- Email: [email protected]
Quality assurance issues in radical prostatectomy
EJSO (2005) 31, 650–655
www.ejso.com
Quality assurance issues in radical prostatectomy H. Van Poppel*, S.F.F. Boulanger, S. Joniau Department of Urology, University Hospital Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Accepted for publication 10 February 2005 Available online 15 April 2005
KEYWORDS Radical prostatectomy; Quality; Hospital-volume; Surgical-volume
Abstract Aims: Quality of surgery is a controversial issue and no studies are reporting on the standard of surgical quality in the treatment of urological cancer. The question is whether quality can be evaluated and whether there is a standard for a qualitatively well performed radical retropubic prostatectomy. Methods: We reviewed the literature on this topic. Data of four large studies based on Medicare claims and an EORTC report were analysed. Results: Two studies reflect hospital-volume rather than surgeon-volume. Two compared hospital-volume and surgeon-volume and in both studies there was no clear relationship between surgeon-volume and the parameters reviewed. Similarly, the EORTC study concluded that there is a variation in outcome that is not related to the caseload and proposed minimal quality standards for radical prostatectomy. Conclusions: There is no clear relationship between surgeon-volume and surgical quality. Since radical prostatectomy is the standard treatment for the most frequent male malignancy and is offered to many patients that might never even suffer from the disease, the procedure must be performed with the highest guarantee of quality. Although, quality control of radical prostatectomy is feasible, its implementation will still require an enormous effort from the urological community. Q 2005 Elsevier Ltd. All rights reserved.
Introduction Most research on the outcome of surgery in relation to the hospital or the surgeon volume has rather been limited to exploring these procedures with a high risk of mortality. For procedures with low mortality such as radical prostatectomy, the morbidity is the most important parameter to be
* Corresponding author. Tel.: C32 16 34 69 30; fax: C32 16 34 69 31. E-mail address: [email protected] (H. Van Poppel).
analysed.1 Indeed, prostate cancer is the commonest malignancy in men and radical prostatectomy (RPr) is the most commonly used treatment. Quality of surgery is a controversial issue and no studies are reporting on the standard of surgical quality in the treatment of urological cancer. Radiotherapists have preceded surgeons when they developed detailed technical guidelines for radiation therapy.2 Numerous trials are carried out on the value of adjunctive treatments to surgery, such as radiotherapy or hormonal therapy, while neglecting the important function of the surgeon and surgical procedure as a prognostic factor.3 For lung, liver, pancreatic, oesophageal and
0748-7983/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2005.02.017
Quality assurance issues in radical prostatectomy bladder cancer the mortality and morbidity has been shown to be significantly lower in high volume centers.4–6 Radical prostatectomy is the treatment of choice for localized prostate cancer and many patients undergo the procedure although their disease might not even affect their life expectancy and quality of life if left untreated. The question rises to what extent the quality of the surgical act is an important prognostic factor in the urological and oncological outcome after a radical prostatectomy. Can quality be evaluated and is there a standard for a qualitatively well performed radical retropubic prostatectomy (RRP)?
Literature review Yao and Yao7 collected data on more then 100,000 prostatectomies from Medicare claims filed from 1991 through 1994. He found an inverse relationship between the hospital-volume of radical prostatectomies and the relative risk of readmission, serious complications and mortality. The mean length of hospital stay was longer in low-volume hospitals (Table 1). Surgical-volume was associated with both the outcome and the length of hospital stay. A decrease in the length of stay, with equivalent or improved outcomes, might be achieved by increasing the number of operations performed. The application of this principle could greatly reduce the costs while maintaining or improving the quality of care. Ellison et al.8 identified over 60,000 men who underwent a radical prostatectomy between 1989 and 1995. They compared hospital-volume with inhospital mortality, overall length of stay and total hospital charges. The primary outcome measure of Table 1
651 the study was the in-hospital mortality, the secondary outcome measures were length of stay and hospital charges (Table 2). The overall adjusted in-hospital mortality after radical prostatectomy was relatively low (0.25%), but patients at lowvolume hospitals were 78% more likely to have inhospital mortality. The overall length of stay decreased over-time in all hospitals but the overall length of stay and the total hospital charges where higher in the low-volume-hospitals. The authors quoted that further study regarding other important outcome measures, including incontinence, impotence, and long-term patient survival, is necessary. Begg et al.9 also used a Medicare linked database to evaluate the mortality rate, post-operative complication rate, late urological complications (strictures, fistulas) and long-term incontinence after the operation in over 10,000 patients between 1992 and 1996. They compared to hospital-volume and to surgeon-volume. Neither hospital- nor surgeon-volume was significantly associated with surgery-related death. There was a difference in the post-operative complication rate and the late urinary complication rate. These complications were lower in very-high-volume-hospitals (114–252 pts per year) and were also lower when the prostatectomy was performed by very-high-volume surgeons (33–121 pts per year). There was no difference in long-term incontinence, nor between hospital nor between surgeon-volume (Table 3). Although, for all parameters there seems to be a trend for better quality (lower complication rate, fewer incontinence and lower mortality) none of these differences were statistically significant. They also studied the outcomes associated with the 159 surgeons in the highest-volume categories
Data collected on 101,604 radical prostatectomies between 1991 and 1994 by Yao and Yao et al.7
Hospital volume
Length of stay (days)
Serious complications (%)
Readmissions in 30 days
Mortality
Low Medium–low Medium–high High
8.50 8.18 7.70 7.81
31.3 28.7 27.8 26.3
5.0 4.5 4.3 4.1
0.63 0.59 0.56 0.39
Compared with high volume hospitals
Relative risk of readmission in 30 days (%)
Relative risk of serious complication (%)
Relative risk of mortality (%)
Low Medium–low Medium–high
30 16 8
43 25 9
51 43 42
Hospitals were classified into four categories of volume on the number of prostatectomies performed during the studied period: low (!38), median low (39–74), median high (74–140), high volume hospitals (O140 prostatectomies). In the four groups data on length of stay, serious complications, readmissions in 30 days and the mortality rates were collected. The relative risk compared with high volume hospitals.
652
H. Van Poppel et al.
Table 2 Data collected on 66,693 radical prostatectomies between 1989 and 1995 by Ellison et al.8
Table 4 Data collected on 2292 radical prostatectomies between 1992 and 1996 by Hu et al.10
Hospital volume
Length of stay (days)
Charges ($)
Mortality (%)
Odds ratio mortality
Hospital volume
Complication rate (%)
Stricture of the anastomosis (%)
Hospitals stay (days)
Low Medium High
5.4 4.80 4.20
15,600 15,100 13,500
0.30 0.20 0.17
1.78 1.71 1.00
Low High
21.6 16.8
26.8 19.8
5.2 4.4
Surgeon volume
Complication rate (%)
Stricture of the anastomosis (%)
Hospitals stay (days)
Low High
21.9 11.8
27.7 22.0
5.2 4.1
Hospitals were classified into three categories of prostatectomy volume: low (!25), medium (25–54), high volume hospitals (O54 radical prostatectomies performed).
(33–121 pts). There were statistically significant variations among those surgeons in the rates of post-operative complications, late urinary complications and long-term incontinence. This evidence suggests that there is a genuine variation in surgical performance which is not reflected by the quantity of operations performed. Hu et al.,10 used a national 5% random sample of claims data of Medicare about more than 2000 procedures. They analysed in-hospital complications, anastomotic strictures and the length of stay compared with hospital-volume and surgeonvolume (Table 4). Hospital-volume was not significantly associated with complications. High-volume hospitals (HVH) tended to have fewer anastomotic strictures, whereas surgeon-volume had no impact on their incidence. There are two explanations for this counterintuitive finding. First, HVHs had younger patients and age is a cofactor in developing strictures. Second, HVHs may disseminate advances Table 3
Data on hospital volume. Hospitals were classified into two categories of prostatectomy volume: low (!60) and high volume hospitals (O60). Data on surgeon volume. Surgeon were classified into two categories of prostatectomy volume: low (!40) and high volume surgeons (O 40).
in surgical technique between the surgeons more readily (e.g. bladder mucosa eversion, mucosa– mucosa apposition). The authors concluded that additional study on potency, continence and longterm survival is warranted. As in the former study, there was a trend, but not a statistically significant difference on complication rates, number of strictures and length of stay in favor of higher hospital and surgeon volumes. In a study of the EORTC GU group,11 the authors tried to assess whether the quality of the surgical act can be assessed by any means. Questionnaires were collected from 23 different institutes
Data collected on 11,522 radical prostatectomies between 1992 and 1996 by Begg et al.9
Hospital Volume
Complication rate (%)
Late urinary complication rate (symptoms/treatment) (%)
Long-term incontinence (symptoms/ treatment) (%)
Mortality
Low Medium High Very-high
32 31 30 27
28/18 29/19 23/16 20/13
19/6.5 19/6.4 18/7.0 18/7.6
0.5 0.5 0.5 0.5
Surgeon volume
Complication rate (%)
Late urinary complication rate (symptoms/treatment) (%)
Long-term incontinence (symptoms/ treatment) (%)
Mortality
Low Medium High Very-high
32 31 30 26
28/19 26/18 27/17 20/14
20/7.3 20/7.2 19/6.7 16/6.6
0.4 0.5 0.5 0.4
Data on hospital volume. Hospitals were classified into four categories of prostatectomy volume: low (1–33), medium (34–61), high (62–107) and very high volume hospitals (114–252). Data on surgeon volume. Surgeon were classified into four categories of prostatectomy volume: low (1–10), medium (11–19), high (20–32) and very high volume surgeons (33–121).
Quality assurance issues in radical prostatectomy Table 5
653
Data collected on 232 radical prostatectomies from 27 urologists by Van Poppel.11
Surgeon volume Surgical results Low Medium High Surgeon volume Oncological results Low Medium High Surgeon volume Urological results Low Medium High
Blood transfusion none (%)
Blood transfusion 1– 3 units (%)
Blood transfusion R3 units (%)
35.1 26.4 61.8
47.5 55.1 38.2
17.4 18.5 0.0
Surgical margin negative (%)
Surgical margin positive (%)
PSA !detection level (%)
PSA Rdetection level (%)
63.3 62.6 69.4
36.8 37.4 30.6
49.6 65.0 55.2
50.4 35.0 44.8
Incontinence at 3 months none (%)
Incontinence at 3 months drops (%)
Incontinence at 3 months R1 pad (%)
7.6 2.9 2.0
82.4 76.0 75.8
10.0 21.1 22.2
Low (!12), medium (12–25), high (O50) volume surgeons.
collecting 232 radical prostatectomies performed for T1, T2 prostate cancer (Table 5). Blood loss, duration of surgery, surgical margin status, post-operative prostate specific antigen (PSA) and urinary incontinence after 3 months were recorded and analysed with respect to surgical-volume 27 urologists were asked to fill out these data on 10 consecutive radical prostatectomies. They were asked how many radical prostatectomies they were used to perform per year. From these 10 files the mean for each parameter was calculated and used in the analysis. The blood loss was analysed by the blood transfusion need, although this is not a very good parameter to asses the quality of surgery. Some urologists and anaesthesiologists are used to replace any blood loss, while others give transfusions only for a hemoglobin level below 100 g/l. The duration of surgery may not be a good parameter to assess quality. Urologists with a large caseload are not always faster. The authors found no relationship between the duration of surgery and any other parameter such as hemorrhage, margin status, PSA level or incontinence. The oncological result was related to the surgical margin status and the 3-month post-operative PSA level. It was recognized that the number of positive margins is related to the initial stage of the disease and the initial PSA, which were not equally balanced in the different institutes. It was acknowledged that the PSA level at 3 months is not an optimal parameter of surgical quality since PSA
persistence or PSA recurrence shortly after surgery often means systemic disease and can therefore be completely independent of the quality of surgery. Nevertheless, this study was able to demonstrate some important issues. Urologists who performed fewer radical prostatectomies tended to have significantly more hemorrhage. Margin positivity varied between the different surgeons but the margin status proved to be comparable for all participants without any correlation with the number of radical prostatectomies done. The same held for the PSA-free status at 3 months emphasizing on the fact that the oncological result was not related to the number of surgeries performed but to the individual surgical skill. In respect of urinary continence, again the percentage of completely continent patients at 3 months was not correlated to the surgeon’s experience and the number of cases performed. Using this simple retrospective analysis of 10 consecutive cases done by one urological surgeon, the authors were able to distinguish whether the quality of the surgery of a given urologist was good, fair or poor for each of the points of technique assessed. None of the participating surgeons was good or bad for all items. When ranking for oncological (margins and PSA) and urological (incontinence) outcome one participant was quoted as best and four other participants as poor performers of radical prostatectomy. The authors proposed what a ‘standard radical retropubic prostatectomy’ should mean. The
654 analysis of 10 consecutive radical prostatectomies for T1T2 prostate cancer should reveal a mean blood transfusion need of less than three units, a mean operative time of less than 3 h, a mean occurrence of positive surgical margins in less than 20%, a PSA below detection level at 3 months in more than 80% and a complete urinary continence at 3 months in more than 50%. These standards proposed need further validation through a more extensive surgical quality survey. From this study, however, it was clear that a higher number of radical prostatectomies performed each year did not in itself correlate with better results in terms of the oncological or urological issues analysed.
Comments The first four studies7–10 all based their conclusions on data from Medicare claims that are designed primarily to provide billing and not clinical information. The first two7,8 do reflect hospital-volume rather then surgeon-volume. They indicate that a higher hospital-volume is correlated with a lower complication rate, lower morbidity and mortality, a shorter hospital stay and lower hospital charges. But as shown by Birkmeyer et al.,12 who examined 474,108 surgical procedures on operative mortality, the association between hospital-volume and operative mortality is largely mediated by surgeon-volume. Begg et al.9 and Hu et al.10 compared hospitalvolume and surgeon-volume; in both studies there was no clear relationship between surgeon-volume and the parameters reviewed. Begg et al.9 compared the very-high-volume-surgeons and even amongst them there was a statistically significant variation. Similarly, Van Poppel et al.11 concluded that there is a variation in outcome that is not related to the caseload but caused by individual differences in surgical quality and skill. Unlike procedures for treating highly aggressive diseases, the quality of RPr may best be evaluated by long-term rather then short-term morbidity and mortality. Further evaluation of surgical-volume is needed, especially, as it relates to continence, potency and long-term survival.13,14 Quality of surgery is a very controversial issue and no studies have actually reported on the standard of surgical quality in the treatment of cancer. For radical prostatectomy, as for any other cancer, survival and local control are the treatment endpoints. Prostate cancer, however, is not killing the majority of patients who have the disease.
H. Van Poppel et al. Therefore, its surgical treatment should be feasible with the lowest complication rate. The quality assessment proposed11 might be useful to distinguish better and poorer performance of radical prostatectomy. Not the number of cases performed but the mean quality of 10 consecutive cases performed by the same surgeon can indicate which surgeon is indeed qualified to surgically treat patients with localized prostate cancer with good oncological and urological outcome. Further investigation on this issue will have an impact on the ongoing competition between different therapeutic modalities for early prostate cancer, being the retropubic, perineal and laparoscopic prostatectomy and the external beam and interstitial radiotherapy. Quality assessment of surgical care can improve the outcomes of surgical teams and individual surgeons that modify their techniques and can consider retraining in a highly experienced centers. All this may also have a political impact.15 In the USA the American Urological Association is exploring the possibility of a two-tiered system whereby ‘urological surgeons’ are allowed to perform major operative procedures while ‘office-based urologist’ can only handle diagnosis and follow-up. One way of evaluation might consist in the analysis of the records of health insurance instances that have to reimburse supplementary medical care as readmission, urethrotomies, and adjuvant oncological treatment (radiotherapy, hormone treatment.).
Conclusion Several reports about radical prostatectomy have indicated that a higher hospital-volume is correlated with a lower complication rate, lower morbidity and mortality, a shorter hospital stay and lower hospital charges. The same does not seem to hold for the surgical-volume of an individual surgeon. An excellent lower-volume urologist can do as well as a high-volume colleague. Since, radical prostatectomy is the standard treatment for the most frequent male malignancy and is offered to many patients that might never ever suffer from the disease, the procedure must be performed with the highest guarantee of quality. The EORTC study has shown that quality can be assessed even by retrospective sampling looking at just a few parameters. This assessment can evaluate an individual surgeon’s skill related to morbidity and oncologic result. The surgeon’s caseload
Quality assurance issues in radical prostatectomy cannot in itself be considered as a determinant of surgical quality.
References 1. Bianco FJ, Begg CB, Scardino PT. Hospital and surgeon volume: impact on outcomes after radical prostatectomy. Am J Urol Rev 2004;2:246–51. 2. Horiot JC. Rationale for a quality assurance programme in clinical trials of conservative management of breast carcinoma. Radiother Oncol 1991;22:222–5. 3. Van de Velde CJH. Lessons from clinical trials in surgical oncology: time for a new paradigm? Eur J Cancer 1999; (Suppl):S299 [Abstract 1200]. 4. Birkmeyer JD, Siewers AE, Finlayson EVA, Stukel TA, Lucas FL, Batista I, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: 1128–36. 5. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998;280:1747–51. 6. Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature Ann Intern Med 2002;137:511–20.
655 7. Yao SL, Lu-Yao G. Population based study of the relationships between hospital-volume of prostatectomies, patient outcomes and length of hospital stay. J Natl Cancer Inst 1999;91:1950–6. 8. Ellison LM, Heany JA, Brikmeyer JD. The effect of hospitalvolume on mortality and resource use after radical prostatectomy. J Urol 2000;163:867–9. 9. Begg CB, Riedel ER, Bach PB, Kattan MW, Schrag D, Warren JL, et al. Variations in morbidity after radical prostatectomy. N Engl J Med 2002;346:1138–44. 10. Hu JC, Gold KF, Pashos CL, Metha SS, Litwin MS. Role of surgeon-volume in radical prostatectomy outcomes. J Clin Oncol 2003;21:401–5. 11. Van Poppel H, Collette L, Kirkali Z, Brausi M, Hoekstra W, Newling DW, et al. Quality control of radical prostatectomy: a feasibility study. Eur J Cancer 2001;37:884–91. 12. Birkmeyer JD, Stuckel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med 2003;349: 2117–27. 13. Potosky AL, Warren JL. Radical prostatectomy: does highervolume lead to a better quality? J Natl Cancer Inst 1999;91: 1906–7. 14. Begg CB, Scardino PT. Taking stock of-volume-outcome studies. J Clin Oncol 2003;21:393–4. 15. McCullough DL. Manpower needs in urology in the twentyfirst century. Urol Clin North Am 1998;25:15–22.
www.ejso.com
Quality assurance issues in radical prostatectomy H. Van Poppel*, S.F.F. Boulanger, S. Joniau Department of Urology, University Hospital Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Accepted for publication 10 February 2005 Available online 15 April 2005
KEYWORDS Radical prostatectomy; Quality; Hospital-volume; Surgical-volume
Abstract Aims: Quality of surgery is a controversial issue and no studies are reporting on the standard of surgical quality in the treatment of urological cancer. The question is whether quality can be evaluated and whether there is a standard for a qualitatively well performed radical retropubic prostatectomy. Methods: We reviewed the literature on this topic. Data of four large studies based on Medicare claims and an EORTC report were analysed. Results: Two studies reflect hospital-volume rather than surgeon-volume. Two compared hospital-volume and surgeon-volume and in both studies there was no clear relationship between surgeon-volume and the parameters reviewed. Similarly, the EORTC study concluded that there is a variation in outcome that is not related to the caseload and proposed minimal quality standards for radical prostatectomy. Conclusions: There is no clear relationship between surgeon-volume and surgical quality. Since radical prostatectomy is the standard treatment for the most frequent male malignancy and is offered to many patients that might never even suffer from the disease, the procedure must be performed with the highest guarantee of quality. Although, quality control of radical prostatectomy is feasible, its implementation will still require an enormous effort from the urological community. Q 2005 Elsevier Ltd. All rights reserved.
Introduction Most research on the outcome of surgery in relation to the hospital or the surgeon volume has rather been limited to exploring these procedures with a high risk of mortality. For procedures with low mortality such as radical prostatectomy, the morbidity is the most important parameter to be
* Corresponding author. Tel.: C32 16 34 69 30; fax: C32 16 34 69 31. E-mail address: [email protected] (H. Van Poppel).
analysed.1 Indeed, prostate cancer is the commonest malignancy in men and radical prostatectomy (RPr) is the most commonly used treatment. Quality of surgery is a controversial issue and no studies are reporting on the standard of surgical quality in the treatment of urological cancer. Radiotherapists have preceded surgeons when they developed detailed technical guidelines for radiation therapy.2 Numerous trials are carried out on the value of adjunctive treatments to surgery, such as radiotherapy or hormonal therapy, while neglecting the important function of the surgeon and surgical procedure as a prognostic factor.3 For lung, liver, pancreatic, oesophageal and
0748-7983/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2005.02.017
Quality assurance issues in radical prostatectomy bladder cancer the mortality and morbidity has been shown to be significantly lower in high volume centers.4–6 Radical prostatectomy is the treatment of choice for localized prostate cancer and many patients undergo the procedure although their disease might not even affect their life expectancy and quality of life if left untreated. The question rises to what extent the quality of the surgical act is an important prognostic factor in the urological and oncological outcome after a radical prostatectomy. Can quality be evaluated and is there a standard for a qualitatively well performed radical retropubic prostatectomy (RRP)?
Literature review Yao and Yao7 collected data on more then 100,000 prostatectomies from Medicare claims filed from 1991 through 1994. He found an inverse relationship between the hospital-volume of radical prostatectomies and the relative risk of readmission, serious complications and mortality. The mean length of hospital stay was longer in low-volume hospitals (Table 1). Surgical-volume was associated with both the outcome and the length of hospital stay. A decrease in the length of stay, with equivalent or improved outcomes, might be achieved by increasing the number of operations performed. The application of this principle could greatly reduce the costs while maintaining or improving the quality of care. Ellison et al.8 identified over 60,000 men who underwent a radical prostatectomy between 1989 and 1995. They compared hospital-volume with inhospital mortality, overall length of stay and total hospital charges. The primary outcome measure of Table 1
651 the study was the in-hospital mortality, the secondary outcome measures were length of stay and hospital charges (Table 2). The overall adjusted in-hospital mortality after radical prostatectomy was relatively low (0.25%), but patients at lowvolume hospitals were 78% more likely to have inhospital mortality. The overall length of stay decreased over-time in all hospitals but the overall length of stay and the total hospital charges where higher in the low-volume-hospitals. The authors quoted that further study regarding other important outcome measures, including incontinence, impotence, and long-term patient survival, is necessary. Begg et al.9 also used a Medicare linked database to evaluate the mortality rate, post-operative complication rate, late urological complications (strictures, fistulas) and long-term incontinence after the operation in over 10,000 patients between 1992 and 1996. They compared to hospital-volume and to surgeon-volume. Neither hospital- nor surgeon-volume was significantly associated with surgery-related death. There was a difference in the post-operative complication rate and the late urinary complication rate. These complications were lower in very-high-volume-hospitals (114–252 pts per year) and were also lower when the prostatectomy was performed by very-high-volume surgeons (33–121 pts per year). There was no difference in long-term incontinence, nor between hospital nor between surgeon-volume (Table 3). Although, for all parameters there seems to be a trend for better quality (lower complication rate, fewer incontinence and lower mortality) none of these differences were statistically significant. They also studied the outcomes associated with the 159 surgeons in the highest-volume categories
Data collected on 101,604 radical prostatectomies between 1991 and 1994 by Yao and Yao et al.7
Hospital volume
Length of stay (days)
Serious complications (%)
Readmissions in 30 days
Mortality
Low Medium–low Medium–high High
8.50 8.18 7.70 7.81
31.3 28.7 27.8 26.3
5.0 4.5 4.3 4.1
0.63 0.59 0.56 0.39
Compared with high volume hospitals
Relative risk of readmission in 30 days (%)
Relative risk of serious complication (%)
Relative risk of mortality (%)
Low Medium–low Medium–high
30 16 8
43 25 9
51 43 42
Hospitals were classified into four categories of volume on the number of prostatectomies performed during the studied period: low (!38), median low (39–74), median high (74–140), high volume hospitals (O140 prostatectomies). In the four groups data on length of stay, serious complications, readmissions in 30 days and the mortality rates were collected. The relative risk compared with high volume hospitals.
652
H. Van Poppel et al.
Table 2 Data collected on 66,693 radical prostatectomies between 1989 and 1995 by Ellison et al.8
Table 4 Data collected on 2292 radical prostatectomies between 1992 and 1996 by Hu et al.10
Hospital volume
Length of stay (days)
Charges ($)
Mortality (%)
Odds ratio mortality
Hospital volume
Complication rate (%)
Stricture of the anastomosis (%)
Hospitals stay (days)
Low Medium High
5.4 4.80 4.20
15,600 15,100 13,500
0.30 0.20 0.17
1.78 1.71 1.00
Low High
21.6 16.8
26.8 19.8
5.2 4.4
Surgeon volume
Complication rate (%)
Stricture of the anastomosis (%)
Hospitals stay (days)
Low High
21.9 11.8
27.7 22.0
5.2 4.1
Hospitals were classified into three categories of prostatectomy volume: low (!25), medium (25–54), high volume hospitals (O54 radical prostatectomies performed).
(33–121 pts). There were statistically significant variations among those surgeons in the rates of post-operative complications, late urinary complications and long-term incontinence. This evidence suggests that there is a genuine variation in surgical performance which is not reflected by the quantity of operations performed. Hu et al.,10 used a national 5% random sample of claims data of Medicare about more than 2000 procedures. They analysed in-hospital complications, anastomotic strictures and the length of stay compared with hospital-volume and surgeonvolume (Table 4). Hospital-volume was not significantly associated with complications. High-volume hospitals (HVH) tended to have fewer anastomotic strictures, whereas surgeon-volume had no impact on their incidence. There are two explanations for this counterintuitive finding. First, HVHs had younger patients and age is a cofactor in developing strictures. Second, HVHs may disseminate advances Table 3
Data on hospital volume. Hospitals were classified into two categories of prostatectomy volume: low (!60) and high volume hospitals (O60). Data on surgeon volume. Surgeon were classified into two categories of prostatectomy volume: low (!40) and high volume surgeons (O 40).
in surgical technique between the surgeons more readily (e.g. bladder mucosa eversion, mucosa– mucosa apposition). The authors concluded that additional study on potency, continence and longterm survival is warranted. As in the former study, there was a trend, but not a statistically significant difference on complication rates, number of strictures and length of stay in favor of higher hospital and surgeon volumes. In a study of the EORTC GU group,11 the authors tried to assess whether the quality of the surgical act can be assessed by any means. Questionnaires were collected from 23 different institutes
Data collected on 11,522 radical prostatectomies between 1992 and 1996 by Begg et al.9
Hospital Volume
Complication rate (%)
Late urinary complication rate (symptoms/treatment) (%)
Long-term incontinence (symptoms/ treatment) (%)
Mortality
Low Medium High Very-high
32 31 30 27
28/18 29/19 23/16 20/13
19/6.5 19/6.4 18/7.0 18/7.6
0.5 0.5 0.5 0.5
Surgeon volume
Complication rate (%)
Late urinary complication rate (symptoms/treatment) (%)
Long-term incontinence (symptoms/ treatment) (%)
Mortality
Low Medium High Very-high
32 31 30 26
28/19 26/18 27/17 20/14
20/7.3 20/7.2 19/6.7 16/6.6
0.4 0.5 0.5 0.4
Data on hospital volume. Hospitals were classified into four categories of prostatectomy volume: low (1–33), medium (34–61), high (62–107) and very high volume hospitals (114–252). Data on surgeon volume. Surgeon were classified into four categories of prostatectomy volume: low (1–10), medium (11–19), high (20–32) and very high volume surgeons (33–121).
Quality assurance issues in radical prostatectomy Table 5
653
Data collected on 232 radical prostatectomies from 27 urologists by Van Poppel.11
Surgeon volume Surgical results Low Medium High Surgeon volume Oncological results Low Medium High Surgeon volume Urological results Low Medium High
Blood transfusion none (%)
Blood transfusion 1– 3 units (%)
Blood transfusion R3 units (%)
35.1 26.4 61.8
47.5 55.1 38.2
17.4 18.5 0.0
Surgical margin negative (%)
Surgical margin positive (%)
PSA !detection level (%)
PSA Rdetection level (%)
63.3 62.6 69.4
36.8 37.4 30.6
49.6 65.0 55.2
50.4 35.0 44.8
Incontinence at 3 months none (%)
Incontinence at 3 months drops (%)
Incontinence at 3 months R1 pad (%)
7.6 2.9 2.0
82.4 76.0 75.8
10.0 21.1 22.2
Low (!12), medium (12–25), high (O50) volume surgeons.
collecting 232 radical prostatectomies performed for T1, T2 prostate cancer (Table 5). Blood loss, duration of surgery, surgical margin status, post-operative prostate specific antigen (PSA) and urinary incontinence after 3 months were recorded and analysed with respect to surgical-volume 27 urologists were asked to fill out these data on 10 consecutive radical prostatectomies. They were asked how many radical prostatectomies they were used to perform per year. From these 10 files the mean for each parameter was calculated and used in the analysis. The blood loss was analysed by the blood transfusion need, although this is not a very good parameter to asses the quality of surgery. Some urologists and anaesthesiologists are used to replace any blood loss, while others give transfusions only for a hemoglobin level below 100 g/l. The duration of surgery may not be a good parameter to assess quality. Urologists with a large caseload are not always faster. The authors found no relationship between the duration of surgery and any other parameter such as hemorrhage, margin status, PSA level or incontinence. The oncological result was related to the surgical margin status and the 3-month post-operative PSA level. It was recognized that the number of positive margins is related to the initial stage of the disease and the initial PSA, which were not equally balanced in the different institutes. It was acknowledged that the PSA level at 3 months is not an optimal parameter of surgical quality since PSA
persistence or PSA recurrence shortly after surgery often means systemic disease and can therefore be completely independent of the quality of surgery. Nevertheless, this study was able to demonstrate some important issues. Urologists who performed fewer radical prostatectomies tended to have significantly more hemorrhage. Margin positivity varied between the different surgeons but the margin status proved to be comparable for all participants without any correlation with the number of radical prostatectomies done. The same held for the PSA-free status at 3 months emphasizing on the fact that the oncological result was not related to the number of surgeries performed but to the individual surgical skill. In respect of urinary continence, again the percentage of completely continent patients at 3 months was not correlated to the surgeon’s experience and the number of cases performed. Using this simple retrospective analysis of 10 consecutive cases done by one urological surgeon, the authors were able to distinguish whether the quality of the surgery of a given urologist was good, fair or poor for each of the points of technique assessed. None of the participating surgeons was good or bad for all items. When ranking for oncological (margins and PSA) and urological (incontinence) outcome one participant was quoted as best and four other participants as poor performers of radical prostatectomy. The authors proposed what a ‘standard radical retropubic prostatectomy’ should mean. The
654 analysis of 10 consecutive radical prostatectomies for T1T2 prostate cancer should reveal a mean blood transfusion need of less than three units, a mean operative time of less than 3 h, a mean occurrence of positive surgical margins in less than 20%, a PSA below detection level at 3 months in more than 80% and a complete urinary continence at 3 months in more than 50%. These standards proposed need further validation through a more extensive surgical quality survey. From this study, however, it was clear that a higher number of radical prostatectomies performed each year did not in itself correlate with better results in terms of the oncological or urological issues analysed.
Comments The first four studies7–10 all based their conclusions on data from Medicare claims that are designed primarily to provide billing and not clinical information. The first two7,8 do reflect hospital-volume rather then surgeon-volume. They indicate that a higher hospital-volume is correlated with a lower complication rate, lower morbidity and mortality, a shorter hospital stay and lower hospital charges. But as shown by Birkmeyer et al.,12 who examined 474,108 surgical procedures on operative mortality, the association between hospital-volume and operative mortality is largely mediated by surgeon-volume. Begg et al.9 and Hu et al.10 compared hospitalvolume and surgeon-volume; in both studies there was no clear relationship between surgeon-volume and the parameters reviewed. Begg et al.9 compared the very-high-volume-surgeons and even amongst them there was a statistically significant variation. Similarly, Van Poppel et al.11 concluded that there is a variation in outcome that is not related to the caseload but caused by individual differences in surgical quality and skill. Unlike procedures for treating highly aggressive diseases, the quality of RPr may best be evaluated by long-term rather then short-term morbidity and mortality. Further evaluation of surgical-volume is needed, especially, as it relates to continence, potency and long-term survival.13,14 Quality of surgery is a very controversial issue and no studies have actually reported on the standard of surgical quality in the treatment of cancer. For radical prostatectomy, as for any other cancer, survival and local control are the treatment endpoints. Prostate cancer, however, is not killing the majority of patients who have the disease.
H. Van Poppel et al. Therefore, its surgical treatment should be feasible with the lowest complication rate. The quality assessment proposed11 might be useful to distinguish better and poorer performance of radical prostatectomy. Not the number of cases performed but the mean quality of 10 consecutive cases performed by the same surgeon can indicate which surgeon is indeed qualified to surgically treat patients with localized prostate cancer with good oncological and urological outcome. Further investigation on this issue will have an impact on the ongoing competition between different therapeutic modalities for early prostate cancer, being the retropubic, perineal and laparoscopic prostatectomy and the external beam and interstitial radiotherapy. Quality assessment of surgical care can improve the outcomes of surgical teams and individual surgeons that modify their techniques and can consider retraining in a highly experienced centers. All this may also have a political impact.15 In the USA the American Urological Association is exploring the possibility of a two-tiered system whereby ‘urological surgeons’ are allowed to perform major operative procedures while ‘office-based urologist’ can only handle diagnosis and follow-up. One way of evaluation might consist in the analysis of the records of health insurance instances that have to reimburse supplementary medical care as readmission, urethrotomies, and adjuvant oncological treatment (radiotherapy, hormone treatment.).
Conclusion Several reports about radical prostatectomy have indicated that a higher hospital-volume is correlated with a lower complication rate, lower morbidity and mortality, a shorter hospital stay and lower hospital charges. The same does not seem to hold for the surgical-volume of an individual surgeon. An excellent lower-volume urologist can do as well as a high-volume colleague. Since, radical prostatectomy is the standard treatment for the most frequent male malignancy and is offered to many patients that might never ever suffer from the disease, the procedure must be performed with the highest guarantee of quality. The EORTC study has shown that quality can be assessed even by retrospective sampling looking at just a few parameters. This assessment can evaluate an individual surgeon’s skill related to morbidity and oncologic result. The surgeon’s caseload
Quality assurance issues in radical prostatectomy cannot in itself be considered as a determinant of surgical quality.
References 1. Bianco FJ, Begg CB, Scardino PT. Hospital and surgeon volume: impact on outcomes after radical prostatectomy. Am J Urol Rev 2004;2:246–51. 2. Horiot JC. Rationale for a quality assurance programme in clinical trials of conservative management of breast carcinoma. Radiother Oncol 1991;22:222–5. 3. Van de Velde CJH. Lessons from clinical trials in surgical oncology: time for a new paradigm? Eur J Cancer 1999; (Suppl):S299 [Abstract 1200]. 4. Birkmeyer JD, Siewers AE, Finlayson EVA, Stukel TA, Lucas FL, Batista I, et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: 1128–36. 5. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on operative mortality for major cancer surgery. JAMA 1998;280:1747–51. 6. Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature Ann Intern Med 2002;137:511–20.
655 7. Yao SL, Lu-Yao G. Population based study of the relationships between hospital-volume of prostatectomies, patient outcomes and length of hospital stay. J Natl Cancer Inst 1999;91:1950–6. 8. Ellison LM, Heany JA, Brikmeyer JD. The effect of hospitalvolume on mortality and resource use after radical prostatectomy. J Urol 2000;163:867–9. 9. Begg CB, Riedel ER, Bach PB, Kattan MW, Schrag D, Warren JL, et al. Variations in morbidity after radical prostatectomy. N Engl J Med 2002;346:1138–44. 10. Hu JC, Gold KF, Pashos CL, Metha SS, Litwin MS. Role of surgeon-volume in radical prostatectomy outcomes. J Clin Oncol 2003;21:401–5. 11. Van Poppel H, Collette L, Kirkali Z, Brausi M, Hoekstra W, Newling DW, et al. Quality control of radical prostatectomy: a feasibility study. Eur J Cancer 2001;37:884–91. 12. Birkmeyer JD, Stuckel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med 2003;349: 2117–27. 13. Potosky AL, Warren JL. Radical prostatectomy: does highervolume lead to a better quality? J Natl Cancer Inst 1999;91: 1906–7. 14. Begg CB, Scardino PT. Taking stock of-volume-outcome studies. J Clin Oncol 2003;21:393–4. 15. McCullough DL. Manpower needs in urology in the twentyfirst century. Urol Clin North Am 1998;25:15–22.