- Email: [email protected]
MAKING ENDS MEET: A COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
0022-5347/04/1721-0269/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 172, 269 –274, July 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000128773.99707.5b
MAKING ENDS MEET: A COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY RICHARD E. LINK,* LI-MING SU, SAM B. BHAYANI
AND
CHRISTIAN P. PAVLOVICH
From the James Buchanan Brady Urological Institute, Johns Hopkins Bayview Medical Center, Baltimore, Maryland
ABSTRACT
Purpose: We compared the perioperative costs of laparoscopic radical prostatectomy (LRP) and open radical retropubic prostatectomy (RRP) at a metropolitan hospital by developing a detailed computer model. Materials and Methods: Our predictive model incorporates institutional cost centers for operative time, operating room consumables, professional fees, hospital room and board, oral analgesics, autologous blood banking, blood transfusion and cystography. Versions with and without pelvic lymphadenectomy (PLND) were evaluated using 1 and 2-way sensitivity analyses. Operative times, lengths of stay and transfusion rates were derived from published series. We also reviewed individual hospital charges for 172 consecutive prostatectomy cases for comparison and validation of model predictions. Results: The model predicted cost premiums for LRP of 14.4% (without PLND) and 17.5% (with PLND). The actual hospital charge premium for LRP and PLND was 18.4%, which differed from the predicted cost premium by less than 1%. The most significant cost centers in order of importance were operative time, length of stay and consumables. To achieve cost equivalence with RRP, operative times would need to average 159 minutes (LRP and PLND) and 174 minutes (LRP alone) holding other factors constant. Cost equivalence could not be achieved by shortening hospital stay alone unless LRP were performed as an outpatient procedure. Conclusions: Our model predicts the perioperative costs of LRP to be greater than RRP by a factor of less than 1.2⫻. If disposable instruments and trocars are eliminated, and patients undergoing LRP and PLND are discharged on postoperative day 2, cost equivalence with RRP and PLND can be achieved with operative times of 3.4 hours. KEY WORDS: costs and cost analysis, prostatectomy, prostatic neoplasms
For 2003 the American Cancer Society projects that more than 229,000 new cases of prostate cancer will be diagnosed and more than 29,000 patients will die of this disease in the United States. With an estimated cost per case of $35,000 over all illness stages,1 the projected lifetime cost for these new cases would exceed $8 billion. Clearly the economic impact of prostate cancer treatment is substantial and worth considering when evaluating any new treatment modality. During the last 5 years there has been a resurgence of interest in laparoscopic approaches to radical prostatectomy for clinically localized disease. Driven by patient desire for more minimally invasive approaches to prostatectomy, several centers around the world are now performing a large volume of laparoscopic radical prostatectomy (LRP) surgery. LRP has several potential technical advantages compared to traditional open radical retropubic prostatectomy (RRP). These include improved surgical visualization, optical magnification, less tissue handling and decreased blood loss due to CO2 insufflation pressures. Disadvantages to LRP include a steep learning curve, decreased tactile feedback, little longterm outcome data, longer operative times and potentially greater cost. The morbidity advantages associated with LRP are likely to be subtle compared to RRP. Therefore, it will be critical for LRP to be economically viable for it to remain in our surgical armamentarium. Yet the relative cost of LRP compared to RRP is an area that has received surprisingly little attention
in the literature. In the only published large LRP series addressing the subject of cost, Guillonneau and Vallancien found that LRP was actually cheaper (by $1,237) than RRP at their institution in Paris.2 The cost savings were primarily due to a decrease in hospital stay from 8 to 6 days. Given the much shorter hospital stays for LRP and RRP in the United States, such cost savings are unlikely to be realized here. Subjectively we might predict that LRP would be much more costly than RRP based on operative time and costs of laparoscopic instrumentation. Yet despite its importance, the magnitude of any cost premium for LRP remains unresolved. We have developed a detailed computer model for predicting the perioperative cost of LRP and RRP at our institution. This type of analysis has the potential to identify the magnitude of any cost premium for LRP and single out individual factors that could be improved to make LRP more cost equivalent to RRP. MATERIALS AND METHODS
Cost comparison model. We developed a mathematical model incorporating 9 cost centers from a metropolitan hospital (see Appendix). Our analysis focused on the perioperative costs associated with an uncomplicated LRP or RRP with and without bilateral pelvic lymphadenectomy (PLND). Cost centers include operative time, surgical consumables, professional fees, hospital room and board, oral analgesics, autologous blood banking, blood transfusion and postoperative cystography. In all cases cost data rather than hospital charges were used for model development. We made several assumptions to simplify our model and eliminate components that were unlikely to differ signifi-
Accepted for publication January 30, 2004. * Correspondence: Brady Urological Institute, Johns Hopkins Bayview Medical Center, 4940 Eastern Ave., Room A-345, Baltimore, Maryland 21224 (telephone: 410-550-3506; FAX: 410-550-3341; email: [email protected]). 269
270
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
cantly between procedures. For example the costs of office visits, imaging, laboratory testing and surgical pathology were excluded. Some factors were excluded because they were too dependent on individual physician preference for consistent modeling (ie inhalational anesthetic costs). We included only oral analgesics since parenteral opiate requirements after LRP and RRP are similar.3 Capital equipment costs were excluded since none of these items are used solely for prostatectomy. We also made several assumptions based on our own clinical practice. It is our policy to have patients undergoing RRP but not those undergoing LRP bank 2 units of autologous blood. We also assume that if transfusion is required the patient undergoing RRP will receive their autologous units whereas those undergoing LRP will receive 2 units of heterologous packed red blood cells. Finally, after LRP we perform cystography before early catheter removal but after RRP the catheter is removed at 12 to 20 days without cystography. At our hospital operating room overhead cost for the first hour of surgery is $960 with each subsequent 15-minute block contributing $240. Surgical consumables costs were calculated from a comprehensive list of items included on the surgical field and reflect purchase price. The 6 most expensive disposables include the ultrasonic scalpel ($358.31, Ethicon Endosurgery, Cincinnati, Ohio), Visiport trocar ($77.86, United States Surgical [USS], Norwalk, Connecticut, Endoshear 5 mm cautery ($70.49, USS), Endocatch bag ($78.47, USS), Versaport 5–12 mm trocar ($55.75, USS) and Surgiport 5 mm trocar ($45.90, USS). Surgeon fees were derived from 2003 Maryland Medicare reimbursement rates based on Current Procedural Terminology (CPT) code. For RRP separate codes for procedures with (55845, $1,631.63) and without (55840, $1,328.68) PLND were used. For LRP there is currently no CPT code to reflect LRP and PLND. Therefore, LRP and PLND was coded as 2 procedures (LRP, 55866, $1,650.88 and laparoscopic PLND, 38571, $772.26). Anesthesiologist fees were derived from 2003 Medicare reimbursement rates using a Relative Value Unit (RVU) base value of 6 and the Baltimore county conversion factor ($17.32). Hospital room and board is calculated for a semiprivate room as $735 for the first night and $620 for each subsequent 24-hour period. Oral analgesic prices were calculated from the Bayview pharmacy cost for oxycodone ($0.92 per tablet) and the
number of tablets was derived from published data.3 The cost for autologous blood banking, purchase of heterologous leuko-reduced blood and blood administration were provided by our blood bank. Cystography costs reflect 2003 Medicare reimbursement (74430) and include a professional fee of $16.89 and a technical fee of $41.40. Model simulation and sensitivity analysis. The cost model was programmed into Data 4.0 software (TreeAge Software, Inc., Williamstown, Massachusetts) and simulated with and without PLND. For starting values we performed a comprehensive 10-year MEDLINE search to identify contemporary series of LRP and RRP. Weighted means were calculated from the combined series (table 1). Since European length of stay after radical prostatectomy is generally longer than in the United States, only domestic series were included in length of stay calculations. Since most published series do not stratify operative time with and without PLND, we had to estimate the time contribution of this procedure (20 minutes). Sensitivity analysis is a well established technique for evaluating cost-effectiveness. In 1-way analysis, a single cost factor is altered and its impact on total cost is determined by keeping all other factors fixed. In 2-way analysis 2 factors are altered simultaneously. We performed 1-way and 2-way sensitivity analyses by holding the variables for RRP constant and adjusting 1 or 2 variables for LRP. Model validation. To provide model validation we compared model predictions to financial data from actual clinical cases. Ideally, these data should reflect hospital costs or reimbursement, but this information is difficult to accumulate due to accounting practices that focus on hospital charges. Based on our cost model we calculated a predicted cost premium for LRP compared to RRP, and compared this value to the charge premium for LRP calculated from hospital charges. Detailed hospital financial records were reviewed for all patients undergoing LRP (128) or RRP (44) from June 2001 through July 2003. Since we perform PLND in the majority of our prostatectomy cases, these charges were assumed to reflect prostatectomy and PLND. Our hospital accounting office provided these records, which reflect almost 20,000 individual charges. Statistical analysis was performed using the SPSS 10 statistical package (SPSS, Inc., Chicago, Illinois). Frequency histograms were prepared for total
TABLE 1. Literature review for laparoscopic radical prostatectomy and open radical retropubic prostatectomy References
No. Pts
Operating Room Mins
LRP: 438 Rassweiler et al7 33 Bhayani et al3 120 Guillonneau and Vallancien2 43 Abbou et al8 180 Rassweiler et al9 350 Guillonneau et al10 125 Turk et al11 180 Hoznek et al12 50 Bollens et al13 70 Dahl et al14 Weighted mean Open RRP: 906 Hsu et al15 219 Rassweiler et al7 100 Guillonneau and Vallancien2 418 Hautmann et al16 472 Dillioglugil et al17 — Hammerer et al18 116 Gaylis et al19 — Catalona et al6 638 Arai et al20 24 Bhayani et al3 Weighted mean Note that only studies from the United States were used for prostatectomy in European centers.
253 348 239 320 271 217 255 240 317 274 263 130 196 135 168 182 — 155 — 263 168 180 calculating length
No. Pts
— 33 — — — — — — — 70 975 — — — 472 — 116 — — 24
Length of Stay (days) — 2.97 — — — — — — — 2.50 2.65
No. Pts
Transfusion Rate %
438 — 120 — 180 350 125 200 50 70
2.4 990 — 219 — 100 — — 6.2 472 — 320 3.0 — — 1,870 — 638 3.04 — 3.58 of stay results, reflecting differing practices for inpatient
9.8 — 10.0 — 17.0 5.7 2.0 3.0 13.0 5.7 8.5 8.2 26.9 31.0 — 28.6 27.8 — 9.0 19.1 — 14.9 management of
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
charges per case for LRP and RRP. The KolmogorovSmirnov test of normality confirmed that LRP (Z ⫽ 1.263, p ⫽ 0.082) and RRP (Z ⫽ 0.871, p ⫽ 0.433) data conformed to a normal distribution. Therefore, mean charges were calculated and compared using the Student t test. How-
271
ever, the LRP charge data histogram was asymmetric and significantly skewed toward the right (skewness statistic 1.120 ⫾ 0.214). For this reason median rather than mean charges for LRP and RRP were used to calculate the charge premium for LRP.
TABLE 2. Model predictions for the cost breakdown of laparoscopic and open radical prostatectomy with and without bilateral pelvic lymphadenectomy Bilat PLND
Starting values: Operating room mins (x)* Days of stay (y) Autologous units banked (m) % Transfusion rate (t) No. units transfused (n) Predicted cost: Operating room time (a) Operating room consumables (b) Surgeon fee (c) Anesthesia fee (d) Hospital room ⫹ board (e) Oral pain medications (f) Autologous blood banking (g) Blood transfusion (h) Postop cystogram (i)
$ $ $ $ $ $ $ $ $
No Bilat PLND
LRP
Open RRP
LRP
Open RRP
263 2.65 0 8.5 2
180 3.58 2 14.9 2
243 2.65 0 8.5 2
160 3.58 2 14.9 2
4,320.00 1,132.56 2,423.14 407.60 1,975.00 8.28 0.00 88.68 58.29
$2,880.00 $ 170.36 $1,631.63 $ 311.76 $2,595.00 $ 15.64 $ 979.86 $ 11.99 $ 0.00
$4,080.00 $1,132.56 $1,650.88 $ 384.50 $1,975.00 $ 8.28 $ 0.00 $ 88.68 $ 58.29
$2,640.00 $ 170.36 $1,328.68 $ 288.67 $2,595.00 $ 15.64 $ 979.86 $ 11.99 $ 0.00
$8,596.24
$9,378.19 $1,347.99 (14.4)
$8,030.20
Totals $10,413.55 Cost premium for laparoscopy (%) $ 1,817.30 (17.5) * Lowercase letters correspond to those in Appendix.
FIG. 1. One-way sensitivity analysis for laparoscopic prostatectomy with and without pelvic lymphadenectomy. Variables include operative time (a, b), length of stay (c, d) and transfusion rate (e, f). Intersection of 2 lines represents point of cost equivalence and is indicated with arrow. Lap RP, LRP.
272
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY RESULTS
Review of the literature yielded 10 LRP (1,589 patients) and 10 RRP (2,893 patients) series providing operative times. Length of stay data were scarcer, with only 2 relevant domestic LRP (103 patients) and 4 RRP (1,587 patients) series. For transfusion rate we found 8 and 7 relevant series for LRP (1,533 patients) and RRP (4,609 patients, table 1). Weighted means for these parameters served as starting points for computer modeling. Models were developed for LRP and RRP with and without concurrent PLND (see Appendix). The model predicted a cost of $10,414 for LRP and PLND, and $8,596 for RRP and PLND, giving a cost premium of 17.5% for the laparoscopic procedure. Without PLND the cost premium decreases to 14.4% for LRP ($9,378 vs $8,030), primarily because surgeon fees for laparoscopic PLND and LRP are currently unbundled (unlike for open RRP and PLND) and, therefore, the combined laparoscopic procedure incurs greater surgeon fees (table 2). The most significant factor was operative time, which contributed 42% on average to the cost of LRP and 33% to the cost of RRP. Surgical consumables costs were 11% (LRP) and 2% (RRP). We performed 1-way sensitivity analysis by varying a single variable for LRP while keeping the starting parameters for RRP fixed (fig. 1). The analysis predicts that an operating room time of 159 minutes (2.65 hours) for LRP and PLND would be required to be cost equivalent with RRP and PLND. Without PLND a laparoscopic operating room time of 174 minutes (2.90 hours) yielded equivalent cost to RRP. Length of stay sensitivity analysis identified that equivalence in cost could only be achieved if LRP was performed on an outpatient basis (less than 1 day length of stay). The impact of transfusion rate on cost was more modest and even decreasing the transfusion rate to 0% could not yield cost equivalence between the 2 procedures. Likewise, varying the cost of operating room consumables could not, by itself, equalize the projected cost of LRP and RRP. Two-way sensitivity analysis was used to analyze the interaction of the 2 most significant cost factors, operating room time and length of stay. As expected figure 2 shows that as operating room time increases the length of stay must decrease to achieve a cost advantage for LRP. The stair-step morphology of these curves reflects the threshold effect of length of stay on the model. Using these curves the operative time required for cost equivalence at any particular length of stay can be determined. To validate the model against financial results for clinical cases, we reviewed individual hospital charge records for all patients undergoing radical prostatectomy from July 2001 to June 2003 (128 LRP and 44 RRP, fig. 3). Assuming inclusion of PLND, mean charges for LRP ($15,451 ⫾ $336) and RRP ($12,441 ⫾ $509) were significantly different during this period (p ⬍0.001). Median values for LRP ($14,848) and RRP ($12,120) were then used to calculate the charge premium for LRP (18.4%). The magnitude of this charge premium is remarkably similar to the predicted cost premium for LRP and PLND derived from our computer model (17.5%). DISCUSSION
Laparoscopic radical prostatectomy is a relatively new treatment modality for localized prostate cancer that is generating excitement within the urological community. A variety of centers are now evaluating the oncological and functional outcomes of LRP, and comparing these factors to RRP. However, despite the impact of economics on the viability of any new surgical procedure, little has been written about cost comparisons between LRP and RRP. We have developed a detailed computer model suitable for comparing the anticipated perioperative costs of LRP and RRP at our United States institution. Using the weighted means of published values for operative time, length of stay
FIG. 2. Two-way sensitivity analysis for laparoscopic prostatectomy varying operative time and length of stay. Within dark region of graph LRP is less expensive than RRP. Interface between dark and crosshatched areas represents point of cost equivalence at varying operative times and lengths of stay.
FIG. 3. Comparison of model prediction with financial data derived from 172 consecutive radical prostatectomy cases (128 LRP, 44 RRP).
and transfusion rate, this model predicts that LRP will cost between 14.4% and 17.5% more than RRP. This result represents a worst-case scenario for LRP for several reasons. First, the operative times in several series reflect an experience that includes the learning curve for LRP. As experience with LRP increases, shorter operative times with resultant cost savings are expected. Several groups are now reporting mean operative times approaching 3 hours for these procedures,4 which would reduce the cost premium to only 4%. The second issue relates to CPT coding. Since no CPT code currently exists for a bundled LRP and PLND procedure, it is necessary to code for these procedures separately, which increases the “cost.” Our model likely overestimates this value since as a concurrent procedure coded with the ⫺51 modifier, reimbursement for the laparoscopic PLND would
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
only be a fraction of the $772.26 used for our calculations. If surgeon fees were excluded from the model, the cost premium for LRP and PLND would decrease to 12.8% ($1,025.80), highlighting the importance of CPT unbundling in increasing the relative cost of LRP. Our model also does not factor in costs associated with short and long-term complications of radical prostatectomy. Complications from deep venous thrombosis and pulmonary embolus appear to be infrequent during LRP, possibly due to the steep Trendelenburg position used during surgery and the lack of a pelvic retractor interfering with lower extremity venous return. There is also reason to believe that long-term complications associated with the urethrovesical anastomosis (ie anastomotic stricture and the need for artificial urinary sphincter [AUS] placement) may be less common with the laparoscopic approach due to the improved visualization of the anastomosis during its construction and the standard use of cystography before catheter removal. In our series of 157 LRP cases only 2 (1.2%) evidenced anastomotic stricture and none required postoperative AUS placement. These results compare favorably with published rates of anastomotic stricture (4% to 20%) and AUS placement (0.2% to 2.3%) after open RRP.5, 6 Complications certainly impact the total cost of any surgical procedure. If the lower rates of anastomotic and thromboembolic complications with LRP prove to be reproducible, this may greatly narrow the cost differential between LRP and RRP. This issue is not addressed in the current analysis and deserves further study.
273
As with any modeling exercise there remain limitations to our approach. This model incorporates constants that are specific to our region of the United States and to our medical center. The costs for operative time, hospital room and board, and consumables may vary significantly at different institutions, and the Medicare reimbursement rates for anesthesia services have region specific modifiers. Another criticism could be that we are blurring the boundary between true hospital cost from overhead and Medicare reimbursement for professional fees. Although this is true, we believe that professional fees should be included in any cost analysis of prostatectomy surgery since this factor impacts the global cost of these procedures to society. CONCLUSIONS
We believe that our cost model is comprehensive and functional since its predictions differed by less than 1% from actual hospital charges for prostatectomy at our institution. Ultimately, however, the usefulness of modeling is that it allows various clinical scenarios to be tested for cost equivalence between LRP and RRP. The cost of LRP consumables would decrease to $914.52 if disposable scissors and trocars other than the Visiport were eliminated. Using these disposables and discharging all patients on postoperative day 2, the model predicts that operative time must only decrease to 204 minutes (3.4 hours) for cost equivalence to RRP, which is a reachable goal.
APPENDIX: COST ANALYSIS MODEL FOR COMPARING LAPAROSCOPIC RADICAL PROSTATECTOMY WITH OPEN RADICAL RETROPUBIC PROSTATECTOMY
a OperatingRoomTime ⫹ b ORConsumables ⫹ c SurgeonFee ⫹ d AnesthesiaFee ⫹ e HospitalRoomBoard ⫹ f OralPainMeds ⫹ g AutologousBanking ⫹ h BloodTransfusion ⫹ i PostopCystogram x ⫽ 兩if (PLND) then [ORtime minutes ] else [ORtime minutes ⫺ 20]兩 y ⫽ Length of Stay days z ⫽ 兩if (open) then [m] else [0]兩 m ⫽ numUnits AutologousBloodBanked t ⫽ % PatientsTransfused n ⫽ numUnitsTransfused a OperatingRoomTime ⫽ $960.00 ⫹ 兩if (x ⬍ 61) then (0) else ((roundUP( x ⫺1560))* $240.00)兩 b ORConsumables ⫽ 兩if (laparoscopic) then [$1,132.56 LapRRP ] else [$170.36 OpenRRP ]兩 cSurgeonFeewithnodes ⫽ 兩if (laparoscopic) then ($1,650.88LapRR ⫹ $772.26LymphNodes) else ($1,631.63OpenRRPw/nodes)兩 c SurgeonFee withoutnodes ⫽ 兩if (laparoscopic) then ($1,650.88 LapRRP ) else ($1,328.68 OpenRRP )兩 x d AnesthesiaFee ⫽ [ 15ORtime ⫹ RVU base ]* ConversionFactor Baltimore ⫽ [( 15 ) ⫹ 6] * $17.32 min/unit e HospitalRoomBoard ⫽ 兩if(y ⬎ 0) then ($735.00 ⫹ [roundUP (y ⫺ 1) * $620.00]) else ($0.00)兩 f OralPainMeds ⫽ 兩if (laparoscopic) then [9 * $0.92] else [17 * $0.92]兩 g AutologousBanking ⫽ z * $489.93 hBloodTransfusion ⫽ (t * n * $40.25)TransfusionFee ⫹ 兩if (z ⫽ 0) then [t * n * $481.39] else [0]兩NonAutoUnits i PostopCystogram ⫽ 兩if (laparoscopic) then [$16.89 ProfessionalFee ⫹ $41.40 TechnicalFee ] else [$0.00]兩
REFERENCES
1. Ruchlin, H. S. and Pellissier, J. M.: An economic overview of prostate carcinoma. Cancer, 92: 2796, 2001 2. Guillonneau, B. and Vallancien, G.: Laparoscopic radical prostatectomy: the Montsouris experience. J Urol, 163: 418, 2000 3. Bhayani, S. B., Pavlovich, C. P., Hsu, T. S., Sullivan, W. and Su, L. M.: Prospective comparison of short-term convalescence: laparoscopic radical prostatectomy versus open radical retropubic prostatectomy. Urology, 61: 612, 2003 4. Rozet, F., Cathelineau, X., Widmer, H., Cahill, D., Baumert, H. and Vallancien, G.: Transperitoneal or extraperitoneal approach for laparoscopic radical prostatectomy. A prospective comparison of 200 cases. J Endourol, 17: A111, 2003 5. Benoit, R. M., Naslund, M. J. and Cohen, J. K.: Complications after radical retropubic prostatectomy in the medicare population. Urology, 56: 116, 2000 6. Catalona, W. J., Carvalhal, G. F., Mager, D. E. and Smith, D. S.:
7.
8.
9.
10.
11.
Potency, continence and complication rates in 1,870 consecutive radical retropubic prostatectomies. J Urol, 162: 433, 1999 Rassweiler, J., Seemann, O., Schulze, M., Teber, D., Hatzinger, M. and Frede, T.: Laparoscopic versus open radical prostatectomy: a comparative study at a single institution. J Urol, 169: 1689, 2003 Abbou, C. C., Salomon, L., Hoznek, A., Antiphon, P., Cicco, A., Saint, F. et al: Laparoscopic radical prostatectomy: preliminary results. Urology, 55: 630, 2000 Rassweiler, J., Sentker, L., Seemann, O., Hatzinger, M. and Rumpelt, H. J.: Laparoscopic radical prostatectomy with the Heilbronn technique: an analysis of the first 180 cases. J Urol, 166: 2101, 2001 Guillonneau, B., Cathelineau, X., Doublet, J. D. and Vallancien, G.: Laparoscopic radical prostatectomy: the lessons learned. J Endourol, 15: 441, 2001 Turk, I., Deger, S., Winkelmann, B., Schonberger, B. and
274
12. 13. 14.
15.
16.
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
Loening, S. A.: Laparoscopic radical prostatectomy. Technical aspects and experience with 125 cases. Eur Urol, 40: 46, 2001 Hoznek, A., Salomon, L., Olsson, L. E., Antiphon, P., Saint, F., Cicco, A. et al: Laparoscopic radical prostatectomy. The Creteil experience. Eur Urol, 40: 38, 2001 Bollens, R., Vanden Bossche, M., Roumeguere, T., Damoun, A., Ekane, S., Hoffmann, P. et al: Extraperitoneal laparoscopic radical prostatectomy. Results after 50 cases. Eur Urol, 40: 65, 2001 Dahl, D. M., L’esperance, J. O., Trainer, A. F., Jiang, Z., Gallagher, K., Litwin, D. E. et al: Laparoscopic radical prostatectomy: initial 70 cases at a U.S. university medical center. Urology, 60: 859, 2002 Hsu, E. I., Hong, E. K. and Lepor, H.: Influence of body weight and prostate volume on intraoperative, perioperative, and postoperative outcomes after radical retropubic prostatectomy. Urology, 61: 601, 2003 Hautmann, R. E., Sauter, T. W. and Wenderoth, U. K.: Radical
17.
18.
19. 20.
retropubic prostatectomy: morbidity and urinary continence in 418 consecutive cases. Urology, 43: 47, 1994 ¨ ., Leibman, B. D., Leibman, N. S., Kattan, M. W., Dillioglugil, O Rosas, A. L. and Scardino, P. T.: Risk factors for complications and morbidity after radical retropubic prostatectomy. J Urol, 157: 1760, 1997 Hammerer, P., Hubner, D., Gonnermann, D. and Huland, H.: Perioperative and postoperative complications of pelvic lymphadenectomy and radical prostatectomy in 320 consecutive patients. Urologe A, 34: 334, 1995 Gaylis, F. D., Friedel, W. E. and Armas, O. A.: Radical retropubic prostatectomy outcomes at a community hospital. J Urol, 159: 167, 1998 Arai, Y., Egawa, S., Tobisu, K., Sagiyama, K., Sumiyoshi, Y., Hashine, K. et al: Radical retropubic prostatectomy: time trends, morbidity and mortality in Japan. BJU Int, 85: 287, 2000
Vol. 172, 269 –274, July 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000128773.99707.5b
MAKING ENDS MEET: A COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY RICHARD E. LINK,* LI-MING SU, SAM B. BHAYANI
AND
CHRISTIAN P. PAVLOVICH
From the James Buchanan Brady Urological Institute, Johns Hopkins Bayview Medical Center, Baltimore, Maryland
ABSTRACT
Purpose: We compared the perioperative costs of laparoscopic radical prostatectomy (LRP) and open radical retropubic prostatectomy (RRP) at a metropolitan hospital by developing a detailed computer model. Materials and Methods: Our predictive model incorporates institutional cost centers for operative time, operating room consumables, professional fees, hospital room and board, oral analgesics, autologous blood banking, blood transfusion and cystography. Versions with and without pelvic lymphadenectomy (PLND) were evaluated using 1 and 2-way sensitivity analyses. Operative times, lengths of stay and transfusion rates were derived from published series. We also reviewed individual hospital charges for 172 consecutive prostatectomy cases for comparison and validation of model predictions. Results: The model predicted cost premiums for LRP of 14.4% (without PLND) and 17.5% (with PLND). The actual hospital charge premium for LRP and PLND was 18.4%, which differed from the predicted cost premium by less than 1%. The most significant cost centers in order of importance were operative time, length of stay and consumables. To achieve cost equivalence with RRP, operative times would need to average 159 minutes (LRP and PLND) and 174 minutes (LRP alone) holding other factors constant. Cost equivalence could not be achieved by shortening hospital stay alone unless LRP were performed as an outpatient procedure. Conclusions: Our model predicts the perioperative costs of LRP to be greater than RRP by a factor of less than 1.2⫻. If disposable instruments and trocars are eliminated, and patients undergoing LRP and PLND are discharged on postoperative day 2, cost equivalence with RRP and PLND can be achieved with operative times of 3.4 hours. KEY WORDS: costs and cost analysis, prostatectomy, prostatic neoplasms
For 2003 the American Cancer Society projects that more than 229,000 new cases of prostate cancer will be diagnosed and more than 29,000 patients will die of this disease in the United States. With an estimated cost per case of $35,000 over all illness stages,1 the projected lifetime cost for these new cases would exceed $8 billion. Clearly the economic impact of prostate cancer treatment is substantial and worth considering when evaluating any new treatment modality. During the last 5 years there has been a resurgence of interest in laparoscopic approaches to radical prostatectomy for clinically localized disease. Driven by patient desire for more minimally invasive approaches to prostatectomy, several centers around the world are now performing a large volume of laparoscopic radical prostatectomy (LRP) surgery. LRP has several potential technical advantages compared to traditional open radical retropubic prostatectomy (RRP). These include improved surgical visualization, optical magnification, less tissue handling and decreased blood loss due to CO2 insufflation pressures. Disadvantages to LRP include a steep learning curve, decreased tactile feedback, little longterm outcome data, longer operative times and potentially greater cost. The morbidity advantages associated with LRP are likely to be subtle compared to RRP. Therefore, it will be critical for LRP to be economically viable for it to remain in our surgical armamentarium. Yet the relative cost of LRP compared to RRP is an area that has received surprisingly little attention
in the literature. In the only published large LRP series addressing the subject of cost, Guillonneau and Vallancien found that LRP was actually cheaper (by $1,237) than RRP at their institution in Paris.2 The cost savings were primarily due to a decrease in hospital stay from 8 to 6 days. Given the much shorter hospital stays for LRP and RRP in the United States, such cost savings are unlikely to be realized here. Subjectively we might predict that LRP would be much more costly than RRP based on operative time and costs of laparoscopic instrumentation. Yet despite its importance, the magnitude of any cost premium for LRP remains unresolved. We have developed a detailed computer model for predicting the perioperative cost of LRP and RRP at our institution. This type of analysis has the potential to identify the magnitude of any cost premium for LRP and single out individual factors that could be improved to make LRP more cost equivalent to RRP. MATERIALS AND METHODS
Cost comparison model. We developed a mathematical model incorporating 9 cost centers from a metropolitan hospital (see Appendix). Our analysis focused on the perioperative costs associated with an uncomplicated LRP or RRP with and without bilateral pelvic lymphadenectomy (PLND). Cost centers include operative time, surgical consumables, professional fees, hospital room and board, oral analgesics, autologous blood banking, blood transfusion and postoperative cystography. In all cases cost data rather than hospital charges were used for model development. We made several assumptions to simplify our model and eliminate components that were unlikely to differ signifi-
Accepted for publication January 30, 2004. * Correspondence: Brady Urological Institute, Johns Hopkins Bayview Medical Center, 4940 Eastern Ave., Room A-345, Baltimore, Maryland 21224 (telephone: 410-550-3506; FAX: 410-550-3341; email: [email protected]). 269
270
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
cantly between procedures. For example the costs of office visits, imaging, laboratory testing and surgical pathology were excluded. Some factors were excluded because they were too dependent on individual physician preference for consistent modeling (ie inhalational anesthetic costs). We included only oral analgesics since parenteral opiate requirements after LRP and RRP are similar.3 Capital equipment costs were excluded since none of these items are used solely for prostatectomy. We also made several assumptions based on our own clinical practice. It is our policy to have patients undergoing RRP but not those undergoing LRP bank 2 units of autologous blood. We also assume that if transfusion is required the patient undergoing RRP will receive their autologous units whereas those undergoing LRP will receive 2 units of heterologous packed red blood cells. Finally, after LRP we perform cystography before early catheter removal but after RRP the catheter is removed at 12 to 20 days without cystography. At our hospital operating room overhead cost for the first hour of surgery is $960 with each subsequent 15-minute block contributing $240. Surgical consumables costs were calculated from a comprehensive list of items included on the surgical field and reflect purchase price. The 6 most expensive disposables include the ultrasonic scalpel ($358.31, Ethicon Endosurgery, Cincinnati, Ohio), Visiport trocar ($77.86, United States Surgical [USS], Norwalk, Connecticut, Endoshear 5 mm cautery ($70.49, USS), Endocatch bag ($78.47, USS), Versaport 5–12 mm trocar ($55.75, USS) and Surgiport 5 mm trocar ($45.90, USS). Surgeon fees were derived from 2003 Maryland Medicare reimbursement rates based on Current Procedural Terminology (CPT) code. For RRP separate codes for procedures with (55845, $1,631.63) and without (55840, $1,328.68) PLND were used. For LRP there is currently no CPT code to reflect LRP and PLND. Therefore, LRP and PLND was coded as 2 procedures (LRP, 55866, $1,650.88 and laparoscopic PLND, 38571, $772.26). Anesthesiologist fees were derived from 2003 Medicare reimbursement rates using a Relative Value Unit (RVU) base value of 6 and the Baltimore county conversion factor ($17.32). Hospital room and board is calculated for a semiprivate room as $735 for the first night and $620 for each subsequent 24-hour period. Oral analgesic prices were calculated from the Bayview pharmacy cost for oxycodone ($0.92 per tablet) and the
number of tablets was derived from published data.3 The cost for autologous blood banking, purchase of heterologous leuko-reduced blood and blood administration were provided by our blood bank. Cystography costs reflect 2003 Medicare reimbursement (74430) and include a professional fee of $16.89 and a technical fee of $41.40. Model simulation and sensitivity analysis. The cost model was programmed into Data 4.0 software (TreeAge Software, Inc., Williamstown, Massachusetts) and simulated with and without PLND. For starting values we performed a comprehensive 10-year MEDLINE search to identify contemporary series of LRP and RRP. Weighted means were calculated from the combined series (table 1). Since European length of stay after radical prostatectomy is generally longer than in the United States, only domestic series were included in length of stay calculations. Since most published series do not stratify operative time with and without PLND, we had to estimate the time contribution of this procedure (20 minutes). Sensitivity analysis is a well established technique for evaluating cost-effectiveness. In 1-way analysis, a single cost factor is altered and its impact on total cost is determined by keeping all other factors fixed. In 2-way analysis 2 factors are altered simultaneously. We performed 1-way and 2-way sensitivity analyses by holding the variables for RRP constant and adjusting 1 or 2 variables for LRP. Model validation. To provide model validation we compared model predictions to financial data from actual clinical cases. Ideally, these data should reflect hospital costs or reimbursement, but this information is difficult to accumulate due to accounting practices that focus on hospital charges. Based on our cost model we calculated a predicted cost premium for LRP compared to RRP, and compared this value to the charge premium for LRP calculated from hospital charges. Detailed hospital financial records were reviewed for all patients undergoing LRP (128) or RRP (44) from June 2001 through July 2003. Since we perform PLND in the majority of our prostatectomy cases, these charges were assumed to reflect prostatectomy and PLND. Our hospital accounting office provided these records, which reflect almost 20,000 individual charges. Statistical analysis was performed using the SPSS 10 statistical package (SPSS, Inc., Chicago, Illinois). Frequency histograms were prepared for total
TABLE 1. Literature review for laparoscopic radical prostatectomy and open radical retropubic prostatectomy References
No. Pts
Operating Room Mins
LRP: 438 Rassweiler et al7 33 Bhayani et al3 120 Guillonneau and Vallancien2 43 Abbou et al8 180 Rassweiler et al9 350 Guillonneau et al10 125 Turk et al11 180 Hoznek et al12 50 Bollens et al13 70 Dahl et al14 Weighted mean Open RRP: 906 Hsu et al15 219 Rassweiler et al7 100 Guillonneau and Vallancien2 418 Hautmann et al16 472 Dillioglugil et al17 — Hammerer et al18 116 Gaylis et al19 — Catalona et al6 638 Arai et al20 24 Bhayani et al3 Weighted mean Note that only studies from the United States were used for prostatectomy in European centers.
253 348 239 320 271 217 255 240 317 274 263 130 196 135 168 182 — 155 — 263 168 180 calculating length
No. Pts
— 33 — — — — — — — 70 975 — — — 472 — 116 — — 24
Length of Stay (days) — 2.97 — — — — — — — 2.50 2.65
No. Pts
Transfusion Rate %
438 — 120 — 180 350 125 200 50 70
2.4 990 — 219 — 100 — — 6.2 472 — 320 3.0 — — 1,870 — 638 3.04 — 3.58 of stay results, reflecting differing practices for inpatient
9.8 — 10.0 — 17.0 5.7 2.0 3.0 13.0 5.7 8.5 8.2 26.9 31.0 — 28.6 27.8 — 9.0 19.1 — 14.9 management of
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
charges per case for LRP and RRP. The KolmogorovSmirnov test of normality confirmed that LRP (Z ⫽ 1.263, p ⫽ 0.082) and RRP (Z ⫽ 0.871, p ⫽ 0.433) data conformed to a normal distribution. Therefore, mean charges were calculated and compared using the Student t test. How-
271
ever, the LRP charge data histogram was asymmetric and significantly skewed toward the right (skewness statistic 1.120 ⫾ 0.214). For this reason median rather than mean charges for LRP and RRP were used to calculate the charge premium for LRP.
TABLE 2. Model predictions for the cost breakdown of laparoscopic and open radical prostatectomy with and without bilateral pelvic lymphadenectomy Bilat PLND
Starting values: Operating room mins (x)* Days of stay (y) Autologous units banked (m) % Transfusion rate (t) No. units transfused (n) Predicted cost: Operating room time (a) Operating room consumables (b) Surgeon fee (c) Anesthesia fee (d) Hospital room ⫹ board (e) Oral pain medications (f) Autologous blood banking (g) Blood transfusion (h) Postop cystogram (i)
$ $ $ $ $ $ $ $ $
No Bilat PLND
LRP
Open RRP
LRP
Open RRP
263 2.65 0 8.5 2
180 3.58 2 14.9 2
243 2.65 0 8.5 2
160 3.58 2 14.9 2
4,320.00 1,132.56 2,423.14 407.60 1,975.00 8.28 0.00 88.68 58.29
$2,880.00 $ 170.36 $1,631.63 $ 311.76 $2,595.00 $ 15.64 $ 979.86 $ 11.99 $ 0.00
$4,080.00 $1,132.56 $1,650.88 $ 384.50 $1,975.00 $ 8.28 $ 0.00 $ 88.68 $ 58.29
$2,640.00 $ 170.36 $1,328.68 $ 288.67 $2,595.00 $ 15.64 $ 979.86 $ 11.99 $ 0.00
$8,596.24
$9,378.19 $1,347.99 (14.4)
$8,030.20
Totals $10,413.55 Cost premium for laparoscopy (%) $ 1,817.30 (17.5) * Lowercase letters correspond to those in Appendix.
FIG. 1. One-way sensitivity analysis for laparoscopic prostatectomy with and without pelvic lymphadenectomy. Variables include operative time (a, b), length of stay (c, d) and transfusion rate (e, f). Intersection of 2 lines represents point of cost equivalence and is indicated with arrow. Lap RP, LRP.
272
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY RESULTS
Review of the literature yielded 10 LRP (1,589 patients) and 10 RRP (2,893 patients) series providing operative times. Length of stay data were scarcer, with only 2 relevant domestic LRP (103 patients) and 4 RRP (1,587 patients) series. For transfusion rate we found 8 and 7 relevant series for LRP (1,533 patients) and RRP (4,609 patients, table 1). Weighted means for these parameters served as starting points for computer modeling. Models were developed for LRP and RRP with and without concurrent PLND (see Appendix). The model predicted a cost of $10,414 for LRP and PLND, and $8,596 for RRP and PLND, giving a cost premium of 17.5% for the laparoscopic procedure. Without PLND the cost premium decreases to 14.4% for LRP ($9,378 vs $8,030), primarily because surgeon fees for laparoscopic PLND and LRP are currently unbundled (unlike for open RRP and PLND) and, therefore, the combined laparoscopic procedure incurs greater surgeon fees (table 2). The most significant factor was operative time, which contributed 42% on average to the cost of LRP and 33% to the cost of RRP. Surgical consumables costs were 11% (LRP) and 2% (RRP). We performed 1-way sensitivity analysis by varying a single variable for LRP while keeping the starting parameters for RRP fixed (fig. 1). The analysis predicts that an operating room time of 159 minutes (2.65 hours) for LRP and PLND would be required to be cost equivalent with RRP and PLND. Without PLND a laparoscopic operating room time of 174 minutes (2.90 hours) yielded equivalent cost to RRP. Length of stay sensitivity analysis identified that equivalence in cost could only be achieved if LRP was performed on an outpatient basis (less than 1 day length of stay). The impact of transfusion rate on cost was more modest and even decreasing the transfusion rate to 0% could not yield cost equivalence between the 2 procedures. Likewise, varying the cost of operating room consumables could not, by itself, equalize the projected cost of LRP and RRP. Two-way sensitivity analysis was used to analyze the interaction of the 2 most significant cost factors, operating room time and length of stay. As expected figure 2 shows that as operating room time increases the length of stay must decrease to achieve a cost advantage for LRP. The stair-step morphology of these curves reflects the threshold effect of length of stay on the model. Using these curves the operative time required for cost equivalence at any particular length of stay can be determined. To validate the model against financial results for clinical cases, we reviewed individual hospital charge records for all patients undergoing radical prostatectomy from July 2001 to June 2003 (128 LRP and 44 RRP, fig. 3). Assuming inclusion of PLND, mean charges for LRP ($15,451 ⫾ $336) and RRP ($12,441 ⫾ $509) were significantly different during this period (p ⬍0.001). Median values for LRP ($14,848) and RRP ($12,120) were then used to calculate the charge premium for LRP (18.4%). The magnitude of this charge premium is remarkably similar to the predicted cost premium for LRP and PLND derived from our computer model (17.5%). DISCUSSION
Laparoscopic radical prostatectomy is a relatively new treatment modality for localized prostate cancer that is generating excitement within the urological community. A variety of centers are now evaluating the oncological and functional outcomes of LRP, and comparing these factors to RRP. However, despite the impact of economics on the viability of any new surgical procedure, little has been written about cost comparisons between LRP and RRP. We have developed a detailed computer model suitable for comparing the anticipated perioperative costs of LRP and RRP at our United States institution. Using the weighted means of published values for operative time, length of stay
FIG. 2. Two-way sensitivity analysis for laparoscopic prostatectomy varying operative time and length of stay. Within dark region of graph LRP is less expensive than RRP. Interface between dark and crosshatched areas represents point of cost equivalence at varying operative times and lengths of stay.
FIG. 3. Comparison of model prediction with financial data derived from 172 consecutive radical prostatectomy cases (128 LRP, 44 RRP).
and transfusion rate, this model predicts that LRP will cost between 14.4% and 17.5% more than RRP. This result represents a worst-case scenario for LRP for several reasons. First, the operative times in several series reflect an experience that includes the learning curve for LRP. As experience with LRP increases, shorter operative times with resultant cost savings are expected. Several groups are now reporting mean operative times approaching 3 hours for these procedures,4 which would reduce the cost premium to only 4%. The second issue relates to CPT coding. Since no CPT code currently exists for a bundled LRP and PLND procedure, it is necessary to code for these procedures separately, which increases the “cost.” Our model likely overestimates this value since as a concurrent procedure coded with the ⫺51 modifier, reimbursement for the laparoscopic PLND would
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
only be a fraction of the $772.26 used for our calculations. If surgeon fees were excluded from the model, the cost premium for LRP and PLND would decrease to 12.8% ($1,025.80), highlighting the importance of CPT unbundling in increasing the relative cost of LRP. Our model also does not factor in costs associated with short and long-term complications of radical prostatectomy. Complications from deep venous thrombosis and pulmonary embolus appear to be infrequent during LRP, possibly due to the steep Trendelenburg position used during surgery and the lack of a pelvic retractor interfering with lower extremity venous return. There is also reason to believe that long-term complications associated with the urethrovesical anastomosis (ie anastomotic stricture and the need for artificial urinary sphincter [AUS] placement) may be less common with the laparoscopic approach due to the improved visualization of the anastomosis during its construction and the standard use of cystography before catheter removal. In our series of 157 LRP cases only 2 (1.2%) evidenced anastomotic stricture and none required postoperative AUS placement. These results compare favorably with published rates of anastomotic stricture (4% to 20%) and AUS placement (0.2% to 2.3%) after open RRP.5, 6 Complications certainly impact the total cost of any surgical procedure. If the lower rates of anastomotic and thromboembolic complications with LRP prove to be reproducible, this may greatly narrow the cost differential between LRP and RRP. This issue is not addressed in the current analysis and deserves further study.
273
As with any modeling exercise there remain limitations to our approach. This model incorporates constants that are specific to our region of the United States and to our medical center. The costs for operative time, hospital room and board, and consumables may vary significantly at different institutions, and the Medicare reimbursement rates for anesthesia services have region specific modifiers. Another criticism could be that we are blurring the boundary between true hospital cost from overhead and Medicare reimbursement for professional fees. Although this is true, we believe that professional fees should be included in any cost analysis of prostatectomy surgery since this factor impacts the global cost of these procedures to society. CONCLUSIONS
We believe that our cost model is comprehensive and functional since its predictions differed by less than 1% from actual hospital charges for prostatectomy at our institution. Ultimately, however, the usefulness of modeling is that it allows various clinical scenarios to be tested for cost equivalence between LRP and RRP. The cost of LRP consumables would decrease to $914.52 if disposable scissors and trocars other than the Visiport were eliminated. Using these disposables and discharging all patients on postoperative day 2, the model predicts that operative time must only decrease to 204 minutes (3.4 hours) for cost equivalence to RRP, which is a reachable goal.
APPENDIX: COST ANALYSIS MODEL FOR COMPARING LAPAROSCOPIC RADICAL PROSTATECTOMY WITH OPEN RADICAL RETROPUBIC PROSTATECTOMY
a OperatingRoomTime ⫹ b ORConsumables ⫹ c SurgeonFee ⫹ d AnesthesiaFee ⫹ e HospitalRoomBoard ⫹ f OralPainMeds ⫹ g AutologousBanking ⫹ h BloodTransfusion ⫹ i PostopCystogram x ⫽ 兩if (PLND) then [ORtime minutes ] else [ORtime minutes ⫺ 20]兩 y ⫽ Length of Stay days z ⫽ 兩if (open) then [m] else [0]兩 m ⫽ numUnits AutologousBloodBanked t ⫽ % PatientsTransfused n ⫽ numUnitsTransfused a OperatingRoomTime ⫽ $960.00 ⫹ 兩if (x ⬍ 61) then (0) else ((roundUP( x ⫺1560))* $240.00)兩 b ORConsumables ⫽ 兩if (laparoscopic) then [$1,132.56 LapRRP ] else [$170.36 OpenRRP ]兩 cSurgeonFeewithnodes ⫽ 兩if (laparoscopic) then ($1,650.88LapRR ⫹ $772.26LymphNodes) else ($1,631.63OpenRRPw/nodes)兩 c SurgeonFee withoutnodes ⫽ 兩if (laparoscopic) then ($1,650.88 LapRRP ) else ($1,328.68 OpenRRP )兩 x d AnesthesiaFee ⫽ [ 15ORtime ⫹ RVU base ]* ConversionFactor Baltimore ⫽ [( 15 ) ⫹ 6] * $17.32 min/unit e HospitalRoomBoard ⫽ 兩if(y ⬎ 0) then ($735.00 ⫹ [roundUP (y ⫺ 1) * $620.00]) else ($0.00)兩 f OralPainMeds ⫽ 兩if (laparoscopic) then [9 * $0.92] else [17 * $0.92]兩 g AutologousBanking ⫽ z * $489.93 hBloodTransfusion ⫽ (t * n * $40.25)TransfusionFee ⫹ 兩if (z ⫽ 0) then [t * n * $481.39] else [0]兩NonAutoUnits i PostopCystogram ⫽ 兩if (laparoscopic) then [$16.89 ProfessionalFee ⫹ $41.40 TechnicalFee ] else [$0.00]兩
REFERENCES
1. Ruchlin, H. S. and Pellissier, J. M.: An economic overview of prostate carcinoma. Cancer, 92: 2796, 2001 2. Guillonneau, B. and Vallancien, G.: Laparoscopic radical prostatectomy: the Montsouris experience. J Urol, 163: 418, 2000 3. Bhayani, S. B., Pavlovich, C. P., Hsu, T. S., Sullivan, W. and Su, L. M.: Prospective comparison of short-term convalescence: laparoscopic radical prostatectomy versus open radical retropubic prostatectomy. Urology, 61: 612, 2003 4. Rozet, F., Cathelineau, X., Widmer, H., Cahill, D., Baumert, H. and Vallancien, G.: Transperitoneal or extraperitoneal approach for laparoscopic radical prostatectomy. A prospective comparison of 200 cases. J Endourol, 17: A111, 2003 5. Benoit, R. M., Naslund, M. J. and Cohen, J. K.: Complications after radical retropubic prostatectomy in the medicare population. Urology, 56: 116, 2000 6. Catalona, W. J., Carvalhal, G. F., Mager, D. E. and Smith, D. S.:
7.
8.
9.
10.
11.
Potency, continence and complication rates in 1,870 consecutive radical retropubic prostatectomies. J Urol, 162: 433, 1999 Rassweiler, J., Seemann, O., Schulze, M., Teber, D., Hatzinger, M. and Frede, T.: Laparoscopic versus open radical prostatectomy: a comparative study at a single institution. J Urol, 169: 1689, 2003 Abbou, C. C., Salomon, L., Hoznek, A., Antiphon, P., Cicco, A., Saint, F. et al: Laparoscopic radical prostatectomy: preliminary results. Urology, 55: 630, 2000 Rassweiler, J., Sentker, L., Seemann, O., Hatzinger, M. and Rumpelt, H. J.: Laparoscopic radical prostatectomy with the Heilbronn technique: an analysis of the first 180 cases. J Urol, 166: 2101, 2001 Guillonneau, B., Cathelineau, X., Doublet, J. D. and Vallancien, G.: Laparoscopic radical prostatectomy: the lessons learned. J Endourol, 15: 441, 2001 Turk, I., Deger, S., Winkelmann, B., Schonberger, B. and
274
12. 13. 14.
15.
16.
COST COMPARISON OF LAPAROSCOPIC AND OPEN RADICAL RETROPUBIC PROSTATECTOMY
Loening, S. A.: Laparoscopic radical prostatectomy. Technical aspects and experience with 125 cases. Eur Urol, 40: 46, 2001 Hoznek, A., Salomon, L., Olsson, L. E., Antiphon, P., Saint, F., Cicco, A. et al: Laparoscopic radical prostatectomy. The Creteil experience. Eur Urol, 40: 38, 2001 Bollens, R., Vanden Bossche, M., Roumeguere, T., Damoun, A., Ekane, S., Hoffmann, P. et al: Extraperitoneal laparoscopic radical prostatectomy. Results after 50 cases. Eur Urol, 40: 65, 2001 Dahl, D. M., L’esperance, J. O., Trainer, A. F., Jiang, Z., Gallagher, K., Litwin, D. E. et al: Laparoscopic radical prostatectomy: initial 70 cases at a U.S. university medical center. Urology, 60: 859, 2002 Hsu, E. I., Hong, E. K. and Lepor, H.: Influence of body weight and prostate volume on intraoperative, perioperative, and postoperative outcomes after radical retropubic prostatectomy. Urology, 61: 601, 2003 Hautmann, R. E., Sauter, T. W. and Wenderoth, U. K.: Radical
17.
18.
19. 20.
retropubic prostatectomy: morbidity and urinary continence in 418 consecutive cases. Urology, 43: 47, 1994 ¨ ., Leibman, B. D., Leibman, N. S., Kattan, M. W., Dillioglugil, O Rosas, A. L. and Scardino, P. T.: Risk factors for complications and morbidity after radical retropubic prostatectomy. J Urol, 157: 1760, 1997 Hammerer, P., Hubner, D., Gonnermann, D. and Huland, H.: Perioperative and postoperative complications of pelvic lymphadenectomy and radical prostatectomy in 320 consecutive patients. Urologe A, 34: 334, 1995 Gaylis, F. D., Friedel, W. E. and Armas, O. A.: Radical retropubic prostatectomy outcomes at a community hospital. J Urol, 159: 167, 1998 Arai, Y., Egawa, S., Tobisu, K., Sagiyama, K., Sumiyoshi, Y., Hashine, K. et al: Radical retropubic prostatectomy: time trends, morbidity and mortality in Japan. BJU Int, 85: 287, 2000