Available online at www.sciencedirect.com
ScienceDirect EJSO 41 (2015) 1540e1546
www.ejso.com
Salvage robot assisted radical prostatectomy: A propensity matched study of perioperative, oncological and functional outcomes A.S. Bates a,*, S. Samavedi a, A. Kumar a, V. Mouraviev a, B. Rocco b, R. Coelho c, K. Palmer a, V.R. Patel a a
Global Robotics Institute, Florida Hospital-Celebration Health, University of Central Florida College of Medicine, USA b Department of Urology, Fondazione Ospedale Maggiore Policlinico, University of Milan, Milan, Italy c University of Sao Paulo State Cancer Institute, Sao Paulo, Brazil Accepted 4 June 2015 Available online 27 June 2015
Abstract Background: To report the perioperative, functional and oncological outcomes of salvage robot-assisted laparoscopic prostatectomy (sRARP) in a propensity score-matched analysis. Study design: 53 patients underwent s-RARP at our institution. Perioperative, functional and oncological outcomes were compared between propensity matched cohorts. Results: Patients in the s-RARP group were at significantly higher risk based on the D’Amico classification system ( p ¼ 0.010). Estimated blood loss, complication rate, hospital stay, BCR risk, persistent cancer and time to return of potency were similar between groups (full nerve spare [NS] n ¼ 22). In the s-RARP cohort there was a higher prevalence of lymphovascular invasion (26.4% versus 13.2%; p ¼ 0.032), time to catheter removal and a higher prevalence of anastomotic leaks in the postoperative period (34.0% vs 5.7%, p < 0.010). The hazard ratio for return to potency regardless of nerve sparing in the s-RARP group was 0.47 (95% CI 0.25e0.88). Significantly more patients undergoing primary RARP with partial nerve sparing (NS) recovered continence (p < 0.001) and potency (p ¼ 0.043) compared to partial NS s-RARP patients. The return to continence and potency did not differ between full NS cases (n ¼ 22; p ¼ 0.616). Conclusions: Salvage RARP patients undergoing surgery have more high risk disease. Patients should be counseled that they are more likely to demonstrate anastomotic leakage on cystogram, and prolonged catheterization times. The time to potency and continence in sRARP undergoing partial and no NS was significantly delayed (n ¼ 49). The proportion of patients returning to potency and continence was also lower in our s-RARP group. Ó 2015 Elsevier Ltd. All rights reserved.
Keywords: Salvage prostatectomy; Robotic assisted radical prostatectomy; Oncological outcomes; Potency; Urinary incontinence
Introduction Prostate cancer is the third most common cause of cancer related death in the United States and is the most prevalent non-cutaneous cancer.1 The Cancer of the Prostate Strategic Urologic Research Endeavour (CaPSURE) database suggests 24.0% of men undergo external beam radiotherapy (EBRT) or brachytherapy (BT) as primary * Corresponding author. Tel.: þ44 7512411978. E-mail address: [email protected] (A.S. Bates). http://dx.doi.org/10.1016/j.ejso.2015.06.002 0748-7983/Ó 2015 Elsevier Ltd. All rights reserved.
treatment for prostate cancer.2 Following treatment of prostate cancer with EBRT, 72.0% of patients with a rising prostate-specific antigen (PSA) are evidenced to have localized disease recurrence.3e5 Local relapse is a significant risk factor for metastatic progression and cancer-specific mortality.3e5 Treatments for patients with local recurrence without metastasis include surgery, cryotherapy and high intensity focused ultrasound.6 Two percent of these patients are subsequently treated with salvage open radical prostatectomy (open SRP),7 and even fewer undergo salvage robotic assisted
A.S. Bates et al. / EJSO 41 (2015) 1540e1546
laparoscopic prostatectomy (s-RARP), due to the technically demanding nature of the procedure. Open salvage radical prostatectomy (SRP) has previously been underutilized due to iatrogenic morbidity directly sustained from the procedure. In recent open SRP series, outcomes have improved and recent open SRP series have reported rates of rectal injury in 15.0%, urinary extravasation in 5.0%, bladder neck contracture in 10.0% and persistent post-prostatectomy incontinence in 20.0%.8e10 Open SRP has been documented to offer biochemical recurrence (BCR) free survival rates from 37.0 to 87.0% and metastasis free survival rate from 64.0 to 95.0%.11 However, there is a wide range of follow up from 18 to 120 months between studies examined.11e13 At the present, a small number of salvage robotic assisted laparoscopic prostatectomy (s-RARP) series are published. These series describe oncological and functional outcomes, yet are limited by relatively short-term follow up.14e20 Rates of oncological control with s-RARP are comparable to those of open SRP,11 while functional outcomes and complication rates are similar to the most favorable outcomes in published open SRP series.11,19 In the present study we provide a comparative analysis of the functional and oncological outcomes of s-RARP cases undertaken at our institution. This comparison between groups is driven by the demand to better inform patients and clinicians of the surgical options currently available for the management of locoregional recurrence of prostate cancer. Methods Patient population From January 2008 through April 2014, 5279 consecutive patients underwent RARP for prostate cancer. All procedures were performed by a single surgeon using the transperitoneal, six-port technique with the DaVinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA). There were 53 patients in this series of s-RARP, all of whom received any form of non-surgical intervention treatment with a PSA concentration of >0.2 ng/mL and local recurrence confirmed. Twenty eight patients (52.8%) had external beam radiation, 14 (26.4%) brachytherapy, 5 (9.4%) intensity modulated radiotherapy, 3 (5.6%) cryotherapy, and 3 (5.6%) high intensity focused ultrasound. Thirteen of these patients (24.5%) received external beam radiation and brachytherapy. The study cohort (n ¼ 53) was then computer-matched using an optimal matching algorithm to 53 non s-RARP patients. Data collection and management Perioperative data were collected retrospectively and entered into an Institutional Review Board approved database. Data collected are detailed (Supplement 1). Followup information was obtained through comprehensive
1541
validated questionnaires and entered into the follow-up component of the database by an interviewer (blinded to primary treatment). Follow-up data were collected at 6 weeks, 3-, 6-, 9-, 12-, 18-months and at 1 year intervals to assess functional and oncologic outcomes. Pathologic specimen analyses are detailed (Supplement 1).21,22 Statistical analysis Demographic and clinical data are presented as frequency distribution and simple percentages. Continuous variables are expressed as the mean the standard deviation and median. Univariate analysis of selected perioperative discrete variables was accomplished by the chi-square test with the appropriate degrees of freedom or the Fisher’s Exact test to assess the equality of proportions. Two-sample t-tests were used to test the equality of means in continuous variables. The within-group comparisons of patient pain scores were accomplished using one way analysis of variance. Each s-RARP patient was assigned for comparison against a RARP case using propensity matching.23,24 Propensity score matching To control for measured potential confounders in the data set, a propensity score was generated for each patient from a multivariable logistic regression model based on 9 preoperative clinical covariates: age, race, body mass index, hypertension, diabetes, Sexual Health Inventory for Men (SHIM) score, American Urological Association (AUA) Symptom score, PSA level, clinical stage and Gleason score as independent variables, with s-RALP versus RALP as the binary dependent variable (Further information in Supplement). Results Preoperative data Propensity score-matched s-RARP (n ¼ 53) and RARP (n ¼ 53) preoperative data are presented in Table 1. Mean age, BMI and ethnicity did not differ between groups. Further preoperative characteristics are provided (Supplement 2). Intraoperative and pathological data In the s-RARP group, 7 patients (13.2%) did not undergo nerve sparing, 42 (79.3%) underwent partial nerve sparing and 4 (7.6%) full nerve sparing. In the primary RARP group, 35 patients (66.0%) underwent partial nerve sparing and 18 (34.0%) full nerve sparing ( p < 0.001), (Table 4; Supplemental file). Total operative time in the s-RARP group was slightly but not clinically longer at 130.0 17.0 min (median 128.0) versus 121.7 18.7 min (median 118.0) in the RARP group ( p ¼ 0.022). Mean estimated blood loss (EBL) was slightly
1542
A.S. Bates et al. / EJSO 41 (2015) 1540e1546
Table 1 Comparison of clinical characteristics in propensity score-matched salvage and nonsalvage patients undergoing robot assisted radical prostatectomy. Variable
s-RARP
Total Patients 53 (100.0) Age (y) Mean SD 65.4 6.9 Median 67.0 Race Caucasian 44 (83.0) Other 9 (17.0) Body Mass Index, kg/m2 Mean SD 28.6 4.4 Median 28.3 Hypertension 23 (56.6) Diabetes 9 (17.0) SHIM Score Mean SD 14.0 8.7 Median 15.0 Preoperative AUA Score Mean SD 9.5 7.5 Median 8.0 Prostate Specific Antigen Mean SD 4.7 3.2 Median 3.7 Clinical Stage T Stage T1eT2 44 (83.0) Stage T3eT4 9 (17.0) Preoperative Gleason Score 6 14 (26.4) 7 23 (43.4) 8 16 (30.2) D’Amico Risk Classification Low 8 (15.1) Intermediate 15 (28.3) High 14 (26.4) Unavailable 16 (30.2)
RARP 53 (100.0)
p Value 0.693
65.9 6.4 66.0 50 (94.3) 3 (5.7)
0.123
29.5 4.2 28.8 26 (49.1) 4 (7.6)
0.267 0.436 0.236
13.0 8.8 14.0
0.536
10.5 7.3 9.0
0.472
5.2 2.1 4.7
0.306
53 (100.0) 0 (0.0)
0.002
14 (26.4) 24 (45.3) 15 (28.3)
0.936
13 (24.5) 25 (47.2) 15 (28.3) 0 (0.0)
0.05). There were no blood transfusions administered in either group. There were no intraoperative or anesthesiarelated complications in either group. There were no open conversions in the series, and no rectal or bowel injuries. Histopathological findings are detailed (Supplement 2) (Table 2).
Postoperative complications and hospital events The overall incidence of significant postoperative morbidity for the two groups was low with only one patient (1.9%) in the s-RARP group reaching Grade II according to the Clavien Classification System. The patient developed a pulmonary embolism on the third postoperative day, from which he fully recovered. No complications were recorded in the RARP match group.
Table 2 Comparison of intraoperative and pathologic outcomes in propensity score-matched salvage and nonsalvage patients undergoing robot assisted radical prostatectomy. Variable Operative Time (min) Mean SD Median Console Time (min) Mean SD Median Estimated Blood Loss (mL) Mean SD Median Nerve Sparing Non nerve sparing Partial Full Difficulty of Anastomosis No difficulty Partial Full Prostate Weight (g) Mean SD Median Prostate Density Mean SD Median Pathologic Stage T T2 T3eT4 Postoperative Gleason Score Not available 6 7 8 Positive Surgical Margins Extracapsular Extension Seminal Vesicle Invasion Lymphovascular Invasion Perineural Invasion Seminal Vesicle Invasion Tumor Volume (%) Mean SD Median Tumor Dimension (cm) Mean SD Median
s-RARP
RARP
p Value
130.0 17.8 128.0
121.7 18.7 118.0
0.022
82.7 13.5 80.0
78.1 25.7 80.0
0.261
120.4 57.2 100.0
132.4 67.1 100.0
0.325
7 (13.2) 42 (79.3) 4 (7.6)
0 (0.0) 35 (66.0) 18 (34.0)
ScienceDirect EJSO 41 (2015) 1540e1546
www.ejso.com
Salvage robot assisted radical prostatectomy: A propensity matched study of perioperative, oncological and functional outcomes A.S. Bates a,*, S. Samavedi a, A. Kumar a, V. Mouraviev a, B. Rocco b, R. Coelho c, K. Palmer a, V.R. Patel a a
Global Robotics Institute, Florida Hospital-Celebration Health, University of Central Florida College of Medicine, USA b Department of Urology, Fondazione Ospedale Maggiore Policlinico, University of Milan, Milan, Italy c University of Sao Paulo State Cancer Institute, Sao Paulo, Brazil Accepted 4 June 2015 Available online 27 June 2015
Abstract Background: To report the perioperative, functional and oncological outcomes of salvage robot-assisted laparoscopic prostatectomy (sRARP) in a propensity score-matched analysis. Study design: 53 patients underwent s-RARP at our institution. Perioperative, functional and oncological outcomes were compared between propensity matched cohorts. Results: Patients in the s-RARP group were at significantly higher risk based on the D’Amico classification system ( p ¼ 0.010). Estimated blood loss, complication rate, hospital stay, BCR risk, persistent cancer and time to return of potency were similar between groups (full nerve spare [NS] n ¼ 22). In the s-RARP cohort there was a higher prevalence of lymphovascular invasion (26.4% versus 13.2%; p ¼ 0.032), time to catheter removal and a higher prevalence of anastomotic leaks in the postoperative period (34.0% vs 5.7%, p < 0.010). The hazard ratio for return to potency regardless of nerve sparing in the s-RARP group was 0.47 (95% CI 0.25e0.88). Significantly more patients undergoing primary RARP with partial nerve sparing (NS) recovered continence (p < 0.001) and potency (p ¼ 0.043) compared to partial NS s-RARP patients. The return to continence and potency did not differ between full NS cases (n ¼ 22; p ¼ 0.616). Conclusions: Salvage RARP patients undergoing surgery have more high risk disease. Patients should be counseled that they are more likely to demonstrate anastomotic leakage on cystogram, and prolonged catheterization times. The time to potency and continence in sRARP undergoing partial and no NS was significantly delayed (n ¼ 49). The proportion of patients returning to potency and continence was also lower in our s-RARP group. Ó 2015 Elsevier Ltd. All rights reserved.
Keywords: Salvage prostatectomy; Robotic assisted radical prostatectomy; Oncological outcomes; Potency; Urinary incontinence
Introduction Prostate cancer is the third most common cause of cancer related death in the United States and is the most prevalent non-cutaneous cancer.1 The Cancer of the Prostate Strategic Urologic Research Endeavour (CaPSURE) database suggests 24.0% of men undergo external beam radiotherapy (EBRT) or brachytherapy (BT) as primary * Corresponding author. Tel.: þ44 7512411978. E-mail address: [email protected] (A.S. Bates). http://dx.doi.org/10.1016/j.ejso.2015.06.002 0748-7983/Ó 2015 Elsevier Ltd. All rights reserved.
treatment for prostate cancer.2 Following treatment of prostate cancer with EBRT, 72.0% of patients with a rising prostate-specific antigen (PSA) are evidenced to have localized disease recurrence.3e5 Local relapse is a significant risk factor for metastatic progression and cancer-specific mortality.3e5 Treatments for patients with local recurrence without metastasis include surgery, cryotherapy and high intensity focused ultrasound.6 Two percent of these patients are subsequently treated with salvage open radical prostatectomy (open SRP),7 and even fewer undergo salvage robotic assisted
A.S. Bates et al. / EJSO 41 (2015) 1540e1546
laparoscopic prostatectomy (s-RARP), due to the technically demanding nature of the procedure. Open salvage radical prostatectomy (SRP) has previously been underutilized due to iatrogenic morbidity directly sustained from the procedure. In recent open SRP series, outcomes have improved and recent open SRP series have reported rates of rectal injury in 15.0%, urinary extravasation in 5.0%, bladder neck contracture in 10.0% and persistent post-prostatectomy incontinence in 20.0%.8e10 Open SRP has been documented to offer biochemical recurrence (BCR) free survival rates from 37.0 to 87.0% and metastasis free survival rate from 64.0 to 95.0%.11 However, there is a wide range of follow up from 18 to 120 months between studies examined.11e13 At the present, a small number of salvage robotic assisted laparoscopic prostatectomy (s-RARP) series are published. These series describe oncological and functional outcomes, yet are limited by relatively short-term follow up.14e20 Rates of oncological control with s-RARP are comparable to those of open SRP,11 while functional outcomes and complication rates are similar to the most favorable outcomes in published open SRP series.11,19 In the present study we provide a comparative analysis of the functional and oncological outcomes of s-RARP cases undertaken at our institution. This comparison between groups is driven by the demand to better inform patients and clinicians of the surgical options currently available for the management of locoregional recurrence of prostate cancer. Methods Patient population From January 2008 through April 2014, 5279 consecutive patients underwent RARP for prostate cancer. All procedures were performed by a single surgeon using the transperitoneal, six-port technique with the DaVinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA). There were 53 patients in this series of s-RARP, all of whom received any form of non-surgical intervention treatment with a PSA concentration of >0.2 ng/mL and local recurrence confirmed. Twenty eight patients (52.8%) had external beam radiation, 14 (26.4%) brachytherapy, 5 (9.4%) intensity modulated radiotherapy, 3 (5.6%) cryotherapy, and 3 (5.6%) high intensity focused ultrasound. Thirteen of these patients (24.5%) received external beam radiation and brachytherapy. The study cohort (n ¼ 53) was then computer-matched using an optimal matching algorithm to 53 non s-RARP patients. Data collection and management Perioperative data were collected retrospectively and entered into an Institutional Review Board approved database. Data collected are detailed (Supplement 1). Followup information was obtained through comprehensive
1541
validated questionnaires and entered into the follow-up component of the database by an interviewer (blinded to primary treatment). Follow-up data were collected at 6 weeks, 3-, 6-, 9-, 12-, 18-months and at 1 year intervals to assess functional and oncologic outcomes. Pathologic specimen analyses are detailed (Supplement 1).21,22 Statistical analysis Demographic and clinical data are presented as frequency distribution and simple percentages. Continuous variables are expressed as the mean the standard deviation and median. Univariate analysis of selected perioperative discrete variables was accomplished by the chi-square test with the appropriate degrees of freedom or the Fisher’s Exact test to assess the equality of proportions. Two-sample t-tests were used to test the equality of means in continuous variables. The within-group comparisons of patient pain scores were accomplished using one way analysis of variance. Each s-RARP patient was assigned for comparison against a RARP case using propensity matching.23,24 Propensity score matching To control for measured potential confounders in the data set, a propensity score was generated for each patient from a multivariable logistic regression model based on 9 preoperative clinical covariates: age, race, body mass index, hypertension, diabetes, Sexual Health Inventory for Men (SHIM) score, American Urological Association (AUA) Symptom score, PSA level, clinical stage and Gleason score as independent variables, with s-RALP versus RALP as the binary dependent variable (Further information in Supplement). Results Preoperative data Propensity score-matched s-RARP (n ¼ 53) and RARP (n ¼ 53) preoperative data are presented in Table 1. Mean age, BMI and ethnicity did not differ between groups. Further preoperative characteristics are provided (Supplement 2). Intraoperative and pathological data In the s-RARP group, 7 patients (13.2%) did not undergo nerve sparing, 42 (79.3%) underwent partial nerve sparing and 4 (7.6%) full nerve sparing. In the primary RARP group, 35 patients (66.0%) underwent partial nerve sparing and 18 (34.0%) full nerve sparing ( p < 0.001), (Table 4; Supplemental file). Total operative time in the s-RARP group was slightly but not clinically longer at 130.0 17.0 min (median 128.0) versus 121.7 18.7 min (median 118.0) in the RARP group ( p ¼ 0.022). Mean estimated blood loss (EBL) was slightly
1542
A.S. Bates et al. / EJSO 41 (2015) 1540e1546
Table 1 Comparison of clinical characteristics in propensity score-matched salvage and nonsalvage patients undergoing robot assisted radical prostatectomy. Variable
s-RARP
Total Patients 53 (100.0) Age (y) Mean SD 65.4 6.9 Median 67.0 Race Caucasian 44 (83.0) Other 9 (17.0) Body Mass Index, kg/m2 Mean SD 28.6 4.4 Median 28.3 Hypertension 23 (56.6) Diabetes 9 (17.0) SHIM Score Mean SD 14.0 8.7 Median 15.0 Preoperative AUA Score Mean SD 9.5 7.5 Median 8.0 Prostate Specific Antigen Mean SD 4.7 3.2 Median 3.7 Clinical Stage T Stage T1eT2 44 (83.0) Stage T3eT4 9 (17.0) Preoperative Gleason Score 6 14 (26.4) 7 23 (43.4) 8 16 (30.2) D’Amico Risk Classification Low 8 (15.1) Intermediate 15 (28.3) High 14 (26.4) Unavailable 16 (30.2)
RARP 53 (100.0)
p Value 0.693
65.9 6.4 66.0 50 (94.3) 3 (5.7)
0.123
29.5 4.2 28.8 26 (49.1) 4 (7.6)
0.267 0.436 0.236
13.0 8.8 14.0
0.536
10.5 7.3 9.0
0.472
5.2 2.1 4.7
0.306
53 (100.0) 0 (0.0)
0.002
14 (26.4) 24 (45.3) 15 (28.3)
0.936
13 (24.5) 25 (47.2) 15 (28.3) 0 (0.0)
0.05). There were no blood transfusions administered in either group. There were no intraoperative or anesthesiarelated complications in either group. There were no open conversions in the series, and no rectal or bowel injuries. Histopathological findings are detailed (Supplement 2) (Table 2).
Postoperative complications and hospital events The overall incidence of significant postoperative morbidity for the two groups was low with only one patient (1.9%) in the s-RARP group reaching Grade II according to the Clavien Classification System. The patient developed a pulmonary embolism on the third postoperative day, from which he fully recovered. No complications were recorded in the RARP match group.
Table 2 Comparison of intraoperative and pathologic outcomes in propensity score-matched salvage and nonsalvage patients undergoing robot assisted radical prostatectomy. Variable Operative Time (min) Mean SD Median Console Time (min) Mean SD Median Estimated Blood Loss (mL) Mean SD Median Nerve Sparing Non nerve sparing Partial Full Difficulty of Anastomosis No difficulty Partial Full Prostate Weight (g) Mean SD Median Prostate Density Mean SD Median Pathologic Stage T T2 T3eT4 Postoperative Gleason Score Not available 6 7 8 Positive Surgical Margins Extracapsular Extension Seminal Vesicle Invasion Lymphovascular Invasion Perineural Invasion Seminal Vesicle Invasion Tumor Volume (%) Mean SD Median Tumor Dimension (cm) Mean SD Median
s-RARP
RARP
p Value
130.0 17.8 128.0
121.7 18.7 118.0
0.022
82.7 13.5 80.0
78.1 25.7 80.0
0.261
120.4 57.2 100.0
132.4 67.1 100.0
0.325
7 (13.2) 42 (79.3) 4 (7.6)
0 (0.0) 35 (66.0) 18 (34.0)
