Assessment of Lymph Node Yield After Pelvic Lymph Node Dissection in Men with Prostate Cancer: A Comparison Between Robot-Assisted Radical Prostatectomy and Open Radical Prostatectomy in the Modern Era

Abstract

Background and Purpose:

Studies of radical prostatectomy (RP) suggest that higher lymph node yield (LNY) improves tumor staging. Robot-assisted radical prostatectomy (RARP) is becoming increasingly popular, yet LNY data are not well reported. We compare LNY from contemporary open RP (ORP) with RARP at an academic center.

Patients and Methods:

A retrospective study was conducted of an Urologic Oncology Database. Between January 2005 and November 2009, 217 men underwent ORP with pelvic lymph node dissection (PLND); 99 underwent RARP with PLND by a single surgeon during the same period. Men were stratified according to the D'Amico risk criteria. For intermediate and high-risk disease, an extended PLND was performed. Patient demographic, operative, and pathologic variables were measured, and LNY was compared across groups.

Results:

No significant differences were seen between groups for race, body mass index, preoperative prostate-specific antigen level or biopsy Gleason score. Patients were younger for RARP vs ORP (P = 0.003) and had higher clinical tumor stage (P = 0.02). Operative time was longer (P = 0.03) and estimated blood loss was greater (P < 0.001) in the ORP group. Overall, only a borderline significant difference was seen in LNY between ORP and RARP (7.49 vs 6.35 nodes, respectively, P = 0.06). No difference was seen for intermediate and high-risk patients, with 7.7 vs 6.8 nodes for ORP and RARP, respectively (P = 0.27). The lymph node metastasis rate was 6.3%, with more positive nodes detected during ORP vs RARP: 19/217 (8.8%) vs 1/99 (1.0%), P = 0.009.

Conclusions:

No significant differences were seen in LNY during RARP and ORP for intermediate and high-risk men. For experienced surgeons, RARP can achieve equivalent LNY as ORP. A future study with a larger sample size is necessary to make a definitive statement of equivalence.

Introduction

P elvic lymph node dissection (PLND) can be used as a staging tool for men who are undergoing radical prostatectomy (RP) for clinically localized prostate cancer. This procedure is especially important for patients with intermediate and high-risk prostate cancer, given their higher risk of lymph node involvement.¹ Several open radical prostatectomy (ORP) series suggest that higher lymph node yield (LNY) improves the detection of lymph node involvement (LNI), thus enhancing the accuracy of tumor staging.^2

–5 A study of patients who were undergoing radical retropubic prostatectomy found that increasing the number of nodes sampled from 10 to 20 improved the probability of detecting LNI from 10% to 65%, respectively.² In addition, a contemporary study on patients who were undergoing laparoscopic PLND for prostate cancer found that increasing the mean number of nodes removed from 10.1 to 13.1 resulted in an improved LNI detection rate from 4.1% to 11.4%.⁶ Such studies suggest that when PLND is indicated, the greater number of nodes taken significantly improves nodal staging accuracy.

Although studies seem to conclusively demonstrate that staging accuracy and detection of LNI improves with the greater number of nodes sampled, there is less agreement about the impact of nodal yield on prostate cancer-specific survival. A study by Joslyn and Konety⁷ in 2006 of 13,020 men who underwent RP as primary treatment for prostate cancer found that men who had 4 to 6 or ≥10 nodes removed had improved cancer-specific survival at 10 years compared with men without PLND, even if the findings for the removed nodes were negative. In contrast, a study by DiMarco and associates⁸ in 2005 of 7,036 men with clinical T₁ to T₃ disease who underwent radical retropubic prostatectomy and PLND with negative nodes on pathologic examination found no cancer-specific survival difference at 10 years for those with greater lymph nodes sampled. Such findings call into question the cancer-specific survival advantage of extended PLND in prostate cancer patients.

Given that PLND remains an important method of staging in intermediate to high-risk prostate cancer patients, it is important that sufficient lymph nodes can be sampled regardless of surgical approach. Unfortunately, the PLND yield data for robot-assisted radical prostatectomy (RARP) are not well reported, leading some experts to question whether surgeons who are performing RARP retrieve an adequate LNY. Research has shown that PLND during RARP has no significant impact on length of hospital stay, estimated blood loss, or complications.^9,10 LNY from RARP has been reported to range from 3.3 to 14.1, although these studies have been of relatively small samples that range from 40 to 62 patients.^9,11,12 A larger study of LNY during RARP of 296 patients found a mean number of 12.5 nodes on a bilateral, standard template PLND.¹⁰

To date, the three studies that compare LNY between ORP and RARP have resulted in conflicting findings, with two studies reporting a statistically significant higher yield in ORP (7.3 vs 3.3, P < 0.001)¹² and (15 vs 12.5, P < 0.01)¹⁰ and the third reporting no significant difference with yields of 8.6, 7.3, and 14.8 (P = 0.84) for RARP, ORP (standard PLND), and ORP (extended PLND [ePLND]), respectively.¹¹

Given that PLND can be a challenging procedure, requiring skilled surgeons,^13,14 we sought to evaluate and compare LNY from contemporary ORP with the yield from RARP by high-volume surgeons at an academic institution. We hypothesize that LNY is equivalent for intermediate- and high-risk patients who are undergoing ORP and RARP in the contemporary era.

Patients and Methods

A retrospective review was conducted of an Urologic Oncology Database that was approved by the Institutional Review Board at a single academic institution. Patients who had undergone RP via either the open or robot-assisted approach with concurrent PLND for histologically proven, clinically localized prostate cancer were included. RARP cases were limited to a single, high-volume surgeon (KKB). ORP cases were limited to those performed at a single institution by four high-volume surgeons (JMM, MB, CO, and EG) between January 2005 and November 2009 to ensure a temporally related comparison. For ORP, PLND was performed via the lower abdominal midline incision for prostatectomy. For RARP, the PLND was performed via the identical port site configuration used for prostatectomy.

At the time of surgery, all low-risk patients according to the D'Amico risk group classification¹⁵ underwent a standard PLND defined as dissection anteriorly to the external iliac artery, posteriorly to the obturator nerve, distally to the node of Cloquet, and proximally to the bifurcation of the iliac vessels. In men with intermediate or high-risk disease, an ePLND was performed in which nodes were also sampled above the bifurcation of the common iliac artery in addition to the external iliac, obturator and hypogastric lymph nodes.¹⁶ Lymph nodes were dissected en bloc as a packet from each node region and submitted for histopathologic examination.

En bloc dissected “nodal” tissue specimens that were submitted for pathologic evaluation were dissected, fixed in 10% neutral buffered formalin, and embedded in paraffin. Slides were prepared from paraffin-embedded tissue and were stained with hematoxylin and eosin, and submitted for analysis under light microscopy by an individual genitourinary (GU) pathologist from a team consisting of seven dedicated GU pathologists at a single academic institution. Specimens were prepared according to the protocol described in our institution's Surgical Pathology Laboratory Manual.

Lymph nodes were counted and reported as positive or negative for adenocarcinoma. For positive nodes, degree of involvement and occurrence of extranodal extension were noted. The numbers of positive and negative nodes were reported with respect to their anatomic location, as described by the attending urologist. The data was used to determine the pathologic stage of each cancer, per the American Joint Committee on Cancer Staging Handbook, 6th edition.

Patient demographic data, preoperative serum prostate-specific antigen (PSA) level, preoperative biopsy Gleason sum, and clinical stage, assessed by digital rectal examination (DRE) and preoperative transrectal ultrasonography (TRUS)-guided prostate biopsy, were measured. In addition, intraoperative variables, including operative time and estimated blood loss, were recorded. Nodal yield outcomes were evaluated across surgical approach groups according to the D'Amico risk group classification.

Chi-square analysis and Student t test were used to compare categorical and continuous variables across surgical approach subgroups. Univariate and multivariate logistic regression analyses were conducted to determine the association of baseline and preoperative variables with LNY. A two-sided P value of < 0.05 was used as the threshold for statistical significance. All statistical analyses were conducted using STATA statistical software. We conducted a power calculation for this study to determine if the study is adequately powered (0.80) to detect a clinically significant difference, if one exists. Using our sample size (217/99), alpha = 0.05, and LNY mean ratio of 7.5/6.4, we calculated the power of our study to be 0.5166.

Results

A total of 316 men with prostate cancer underwent PLND at our institution with 217 having ORP and 99 having RARP. Patient demographic information is summarized in Table 1. Patients were significantly younger at the time of surgery for RARP vs ORP, with mean ages of 59.2 and 61.7, respectively (P = 0.003). In addition, 70.5% of ORP patients were of intermediate or high-risk disease vs only 56.6% of RARP patients (P = 0.05). No significant differences were seen between groups for race, preoperative PSA level, preoperative biopsy Gleason sum, or body mass index. Men who were undergoing RARP had a slightly higher clinical tumor stage according to preoperative DRE and TRUS prostate biopsy results (P = 0.02).

Table 1.

Baseline Demographic Information Stratified by Surgical Approach

	ORP	RARP	Total	P-value
Sample size	217	99	316
Age	61.7 ± 6.8	59.2 ± 7.1	60.9 ± 7.0	0.003^a
Race				0.12^b
Caucasian	107 (49.3%)	49 (49.5%)	156 (49.4%)
Black	45 (20.7%)	22 (22.2%)	67 (21.2%)
Hispanic	43 (19.8%)	11 (11.1%)	54 (17.1%)
Other	22 (10.1%)	17 (17.2%)	39 (12.3%)
D'Amico risk group				0.05^b
Low	64 (29.5%)	43 (43.4%)	107 (33.9%)
Intermediate	94 (43.3%)	36 (36.4%)	130 (41.1%)
High	59 (27.2%)	20 (20.2%)	79 (25.0%)
Preoperative PSA (ng/mL)	8.35 ± 7.62	7.04 ± 7.5	7.93 ± 7.60	0.17^a
Preoperative Gleason sum				0.15^b
≤ 6	63 (29.0%)	28 (28.3%)	91 (28.8%)
7	95 (43.8%)	34 (34.3%)	129 (40.8%)
≥ 8	59 (27.2%)	37 (37.4%)	96 (30.4%)
Clinical tumor stage				0.02^b
≤ T_2a	155 (71.4%)	57 (57.6%)	212 (67.1%)
T_2b	12 (5.5%)	4 (4.0%)	16 (5.1%)
≥ T_2c	50 (23.0%)	38 (38.4%)	88 (27.8%)
Body mass index (kg/m²)	23.1 ± 9.1	24.6 ± 8.3	23.6 ± 8.8	0.26^a
Median year of surgery (range)	2006 (2005–2009)	2008 (2007–2009)	2007 (2005–2009)	< 0.001^b

Student's t test.

Chi-square test.

ORP = open radical prostatectomy; RARP = robot-assisted radical prostatectomy; PSA = prostate-specific antigen.

Operative and postoperative details are summarized in Table 2. Operative time was shorter for RARP compared with ORP with mean times of 153.4 min vs 204.0 min (P = 0.03). In addition, estimated blood loss was significantly higher for ORP vs RARP with mean volumes reported of 940.5 mL and 157.7 mL, respectively (P < 0.001). There was no significant difference seen across surgical approach groups for postoperative Gleason sum or pathologic tumor stage.

Table 2.

Postoperative Pathologic Data Stratified by Surgical Approach

	ORP	RARP	Total	P value
LNY (total sample)	7.49 ± 5.10	6.35 ± 4.52	7.13 ± 4.95	0.06^a
LNY (intermediate and high risk patients)	7.67 ± 5.41	6.79 ± 4.37	7.43 ± 5.15	0.27^a
Postoperative Gleason sum				0.07^b
≤ 6	26 (12.0%)	14 (14.1%)	40 (12.7%)
7	135 (62.2%)	71 (71.7%)	206 (65.2%)
≥ 8	56 (25.8%)	14 (14.1%)	70 (22.2%)
Pathologic tumor stage				0.24^b
T₂	136 (63.8%)	71 (72.4%)	207 (66.6%)
T₃	70 (32.9%)	23 (23.5%)	93 (29.9%)
T₄	7 (3.3%)	4 (4.1%)	11 (3.5%)
Pathologic nodal stage				0.009^b
N₀	198 (91.2%)	98 (99.0%)	296 (93.7%)
N₁	19 (8.8%)	1 (1.0%)	20 (6.3%)
Operative time (min)	204.0 ± 32.9	153.4 ± 51.3	156.5 ± 51.7	0.03^a
Estimated blood loss (mL)	940.5 ± 615.0	157.7 ± 105.1	636.1 ± 617.2	< 0.001^a

Student t test.

Chi-quare test.

ORP = open radical prostatectomy; RARP = robot-assisted radical prostatectomy; LNY = lymph node yield.

Overall, a mean of 7.13 ± 4.95 nodes were removed at each surgery with borderline significant difference seen between surgical approaches with individual means of 7.49 and 6.35 nodes for ORP and RARP, respectively (P = 0.06). When stratified by preoperative D'Amico risk criteria, no significant difference was seen for intermediate and high-risk patients with means for ORP and RARP of 7.67 ± 5.41 and 6.79 ± 4.37, respectively (P = 0.27). Significantly more nodes were removed for the D'Amico high-risk patients in which an ePLND was performed compared with D'Amico low-risk patients in whom a standard PLND was conducted across all patients, with mean LNY of 8.0 ± 5.3 vs 6.5 ± 4.5 nodes (P = 0.04). When stratified by surgical approach, no significant difference was seen for LNY between high and low-risk patients (ie, ePLND vs standard PLND). Mean LNY for RALP high- vs low-risk patients was 7.4 vs 5.8 (P = 0.2). Mean LNY for ORP high- vs low-risk patients was 8.2 vs 7.0 (P = 0.2). The overall lymph node positivity rate was 6.3%, with a greater number of positive nodes detected during ORP vs RARP: 19/217 (8.8%) vs 1/99 (1.0%), P = 0.009.

Discussion

In recent years, the number of RARPs performed for the treatment of prostate cancer has increased at a rapid pace. Measurable benefits, including decreased operative time, lower blood loss, smaller incisions, and decreased length of hospital stays have made this surgical approach appealing for both surgeons and patients alike. For patients with more aggressive disease, PLND has been recommended at the time of surgery to ensure accurate staging, thus facilitating optimal risk assessment and postoperative treatment planning. Studies of RP series suggest that greater LNY improves accuracy of staging, although the yield in RARP is not well characterized. Given that surgeons are continuing to become more technically skilled at performing RARP, we sought to compare LNY between contemporary RARP and ORP at a single, high-volume academic surgical center to assess whether differences exist in the number of nodes sampled for the two operative approaches.

Overall, a LNY of 7.1 was seen in our series, with yields of 7.5 and 6.4 for ORP and RARP, respectively. Although these numbers fall within the range for those previously reported for RARP,^9,11,12 they are slightly lower than those reported in the literature for ORP, which have been documented at 15.5¹⁴ and 19.5.³ Part of this discrepancy in the ORP LNY seen in our study may be explained by the ongoing debate in the field about what exactly constitutes an ePLND. Some surgeons define an ePLND as removal of the obturator, external iliac, and hypogastric nodes,^4,6,17 while others also consider it necessary to remove presacral nodes.^3,18 The variations in the definition of ePLND can contribute to the variations seen across studies. Furthermore, studies that looked at LNY in cystectomy specimens found that greater nodes were retrieved with a greater number of nodal specimen packets compared with a single en bloc specimen.¹⁹ Given that the number of packets included for pathologic examination is surgeon dependent, this, too, can contribute to overall LNY and may help explain the differences seen in our sample.

The nodal yield is not necessarily the total number of nodes taken at the time of surgery but rather the total number of nodes identified during pathologic examination. Because of this fact, absolute nodal yield is difficult to extrapolate across studies where different laboratories were used to identify nodal tissue within extirpated tissue. Because the same group of pathologists examined lymph node specimens for both the RARP and ORP in our study, possible variation that is introduced when multiple pathologic laboratories are included should be reduced.

Only a borderline significant difference was seen in our study for overall LNY across surgical groups (P = 0.06). This lack of statistically significant difference was also seen when patients were substratified according to the D'Amico risk group classification. There was, however, a significant difference seen in the lymph node positivity rate, with a greater number of positive nodes detected during ORP vs RARP: 19/217 (8.8%) vs 1/99 (1.0%), P = 0.009. This finding can be partially explained by the fact that a significantly larger proportion of ORP patients were D'Amico intermediate and high risk, putting them at increased risk for lymph node involvement.

The effectiveness of lymph node dissection in low-risk patients is still under debate, given the low rate of lymph node metastasis found.²⁰ The largest series to data that evaluated lymph node metastasis rates in low-risk patients (defined as cT₁ and PSA ≤6 ng/mL) found rates as low as 0.7%.²¹ This low rate of 0.7% lymph node involvement was even seen in a second study that expanded the low-risk criteria to include cT_2a, Gleason score ≤6, and PSA ≤10 ng/mL.²² Even when looking at long-term (10-year) biochemical recurrence-free survival, no benefit was seen for performing PLND during RP in low-risk patients.²³ Taken together, these results call into question the clinical advantage of performing PLND in patients with low-risk prostate cancer.

When determining the extent of PLND at the time of surgery, it is important to balance the ability to accurately detect carcinomatous LNI with the complications of PLND. A study by Briganti and coworkers² noted that increasing the number of nodes removed from 10 to 25 results in a “virtually linear increase” in the probability of detecting LNI from around 10% to more than 80%.

This improved detection rate of LNI is not without its consequences. Some of the adverse effects reported as a result of PLND include lymphocele, lower extremity edema, deep venous thrombosis, ureteral injury and pelvic abscess.¹³ Studies that compared outcomes as a result of the extent of dissection have found that extended dissection may be associated with an increased complication rate. One study of men who were undergoing laparoscopic PLND found significantly higher complication rates for extended dissection vs limited dissections with rates of 35.9% and 2%, respectively.¹⁶ A larger study of 963 men who underwent PLND before radical retropubic prostatectomy for prostate cancer also found increased rates of complications with extended dissection (defined in this study as ≥10 nodes) compared with limited dissection with rates of 19.8% vs 8.2%, respectively.¹⁴ Given these increased complication rates with ePLND, it is important that surgeons balance the extent of dissection with the potential benefits of accurate nodal staging.

In our study, patients were significantly younger at the time of surgery for RARP vs ORP, with mean ages of 59.2 and 61.7 years, respectively (P = 0.003). Older age has been shown to be associated with increased rates of biochemical failure after RP (P = 0.005).²⁴ In addition, younger patients have been shown to have lower grade disease (P < 0.001), lower rates of positive surgical margins (P = 0.035), and lower rates of extraprostatic extension (P < 0.001) after RP.²⁵ No differences have been seen in rates of LNI (P = 0.85) or seminal vesicle invasion (P = 0.56) in men after RP for men based on age differences.²⁵ Although the mean ages were significantly higher in the ORP group in our study, the magnitude of this difference was only 2.5 years. Previous studies generally use 10-year increments to delineate age groups with significantly disparate clinical outcomes, suggesting that our age difference of only 2.5 years is likely below the threshold for clinical significance.²⁵ For this reason, it is unlikely that the differences in age contributed to LNY.

An important critique of our study is its relatively small sample size. Although our analysis resulted in an insignificant P value for the comparison of LNY between RARP and ORP, we cannot necessarily conclude that no difference exists, only that we were not able to detect a difference. Because our calculated power of 0.5166 is well below the threshold of significant power (0.80), we cannot conclude that the lack of a significant P value represents equivalence between our two groups. It is important to note, however, that the difference between LNY across surgical approaches in our study was only 1.1 nodes.

A study by Briganti and associates² defined the probability of finding LNI according to a given number of nodes examined. In the study results, probabilities for detecting LNI were given in groups of four node increments, indicating that a difference of ≤4 nodes was clinically insignificant. Because our study detected a difference that is much smaller than this threshold of clinical significance (ie, 1.1 vs 4), we can infer that no clinically significant difference exists for LNY at our institution for the defined study period for either ORP or RARP. A future larger, well-powered study will be necessary to make a conclusive statement of equivalence. Our series, however, represents the largest to date that compares LNY between RARP and ORP.

Another important limitation in our study is that while only one surgeon was included for the RARP cases, four surgeons were included in the ORP cases. Although this was done to ensure that only those performing a high volume of surgery were included in the analysis, it represents an important limitation in the interpretation of the study results, especially given that the boundaries of lymph node dissection are often subject to surgeon discretion.

Conclusion

No significant differences were seen in LNY across all subjects and specifically among intermediate- and high-risk men who are at greatest risk for lymph node metastasis. In the hands of an experienced, surgeon performing a high volume of surgery, RARP can achieve equivalent LNY compared with ORP. A future study with a larger sample size will be needed to make a definitive statement of equivalence between these two surgical approaches.

Footnotes

Acknowledgments

Matthew Truesdale is supported by a research fellowship sponsored by the Doris Duke Charitable Foundation. The authors would also like to thank the contributions of Alison Conforto and Amy Cooper for their help with data gathering.

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Thompson

, Thrasher

, Aus

et al. Guideline for the management of clinically localized prostate cancer: 2007 update. J Urol, 2007; 177:2106–2131.

Briganti

, Chun

, Salonia

et al. Critical assessment of ideal nodal yield at pelvic lymphadenectomy to accurately diagnose prostate cancer nodal metastasis in patients undergoing radical retropubic prostatectomy. Urology, 2007; 69:147–151.

Heidenreich

, Varga

, Von Knobloch

. Extended pelvic lymphadenectomy in patients undergoing radical prostatectomy: High incidence of lymph node metastasis. J Urol, 2002; 167:1681–1686.

Allaf

, Palapattu

, Trock

et al. Anatomical extent of lymph node dissection: impact on men with clinically localized prostate cancer. J Urol, 2004; 172:1840–1844.

Masterson

, Bianco

Jr , Vickers

et al. The association between total and positive lymph node counts, and disease progression in clinically localized prostate cancer. J Urol, 2006; 175:1320–1325.

Touijer

, Rabbani

, Otero

et al.

Standard versus limited pelvic lymph node dissection for prostate cancer in patients with a predicted probability of nodal metastasis greater than 1%

J Urol, 2007; 178:120–124.

Joslyn

, Konety

. Impact of extent of lymphadenectomy on survival after radical prostatectomy for prostate cancer. Urology, 2006; 68:121–125.

DiMarco

, Zincke

, Sebo

et al. The extent of lymphadenectomy for pTXNO prostate cancer does not affect prostate cancer outcome in the prostate specific antigen era. J Urol, 2005; 173:1121–1125.

Atug

, Castle

, Srivastav

et al. Prospective evaluation of concomitant lymphadenectomy in robot-assisted radical prostatectomy: Preliminary analysis of outcomes. J Endourol, 2006; 20:514–518.

10.

Zorn

, Katz

, Bernstein

et al. Pelvic lymphadenectomy during robot-assisted radical prostatectomy: Assessing nodal yield, perioperative outcomes, and complications. Urology, 2009; 74:296–302.

11.

Polcari

, Hugen

, Sivarajan

et al. Comparison of open and robot-assisted pelvic lymphadenectomy for prostate cancer. J Endourol, 2009; 23:1313–1317.

12.

Yates

, Haleblian

, Stein

et al. The impact of robotic surgery on pelvic lymph node dissection during radical prostatectomy for localized prostate cancer: The Brown University early robotic experience. Can J Urol, 2009; 16:4842–4846.

13.

Clark

, Parekh

, Cookson

et al. Randomized prospective evaluation of extended versus limited lymph node dissection in patients with clinically localized prostate cancer. J Urol, 2003; 169:145–148.

14.

Briganti

, Chun

, Salonia

et al. Complications and other surgical outcomes associated with extended pelvic lymphadenectomy in men with localized prostate cancer. Eur Urol, 2006; 50:1006–1013.

15.

D'Amico

, Whittington

, Malkowicz

et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA, 1998; 280:969–974.

16.

Stone

, Stock

, Unger

. Laparoscopic pelvic lymph node dissection for prostate cancer: Comparison of the extended and modified techniques. J Urol, 1997; 158:1891–1894.

17.

Briganti

, Chun

, Salonia

et al. Validation of a nomogram predicting the probability of lymph node invasion among patients undergoing radical prostatectomy and an extended pelvic lymphadenectomy. Eur Urol, 2006; 49:1019–1027.

18.

Golimbu

, Morales

, Al-Askari

, Brown

. Extended pelvic lymphadenectomy for prostatic cancer. J Urol, 1979; 121:617–620.

19.

Bochner

, Herr

, Reuter

. Impact of separate versus en bloc pelvic lymph node dissection on the number of lymph nodes retrieved in cystectomy specimens. J Urol, 2001; 166:2295–2296.

20.

Briganti

, Blute

, Eastham

et al. Pelvic lymph node dissection in prostate cancer. Eur Urol, 2009; 55:1251–1265.

21.

Makarov

, Humphreys

, Mangold

et al. Pathological outcomes and biochemical progression in men with T1c prostate cancer undergoing radical prostatectomy with prostate specific antigen 2.6 to 4.0 vs 4.1 to 6.0 ng/ml. J Urol, 2006; 176:554–558.

22.

Berglund

, Sadetsky

, DuChane

et al. Limited pelvic lymph node dissection at the time of radical prostatectomy does not affect 5-year failure rates for low, intermediate and high risk prostate cancer: Results from CaPSURE. J Urol, 2007; 177:526–530.

23.

Weight

, Reuther

, Gunn

et al. Limited pelvic lymph node dissection does not improve biochemical relapse-free survival at 10 years after radical prostatectomy in patients with low-risk prostate cancer. Urology, 2008; 71:141–145.

24.

Kane

, Im

, Amling

et al. Outcomes after radical prostatectomy among men who are candidates for active surveillance: Results from the SEARCH database. Urology, 2010Epub ahead of print.

25.

Magheli

, Rais-Bahrami

, Humphreys

et al. Impact of patient age on biochemical recurrence rates following radical prostatectomy. J Urol, 2007; 178:1933–1938.