- Email: [email protected]
Post-Radical Retropubic Prostatectomy Inguinal Hernia: An Analysis of Risk Factors With Special Reference to Preoperative Inguinal Hernia Morbidity and Pelvic Lymph Node Dissection
Trauma/Reconstruction/Diversion Post-Radical Retropubic Prostatectomy Inguinal Hernia: An Analysis of Risk Factors With Special Reference to Preoperative Inguinal Hernia Morbidity and Pelvic Lymph Node Dissection Johan Stranne,* Jonas Hugosson and Pär Lodding From the Department of Urology, Sahlgrenska University Hospital, Göteborg, Sweden
Purpose: Inguinal hernia after radical retropubic prostatectomy has been reported to occur in 7% to 21% of patients. We analyzed the impact of simultaneous pelvic lymph node dissection, preoperative inguinal hernia morbidity, postoperative anastomotic stricture, duration of surgery and patient age. We also compared the detection rate of inguinal hernia events in a retrospective patient file survey to that in a prospective patient administered questionnaire. Materials and Methods: A total of 498 patients underwent radical retropubic prostatectomy plus pelvic lymph node dissection and 166 underwent radical retropubic prostatectomy only. Mean followup was 40 months (median 37, range 3 to 85). All 664 patients were analyzed in the patient file survey. The patient administered questionnaire was mailed preoperatively, and after 3, 6, 12, 18, 24 and 36 months to 271 patients who underwent operation between 2001 and 2002. A total of 207 patients (76.4%) completed the preoperative questionnaire. Results: The cumulative incidence of inguinal hernia after 24 months was 11.6% in the patient file survey and 15.7% in the patient administered questionnaire. In the patient file survey patient age was the only studied factor that significantly influenced risk. The patient file survey failed to detect half of the men with preoperative inguinal hernia morbidity and a third of post-radical retropubic prostatectomy inguinal hernias compared to the patient administered questionnaire. On patient administered questionnaire analysis preoperative inguinal hernia morbidity was a significant risk factor for postoperative inguinal hernia (log rank Mantel-Cox test p ⫽ 0.010). Conclusions: Previous inguinal hernia morbidity and age increase the risk of post-radical retropubic prostatectomy inguinal hernia. Simultaneous pelvic lymph node dissection, postoperative anastomotic stricture and duration of surgery were not significant risk factors in this study. The patient file survey is inferior to the patient administered questionnaire for detecting inguinal hernia events. Key Words: prostate; prostatic neoplasms; prostatectomy; hernia, inguinal; complications
adical retropubic prostatectomy is one of the most common procedures in urological surgery. Incontinence, impotence and postoperative anastomotic stricture are complications well described in the literature.1 IH as a complication of RRP was first recognized in 1996 by Regan et al, who reported a 12% incidence within 6 months of RRP.2 In the following year Fischer and Wantz reported a case-control study of 1,504 hernioplasties in males showing an over representation of patients who had previously undergone RRP (6.1%).3 In 2001 we reported a crude incidence of 13.6% in 375 patients who underwent RRP with simultaneous PLND at a mean followup of 39 months for a cumulative incidence of 11% at 24 months.4 The study indicated that the risk of postoperative IH was increased in patients who had previously undergone IH surgery and also increased according to higher patient age. The high incidence of post-RRP IH was further confirmed in at least 5 recent studies describing an incidence of between 7% and 21%.5–9 We also reported that the crude IH incidence in a large group of men with nonmetastatic prostate cancer treated
R
Submitted for publication October 26, 2005. * Correspondence: Department of Urology, Sahlgrenska University Hospital, 413 45 Göteborg, Sweden (telephone: ⫹46-31-3423315; Fax: ⫹46-31-821740; e-mail: [email protected]).
0022-5347/06/1765-2072/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
without surgery was only 2.4% at a median followup of 39 months.8 The cumulative incidence was 1% at 24 months. Similar values were reported by Tsai et al.6 These studies confirm that IH truly represents a complication in patients who undergo to RRP. We studied the impact of 5 potential risk factors for post-RRP IH development, including simultaneous PLND, preoperative IH morbidity, postoperative anastomotic stricture, patient age at RRP and surgical procedure duration. In January 2001 a quality control instrument was introduced at our clinic in the form of a prospective PAQ, which is completed preoperatively and postoperatively. Among other factors we inquired about preoperative and postoperative IH morbidity. We also compared the detection rate of preoperative and postoperative IH morbidity in the retrospective PFS to that in the PAQ.
MATERIALS AND METHODS All 675 patients who underwent RRP at our institution from January 1998 to December 2002 were primarily included in this study. A total of 11 patients were excluded from analysis due to a followup shorter than 3 months. The remaining 664 patients had a mean followup of 40 months (median 37,
2072
Vol. 176, 2072-2076, November 2006 Printed in U.S.A. DOI:10.1016/j.juro.2006.07.007
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA Cum. IH free Survival
1,0 0,9 0,8 0,7 0,6
0
20 10 30 40 50 60 70 80 (n=373) (n=161) (n=37) (n=604) (n=258) (n=516) (n=100)
90
Follow up duration (months) (Number of eligible patients)
FIG. 1. Kaplan-Meier plot of cumulative (Cum.) IH-free survival in PFS in 664 patients (89 events).
range 3 to 85). Simultaneous RRP plus PLND was performed in 498 patients in whom mean followup was 44 months (median 43, range 3 to 85). The remaining 166 patients underwent RRP only and had a mean followup of 27 months (median 25, range 3 to 60). In 1998 and 1999 almost all patients underwent nonnerve sparing RRP plus PLND. However, as of 2000, most patients with Gleason score 6 or less, prostate specific antigen less than 10 ng/ml and 2 or fewer positive biopsy cores underwent nerve sparing RRP without simultaneous PLND. Therefore, followup was somewhat shorter in patients who underwent RRP alone. All procedures were performed basically using the technique of anatomical RRP described by Walsh.10 When a prevalent IH was repaired during RRP, as in 11 men, the preperitoneal technique described by Schlegel and Walsh was used.11 For the PFS clinical files from the time of surgery and thereafter were examined for postoperative IH events. We also recorded preoperative IH morbidity, ie previous IH surgery and/or a prevalent IH at RRP, as reported in the files, as well as patient age at RRP, procedure duration and a diagnosis of postoperative anastomotic stricture. Patients were followed postoperatively at 3, 6 and 12 months, and thereafter yearly. In the PFS postoperative IH was defined as de novo or recurrent IH after previous IH surgery, as mentioned in the patient file. Patients were not systematically asked about or examined for IH. Consequently only clinically apparent IHs that had been noted in the files were detected in the PFS. As part of the postoperative routine, patients were asked about voiding patterns. If an anastomotic stricture was suspected, cystoscopy was done to confirm or rule out this complication. Thus, subclinical strictures may have remained undetected. The 271 patients who underwent RRP between January 2001 and December 2002 were mailed a preoperative questionnaire immediately before being hospitalized for surgery. Among other things, the questionnaire inquired about previous IH surgery and prevalent IH. Further questionnaires inquiring about any new or recurrent IH noticed by the patient during followup were subsequently mailed 3, 6, 12, 18, 24 and 36 months postoperatively. Of the patients 207 (76.4%) completed the preoperative and at least 1 postoper-
2073
ative questionnaire, and were included in this analysis. RRP plus PLND was performed in 94 patients and RRP only in 113. Mean followup time in the questionnaire based group was 24.6 months (median 24.0, range 3 to 36). PAQ data were compared with those obtained in the PFS on the corresponding 207 patients. For this comparison the definition of postoperative IH was a patient reported event in response to a direct question in the PAQ, while in the PFS it was defined as a physician recorded event. A multivariable Cox proportional hazard model was used to calculate the relative risks of studied potential risk factors for PFS and PAQ data. A Kaplan-Meier cumulative survival plot was used to evaluate cumulative herniafree survival in the retrospective PFS. A life table cumulative survival plot and log rank Mantel-Cox test were used to evaluate the influence of preoperative IH morbidity in the PAQ based study. RESULTS Retrospective PFS in 664 Patients Mean patient age at RRP was 63 years. An IH developed postoperatively in 89 patients (13%), of which 30 were on the left side, 37 were on the right side, 3 were bilateral and 19 were on an unknown side. Mean time to IH was 16 months (median 11, range 3 to 71). Figure 1 shows cumulative postoperative IH-free survival. Table 1 shows the cumulative incidence of IH. Of the 498 patients who underwent RRP plus PLND 66 (13%) had an IH postoperatively compared to 23 of the 166 (14%) who underwent RRP only. A total of 56 patients had undergone previous IH surgery, of whom 8 experienced an IH postoperatively, including 4 on the previously operated side. Thus, the recurrence rate of IH after RRP was 7.1% (4 of 56 cases). Of the 25 patients who presented with a prevalent IH at RRP 11 had also undergone previous IH surgery. Eight of the 25 prevalent IHs were repaired during RRP. Thus, 70 patients (56 ⫹ 25 ⫺ 11 or 11%) had any form of preoperative IH morbidity, of whom 9 (13%) had an IH postoperatively. A total of 66 patients had an anastomotic stricture postoperatively, of whom 11 (16.7%) also had an IH postoperatively. The mean duration of surgery was 129 minutes. The RRP only group underwent nerve sparing procedures. The time gained by not performing PLND was balanced by extra time required to dissect the neurovascular bundles. Therefore, the duration of surgery was identical in the 2 groups. Table 2 lists Cox proportional HRs for the potential risk factors on PFS based analysis. Only patient age was a significant risk factor. PLND, previous IH morbidity, postoperative anastomotic stricture and surgery duration were not significant risk factors for an IH postoperatively.
TABLE 1. Cumulative postoperative IH incidence 12, 24 and 36 months postoperatively in retrospective PFS and PAQ
No. pts % Incidence: 12 Mos 24 Mos 36 Mos
PFS
PAQ
664
207
7.6 11.6 13.1
10.8 15.7 19.5
2074
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA TABLE 2. Cox proportional HRs of potential risk factors in PFS
PLND Preop IH morbidity Postop stricture Surgery duration Age
No. Pts
No. IH (%)
HR (CI)
p Value
498 70 66 664 664
66 (13.3) 9 (12.9) 11 (16.7) Not applicable Not applicable
0.730 (0.449–1.187) 0.951 (0.475–1.907) 1.205 (0.638–2.275) 0.997 (0.991–1.004) 1.051 (1.007–1.097)
0.205 0.888 0.565 0.448 0.0220*
* Significant.
Prospective PAQ in 207 Patients and Comparison to Corresponding PFS Data Mean patient age at RRP was 63 years. Of the 207 patients who were evaluated with the PAQ 33 (16%) reported an IH postoperatively at a mean of 13 months (median 12, range 3 to 36). Of these IHs 8 were on the left side, 23 were on the right side and 2 were bilateral. Table 1 shows the cumulative incidence of IH. A total of 28 patients (14%) reported previous IH surgery, of whom 7 reported an IH postoperatively, including 5 that developed on the previously operated side. Of the patients 12 presented with prevalent IH at RRP, of whom 5 had also undergone previous IH surgery. Three of the 12 prevalent IHs were repaired during RRP. Of these patients 2 subsequently had a recurrent IH. Thus, the recurrence rate of IH after RRP was 23% ([5 ⫹ 2]/[28 ⫹ 3]). Of the patients 35 (28 ⫹ 12 ⫺ 5 or 17%) had any form of preoperative IH morbidity, of whom 11 (31.4%) had an IH postoperatively. PFS was inferior to PAQ, especially for identifying a history of IH surgery and also for post-RRP IH (table 3). Table 4shows Cox proportional HRs for the studied potential risk factors on PAQ based analysis. For comparison the corresponding HRs in the same 207 men are also shown, as determined based on the PFS. HRs were similar except for preoperative IH morbidity, which emerged as a statistically significant risk factor on PAQ analysis (p ⫽ 0.010). Figure 2 also shows this risk factor. DISCUSSION As postoperative IH becomes an increasingly well recognized phenomenon, more literature concerning its incidence, risk factors and prophylactic measures is appearing. When comparing data, it is useful to look at the cumulative incidence at specified intervals, for instance 12, 24 and 36 months (table 1). Moreover, the method of data collection
also must be considered. Our study indicates that retrospective PFSs are inferior to PAQs for detecting IH events. Thus, previous patient file based studies may have underestimated the problem and it has been difficult to verify the significance of potential risk factors for postoperative IH. A possible weakness of our study is that the PAQ is a nonvalidated instrument. However, questions about IH in the PAQ were simple and straightforward, and it was not possible to validate the PAQ when the database was retrospectively evaluated for study purposes. The PFS detected only two-thirds of the IHs in the 207 patients who were investigated using the PFS and PAQ (table 3). Furthermore, PFS identified only half of the patients with previous IH surgery (table 3). Therefore, this could not be identified as a risk factor for post-RRP IH on PFS analysis. However, on PAQ analysis 40 of 44 instances (91%) of preoperative IH morbidity were identified and a statistically significant impact was demonstrable (tables 3 and 4, and fig. 2). Preoperative IH surgery has previously been suspected to be a risk factor for post-RRP IH.4,5,9 In this study we noted that any preoperative IH morbidity increased the risk of IH after RRP. While a 5-year 5.6% cumulative risk of recurrence after IH surgery was reported,12 our crude incidence from the PAQ material was 23%. Although we could not extract the cumulative recurrence risk at 5 years from our material, the post-RRP recurrence rate for previously repaired IHs appeared to be high. However, most post-RRP IHs occurred on the previously unoperated side. Prophylactic measures should be considered in these patients. Number of events at each interval ( (3/1)
/
(1/0)
(1/0) (6/2) (5/3) (1/0)
(4/3)
): (1/1)
(0/1)
Cum. IH free Survival 1,0 0,8 0,6
TABLE 3. Detection rate of previously operated, prevalent and post-RRP IH in 207 patients in PFS and PAQ Event Type (detection method) Previous IH surgery: PFS only PAQ only PFS ⫹ PAQ Prevalent IH at RRP: PFS only PAQ only PFS ⫹ PAQ Post-RRP IH: PFS only PAQ only PFS ⫹ PAQ
0,4
No pre-operative IH morbidity
0,2
Log rank (Mantel-Cox) test: p=0.0103
Pre-operative IH morbidity
No. Detected Events (%) 15 (52) 28 (97) 29 (100) 11 (73) 12 (80) 15 (100) 24 (66) 33 (92) 36 (100)
Event could be detected by 1 or 2 methods and, therefore, total number of each event for PFS plus PAQ is less than the sum of the 2 methods.
0 0
5
10
15
20
25
30
35
40
Follow up duration (months)
FIG. 2. Life table plot shows cumulative (Cum.) IH-free survival in PAQ study in 207 patients (33 events) with and without preoperative IH morbidity. Events were attributed only after 1.5, 4.5, 6, 7.5, 9, 15, 18, 21 and 30 months due to questionnaire documentation and, thus, they are superimposed. Values in brackets indicate number of events occurring at each documentation point, for example (6/2) indicates that at 9 months 6 patients without and 2 with preoperative IH morbidity had post-RRP IH.
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA
2075
TABLE 4. Cox proportional HRs and CIs of potential risk factors calculated from PFS and prospective PAQ for the 207 patients evaluated with both methods Post-RRP IH According to PFS
PLND Preop IH morbidity Postop stricture Surgery duration Age
Post-RRP IH According to PAQ
No. Pts
No. IH (%)
HR (CI)
p Value
No. Pts
No. IH (%)
HR (CI)
p Value
94 22 11 207 207
10 (10.6) 4 (18.2) 2 (18.2) Not applicable Not applicable
0.707 (0.305–1.636) 1.348 (0.438–4.148) 1.282 (0.285–5.768) 1.009 (0.997–1.021) 1.007 (0.930–1.091)
0.417 0.603 0.746 0.147 0.8604
94 35 11 207 207
16 (17.0) 11 (31.4) 2 (18.2) Not applicable Not applicable
1.154 (0.568–2.345) 2.688 (1.264–5.715) 0.904 (0.211–3.875) 0.999 (0.988–1.010) 0.972 (0.910–1.038)
0.692 0.010* 0.892 0.826 0.3936
* Significant.
A possible factor contributing to post-RRP IH could be subclinical IHs. Various studies indicate that subclinical defects of the internal orifice occur in between 10% and 20% of patients.11,13–16 Such lesions might be more frequent in patients with previous IH morbidity and RRP seems able to transform them into clinically significant IHs. We found 15 of 207 clinical IHs (7%) at surgery when combining the results of PFS and PAQ analyses in the subgroup of 207 patients from PAQ analysis (table 3). However, no specific screening was done to identify subclinical lesions. Conflicting data have been reported on whether stricture of the anastomotic area after RRP is a risk factor for IH.4 –7,9,17 Approximately 10% of our patients had an anastomotic stricture, which is a result that is well in line with the literature.1 Contrary to our previous findings,4 stricture was not a risk factor in this large patient group. We have no explanation of this discrepancy because the frequency of anastomotic stricture was the same in the 2 populations. However, in view of the fact that the current population was almost twice as large as in our first study, it seems unlikely that any significant influence would have been undetected. As in our previous study, patient age was a risk factor for IHs postoperatively.4 However, the influence seemed to be small and of little practical importance in individuals. Our previous study4 and that of Ichioka et al5 suggest that PLND alone can induce postoperative IH but perhaps at a somewhat lower frequency than after RRP plus PLND. Therefore, a hypothesis in the current study was that PLND performed in addition to RRP might add to the risk of postoperative IH. However, our data indicate that simultaneous PLND during RRP does not influence the incidence of postoperative IH. This finding indicates that postoperative hernia formation probably is not due to the unique maneuvers of the respective procedures, but rather to the maneuvers that they have in common. PLND and RRP are extraperitoneal procedures of the pelvic visceral organs that are performed through a lower midline incision. A self-retaining retractor is usually used in each procedure. Potential ischemia caused by the self-retaining retractor has been suggested as a factor contributing to post-RRP IH formation.2,4,5 Therefore, procedure duration, during which traction induced ischemia of tissue may occur, might be an important factor. The somewhat lower incidence of postoperative IH seen after PLND compared to after RRP4,5 is not yet satisfactorily verified but it could theoretically depend on the different durations of the 2 procedures. However, we could not identify any influence of surgery duration on the subsequent development of post-RRP IH. It has been suggested that IH formation in general is due to a deficiency in the collaboration between the transversalis
fascia and external oblique aponeurosis in the area around the internal orifice.18 The disruption of the transversalis fascia caused by the lower abdominal incision has been proposed as a factor contributing to post-RRP IH because it assaults the anatomical-physiological balance in the abdominal wall.3 In a recent small series Nomura et al reported a decreased incidence of postoperative IH after minilaparotomy RRP compared to when a conventional incision was used.7 The same observation was reported by Walsh, who now makes an 8 cm incision.19 Prospective, randomized studies of the impact of various incision techniques are warranted. CONCLUSIONS Previous IH morbidity and age are risk factors for post-RRP IH. PLND, anastomotic stricture and duration of surgery do not add significantly to the risk. Patient constitution and the surgical incision currently seem to be the most probable explanatory factors. Preoperative identification and repair of incidental IH is warranted and prophylactic measures should be considered, particularly in patients with a history of IH. Retrospective PFS grossly underestimates the incidence of preoperative IH morbidity and postoperative IH development compared to a prospective PAQ. Therefore, the true post-RRP IH incidence is likely to be higher than previous studies have suggested.
Abbreviations and Acronyms IH PAQ PFS PLND RRP
⫽ ⫽ ⫽ ⫽ ⫽
inguinal hernia patient administered questionnaire patient file survey pelvic lymph node dissection radical retropubic prostatectomy
REFERENCES 1.
Besarani, D., Amoroso, P. and Kirby, R.: Bladder neck contracture after radical retropubic prostatectomy. BJU Int, 94: 1245, 2004 2. Regan, T. C., Mordkin, R. M., Constantinople, N. L., Spence, I. J. and Dejter, S. W., Jr.: Incidence of inguinal hernias following radical retropubic prostatectomy. Urology, 47: 536, 1996 3. Fischer, E. and Wantz G. E.: Radical retropubic prostatectomy and groin hernia— cause and effect? Hernia, 1: 67, 1997 4. Lodding, P., Bergdahl, C., Nyberg, M., Pileblad, E., Stranne, J. and Hugosson, J.: Inguinal hernia after radical retropubic prostatectomy for prostate cancer: a study of incidence and
2076
5.
6.
7.
8.
9.
10.
11.
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA
risk factors in comparison to no operation and lymphadenectomy. J Urol, 166: 964, 2001 Ichioka, K., Yoshimura, K., Utsunomiya, N., Ueda, N., Matsui, Y., Terai, A. et al: High incidence of inguinal hernia after radical retropubic prostatectomy. Urology, 63: 278, 2004 Tsai, P. J., Yu, C. C., Lee, Y. H. and Huang, J. K.: Inguinal hernia after radical retropubic prostatectomy— experience of Kaohsiung Veterans General Hospital. J Chin Med Assoc, 67: 141, 2004 Nomura, T., Mimata, H., Kitamura, H., Fujikura, Y., Akita, Y., Yamasaki, M. et al: Lower incidence of inguinal hernia: minilaparotomy radical retropubic prostatectomy compared with conventional technique. A preliminary report. Urol Int, 74: 32, 2005 Stranne, J., Hugosson, J., Iversen, P., Morris, T. and Lodding, P.: Inguinal hernia in stage M0 prostate cancer: a comparison of incidence in men treated with and without radical retropubic prostatectomy—an analysis of 1105 patients. Urology, 65: 847, 2005 Twu, C. M., Ou, Y. C., Yang, C. R., Cheng, C. L. and Ho, H. C.: Predicting risk factors for inguinal hernia after radical retropubic prostatectomy. Urology, 66: 814, 2005 Walsh, P.: Anatomic radical retropubic prostatectomy. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, pp. 3107–3129, 2002 Schlegel, P. N. and Walsh, P. C.: The use of the preperitoneal approach for the simultaneous repair of inguinal hernia
12.
13.
14. 15.
16.
17.
18.
19.
during surgery on the bladder and prostate. World J Surg, 13: 555, 1989 Nilsson E., Haapaniemi, S., Gruber, G. and Sandblom, G.: Methods of repair and risk for reoperation in Swedish hernia surgery from 1992 to 1996. Br J Surg, 85: 1686, 1998 Watson, D. S., Sharp, K. W., Vasquez, J. M. and Richards, W. O.: Incidence of inguinal hernias diagnosed during laparoscopy. South Med J, 87: 23, 1994 Ajmani, M. L. and Ajmani, K.: The anatomical basis for the inguinal hernia. Anat AnZ, 153: 245, 1983 Lepor, H., Nieder, A. M. and Ferrandino M. N.: Intraoperative and postoperative complications of radical retropubic prostatectomy in a consecutive series of 1,000 cases. J Urol, 166: 1729, 2001 Nielsen, M. E. and Walsh, P. C.: Inguinal hernia repair at the time of radical retropubic prostatectomy. J Urol, 173: 278, 2005 Anderson, W. R. and Holmes, S. A.: Re: Inguinal hernia after radical retropubic prostatectomy for prostate cancer: a study of incidence and risk factors in comparison to no operation and lymphadenectomy (letter to the editor). J Urol, 167: 1800, 2002 Kux, M.: Anatomy of the groin: a view from the surgeon. In: Nyhus and Condon’s Hernia, 5th ed. Edited by R. J. J. Fitzgibbons and A. G. Greenburg. Philadelphia: Lippincott Williams & Wilkins, pp. 45–53, 2002 Walsh, P. C.: Personal communication, 2005
Purpose: Inguinal hernia after radical retropubic prostatectomy has been reported to occur in 7% to 21% of patients. We analyzed the impact of simultaneous pelvic lymph node dissection, preoperative inguinal hernia morbidity, postoperative anastomotic stricture, duration of surgery and patient age. We also compared the detection rate of inguinal hernia events in a retrospective patient file survey to that in a prospective patient administered questionnaire. Materials and Methods: A total of 498 patients underwent radical retropubic prostatectomy plus pelvic lymph node dissection and 166 underwent radical retropubic prostatectomy only. Mean followup was 40 months (median 37, range 3 to 85). All 664 patients were analyzed in the patient file survey. The patient administered questionnaire was mailed preoperatively, and after 3, 6, 12, 18, 24 and 36 months to 271 patients who underwent operation between 2001 and 2002. A total of 207 patients (76.4%) completed the preoperative questionnaire. Results: The cumulative incidence of inguinal hernia after 24 months was 11.6% in the patient file survey and 15.7% in the patient administered questionnaire. In the patient file survey patient age was the only studied factor that significantly influenced risk. The patient file survey failed to detect half of the men with preoperative inguinal hernia morbidity and a third of post-radical retropubic prostatectomy inguinal hernias compared to the patient administered questionnaire. On patient administered questionnaire analysis preoperative inguinal hernia morbidity was a significant risk factor for postoperative inguinal hernia (log rank Mantel-Cox test p ⫽ 0.010). Conclusions: Previous inguinal hernia morbidity and age increase the risk of post-radical retropubic prostatectomy inguinal hernia. Simultaneous pelvic lymph node dissection, postoperative anastomotic stricture and duration of surgery were not significant risk factors in this study. The patient file survey is inferior to the patient administered questionnaire for detecting inguinal hernia events. Key Words: prostate; prostatic neoplasms; prostatectomy; hernia, inguinal; complications
adical retropubic prostatectomy is one of the most common procedures in urological surgery. Incontinence, impotence and postoperative anastomotic stricture are complications well described in the literature.1 IH as a complication of RRP was first recognized in 1996 by Regan et al, who reported a 12% incidence within 6 months of RRP.2 In the following year Fischer and Wantz reported a case-control study of 1,504 hernioplasties in males showing an over representation of patients who had previously undergone RRP (6.1%).3 In 2001 we reported a crude incidence of 13.6% in 375 patients who underwent RRP with simultaneous PLND at a mean followup of 39 months for a cumulative incidence of 11% at 24 months.4 The study indicated that the risk of postoperative IH was increased in patients who had previously undergone IH surgery and also increased according to higher patient age. The high incidence of post-RRP IH was further confirmed in at least 5 recent studies describing an incidence of between 7% and 21%.5–9 We also reported that the crude IH incidence in a large group of men with nonmetastatic prostate cancer treated
R
Submitted for publication October 26, 2005. * Correspondence: Department of Urology, Sahlgrenska University Hospital, 413 45 Göteborg, Sweden (telephone: ⫹46-31-3423315; Fax: ⫹46-31-821740; e-mail: [email protected]).
0022-5347/06/1765-2072/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION
without surgery was only 2.4% at a median followup of 39 months.8 The cumulative incidence was 1% at 24 months. Similar values were reported by Tsai et al.6 These studies confirm that IH truly represents a complication in patients who undergo to RRP. We studied the impact of 5 potential risk factors for post-RRP IH development, including simultaneous PLND, preoperative IH morbidity, postoperative anastomotic stricture, patient age at RRP and surgical procedure duration. In January 2001 a quality control instrument was introduced at our clinic in the form of a prospective PAQ, which is completed preoperatively and postoperatively. Among other factors we inquired about preoperative and postoperative IH morbidity. We also compared the detection rate of preoperative and postoperative IH morbidity in the retrospective PFS to that in the PAQ.
MATERIALS AND METHODS All 675 patients who underwent RRP at our institution from January 1998 to December 2002 were primarily included in this study. A total of 11 patients were excluded from analysis due to a followup shorter than 3 months. The remaining 664 patients had a mean followup of 40 months (median 37,
2072
Vol. 176, 2072-2076, November 2006 Printed in U.S.A. DOI:10.1016/j.juro.2006.07.007
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA Cum. IH free Survival
1,0 0,9 0,8 0,7 0,6
0
20 10 30 40 50 60 70 80 (n=373) (n=161) (n=37) (n=604) (n=258) (n=516) (n=100)
90
Follow up duration (months) (Number of eligible patients)
FIG. 1. Kaplan-Meier plot of cumulative (Cum.) IH-free survival in PFS in 664 patients (89 events).
range 3 to 85). Simultaneous RRP plus PLND was performed in 498 patients in whom mean followup was 44 months (median 43, range 3 to 85). The remaining 166 patients underwent RRP only and had a mean followup of 27 months (median 25, range 3 to 60). In 1998 and 1999 almost all patients underwent nonnerve sparing RRP plus PLND. However, as of 2000, most patients with Gleason score 6 or less, prostate specific antigen less than 10 ng/ml and 2 or fewer positive biopsy cores underwent nerve sparing RRP without simultaneous PLND. Therefore, followup was somewhat shorter in patients who underwent RRP alone. All procedures were performed basically using the technique of anatomical RRP described by Walsh.10 When a prevalent IH was repaired during RRP, as in 11 men, the preperitoneal technique described by Schlegel and Walsh was used.11 For the PFS clinical files from the time of surgery and thereafter were examined for postoperative IH events. We also recorded preoperative IH morbidity, ie previous IH surgery and/or a prevalent IH at RRP, as reported in the files, as well as patient age at RRP, procedure duration and a diagnosis of postoperative anastomotic stricture. Patients were followed postoperatively at 3, 6 and 12 months, and thereafter yearly. In the PFS postoperative IH was defined as de novo or recurrent IH after previous IH surgery, as mentioned in the patient file. Patients were not systematically asked about or examined for IH. Consequently only clinically apparent IHs that had been noted in the files were detected in the PFS. As part of the postoperative routine, patients were asked about voiding patterns. If an anastomotic stricture was suspected, cystoscopy was done to confirm or rule out this complication. Thus, subclinical strictures may have remained undetected. The 271 patients who underwent RRP between January 2001 and December 2002 were mailed a preoperative questionnaire immediately before being hospitalized for surgery. Among other things, the questionnaire inquired about previous IH surgery and prevalent IH. Further questionnaires inquiring about any new or recurrent IH noticed by the patient during followup were subsequently mailed 3, 6, 12, 18, 24 and 36 months postoperatively. Of the patients 207 (76.4%) completed the preoperative and at least 1 postoper-
2073
ative questionnaire, and were included in this analysis. RRP plus PLND was performed in 94 patients and RRP only in 113. Mean followup time in the questionnaire based group was 24.6 months (median 24.0, range 3 to 36). PAQ data were compared with those obtained in the PFS on the corresponding 207 patients. For this comparison the definition of postoperative IH was a patient reported event in response to a direct question in the PAQ, while in the PFS it was defined as a physician recorded event. A multivariable Cox proportional hazard model was used to calculate the relative risks of studied potential risk factors for PFS and PAQ data. A Kaplan-Meier cumulative survival plot was used to evaluate cumulative herniafree survival in the retrospective PFS. A life table cumulative survival plot and log rank Mantel-Cox test were used to evaluate the influence of preoperative IH morbidity in the PAQ based study. RESULTS Retrospective PFS in 664 Patients Mean patient age at RRP was 63 years. An IH developed postoperatively in 89 patients (13%), of which 30 were on the left side, 37 were on the right side, 3 were bilateral and 19 were on an unknown side. Mean time to IH was 16 months (median 11, range 3 to 71). Figure 1 shows cumulative postoperative IH-free survival. Table 1 shows the cumulative incidence of IH. Of the 498 patients who underwent RRP plus PLND 66 (13%) had an IH postoperatively compared to 23 of the 166 (14%) who underwent RRP only. A total of 56 patients had undergone previous IH surgery, of whom 8 experienced an IH postoperatively, including 4 on the previously operated side. Thus, the recurrence rate of IH after RRP was 7.1% (4 of 56 cases). Of the 25 patients who presented with a prevalent IH at RRP 11 had also undergone previous IH surgery. Eight of the 25 prevalent IHs were repaired during RRP. Thus, 70 patients (56 ⫹ 25 ⫺ 11 or 11%) had any form of preoperative IH morbidity, of whom 9 (13%) had an IH postoperatively. A total of 66 patients had an anastomotic stricture postoperatively, of whom 11 (16.7%) also had an IH postoperatively. The mean duration of surgery was 129 minutes. The RRP only group underwent nerve sparing procedures. The time gained by not performing PLND was balanced by extra time required to dissect the neurovascular bundles. Therefore, the duration of surgery was identical in the 2 groups. Table 2 lists Cox proportional HRs for the potential risk factors on PFS based analysis. Only patient age was a significant risk factor. PLND, previous IH morbidity, postoperative anastomotic stricture and surgery duration were not significant risk factors for an IH postoperatively.
TABLE 1. Cumulative postoperative IH incidence 12, 24 and 36 months postoperatively in retrospective PFS and PAQ
No. pts % Incidence: 12 Mos 24 Mos 36 Mos
PFS
PAQ
664
207
7.6 11.6 13.1
10.8 15.7 19.5
2074
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA TABLE 2. Cox proportional HRs of potential risk factors in PFS
PLND Preop IH morbidity Postop stricture Surgery duration Age
No. Pts
No. IH (%)
HR (CI)
p Value
498 70 66 664 664
66 (13.3) 9 (12.9) 11 (16.7) Not applicable Not applicable
0.730 (0.449–1.187) 0.951 (0.475–1.907) 1.205 (0.638–2.275) 0.997 (0.991–1.004) 1.051 (1.007–1.097)
0.205 0.888 0.565 0.448 0.0220*
* Significant.
Prospective PAQ in 207 Patients and Comparison to Corresponding PFS Data Mean patient age at RRP was 63 years. Of the 207 patients who were evaluated with the PAQ 33 (16%) reported an IH postoperatively at a mean of 13 months (median 12, range 3 to 36). Of these IHs 8 were on the left side, 23 were on the right side and 2 were bilateral. Table 1 shows the cumulative incidence of IH. A total of 28 patients (14%) reported previous IH surgery, of whom 7 reported an IH postoperatively, including 5 that developed on the previously operated side. Of the patients 12 presented with prevalent IH at RRP, of whom 5 had also undergone previous IH surgery. Three of the 12 prevalent IHs were repaired during RRP. Of these patients 2 subsequently had a recurrent IH. Thus, the recurrence rate of IH after RRP was 23% ([5 ⫹ 2]/[28 ⫹ 3]). Of the patients 35 (28 ⫹ 12 ⫺ 5 or 17%) had any form of preoperative IH morbidity, of whom 11 (31.4%) had an IH postoperatively. PFS was inferior to PAQ, especially for identifying a history of IH surgery and also for post-RRP IH (table 3). Table 4shows Cox proportional HRs for the studied potential risk factors on PAQ based analysis. For comparison the corresponding HRs in the same 207 men are also shown, as determined based on the PFS. HRs were similar except for preoperative IH morbidity, which emerged as a statistically significant risk factor on PAQ analysis (p ⫽ 0.010). Figure 2 also shows this risk factor. DISCUSSION As postoperative IH becomes an increasingly well recognized phenomenon, more literature concerning its incidence, risk factors and prophylactic measures is appearing. When comparing data, it is useful to look at the cumulative incidence at specified intervals, for instance 12, 24 and 36 months (table 1). Moreover, the method of data collection
also must be considered. Our study indicates that retrospective PFSs are inferior to PAQs for detecting IH events. Thus, previous patient file based studies may have underestimated the problem and it has been difficult to verify the significance of potential risk factors for postoperative IH. A possible weakness of our study is that the PAQ is a nonvalidated instrument. However, questions about IH in the PAQ were simple and straightforward, and it was not possible to validate the PAQ when the database was retrospectively evaluated for study purposes. The PFS detected only two-thirds of the IHs in the 207 patients who were investigated using the PFS and PAQ (table 3). Furthermore, PFS identified only half of the patients with previous IH surgery (table 3). Therefore, this could not be identified as a risk factor for post-RRP IH on PFS analysis. However, on PAQ analysis 40 of 44 instances (91%) of preoperative IH morbidity were identified and a statistically significant impact was demonstrable (tables 3 and 4, and fig. 2). Preoperative IH surgery has previously been suspected to be a risk factor for post-RRP IH.4,5,9 In this study we noted that any preoperative IH morbidity increased the risk of IH after RRP. While a 5-year 5.6% cumulative risk of recurrence after IH surgery was reported,12 our crude incidence from the PAQ material was 23%. Although we could not extract the cumulative recurrence risk at 5 years from our material, the post-RRP recurrence rate for previously repaired IHs appeared to be high. However, most post-RRP IHs occurred on the previously unoperated side. Prophylactic measures should be considered in these patients. Number of events at each interval ( (3/1)
/
(1/0)
(1/0) (6/2) (5/3) (1/0)
(4/3)
): (1/1)
(0/1)
Cum. IH free Survival 1,0 0,8 0,6
TABLE 3. Detection rate of previously operated, prevalent and post-RRP IH in 207 patients in PFS and PAQ Event Type (detection method) Previous IH surgery: PFS only PAQ only PFS ⫹ PAQ Prevalent IH at RRP: PFS only PAQ only PFS ⫹ PAQ Post-RRP IH: PFS only PAQ only PFS ⫹ PAQ
0,4
No pre-operative IH morbidity
0,2
Log rank (Mantel-Cox) test: p=0.0103
Pre-operative IH morbidity
No. Detected Events (%) 15 (52) 28 (97) 29 (100) 11 (73) 12 (80) 15 (100) 24 (66) 33 (92) 36 (100)
Event could be detected by 1 or 2 methods and, therefore, total number of each event for PFS plus PAQ is less than the sum of the 2 methods.
0 0
5
10
15
20
25
30
35
40
Follow up duration (months)
FIG. 2. Life table plot shows cumulative (Cum.) IH-free survival in PAQ study in 207 patients (33 events) with and without preoperative IH morbidity. Events were attributed only after 1.5, 4.5, 6, 7.5, 9, 15, 18, 21 and 30 months due to questionnaire documentation and, thus, they are superimposed. Values in brackets indicate number of events occurring at each documentation point, for example (6/2) indicates that at 9 months 6 patients without and 2 with preoperative IH morbidity had post-RRP IH.
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA
2075
TABLE 4. Cox proportional HRs and CIs of potential risk factors calculated from PFS and prospective PAQ for the 207 patients evaluated with both methods Post-RRP IH According to PFS
PLND Preop IH morbidity Postop stricture Surgery duration Age
Post-RRP IH According to PAQ
No. Pts
No. IH (%)
HR (CI)
p Value
No. Pts
No. IH (%)
HR (CI)
p Value
94 22 11 207 207
10 (10.6) 4 (18.2) 2 (18.2) Not applicable Not applicable
0.707 (0.305–1.636) 1.348 (0.438–4.148) 1.282 (0.285–5.768) 1.009 (0.997–1.021) 1.007 (0.930–1.091)
0.417 0.603 0.746 0.147 0.8604
94 35 11 207 207
16 (17.0) 11 (31.4) 2 (18.2) Not applicable Not applicable
1.154 (0.568–2.345) 2.688 (1.264–5.715) 0.904 (0.211–3.875) 0.999 (0.988–1.010) 0.972 (0.910–1.038)
0.692 0.010* 0.892 0.826 0.3936
* Significant.
A possible factor contributing to post-RRP IH could be subclinical IHs. Various studies indicate that subclinical defects of the internal orifice occur in between 10% and 20% of patients.11,13–16 Such lesions might be more frequent in patients with previous IH morbidity and RRP seems able to transform them into clinically significant IHs. We found 15 of 207 clinical IHs (7%) at surgery when combining the results of PFS and PAQ analyses in the subgroup of 207 patients from PAQ analysis (table 3). However, no specific screening was done to identify subclinical lesions. Conflicting data have been reported on whether stricture of the anastomotic area after RRP is a risk factor for IH.4 –7,9,17 Approximately 10% of our patients had an anastomotic stricture, which is a result that is well in line with the literature.1 Contrary to our previous findings,4 stricture was not a risk factor in this large patient group. We have no explanation of this discrepancy because the frequency of anastomotic stricture was the same in the 2 populations. However, in view of the fact that the current population was almost twice as large as in our first study, it seems unlikely that any significant influence would have been undetected. As in our previous study, patient age was a risk factor for IHs postoperatively.4 However, the influence seemed to be small and of little practical importance in individuals. Our previous study4 and that of Ichioka et al5 suggest that PLND alone can induce postoperative IH but perhaps at a somewhat lower frequency than after RRP plus PLND. Therefore, a hypothesis in the current study was that PLND performed in addition to RRP might add to the risk of postoperative IH. However, our data indicate that simultaneous PLND during RRP does not influence the incidence of postoperative IH. This finding indicates that postoperative hernia formation probably is not due to the unique maneuvers of the respective procedures, but rather to the maneuvers that they have in common. PLND and RRP are extraperitoneal procedures of the pelvic visceral organs that are performed through a lower midline incision. A self-retaining retractor is usually used in each procedure. Potential ischemia caused by the self-retaining retractor has been suggested as a factor contributing to post-RRP IH formation.2,4,5 Therefore, procedure duration, during which traction induced ischemia of tissue may occur, might be an important factor. The somewhat lower incidence of postoperative IH seen after PLND compared to after RRP4,5 is not yet satisfactorily verified but it could theoretically depend on the different durations of the 2 procedures. However, we could not identify any influence of surgery duration on the subsequent development of post-RRP IH. It has been suggested that IH formation in general is due to a deficiency in the collaboration between the transversalis
fascia and external oblique aponeurosis in the area around the internal orifice.18 The disruption of the transversalis fascia caused by the lower abdominal incision has been proposed as a factor contributing to post-RRP IH because it assaults the anatomical-physiological balance in the abdominal wall.3 In a recent small series Nomura et al reported a decreased incidence of postoperative IH after minilaparotomy RRP compared to when a conventional incision was used.7 The same observation was reported by Walsh, who now makes an 8 cm incision.19 Prospective, randomized studies of the impact of various incision techniques are warranted. CONCLUSIONS Previous IH morbidity and age are risk factors for post-RRP IH. PLND, anastomotic stricture and duration of surgery do not add significantly to the risk. Patient constitution and the surgical incision currently seem to be the most probable explanatory factors. Preoperative identification and repair of incidental IH is warranted and prophylactic measures should be considered, particularly in patients with a history of IH. Retrospective PFS grossly underestimates the incidence of preoperative IH morbidity and postoperative IH development compared to a prospective PAQ. Therefore, the true post-RRP IH incidence is likely to be higher than previous studies have suggested.
Abbreviations and Acronyms IH PAQ PFS PLND RRP
⫽ ⫽ ⫽ ⫽ ⫽
inguinal hernia patient administered questionnaire patient file survey pelvic lymph node dissection radical retropubic prostatectomy
REFERENCES 1.
Besarani, D., Amoroso, P. and Kirby, R.: Bladder neck contracture after radical retropubic prostatectomy. BJU Int, 94: 1245, 2004 2. Regan, T. C., Mordkin, R. M., Constantinople, N. L., Spence, I. J. and Dejter, S. W., Jr.: Incidence of inguinal hernias following radical retropubic prostatectomy. Urology, 47: 536, 1996 3. Fischer, E. and Wantz G. E.: Radical retropubic prostatectomy and groin hernia— cause and effect? Hernia, 1: 67, 1997 4. Lodding, P., Bergdahl, C., Nyberg, M., Pileblad, E., Stranne, J. and Hugosson, J.: Inguinal hernia after radical retropubic prostatectomy for prostate cancer: a study of incidence and
2076
5.
6.
7.
8.
9.
10.
11.
POST-RADICAL RETROPUBIC PROSTATECTOMY INGUINAL HERNIA
risk factors in comparison to no operation and lymphadenectomy. J Urol, 166: 964, 2001 Ichioka, K., Yoshimura, K., Utsunomiya, N., Ueda, N., Matsui, Y., Terai, A. et al: High incidence of inguinal hernia after radical retropubic prostatectomy. Urology, 63: 278, 2004 Tsai, P. J., Yu, C. C., Lee, Y. H. and Huang, J. K.: Inguinal hernia after radical retropubic prostatectomy— experience of Kaohsiung Veterans General Hospital. J Chin Med Assoc, 67: 141, 2004 Nomura, T., Mimata, H., Kitamura, H., Fujikura, Y., Akita, Y., Yamasaki, M. et al: Lower incidence of inguinal hernia: minilaparotomy radical retropubic prostatectomy compared with conventional technique. A preliminary report. Urol Int, 74: 32, 2005 Stranne, J., Hugosson, J., Iversen, P., Morris, T. and Lodding, P.: Inguinal hernia in stage M0 prostate cancer: a comparison of incidence in men treated with and without radical retropubic prostatectomy—an analysis of 1105 patients. Urology, 65: 847, 2005 Twu, C. M., Ou, Y. C., Yang, C. R., Cheng, C. L. and Ho, H. C.: Predicting risk factors for inguinal hernia after radical retropubic prostatectomy. Urology, 66: 814, 2005 Walsh, P.: Anatomic radical retropubic prostatectomy. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, pp. 3107–3129, 2002 Schlegel, P. N. and Walsh, P. C.: The use of the preperitoneal approach for the simultaneous repair of inguinal hernia
12.
13.
14. 15.
16.
17.
18.
19.
during surgery on the bladder and prostate. World J Surg, 13: 555, 1989 Nilsson E., Haapaniemi, S., Gruber, G. and Sandblom, G.: Methods of repair and risk for reoperation in Swedish hernia surgery from 1992 to 1996. Br J Surg, 85: 1686, 1998 Watson, D. S., Sharp, K. W., Vasquez, J. M. and Richards, W. O.: Incidence of inguinal hernias diagnosed during laparoscopy. South Med J, 87: 23, 1994 Ajmani, M. L. and Ajmani, K.: The anatomical basis for the inguinal hernia. Anat AnZ, 153: 245, 1983 Lepor, H., Nieder, A. M. and Ferrandino M. N.: Intraoperative and postoperative complications of radical retropubic prostatectomy in a consecutive series of 1,000 cases. J Urol, 166: 1729, 2001 Nielsen, M. E. and Walsh, P. C.: Inguinal hernia repair at the time of radical retropubic prostatectomy. J Urol, 173: 278, 2005 Anderson, W. R. and Holmes, S. A.: Re: Inguinal hernia after radical retropubic prostatectomy for prostate cancer: a study of incidence and risk factors in comparison to no operation and lymphadenectomy (letter to the editor). J Urol, 167: 1800, 2002 Kux, M.: Anatomy of the groin: a view from the surgeon. In: Nyhus and Condon’s Hernia, 5th ed. Edited by R. J. J. Fitzgibbons and A. G. Greenburg. Philadelphia: Lippincott Williams & Wilkins, pp. 45–53, 2002 Walsh, P. C.: Personal communication, 2005