous prostate surgery can adversely affect urinary continence and erectile function in patients undergoing open RP [11]. Much of the published data in the literature has looked at patients undergoing LRP or open RP, yet only a few studies have investigated the impact of prior prostate surgery history on RARP outcomes. Hampton and colleagues published the data of 51 men with history of TURP and found statistically significant worse PSM rates (35.3 vs. 17.7 %) when compared with men with no TURP [12]. In addition, Gupta et al. found prolonged OT, increased blood loss, increased biochemical recurrence and PSM rates, as well as worse continence outcomes in a group of 26 patients that had undergone previous TURP [1]. On the contrary, Martin and colleagues disagree with such findings by illustrating no significant difference in perioperative and oncological outcomes of RARP when comparing patients with previous prostate surgery with those that are surgery naı¨ve [3]. Thus, there is a paucity of relevant data for outcomes of patients with prior TURP undergoing RARP, and the present study aims to bridge this gap in the existing literature .
Fig. 1 Positive surgical margin (PSM) locations in group 1 patients with history of previous TURP
123
J Robotic Surg (2015) 9:291–297
Our study compares RARP patients with a history of previous prostate surgery with patients that did not undergo previous prostate surgery. Perioperative data, oncological outcomes, sexual function, continence, and urinary symptoms were evaluated.
Materials and methods We performed a retrospective cohort study of a total of 2693 patients from 2007 to 2014 who underwent RARP at our institution by a single surgeon (DIL). The database was collected prospectively and reviewed under an IRB approved protocol. Informed consent was obtained from all patients for being included in the study. We stratified patients into two groups: 49 patients who had previous TURP prior to RARP (group 1) (Fig. 1) and 2644 who did not have TURP prior to RARP (group 2) (Fig. 2). Perioperative and postoperative variables were recorded and analyzed for significant differences in outcomes between groups. Operative variables analyzed were estimated blood loss (EBL), operative time, and PSM rates. Urinary continence, defined as 0 pads per day (PPD), and social continence, defined as 1-PPD, were also assessed. With regard to urinary symptoms and sexual function outcome variables, American Urological Association Symptoms Score (AUASS) and overall ability to function sexually (score from 1 to 5, highest score indicating highest satisfaction with the ability to achieve and maintain erection for penetration) at 3 and 12 months after RARP were examined. Additionally, we also evaluated the results of #1 through #5 from the Expanded Prostate Cancer Index Composite (EPIC-26 short form) Questionnaire at 3 and 12 months after RARP. Increased surgeon experience was adjusted for by demonstrating no consistent learning curve trend between subgroups of 250 consecutive patients over time for weeks to 0 and 1 pads per day (PPD), EBL, and OR time using ANOVA. A p value of less than 0.05 was considered statistically significant. All calculations were completed with SPSS (IBM SPSS Inc., Chicago, IL, USA).
J Robotic Surg (2015) 9:291–297
293
Fig. 2 Positive surgical margin (PSM) locations in group 2 patients without history of previous TURP
Results PSM rates were not statistically different between the two patient groups (p = 0.110) (Table 1), with group 1 patients with a history of TURP having a higher PSM rate (30.61 %) as compared to group 2 patients that did not have a history of TURP (20.95 %). As illustrated in Table 2, age, BMI, Gleason total score, and PSA values were compared between patients that had previous TURP and patients that did not, and there was no statistically significant difference except for age (p = 0.0002). The PSM locations of group 1 patients with prior TURP were further analyzed (Table 2). The majority of PSM locations were at the posterior, apex, and bladder neck. As seen in Table 3, six patients with previous TURP after RARP had PSM at the posterior location (31.58 %) and five patients had PSM at the apex (26.32 %), while four patients had PSM at the bladder neck (21.06 %) (Fig. 3). In terms of EBL, group 1 was measured to be a mean of 128.13 mL, while group 2 was 115.94 mL; this did not yield significance at a p value of 0.159 (Table 1). Operative time in group 1 patients with prior TURP was at 93.85 min compared to 99.30 min in those without prior TURP; however, this also did not reach significance (p = 0.532). With regard to overall ability to sexually function (Table 4), there was no statistical significance in both groups of patients at 3 and 12 months. With regard to urinary symptom outcomes post-RARP, we found one variable with statistical significance between the two groups (Table 4). Weakness in urinary stream (EPIC Questionnaire #4d) was found to be statistically significant at 12 months. Patients with previous TURP had
Table 1 Perioperative and postoperative outcomes of patients with history of TURP vs. patients without history of TURP
a mean score of 1.00, while patients without previous TURP had a mean score of 0.29 at 12 months. Analysis demonstrated that patients with TURP have worse symptoms by a score difference of 0.71 as compared to those without TURP at 12 months (p = 0.012). There was no statistically significant difference in AUASS between the two patient groups at any time point (Table 4). Additional postoperative variables of weeks-to-1-PPD and weeks-to0-PPD did not demonstrate statistically significant difference between the two patient groups (Table 1). No difference was seen in the remaining EPIC questionnaire at any time point (Table 4).
Discussion As age increases risks for both BPH and prostate cancer, there will be increasing prevalence in patients undergoing RARP with a history of TURP [1]. There has been ongoing debate in the literature regarding the effect of prior prostate surgery on outcomes of prostatectomy. Many studies have illustrated no change of perioperative or postoperative outcomes after LRP and open RP [5–9], while other groups present data that disagree with these conclusions [4, 10, 11]. There is an even greater dearth of relevant literature for RARP, and the lack of consensus brings attention to our present study [1, 3, 12]. We present a series that assesses perioperative variables as well as functional and oncological outcomes of RARP patients with prior TURP by comparing their symptoms to those who are surgery naı¨ve. Our results show that patients with a history of TURP did not have a statistically significant higher PSM rate
Variable name
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value
EBL (mL)
0.159
128.13
115.94
Operative time (min)
93.85
99.30
0.532
Weeks to 1 PPD
11.10
8.31
0.660
Weeks to 0 PPD Positive surgical margin rate
11.13 30.61 %
12.21 20.95 %
0.759 0.110
Group 1 patients with history of TURP, group 2 patients without history of TURP
123
294 Table 2 Baseline demographics, preoperative PSA and Gleason score total of the patient population
J Robotic Surg (2015) 9:291–297
Variable name
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value
Age
63.65
59.77
0.0002
BMI (kg/m2)
28.04
113.92
0.8914
Clinical total Gleason score
6.49
6.56
0.5149
Preop PSA
6.91
5.91
0.1586
Group 1 patients with history of TURP, group 2 patients without history of TURP Bold values indicate statistically significant Table 3 AUASS and sexual function outcomes of patients with history of TURP vs. patients without history of TURP Variable name AUASS (n) Overall ability to function sexually (n)
Months after surgery
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value 0.200
3
9.22 (16)
7.48 (886)
12
4.9 (10)
5.88 (542)
0.530
3
2.5 (6)
1.78 (521)
0.117
12
3 (6)
2.34 (18)
0.270
Group 1 patients with history of TURP, group 2 patients without history of TURP
Fig. 3 Positive surgical margin (PSM) location percentages in group 1 vs. group 2 patients
than those without previous prostate surgery (p = 0.110). This is in agreement with similar data results published by Gacci et al., which showed that men treated with TURP before prostatectomy presented an overall incidence of PSM similar to those without previous TURP [13]. However, other studies in literature such as those by Hampton et al. and Gupta et al. found a different trend of higher PSM rates in patients post-TURP versus those who are TURP naı¨ve [1, 12]. Possible explanations for the different study results of PSM rates have been suggested in the literature. Bandhauer et al. have commented that post-TURP patients’ dissection can be technically difficult due to periprostatic fibrosis [8]. Colombo and colleagues have further explained that capsular perforation and fluid absorption during TURP can cause fibrotic post-inflammatory reaction, ultimately resulting in a more difficult bladder neck dissection to negatively affect surgical
123
margins within this region [11]. This could compromise surgical margin rates in the subset of patients with prior TURP that undergo bladder neck sparing during RARP [11]. Our observation is that the post-TURP RARP is not always more difficult. Some patients have much more observed fibrosis than others. These cases will certainly be more difficult, but these changes are not uniform in these patients. The bladder neck anatomy can certainly be challenging and bladder neck sparing can seldom be performed. However, it is unlikely that the observed error at the bladder neck will be a positive margin, but rather a rather large bladder neck that requires reconstruction. There can be some difficulty when initially identifying and entering the bladder neck, but once entered there is usually a fairly clear demarcation at the limit of the prior resection margin that can greatly aid the proper plane of dissection of the bladder neck.
J Robotic Surg (2015) 9:291–297 Table 4 EPIC Questionnaire outcomes of patients with history of TURP vs. patients without history of TURP
295
EPIC Questionnaire # #1 Leaked urine (n) #2 Urinary control (n)
Months after surgery
#4a Dripping urine (n)
Group 2 (non-TURP) mean
p value
3
2.44 (9)
2.05 (555)
0.406
12
4 (6)
3.31 (366)
0.293
3
2.78 (9)
2.90 (558)
0.596
3 (6)
3.26 (368)
0.349
12 #3 PPD (n)
Group 1 (TURP) mean
3
1.29 (19)
1.20 (922)
0.691
12
0.44 (9)
0.50 (535)
0.850
3
1.63 (8)
1.86 (538)
0.594
12
1.33 (6)
1.14 (363)
0.669 0.435
#4b Pain on urination (n)
3
0 (9)
0.15 (533)
12
0 (6)
0.06 (354)
0.660
#4c Urinary bleeding (n)
3 12
0 (9) 0 (6)
0.03 (532) 0.003 (356)
0.722 0.897
#4d Weak stream (n)
3
0.67 (9)
0.49 (533)
0.579
12
1 (6)
0.29 (354)
0.012
3
1.78 (9)
1.46 (540)
0.450
12
1 (6)
1.00 (355)
0.998
3
3.22 (9)
2.65 (553)
0.171
12
2 (6)
1.91 (364)
0.848
#4e Frequent urination (n) #5 Overall function (n)
Group 1 patients with history of TURP, group 2 patients without history of TURP Bold values indicate statistically significant
We controlled for potential confounders of the difference in PSM rates between two groups by showing that there was no difference in BMI, Gleason total score, or PSA values. Our study shows similar PSM rates between patients with and without prior TURP, which can further guide patient care in supporting the notion that prior prostate surgery is not a contraindication to RARP. Our PSM rate of 30.61 % falls well within the published range of 2–59 % in literature [14]. Surgeons should, however, keep this information in mind during preoperative planning, so both the clinician and the patient can anticipate risk and location of positive surgical margins. Additionally, surgical technique may be altered when wider dissection planes are taken with regard to nerve sparing. With regard to continence and urinary functional outcomes, our findings show that weakness in urinary stream at 12 months (EPIC #4d) was significantly worse in patients with previous TURP compared to those that were surgery naı¨ve. Confirming our findings, the study by Gupta et al. also found worse urinary outcomes in patients with prior TURP [1]. To explain these results, Gupta and colleagues hypothesized that certain periprostatic anatomical changes after TURP could serve to worsen urinary outcomes: (1) the urethrovesical anastomosis can be hindered by anatomical obscuration from cicatrization of bladder neck and difficulty in preserving adequate residual urethral length due to periprostatic fibrosis; (2) TURP renders the internal sphincter mechanism deficient
and also places the external sphincter at risk, and (3) increased risk of stricture formation from urethral manipulation and catheterization [1]. However, our results show that the significant differences were derived from follow-up data consisting of six patients at 12 months follow-up. It is difficult to interpret the significance of the results because many patients were lost to follow-up, and this increases the likelihood that significant results were due to selection bias rather than a true significance. A future study with larger patient population at long-term follow-up is necessary to eliminate the possibility of a false-positive finding in our series. In terms of AUASS, our results demonstrate that there were no statistically significant differences between the two groups at any time point in our study. This points out that although patients with prior TURP may experience worse voiding symptoms postoperatively at 12 months (#4d on EPIC questionnaire), their overall urinary symptom quality of life after RARP may not be affected. Since AUASS utilizes a sum score incorporating all aspects of urinary function and symptom outcomes, the lack of statistical significance in AUASS between the two patient groups suggests that prior TURP may have no bothersome impact on overall urinary outcomes for patients undergoing RARP. This is further supported by our finding from the EPIC questionnaire #5 (survey of patients’ overall urinary function), which shows no statistically significant differences at any time point after RARP (Table 4).
123
296
With regard to sexual function, we found no significant differences in short-term sexual function outcomes at 3 and 12 months on comparing patients with prior TURP to those without prior TURP. This is in agreement with literature, as shown by Palisaar et al. and Hung et al. [15, 16]. However, this is in contrast to Colombo and colleagues who reported a worsening of erectile dysfunction when performing RP as a second operation after previous prostate surgery [11]. We utilized the symptom score for the overall ability to function sexually as a quality measure of patient’s sexual function, which expressed the patient’s overall satisfaction with the ability to achieve and maintain erection for penetration during intercourse. QueryUp to 12 months after RARP, both groups of patients in our study had a symptom score of below ‘3’, indicating that the impact of RARP on sexual dysfunction in both groups of patients in our study were similar regardless of the history of previous TURP. Despite our results, the etiology of sexual dysfunction after prostatectomy is multifactorial, with neurogenic factors and thermal injury playing an important role [17]. Consistent with several studies in literature, the estimated blood loss did not approach statistical significance in patients with prior TURP compared to patients that were surgically naı¨ve (p = 0.159). However, multiple studies in literature have also shown an increased blood loss during prostatectomy in patients with previous prostate surgery [4, 10, 11]. Surprisingly, the operative time of RARP was similar between the two groups of patients in our study. There are several limitations to our study that should be considered when interpreting the results. First, several recent modifications of TURP such as transurethral vaporization resection of the prostate and bipolar TURP, as well as several types of laser prostatectomy such as photoselective vaporization and holmium laser enucleation of the prostate, have become available, with each having its advantages and disadvantages [18]. Each modification could potentially result in different bladder neck or periprostatic changes that may influence surgical outcomes. In our study, we have not distinguished which technology was utilized in TURP patients. Second, there was a significant population discrepancy in the number of patients between groups. Therefore, we cannot exclude the contribution of random error from population discrepancies. Third, the significant decrease in follow-up in both patient populations can undermine the results of urinary and sexual function outcomes. Loss to follow-up can lead to selection bias and this effect is greater, as we analyzed patients who were further away in time point from their RARP date. It is difficult to minimize this limitation, since our institution is a large referral center where many patients have major travel constraints to postoperative follow-up. Other factors such as prostate gland size and the individual’s inflammatory response could affect the anatomical and
123
J Robotic Surg (2015) 9:291–297
inflammatory changes post-TURP, and can subsequently have an impact on outcomes after RARP. In comparison to other surgical approaches such as LRP or open RP, RARP appears to be advantageous for navigating the difficult tissue planes of a previously treated surgical field. RARP is able to achieve the advantages of minimally invasive surgery while enhancing the surgeon’s technical boundaries through visualization of three-dimensional anatomic perspectives, wristed instrumentation that maximizes freedom of motion, and preservation of hand–eye coordination [19]. This study is unique in that we have analyzed over 2693 patients from 2007 to 2014 with a follow-up period of up to 12 months postoperatively. Recognizing the lack of conclusive studies with regard to RARP outcomes for patients with previous prostate surgery, we hope our results can help fill the gap that currently exists in the literature. In addition, our study serves as a reference for future prospective trials comparing outcomes after RARP for patients with and without previous prostate surgery. In the future, biochemical recurrence data and complication rates will be supplemented to existing results.
Conclusion In patients undergoing RARP, a history of previous TURP has similar EBL and operative times as those that are TURP naı¨ve. Having prior TURP did not yield a significant increase in PSM rates. Interestingly, there was no significant increase in bladder neck PSM location in patients with prior TURP. The overall ability to function sexually was similar in patients with previous TURP compared to those without TURP. There was significant increased weakness of urinary stream at 12 months after RARP in patients with prior TURP. However, the overall urinary function outcomes appear to be similar. Although having a prior history of prostate surgery can present as a complicating factor, it should not be a contraindication to RARP in experienced hands. Compliance with ethical standards Conflict of interest Yu-Kai Su, Benjamin F. Katz, Shailen S. Sehgal, Sue-Jean S. Yu, Yu-Chen Su, Andrew Lightfoot, Ziho Lee, Elton Llukani, Kelly Monahan, and David I. Lee declare that they have no conflict of interest. Informed consent Informed consent was obtained from Yu-Kai Su, Benjamin F. Katz, Shailen S. Sehgal, Sue-Jean S. Yu, Yu-Chen Su, Andrew Lightfoot, Ziho Lee, Elton Llukani, Kelly Monahan, and David I. Lee. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
J Robotic Surg (2015) 9:291–297
References 1. Gupta NP, Singh P, Nayyar R (2011) Outcomes of robot-assisted radical prostatectomy in men with previous transurethral resection of prostate. BJU Int 108(9):1501–1505 2. Campbell MF et al (2012) Campbell-Walsh urology, vol 3, 10th edn. Elsevier Saunders, Philadelphia 3. Martin AD et al (2009) Does a history of previous surgery or radiation to the prostate affect outcomes of robot-assisted radical prostatectomy? BJU Int 103(12):1696–1698 4. Bernstein AJ et al (2008) Robotic assisted laparoscopic prostatectomy after TURP: a multi-institutional analysis of oncologic and quality of life outcomes. J Urol 179(4, Supplement):347 5. Ramon J et al (1994) Morbidity of radical retropubic prostatectomy following previous prostate resection. J Surg Oncol 55(1):14–19 6. Katz R et al (2006) Laparoscopic radical prostatectomy in patients following transurethral resection of the prostate. Urol Int 77(3):216–221 7. Bujons Tur A et al (2006) Radical prostatectomy in patients with history of transurethral resection of the prostate. Arch Esp Urol 59(5):473–478 8. Bandhauer K, Senn E (1988) Radical retropubic prostatectomy after transurethral prostatic resection. Eur Urol 15(3–4):180–181 9. Leewansangtong S, Taweemonkongsap T (2006) Is laparoscopic radical prostatectomy after transurethral prostatectomy appropriated? J Med Assoc Thai 89(8):1146–1149 10. Jaffe J et al (2007) Surgical outcomes for men undergoing laparoscopic radical prostatectomy after transurethral resection of the prostate. J Urol 178(2):483–487 (discussion 487)
297 11. Colombo R et al (2006) Radical prostatectomy after previous prostate surgery: clinical and functional outcomes. J Urol 176(6 Pt 1):2459–2463 (discussion 2463) 12. Hampton LJ et al (2008) Patients with prior turp undergoing robotic assisted laparoscopic radical prostatectomy have higher positive surgical margin rates. J Urol 179(4, Supplement):606 13. Gacci M et al (2013) The impact of prior TURP on radical prostatectomy surgical margins: a multicenter analysis. Urol Int 91(1):62–68 14. Ficarra V et al (2007) Evidence from robot-assisted laparoscopic radical prostatectomy: a systematic review. Eur Urol 51(1):45–55 (discussion 56) 15. Palisaar JR et al (2009) Open radical retropubic prostatectomy gives favourable surgical and functional outcomes after transurethral resection of the prostate. BJU Int 104(5):611–615 16. Hung CF, Yang CK, Ou YC (2014) Robotic assisted laparoscopic radical prostatectomy following transurethral resection of the prostate: perioperative, oncologic and functional outcomes. Prostate Int 2(2):82–89 17. Menard J et al (2008) Laparoscopic radical prostatectomy after transurethral resection of the prostate: surgical and functional outcomes. Urology 72(3):593–597 18. Muschter R, Bach T, Seitz M (2013) [Transurethral resection of the prostate and current modifications (bipolar,electrovaporization)]. Urologe A 52(3):331–338 19. Ploussard G et al (2009) Is robot assistance affecting operating room time compared with pure retroperitoneal laparoscopic radical prostatectomy? J Endourol 23(6):939–943
123
ORIGINAL ARTICLE
Does previous transurethral prostate surgery affect oncologic and continence outcomes after RARP? Yu-Kai Su1 • Benjamin F. Katz1 • Shailen S. Sehgal1 • Sue-Jean S. Yu1 Yu-Chen Su2 • Andrew Lightfoot1 • Ziho Lee1 • Elton Llukani1 • Kelly Monahan1 • David I. Lee1
•
Received: 19 February 2015 / Accepted: 17 July 2015 / Published online: 8 August 2015 Ó Springer-Verlag London 2015
Abstract We examined the effect of previous transurethral resection of the prostate (TURP) on multiple oncologic and continence outcomes after robotic-assisted radical prostatectomy (RARP). We performed a retrospective cohort study of a total of 2693 patients from 2007 to 2014 who underwent RARP. Patients were stratified into 49 patients who had previous TURP prior to RARP (group 1) and 2644 patients who had no TURP prior to RARP (group 2). We collected operative variables including
& David I. Lee [email protected]; [email protected] Yu-Kai Su [email protected] Benjamin F. Katz [email protected] Shailen S. Sehgal [email protected] Yu-Chen Su [email protected] Andrew Lightfoot [email protected] Ziho Lee [email protected] Elton Llukani [email protected] Kelly Monahan [email protected] 1
2
Department of Urology, University of Pennsylvania Health System, Penn Presbyterian Medical Center, University of Pennsylvania, 300 Medical Office Building, 51 North 39th Street, Philadelphia, PA 19104, USA Division of Biostatistics, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA
estimated blood loss, operative time, and positive surgical margin (PSM) rates. Urinary continence, defined as 0 pads per day (PPD), and social continence, defined as 1-PPD, were also assessed. American Urological Association Symptoms Score (AUASS), overall ability to function sexually, and Expanded Prostate Cancer Index Composite (EPIC) questionnaire were evaluated at 3 and 12 months after RARP. Weakness of urinary stream (EPIC #4d) at 12 months imposed a greater problem for group 1 patients with prior TURP compared to group 2 patients without prior TURP (p = 0.012). PSM was not statistically significant between the two groups (p = 0.110). Group 1 patients had a greater PSM rate (30.61 %) as compared to group 2 (20.95 %). PSM locations in group 1 patients showed the most common locations at the posterior and apex. The difference between the two groups for AUASS, overall sexual function, estimated blood loss, operative time, urinary continence, and social continence was not statistically significant. We examined the effect of previous TURP on postoperative RARP continence and oncologic outcomes. This data can be used to counsel those with prior TURP before RARP. Keywords Robotic prostatectomy Prostate cancer Transurethral prostate surgery TURP Continence outcomes
Introduction With advancing age being a risk factor for both benign prostatic hyperplasia (BPH) and prostate cancer, it is not uncommon for aging men once treated for BPH to later undergo prostatectomy for prostate cancer [1]. Robot-assisted radical prostatectomy (RARP) is now the dominant
123
292
approach to radical prostatectomy (RP) in the USA [2]. With the increasing prevalence and availability of RARP [2], surgeons are expanding this procedure to patients with prior transurethral resection of the prostate (TURP) [3]. However, concerns have been raised regarding outcomes of these cases due to their potential increased complexity [4]. In the open RP literature, a number of studies have described an increase in surgical difficulty of patients with prior prostate surgery as compared to the surgery naı¨ve. However, the perioperative and postoperative morbidity, complication rates, and oncological outcomes in these patients were not statistically different [5–9]. Conversely, other groups have shown that in those with previous prostate surgery, the outcomes are compromised in several ways in patients undergoing either laparoscopic radical prostatectomy (LRP) or open RP [4, 10, 11]. For example, Jaffe and colleagues recently illustrated that men with prior history of TURP undergoing LRP not only have worse outcomes with regard to positive surgical margin (PSM) rates (21.8 vs. 12.6 %), but also have prolonged operative time (OT), and poorer overall complication rates [10]. Moreover, Colombo et al. have suggested that previous prostate surgery can adversely affect urinary continence and erectile function in patients undergoing open RP [11]. Much of the published data in the literature has looked at patients undergoing LRP or open RP, yet only a few studies have investigated the impact of prior prostate surgery history on RARP outcomes. Hampton and colleagues published the data of 51 men with history of TURP and found statistically significant worse PSM rates (35.3 vs. 17.7 %) when compared with men with no TURP [12]. In addition, Gupta et al. found prolonged OT, increased blood loss, increased biochemical recurrence and PSM rates, as well as worse continence outcomes in a group of 26 patients that had undergone previous TURP [1]. On the contrary, Martin and colleagues disagree with such findings by illustrating no significant difference in perioperative and oncological outcomes of RARP when comparing patients with previous prostate surgery with those that are surgery naı¨ve [3]. Thus, there is a paucity of relevant data for outcomes of patients with prior TURP undergoing RARP, and the present study aims to bridge this gap in the existing literature .
Fig. 1 Positive surgical margin (PSM) locations in group 1 patients with history of previous TURP
123
J Robotic Surg (2015) 9:291–297
Our study compares RARP patients with a history of previous prostate surgery with patients that did not undergo previous prostate surgery. Perioperative data, oncological outcomes, sexual function, continence, and urinary symptoms were evaluated.
Materials and methods We performed a retrospective cohort study of a total of 2693 patients from 2007 to 2014 who underwent RARP at our institution by a single surgeon (DIL). The database was collected prospectively and reviewed under an IRB approved protocol. Informed consent was obtained from all patients for being included in the study. We stratified patients into two groups: 49 patients who had previous TURP prior to RARP (group 1) (Fig. 1) and 2644 who did not have TURP prior to RARP (group 2) (Fig. 2). Perioperative and postoperative variables were recorded and analyzed for significant differences in outcomes between groups. Operative variables analyzed were estimated blood loss (EBL), operative time, and PSM rates. Urinary continence, defined as 0 pads per day (PPD), and social continence, defined as 1-PPD, were also assessed. With regard to urinary symptoms and sexual function outcome variables, American Urological Association Symptoms Score (AUASS) and overall ability to function sexually (score from 1 to 5, highest score indicating highest satisfaction with the ability to achieve and maintain erection for penetration) at 3 and 12 months after RARP were examined. Additionally, we also evaluated the results of #1 through #5 from the Expanded Prostate Cancer Index Composite (EPIC-26 short form) Questionnaire at 3 and 12 months after RARP. Increased surgeon experience was adjusted for by demonstrating no consistent learning curve trend between subgroups of 250 consecutive patients over time for weeks to 0 and 1 pads per day (PPD), EBL, and OR time using ANOVA. A p value of less than 0.05 was considered statistically significant. All calculations were completed with SPSS (IBM SPSS Inc., Chicago, IL, USA).
J Robotic Surg (2015) 9:291–297
293
Fig. 2 Positive surgical margin (PSM) locations in group 2 patients without history of previous TURP
Results PSM rates were not statistically different between the two patient groups (p = 0.110) (Table 1), with group 1 patients with a history of TURP having a higher PSM rate (30.61 %) as compared to group 2 patients that did not have a history of TURP (20.95 %). As illustrated in Table 2, age, BMI, Gleason total score, and PSA values were compared between patients that had previous TURP and patients that did not, and there was no statistically significant difference except for age (p = 0.0002). The PSM locations of group 1 patients with prior TURP were further analyzed (Table 2). The majority of PSM locations were at the posterior, apex, and bladder neck. As seen in Table 3, six patients with previous TURP after RARP had PSM at the posterior location (31.58 %) and five patients had PSM at the apex (26.32 %), while four patients had PSM at the bladder neck (21.06 %) (Fig. 3). In terms of EBL, group 1 was measured to be a mean of 128.13 mL, while group 2 was 115.94 mL; this did not yield significance at a p value of 0.159 (Table 1). Operative time in group 1 patients with prior TURP was at 93.85 min compared to 99.30 min in those without prior TURP; however, this also did not reach significance (p = 0.532). With regard to overall ability to sexually function (Table 4), there was no statistical significance in both groups of patients at 3 and 12 months. With regard to urinary symptom outcomes post-RARP, we found one variable with statistical significance between the two groups (Table 4). Weakness in urinary stream (EPIC Questionnaire #4d) was found to be statistically significant at 12 months. Patients with previous TURP had
Table 1 Perioperative and postoperative outcomes of patients with history of TURP vs. patients without history of TURP
a mean score of 1.00, while patients without previous TURP had a mean score of 0.29 at 12 months. Analysis demonstrated that patients with TURP have worse symptoms by a score difference of 0.71 as compared to those without TURP at 12 months (p = 0.012). There was no statistically significant difference in AUASS between the two patient groups at any time point (Table 4). Additional postoperative variables of weeks-to-1-PPD and weeks-to0-PPD did not demonstrate statistically significant difference between the two patient groups (Table 1). No difference was seen in the remaining EPIC questionnaire at any time point (Table 4).
Discussion As age increases risks for both BPH and prostate cancer, there will be increasing prevalence in patients undergoing RARP with a history of TURP [1]. There has been ongoing debate in the literature regarding the effect of prior prostate surgery on outcomes of prostatectomy. Many studies have illustrated no change of perioperative or postoperative outcomes after LRP and open RP [5–9], while other groups present data that disagree with these conclusions [4, 10, 11]. There is an even greater dearth of relevant literature for RARP, and the lack of consensus brings attention to our present study [1, 3, 12]. We present a series that assesses perioperative variables as well as functional and oncological outcomes of RARP patients with prior TURP by comparing their symptoms to those who are surgery naı¨ve. Our results show that patients with a history of TURP did not have a statistically significant higher PSM rate
Variable name
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value
EBL (mL)
0.159
128.13
115.94
Operative time (min)
93.85
99.30
0.532
Weeks to 1 PPD
11.10
8.31
0.660
Weeks to 0 PPD Positive surgical margin rate
11.13 30.61 %
12.21 20.95 %
0.759 0.110
Group 1 patients with history of TURP, group 2 patients without history of TURP
123
294 Table 2 Baseline demographics, preoperative PSA and Gleason score total of the patient population
J Robotic Surg (2015) 9:291–297
Variable name
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value
Age
63.65
59.77
0.0002
BMI (kg/m2)
28.04
113.92
0.8914
Clinical total Gleason score
6.49
6.56
0.5149
Preop PSA
6.91
5.91
0.1586
Group 1 patients with history of TURP, group 2 patients without history of TURP Bold values indicate statistically significant Table 3 AUASS and sexual function outcomes of patients with history of TURP vs. patients without history of TURP Variable name AUASS (n) Overall ability to function sexually (n)
Months after surgery
Group 1 (TURP) mean
Group 2 (non-TURP) mean
p value 0.200
3
9.22 (16)
7.48 (886)
12
4.9 (10)
5.88 (542)
0.530
3
2.5 (6)
1.78 (521)
0.117
12
3 (6)
2.34 (18)
0.270
Group 1 patients with history of TURP, group 2 patients without history of TURP
Fig. 3 Positive surgical margin (PSM) location percentages in group 1 vs. group 2 patients
than those without previous prostate surgery (p = 0.110). This is in agreement with similar data results published by Gacci et al., which showed that men treated with TURP before prostatectomy presented an overall incidence of PSM similar to those without previous TURP [13]. However, other studies in literature such as those by Hampton et al. and Gupta et al. found a different trend of higher PSM rates in patients post-TURP versus those who are TURP naı¨ve [1, 12]. Possible explanations for the different study results of PSM rates have been suggested in the literature. Bandhauer et al. have commented that post-TURP patients’ dissection can be technically difficult due to periprostatic fibrosis [8]. Colombo and colleagues have further explained that capsular perforation and fluid absorption during TURP can cause fibrotic post-inflammatory reaction, ultimately resulting in a more difficult bladder neck dissection to negatively affect surgical
123
margins within this region [11]. This could compromise surgical margin rates in the subset of patients with prior TURP that undergo bladder neck sparing during RARP [11]. Our observation is that the post-TURP RARP is not always more difficult. Some patients have much more observed fibrosis than others. These cases will certainly be more difficult, but these changes are not uniform in these patients. The bladder neck anatomy can certainly be challenging and bladder neck sparing can seldom be performed. However, it is unlikely that the observed error at the bladder neck will be a positive margin, but rather a rather large bladder neck that requires reconstruction. There can be some difficulty when initially identifying and entering the bladder neck, but once entered there is usually a fairly clear demarcation at the limit of the prior resection margin that can greatly aid the proper plane of dissection of the bladder neck.
J Robotic Surg (2015) 9:291–297 Table 4 EPIC Questionnaire outcomes of patients with history of TURP vs. patients without history of TURP
295
EPIC Questionnaire # #1 Leaked urine (n) #2 Urinary control (n)
Months after surgery
#4a Dripping urine (n)
Group 2 (non-TURP) mean
p value
3
2.44 (9)
2.05 (555)
0.406
12
4 (6)
3.31 (366)
0.293
3
2.78 (9)
2.90 (558)
0.596
3 (6)
3.26 (368)
0.349
12 #3 PPD (n)
Group 1 (TURP) mean
3
1.29 (19)
1.20 (922)
0.691
12
0.44 (9)
0.50 (535)
0.850
3
1.63 (8)
1.86 (538)
0.594
12
1.33 (6)
1.14 (363)
0.669 0.435
#4b Pain on urination (n)
3
0 (9)
0.15 (533)
12
0 (6)
0.06 (354)
0.660
#4c Urinary bleeding (n)
3 12
0 (9) 0 (6)
0.03 (532) 0.003 (356)
0.722 0.897
#4d Weak stream (n)
3
0.67 (9)
0.49 (533)
0.579
12
1 (6)
0.29 (354)
0.012
3
1.78 (9)
1.46 (540)
0.450
12
1 (6)
1.00 (355)
0.998
3
3.22 (9)
2.65 (553)
0.171
12
2 (6)
1.91 (364)
0.848
#4e Frequent urination (n) #5 Overall function (n)
Group 1 patients with history of TURP, group 2 patients without history of TURP Bold values indicate statistically significant
We controlled for potential confounders of the difference in PSM rates between two groups by showing that there was no difference in BMI, Gleason total score, or PSA values. Our study shows similar PSM rates between patients with and without prior TURP, which can further guide patient care in supporting the notion that prior prostate surgery is not a contraindication to RARP. Our PSM rate of 30.61 % falls well within the published range of 2–59 % in literature [14]. Surgeons should, however, keep this information in mind during preoperative planning, so both the clinician and the patient can anticipate risk and location of positive surgical margins. Additionally, surgical technique may be altered when wider dissection planes are taken with regard to nerve sparing. With regard to continence and urinary functional outcomes, our findings show that weakness in urinary stream at 12 months (EPIC #4d) was significantly worse in patients with previous TURP compared to those that were surgery naı¨ve. Confirming our findings, the study by Gupta et al. also found worse urinary outcomes in patients with prior TURP [1]. To explain these results, Gupta and colleagues hypothesized that certain periprostatic anatomical changes after TURP could serve to worsen urinary outcomes: (1) the urethrovesical anastomosis can be hindered by anatomical obscuration from cicatrization of bladder neck and difficulty in preserving adequate residual urethral length due to periprostatic fibrosis; (2) TURP renders the internal sphincter mechanism deficient
and also places the external sphincter at risk, and (3) increased risk of stricture formation from urethral manipulation and catheterization [1]. However, our results show that the significant differences were derived from follow-up data consisting of six patients at 12 months follow-up. It is difficult to interpret the significance of the results because many patients were lost to follow-up, and this increases the likelihood that significant results were due to selection bias rather than a true significance. A future study with larger patient population at long-term follow-up is necessary to eliminate the possibility of a false-positive finding in our series. In terms of AUASS, our results demonstrate that there were no statistically significant differences between the two groups at any time point in our study. This points out that although patients with prior TURP may experience worse voiding symptoms postoperatively at 12 months (#4d on EPIC questionnaire), their overall urinary symptom quality of life after RARP may not be affected. Since AUASS utilizes a sum score incorporating all aspects of urinary function and symptom outcomes, the lack of statistical significance in AUASS between the two patient groups suggests that prior TURP may have no bothersome impact on overall urinary outcomes for patients undergoing RARP. This is further supported by our finding from the EPIC questionnaire #5 (survey of patients’ overall urinary function), which shows no statistically significant differences at any time point after RARP (Table 4).
123
296
With regard to sexual function, we found no significant differences in short-term sexual function outcomes at 3 and 12 months on comparing patients with prior TURP to those without prior TURP. This is in agreement with literature, as shown by Palisaar et al. and Hung et al. [15, 16]. However, this is in contrast to Colombo and colleagues who reported a worsening of erectile dysfunction when performing RP as a second operation after previous prostate surgery [11]. We utilized the symptom score for the overall ability to function sexually as a quality measure of patient’s sexual function, which expressed the patient’s overall satisfaction with the ability to achieve and maintain erection for penetration during intercourse. QueryUp to 12 months after RARP, both groups of patients in our study had a symptom score of below ‘3’, indicating that the impact of RARP on sexual dysfunction in both groups of patients in our study were similar regardless of the history of previous TURP. Despite our results, the etiology of sexual dysfunction after prostatectomy is multifactorial, with neurogenic factors and thermal injury playing an important role [17]. Consistent with several studies in literature, the estimated blood loss did not approach statistical significance in patients with prior TURP compared to patients that were surgically naı¨ve (p = 0.159). However, multiple studies in literature have also shown an increased blood loss during prostatectomy in patients with previous prostate surgery [4, 10, 11]. Surprisingly, the operative time of RARP was similar between the two groups of patients in our study. There are several limitations to our study that should be considered when interpreting the results. First, several recent modifications of TURP such as transurethral vaporization resection of the prostate and bipolar TURP, as well as several types of laser prostatectomy such as photoselective vaporization and holmium laser enucleation of the prostate, have become available, with each having its advantages and disadvantages [18]. Each modification could potentially result in different bladder neck or periprostatic changes that may influence surgical outcomes. In our study, we have not distinguished which technology was utilized in TURP patients. Second, there was a significant population discrepancy in the number of patients between groups. Therefore, we cannot exclude the contribution of random error from population discrepancies. Third, the significant decrease in follow-up in both patient populations can undermine the results of urinary and sexual function outcomes. Loss to follow-up can lead to selection bias and this effect is greater, as we analyzed patients who were further away in time point from their RARP date. It is difficult to minimize this limitation, since our institution is a large referral center where many patients have major travel constraints to postoperative follow-up. Other factors such as prostate gland size and the individual’s inflammatory response could affect the anatomical and
123
J Robotic Surg (2015) 9:291–297
inflammatory changes post-TURP, and can subsequently have an impact on outcomes after RARP. In comparison to other surgical approaches such as LRP or open RP, RARP appears to be advantageous for navigating the difficult tissue planes of a previously treated surgical field. RARP is able to achieve the advantages of minimally invasive surgery while enhancing the surgeon’s technical boundaries through visualization of three-dimensional anatomic perspectives, wristed instrumentation that maximizes freedom of motion, and preservation of hand–eye coordination [19]. This study is unique in that we have analyzed over 2693 patients from 2007 to 2014 with a follow-up period of up to 12 months postoperatively. Recognizing the lack of conclusive studies with regard to RARP outcomes for patients with previous prostate surgery, we hope our results can help fill the gap that currently exists in the literature. In addition, our study serves as a reference for future prospective trials comparing outcomes after RARP for patients with and without previous prostate surgery. In the future, biochemical recurrence data and complication rates will be supplemented to existing results.
Conclusion In patients undergoing RARP, a history of previous TURP has similar EBL and operative times as those that are TURP naı¨ve. Having prior TURP did not yield a significant increase in PSM rates. Interestingly, there was no significant increase in bladder neck PSM location in patients with prior TURP. The overall ability to function sexually was similar in patients with previous TURP compared to those without TURP. There was significant increased weakness of urinary stream at 12 months after RARP in patients with prior TURP. However, the overall urinary function outcomes appear to be similar. Although having a prior history of prostate surgery can present as a complicating factor, it should not be a contraindication to RARP in experienced hands. Compliance with ethical standards Conflict of interest Yu-Kai Su, Benjamin F. Katz, Shailen S. Sehgal, Sue-Jean S. Yu, Yu-Chen Su, Andrew Lightfoot, Ziho Lee, Elton Llukani, Kelly Monahan, and David I. Lee declare that they have no conflict of interest. Informed consent Informed consent was obtained from Yu-Kai Su, Benjamin F. Katz, Shailen S. Sehgal, Sue-Jean S. Yu, Yu-Chen Su, Andrew Lightfoot, Ziho Lee, Elton Llukani, Kelly Monahan, and David I. Lee. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
J Robotic Surg (2015) 9:291–297
References 1. Gupta NP, Singh P, Nayyar R (2011) Outcomes of robot-assisted radical prostatectomy in men with previous transurethral resection of prostate. BJU Int 108(9):1501–1505 2. Campbell MF et al (2012) Campbell-Walsh urology, vol 3, 10th edn. Elsevier Saunders, Philadelphia 3. Martin AD et al (2009) Does a history of previous surgery or radiation to the prostate affect outcomes of robot-assisted radical prostatectomy? BJU Int 103(12):1696–1698 4. Bernstein AJ et al (2008) Robotic assisted laparoscopic prostatectomy after TURP: a multi-institutional analysis of oncologic and quality of life outcomes. J Urol 179(4, Supplement):347 5. Ramon J et al (1994) Morbidity of radical retropubic prostatectomy following previous prostate resection. J Surg Oncol 55(1):14–19 6. Katz R et al (2006) Laparoscopic radical prostatectomy in patients following transurethral resection of the prostate. Urol Int 77(3):216–221 7. Bujons Tur A et al (2006) Radical prostatectomy in patients with history of transurethral resection of the prostate. Arch Esp Urol 59(5):473–478 8. Bandhauer K, Senn E (1988) Radical retropubic prostatectomy after transurethral prostatic resection. Eur Urol 15(3–4):180–181 9. Leewansangtong S, Taweemonkongsap T (2006) Is laparoscopic radical prostatectomy after transurethral prostatectomy appropriated? J Med Assoc Thai 89(8):1146–1149 10. Jaffe J et al (2007) Surgical outcomes for men undergoing laparoscopic radical prostatectomy after transurethral resection of the prostate. J Urol 178(2):483–487 (discussion 487)
297 11. Colombo R et al (2006) Radical prostatectomy after previous prostate surgery: clinical and functional outcomes. J Urol 176(6 Pt 1):2459–2463 (discussion 2463) 12. Hampton LJ et al (2008) Patients with prior turp undergoing robotic assisted laparoscopic radical prostatectomy have higher positive surgical margin rates. J Urol 179(4, Supplement):606 13. Gacci M et al (2013) The impact of prior TURP on radical prostatectomy surgical margins: a multicenter analysis. Urol Int 91(1):62–68 14. Ficarra V et al (2007) Evidence from robot-assisted laparoscopic radical prostatectomy: a systematic review. Eur Urol 51(1):45–55 (discussion 56) 15. Palisaar JR et al (2009) Open radical retropubic prostatectomy gives favourable surgical and functional outcomes after transurethral resection of the prostate. BJU Int 104(5):611–615 16. Hung CF, Yang CK, Ou YC (2014) Robotic assisted laparoscopic radical prostatectomy following transurethral resection of the prostate: perioperative, oncologic and functional outcomes. Prostate Int 2(2):82–89 17. Menard J et al (2008) Laparoscopic radical prostatectomy after transurethral resection of the prostate: surgical and functional outcomes. Urology 72(3):593–597 18. Muschter R, Bach T, Seitz M (2013) [Transurethral resection of the prostate and current modifications (bipolar,electrovaporization)]. Urologe A 52(3):331–338 19. Ploussard G et al (2009) Is robot assistance affecting operating room time compared with pure retroperitoneal laparoscopic radical prostatectomy? J Endourol 23(6):939–943
123
