European Urology

1. Introduction

Radical prostatectomy (RP) represents the most common treatment for prostate cancer (PCa) performed in the United States [1]. It has been reported that RP accounted for 50% of the $1.7 billion annual cost spent on PCa [2]. New technologies, such as laparoscopic RP (LRP) and robotic-assisted LRP (RALP), have been developed. The demand and utilization of these minimally invasive surgical treatments have increased in recent years, whereas utilization of open retropubic RP (RRP) has decreased [3].

Although minimally invasive approaches offer the benefit of decreased blood loss and shorter hospital stay [4], the need for more expensive equipment may have a significant impact on health economics. European studies found increasing costs for PCa care, caused by technological changes in the management of PCa [5], and [6]. We have previously shown that costs for LRP were higher than those for RRP, predominantly due to higher surgical supply and operating room (OR) costs [7]. Another study using a decision-analysis model found that RALP and LRP were more expensive than RRP, based on outcomes of a literature review and meta-analysis [8]. It is important to validate models using real-world data.

A careful evaluation of costs associated with RRP, LRP, and RALP is necessary for reasonable health care resource allocation. Given the lack of comparative studies on the costs of different approaches to RP, we evaluated direct costs using data from 643 consecutive patients recently treated at our institution.

2. Patients and methods

3. Results

The baseline clinical and pathologic data of the patients are shown in Table 1. The ages of the patients in the three groups were similar (median: 61 yr [RALP and RRP groups] vs 59 yr [LRP]), but the slight difference was statistically significant (p = 0.001). BMI, preoperative PSA level, and prostate size were comparable among the three groups (Table 1). Percentage of patients with biopsy Gleason sum 8–10 was similar in the three groups (RALP: 6.1%; LRP: 8.4%; RRP: 8.8%; p = 0.5), but there were more patients with Gleason 7 in the RALP group.

Nerve sparing was performed in 85% of RALP procedures (192 of 225 patients for whom information on the performance of nerve sparing was available), in 96% (210 of 219) of LRP procedures; and in 90% (137 of 152) of RRP procedures (p < 0.001). A lymph node dissection was more commonly performed in conjunction with RRP (100%) than with LRP (22%) and RALP (11%) (p < 0.001). Blood transfusions were utilized in 21% (32 of 154) of patients who underwent RRP compared to 4.6% (12 of 262) of those who underwent RALP and 1.8% (4 of 220) of those who underwent LRP (p < 0.001). The LOS was longer for RRP compared to RALP and LRP (Table 1; p < 0.0001).

The comparison of overall direct costs and the individual cost domains among the different types of RP are reported in Table 2. The median direct cost was highest for RALP ($6752 [interquartile range (IQR): $6283—7369]); cost for LRP was $5687 (IQR: $4941–5905), and RRP was the cheapest procedure ($4437 [IQR: $3989–5141]; overall p < 0.001). The main difference was in surgical supply costs for each procedure ($2015 for RALP, $725 for LRP, $185 for RRP) and OR costs ($2798 for RALP, $2453 for LRP, $1611 for RRP; p < 0.001). Additional costs for purchase and maintenance (total of $340 000 per year when amortized over a 7-yr period) of the robot would result in $2698 per patient undergoing RALP, given an average of 126 cases performed per year.

4. Discussion

Financial resources in national health care systems are limited, and high pressure is constantly applied to make procedures cost effective. Economic considerations are particularly important when new technologies are introduced. These need to be evaluated in terms of new costs and potential benefits for patient care. Demand and utilization of minimally invasive surgical techniques are increasing and open RRP is less commonly performed, but the impact on cost has not been well evaluated. In a previous study, we used decision-analysis models and a meta-analysis of the published literature to compare costs of RALP, LRP, and RRP, and we projected that RALP is the most expensive approach [8]. Using actual cost values from our institution, we now report the largest comparison of RALP, LRP, and RRP costs to date.

We found that RALP resulted in the highest costs, despite excluding the costs of purchasing and maintaining the robot. Costs were significantly higher than for the other approaches and exceeded the median costs for LRP by more than $1000. RRP was the cheapest procedure, with a cost advantage of $1250 and $2315 over LRP and RALP, respectively. The main component leading to the higher cost of RALP was the high surgical supply costs, which were not overcome by reduced room and board costs.

Another cost disadvantage of RALP was the higher OR cost. OR costs are a direct reflection of OR time, and RALP and LRP OR costs were higher than those for RRP due to longer operating times. At the same time, even assuming equivalent operating time, RALP would still cost at least $1000 more than RRP based on high surgical equipment costs. Our previous analysis of RRP, RALP, and LRP reports in the literature found no significant advantage of one approach over another in terms of operative time [8]. If a surgeon is dramatically faster at one approach over another, that could affect the cost effectiveness of that particular surgeon at performing prostatectomy, but such nuances probably are not generalizable.

LRP had an intermediate cost between RALP and RRP. The higher cost compared to RRP was due to higher equipment costs of LRP ($725 vs $185) and a higher OR cost due to longer surgical times. These findings are in line with a prior report from our institution showing that actual LRP costs were significantly higher than the costs for RRP due to higher surgical supply and OR costs [7]. Although LRP is less expensive that RALP, it is a technically challenging procedure, and most laparoscopic prostatectomy procedures in the United States are performed using the robot-assisted approach.

Can the robot become cost competitive with the other approaches? One direct way to lower costs would require a significant decrease in surgical supply costs [8]. Reduction of operative time is possible if resident teaching is eliminated, but it also should be assumed that RRP times would decrease if residents were not involved. As noted above, with equivalent OR times, the equipment costs still drive the higher cost of RALP. It is also unlikely that cost savings can be achieved by lowering the LOS, which was 1.5 d for RALP. Even a half-day reduction will only reduce costs by $250, and it is unlikely that minimally invasive RP will be routinely performed in an outpatient setting. Regarding time to convalescence, one recent study showed that patients who underwent RALP had shorter postoperative hospital stays and less need for paid sick leave than patients who had RRP [9]. Shorter convalescence following RALP may reduce the cost burden to society but needs to be confirmed prospectively.

We excluded the robot purchase costs ($1.5–1.75 million) and maintenance fees ($112 000–150 000 per year, depending on model and year of purchase) from our cost comparison, but based on our clinical volume, this expense would add almost $2700 per case. Many institutions do not have donors who can provide a robot. Thus, purchase and maintenance of a robot increases cost of RALP, especially during the learning curve, and depends on case volume [10]. Annual maintenance cost would add $1000 per case if 150 cases were performed each year and $500 if 300 cases were performed annually, which likely is the logistical limit of cases for one robot.

Only evidence for improved patient care when performing RALP could justify the high costs for this procedure. A recent systematic review of the literature for prostatectomy procedures identified 37 comparative studies [4]. LRP and RALP were more time consuming than RRP, especially in the initial steps of the learning curve, but blood loss, transfusion rates, length of stay, and complication rates all favored minimally invasive approaches (LRP and RALP). The available studies were not sufficient to prove the superiority of any surgical approach in terms of functional and oncologic outcomes. Minimally invasive RP results in lower transfusion rates and more rapid convalescence with shorter LOS [11], [12], [13], [14], [15], and [16]. Transfusion rates for open RRP, however, tend to be low [17], and [18], and the higher transfusion rates associated with RRP had no major impact (<15% of overall costs of RRP) in the present study.

The decrease in pain and LOS has been questioned in several studies that used similar pathways to discharge for RALP and RRP [19], and [20]. Webster et al prospectively compared postoperative pain scores among 314 patients who underwent RALP or RRP and found no differences in pain or analgesic use [20]. Most important, the adoption of new technology should improve continence and potency rates following RP. Unfortunately, no randomized trials compared outcomes for different surgical approaches to RP, and no significant advantages in functional outcomes and cancer control for the RALP approach have been observed to date [17], [18], [21], and [22].

One recent report by Hu et al showed a significantly increased use of salvage radiation or hormone therapy within 6 mo following surgery in patients who underwent minimally invasive RP (LRP or RALP) compared to RRP [23]. This finding was especially true for low-volume surgeons. In contrast, a study from a center of excellence showed that the overall incidence of positive surgical margins was significantly lower among patients who underwent RALP compared to RRP (15% vs 35%, p < 0.001) [24]; however, lower PSA, clinical stage, and Gleason scores were present in the RALP group. Long-term oncologic outcomes are not yet available.

The few studies comparing costs of RALP and other approaches to RP all found that the robotic approach was more expensive [8], [25], and [26]. The high utilization of RALP is partially driven by patient and physician demand, but the cost effectiveness of RALP ultimately will be determined with functional and oncologic outcomes. To date, there is not enough evidence of a significant improvement in patient care with RALP to justify the high added cost. From a patient perspective, it has been reported that those who underwent RALP were more likely to be dissatisfied with their outcome, possibly because of higher expectations for a new and “better” procedure [27]. Prospective studies should be performed to determine whether RALP offers a significant advantage over other approaches to RP to justify the added expense.

Our analysis has several limitations. One limitation is that costs of the procedures can vary among institutions; however, our costs were similar to other publications focusing on the cost of RP [25], [28], and [29]. The procedures in our series were performed by experts in their fields. The costs will be higher for surgeons with less experience and longer operative times. Moreover, costs may vary geographically at different centers and may not be applicable to all institutions. The costs of any salvage therapy are not included in our analysis. Additionally, given the relatively short follow-up period for patients who underwent RALP, we were unable to include data on biochemical recurrence rates and associated costs of any palliative hormonal treatment. Finally, this study was done primarily from a hospital perspective; if there is evidence for other cost advantages such as earlier return to work, then those results may make LRP and RALP more effective from a societal perspective.

5. Conclusions

RALP is associated with significantly higher costs compared to LRP and RRP, predominantly due to increased surgical supply and OR costs. Our clinically derived cost comparison confirmed the computer model–based predictions of high RALP costs. These costs may have a significant impact on overall cost of PCa care. Evidence for functional and oncologic superiority will be needed to justify the high costs of this new technology.