Articles

Perioperative Outcomes of Robot-Assisted Radical Prostatectomy Compared With Open Radical Prostatectomy: Results From the Nationwide Inpatient Sample eulogo1

By: Quoc-Dien Trinha b 1 lowast , Jesse Sammona 1, Maxine Sunb, Praful Ravic, Khurshid R. Ghania, Marco Bianchid, Wooju Jeonga, Shahrokh F. Shariate, Jens Hansenf, Jan Schmitgesf, Claudio Jeldresb, Craig G. Rogersa, James O. Peabodya, Francesco Montorsid, Mani Menona and Pierre I. Karakiewiczb

Published online: 01 April 2012

Keywords: Prostatic neoplasms, Prostatectomy, Minimally invasive, Robotic, Open

Abstract Full Text Full Text PDF (163 KB)

Abstract

Background

Prior to the introduction and dissemination of robot-assisted radical prostatectomy (RARP), population-based studies comparing open radical prostatectomy (ORP) and minimally invasive radical prostatectomy (MIRP) found no clinically significant difference in perioperative complication rates.

Objective

Assess the rate of RARP utilization and reexamine the difference in perioperative complication rates between RARP and ORP in light of RARP's supplanting laparoscopic radical prostatectomy (LRP) as the most common MIRP technique.

Design, setting, and participants

As of October 2008, a robot-assisted modifier was introduced to denote robot-assisted procedures. Relying on the Nationwide Inpatient Sample between October 2008 and December 2009, patients treated with radical prostatectomy (RP) were identified. The robot-assisted modifier (17.4x) was used to identify RARP (n=11 889). Patients with the minimally invasive modifier code (54.21) without the robot-assisted modifier were classified as having undergone LRP and were removed from further analyses. The remainder were classified as ORP patients (n=7389).

Intervention

All patients underwent RARP or ORP.

Measurements

We compared the rates of blood transfusions, intraoperative and postoperative complications, prolonged length of stay (pLOS), and in-hospital mortality. Multivariable logistic regression analyses of propensity score–matched populations, fitted with general estimation equations for clustering among hospitals, further adjusted for confounding factors.

Results and limitations

Of 19 462 RPs, 61.1% were RARPs, 38.0% were ORPs, and 0.9% were LRPs. In multivariable analyses of propensity score–matched populations, patients undergoing RARP were less likely to receive a blood transfusion (odds ratio [OR]: 0.34; 95% confidence interval [CI], 0.28–0.40), to experience an intraoperative complication (OR: 0.47; 95% CI, 0.31–0.71) or a postoperative complication (OR: 0.86; 95% CI, 0.77–0.96), and to experience a pLOS (OR: 0.28; 95% CI, 0.26–0.30). Limitations of this study include lack of adjustment for tumor characteristics, surgeon volume, learning curve effect, and longitudinal follow-up.

Conclusions

RARP has supplanted ORP as the most common surgical approach for RP. Moreover, we demonstrate superior adjusted perioperative outcomes after RARP in virtually all examined outcomes.

Take Home Message

Contemporary data from the United States demonstrate that robot-assisted radical prostatectomy has supplanted open radical prostatectomy as the most common surgical approach for prostate cancer. Adjusted analyses demonstrate superior perioperative outcomes after robot-assisted radical prostatectomy in virtually all of the examined outcomes.

Keywords: Prostatic neoplasms, Prostatectomy, Minimally invasive, Robotic, Open.

1. Introduction

Prostate cancer (PCa) remains the most common solid organ malignancy in men in the United States and the second leading cause of cancer death [1]. Radical prostatectomy (RP) is a standard-of-care option for localized PCa with a demonstrated survival advantage when compared with watchful waiting [2]. In the past decade, we have seen a significant trend toward the utilization of minimally invasive approaches to RP for the treatment of PCa [3], particularly in the United States. This evolution has been controversial and marked by few comparative studies [4] and [5], the preponderance of which are single-institution studies of poor evidentiary quality [6] and [7].

In the absence of randomized controlled trials, population-based analyses provide important evidence to compare competing treatment modalities for men undergoing RP. Previous population-based studies compared minimally invasive RP (MIRP) and open RP (ORP) and found that men undergoing MIRP experienced equivalent rates of postoperative complications and need for adjuvant therapy and had shorter hospitalizations, fewer respiratory and miscellaneous surgical complications, and fewer strictures. These patients did experience more genitourinary complications, incontinence, and erectile dysfunction, however [3] and [8]. These analyses had several limitations [9], [10], and [11], including the combined reporting of minimally invasive approaches, namely, robot-assisted RP (RARP) and laparoscopic RP (LRP).

As of October 1, 2008, a robot-assisted modifier code was introduced and received approval by the US Food and Drug Administration to identify robot-assisted procedures. In this manuscript, we reexamine the impact of surgical approach (RARP compared with ORP) on perioperative outcomes, using contemporary data from the Nationwide Inpatient Sample (NIS).

2. Methods

2.1. Data source

Data from the NIS were abstracted between October 2008 and December 2009. The NIS includes inpatient discharge data collected via federal–state partnerships as part of the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality. As of 2009, the NIS has been cataloging data from approximately 8 million hospital discharges drawn from >1000 community hospitals in 44 states, approximating 20% of the community hospitals within the United States, including public hospitals and academic medical centers. The NIS is the sole hospital database in the United States with charge information on all patients regardless of payer, including persons covered by Medicare, Medicaid, and private insurance, as well as the uninsured.

2.2. Sample population and surgical procedures

Patients with a primary diagnosis of PCa were identified using the International Classification of Disease, 9th revision, clinical modification (ICD-9-CM) diagnostic code 185.0. Relying on the ICD-9 procedure codes, patients who underwent radical prostatectomy (60.5) were abstracted (n=19 462). As recognized by the National Center of Health Statistics and the Centers for Medicare and Medicaid Services, beginning October 1, 2008, the robot-assisted modifier code (ICD-9-CM 17.4x) was used to identify RARP. Patients with the minimally invasive modifier code (ICD-9-CM 54.21) without the robot-assisted modifier were classified as having undergone LRP and were removed from further analyses (n=184). The remaining patients, who did not have RARP or LRP codes, were determined to have undergone ORP.

2.3. Baseline patient and hospital characteristics

For all patients, the following variables were available: age, race, insurance status, and Charlson comorbidity index (CCI). Baseline CCI was calculated according to Charlson et al. [12] and adapted according to Deyo and colleagues [13]. Insurance categories are combined in general groups, namely, private insurance, Medicare, Medicaid, and self-pay. Hospital characteristics include hospital region, location, and academic status. These data were obtained from the American Hospital Association Annual Survey of Hospitals and the US Census Bureau [14].

2.4. Intraoperative complications, postoperative complications, and blood transfusions during hospitalization

The NIS records up to 15 diagnoses and procedures per in-hospital stay. The presence of any complication was defined using ICD-9 diagnoses 2–15 and previously published methodology [3]. Intraoperative complications consisted of surgical laceration of the bowel, ureter, and nerves and/or vessels during a procedure (ICD-9 998.2). Blood transfusion recipients were identified using the ICD-9 procedure codes 99.02, 99.04. Seven groups of postoperative complications were identified consisting of potentially life-threatening cardiac, respiratory, or vascular events; miscellaneous medical and surgical events; and other events, such as genitourinary complications and wound infection; ICD-9-CM codes were selected as previously described [8].

2.5. Length of stay and in-hospital mortality

Length of stay is calculated by subtracting the admission date from the date of discharge. Prolonged length of stay (pLOS) is defined as a hospitalization beyond the 75th percentile cut-off point of 2 d. Perioperative mortality was coded from patient disposition.

2.6. Statistical analysis

Medians and interquartile ranges (IQRs) were generated for continuously coded variables; frequencies and proportions were generated for categorical variables. The Mann-Whitney test and chi-square test were used to assess the statistical significance of differences in medians and proportions, respectively.

To account for the inherent differences between patients undergoing RARP and ORP, and to control for common biases associated with multivariable analysis [15] and [16], we relied on propensity-based matching. The ORP and RARP cohorts were matched by patient characteristics (age, race, CCI, year of surgery, and insurance status) and by hospital characteristics (volume, location, region, and academic status). To adjust for residual differences between populations, separate multivariable logistic regression models tested the effect of operative technique (RARP and ORP) on perioperative mortality, blood transfusion, intraoperative and postoperative complications, and pLOS. Covariates included age, race, CCI, hospital region, hospital location, hospital academic status, annual hospital caseload, pelvic lymphadenectomy (yes vs no), and insurance status. For each model, generalized estimating equations further adjusted for clustering within hospitals.

All tests were two-sided, with a statistical significance set at p<0.05. Analyses were conducted using the R statistical package v.2.13.1 (R Foundation for Statistical Computing).

3. Results

Within the NIS, 19 278 patients underwent RARP or ORP between October 2008 and December 2009. Of this group, 11 889 patients (61.7%) underwent RARP, and 7389 patients (38.3%) underwent ORP. Of 647 institutions performing RP within the study period, 200 performed both approaches, 19 performed RARP only, and 428 performed ORP only. Characteristics of the patient populations are presented in Table 1. Patient populations differed significantly by race and comorbidity status, but not by age. A greater proportion of RARP than ORP patients were white (66.5% compared with 61.8%, respectively), and fewer were black (8.2% compared with 11.2%, respectively). RARP patients had fewer comorbidities: CCI of 0 in 78.8% of RARP patients compared with 76.4% of ORP patients. Several hospital characteristics differed significantly between the two groups. Specifically, a higher proportion of RARPs were performed at teaching institutions (72.3% of RARPs compared with 60.7% of ORPs, p<0.001) in urban locations (98.4% of RARPs compared with 92.1% of ORPs, p<0.001). A higher proportion of RARPs were performed at high-volume centers: the median annual hospital caseload was 159 (IQR: 83,336) for RARPs compared with 64 (IQR: 22,150) for ORPs.

Table 1 Demographic characteristics of patients treated with radical prostatectomy for prostate cancer, stratified according to open or robotic surgery, Nationwide Inpatient Sample, October 2008–December 2009

Total Open Robotic p value
No. of patients 19 278 7389 11 889
Median age, yr (IQR) 62 (56–67) 62 (57–67) 62 (56–67) 0.928
Race, n (%)
White 12 468 (64.7) 4565 (61.8) 7903 (66.5) <0.001
Black 1795 (9.3) 824 (11.2) 971 (8.2)
Hispanic 774 (4.0) 335 (4.5) 439 (3.7)
Asian/Pacific Islander 3547 (18.4) 1439 (19.5) 2108 (17.7)
Native American 188 (1.0) 27 (0.4) 161 (1.4)
Other 506 (2.6) 199 (2.7) 307 (2.6)
CCI,* n (%)
0 15 007 (77.8) 5644 (76.4) 9363 (78.8) <0.001
1 3664 (19.0) 1470 (19.9) 2194 (18.5)
2 463 (2.4) 215 (2.9) 248 (2.1)
≥3 144 (0.7) 60 (0.8) 84 (0.7)
Year of surgery, n (%)
2008 3838 (19.9) 1810 (24.5) 2028 (17.1) <0.001
2009 15 440 (80.1) 5579 (75.5) 9861 (82.9)
Hospital academic status, n (%)
Nonteaching 6108 (32.1) 2881 (39.3) 3227 (27.7) <0.001
Teaching 12 892 (67.9) 4451 (60.7) 8441 (72.3)
Hospital location, n (%)
Rural 768 (4.0) 578 (7.9) 190 (1.6) <0.001
Urban 18 232 (96.0) 6754 (92.1) 11 478 (98.4)
Hospital region, n (%)
Northeast 3197 (16.6) 1113 (15.1) 2084 (17.5) <0.001
Midwest 5061 (26.3) 2042 (27.6) 3019 (25.4)
South 6154 (31.9) 2571 (34.8) 3583 (30.1)
West 4866 (25.2) 1663 (22.5) 3203 (26.9)
Median annual hospital caseload, n (IQR) 127 (58–237) 64 (22–150) 159 (83–336) <0.001
Insurance status, n (%)
Medicare 5976 (31.0) 2343 (31.7) 3633 (30.6) <0.001
Medicaid 360 (1.9) 190 (2.6) 170 (1.4)
Private insurance 12 045 (62.5) 4415 (59.8) 7630 (64.2)
Other 897 (4.7) 441 (6.0) 456 (3.8)

* Based on comorbidity developed by Charlson et al. [12] and adapted by Deyo et al. [13].

Hospital region is defined by the US Census Bureau.

IQR=interquartile range; CCI=Charlson comorbidity index.

Propensity score matching resulted in a cohort of 7598 RARP patients and 7389 ORP patients (Table 2). The standardized differences between the two groups in patient and hospital characteristics were<10% [15]. This cohort represents the focus of all subsequent analyses.

Table 2 Post–propensity score–matched demographic characteristics of patients treated with radical prostatectomy for prostate cancer, stratified according to open or robotic surgery, Nationwide Inpatient Sample, October 2008–December 2009

Total Open Robotic p value
No. of patients 14 987 7389 7598
Median age, yr (IQR) 62 (56–67) 62 (57–67) 62 (56–67) 0.733
Race, n (%)
White 9303 (62.1) 4565 (61.8) 4738 (62.4) <0.001
Black 1566 (10.4) 824 (11.2) 742 (9.8)
Hispanic 623 (4.2) 335 (4.5) 288 (3.8)
Asian/Pacific Islander 3050 (20.4) 1439 (19.5) 1611 (21.2)
Native American 39 (0.3) 27 (0.4) 12 (0.2)
Other 406 (2.7) 199 (2.7) 207 (2.7)
CCI,* n (%)
0 11 557 (77.1) 5644 (76.4) 5913 (77.8) 0.151
1 2902 (19.4) 1470 (19.9) 1432 (18.8)
2 406 (2.7) 215 (2.9) 191 (2.5)
≥3 122 (0.8) 60 (0.8) 62 (0.8)
Year of surgery, n (%)
2008 3604 (24.0) 1810 (24.5) 1794 (23.6) 0.205
2009 11 383 (76.0) 5579 (75.5) 5804 (76.4)
Hospital academic status, n (%)
Nonteaching 5192 (34.8) 2881 (39.3) 2311 (30.4) <0.001
Teaching 9738 (65.2) 4451 (60.7) 5287 (69.6)
Hospital location, n (%)
Rural 768 (5.1) 578 (7.9) 190 (2.5) <0.001
Urban 14 162 (94.9) 6754 (92.1) 7408 (97.5)
Hospital region, n (%)
Northeast 2219 (14.8) 1113 (15.1) 1106 (14.6) <0.001
Midwest 4658 (31.1) 2042 (27.6) 2616 (34.4)
South 4903 (32.7) 2571 (34.8) 2332 (30.7)
West 3207 (21.4) 1663 (22.5) 1544 (20.3)
Median annual hospital caseload, n (IQR) 95 (43–171) 64 (22–150) 121 (66–184) <0.001
Insurance status, n (%)
Medicare 4654 (31.1) 2343 (31.7) 2311 (30.4) <0.001
Medicaid 347 (2.3) 190 (2.6) 157 (2.1)
Private insurance 9188 (61.3) 4415 (59.8) 4773 (62.8)
Other 798 (5.3) 441 (6.0) 357 (4.7)

* Based on comorbidity developed by Charlson et al. [12] and adapted by Deyo et al. [13].

Hospital region is defined by the US Census Bureau.

IQR=interquartile range; CCI=Charlson comorbidity index.

Table 3 and Table 4 show the rate of intraoperative and postoperative outcomes, stratified according to RARP and ORP, before and after propensity score matching. In the cohort after propensity score matching, the rate of intraoperative complications was found to be 0.4% and 1.0% (p<0.001) for RARP and ORP, respectively; postoperative complications were 9.3% and 11.1% (p<0.001); blood transfusion rates were 2.4% and 7.7% (p<0.001); pLOS rates were 14.5% and 39.6% (p<0.001); and in-hospital mortality was 0.0% and 0.1% (p=0.092), respectively.

Table 3 Perioperative outcomes during initial hospitalization following radical prostatectomy, stratified by open or robotic surgical approach

Open, n=7389 Robotic, n=11889 Robotic vs open, odds ratio (95% CI) p value
Homologous blood transfusion, n (%) 572 (7.7) 232 (2.0) 0.24 (0.20–0.28) <0.001
Intraoperative complication, n (%) 73 (1.0) 43 (0.4) 0.36 (0.25–0.53) <0.001
Postoperative complication,* n (%)
Overall 823 (11.1) 975 (8.2) 0.71 (0.65–0.79) <0.001
Cardiac 96 (1.3) 106 (0.9) 0.68 (0.52–0.90) 0.007
Respiratory 191 (2.6) 140 (1.2) 0.45 (0.36–0.56) <0.001
Vascular 45 (0.6) 44 (0.4) 0.61 (0.40–0.92) 0.019
Operative wound 48 (0.6) 50 (0.4) 0.65 (0.43–0.96) 0.031
Genitourinary 86 (1.2) 121 (1.0) 0.87 (0.66–1.15) 0.339
Miscellaneous medical 459 (6.2) 584 (4.9) 0.78 (0.69–0.88) <0.001
Miscellaneous surgical 121 (1.6) 166 (1.4) 0.85 (0.67–1.08) 0.179
Length of stay >2 d, n (%) 2923 (39.6) 1529 (12.9) 0.23 (0.21–0.24) <0.001
In-hospital mortality, n (%) 6 (0.1) 1 (0.0) 0.10 (0.01–0.86) 0.036

* Rates of complication are not additive, as patients may have had multiple complications.

CI=confidence interval.

Table 4 Propensity score–matched intraoperative and postoperative outcomes during hospitalization stratified by open or robotic surgery

Open, n=7389 Robotic, n=7598 Robotic vs open, odds ratio (95% CI) p value
Homologous blood transfusion, n (%) 572 (7.7) 184 (2.4) 0.30 (0.25–0.35) <0.001
Intraoperative complication, n (%) 73 (1.0) 33 (0.4) 0.44 (0.29–0.66) <0.001
Postoperative complication,* n (%)
Overall 823 (11.1) 705 (9.3) 0.82 (0.73–0.91) <0.001
Cardiac 96 (1.3) 68 (0.9) 0.69 (0.5–0.94) 0.018
Respiratory 191 (2.6) 105 (1.4) 0.53 (0.42–0.67) <0.001
Vascular 45 (0.6) 30 (0.4) 0.65 (0.41–1.03) 0.065
Operative wound 48 (0.6) 35 (0.5) 0.71 (0.46–1.1) 0.121
Genitourinary 86 (1.2) 90 (1.2) 1.02 (0.76–1.37) 0.907
Miscellaneous medical 459 (6.2) 432 (5.7) 0.91 (0.79–1.04) 0.173
Miscellaneous surgical 121 (1.6) 122 (1.6) 0.98 (0.76–1.26) 0.877
Length of stay >2 d, n (%) 2923 (39.6) 1105 (14.5) 0.26 (0.24–0.28) <0.001
In-hospital mortality, n (%) 6 (0.1) 1 (0.0) 0.16 (0.02–1.35) 0.092

* Rates of complication are not additive, as patients may have had multiple complications.

CI=confidence interval.

In multivariable analyses of propensity score–matched populations (Table 5), patients undergoing RARP were less likely to receive a blood transfusion (odds ratio [OR]: 0.34; 95% confidence interval [CI], 0.28–0.40), to experience an intraoperative complication (OR: 0.47; 95% CI, 0.31–0.71) or postoperative complication (OR: 0.86; 95% CI, 0.77–0.96), and to experience a pLOS (OR: 0.28; 95% CI, 0.26–0.30).

Table 5 Multivariable analyses of propensity score–matched intraoperative and postoperative outcomes, adjusted for age, race, Charlson comorbidity index, hospital region, hospital location, hospital academic status, annual hospital caseload, pelvic lymphadenectomy (yes vs no), and insurance status

Robotic vs open, odds ratio (95% CI) p value
Homologous blood transfusion 0.34 (0.28–0.4) <0.001
Intraoperative complication 0.47 (0.31–0.71) <0.001
Postoperative complication
Overall 0.86 (0.77–0.96) 0.007
Cardiac 0.73 (0.53–0.99) 0.047
Respiratory 0.54 (0.42–0.69) <0.001
Vascular 0.59 (0.37–0.95) 0.029
Operative wound 0.74 (0.47–1.15) 0.183
Genitourinary 1.06 (0.78–1.45) 0.691
Miscellaneous medical 0.95 (0.83–1.09) 0.478
Miscellaneous surgical 1.07 (0.82–1.39) 0.605
Length of stay >2 d 0.28 (0.26–0.3) <0.001
In-hospital mortality 0.21 (0.02–1.94) 0.168

CI=confidence interval.

4. Discussion

To date, the preponderance of comparative studies between RARP and ORP have been of low evidentiary level, largely consisting of single-institution or single-surgeon series. With the introduction and approval of a robot-assisted modifier code in October 2008, it became possible to directly compare the two most common extirpative treatments for PCa. In this study, we examine and compare short-term perioperative outcomes of RARP and ORP, relying on a contemporary population-based cohort.

Several of our findings are noteworthy. For the first time with the use of a nationally representative dataset, we demonstrate that RARP has supplanted ORP as the most common surgical approach for RP in the United States. In this 20% representative sample of all inpatient admissions between October 2008 and December 2009, 19 462 patients underwent RP for PCa; of these patients, 61.1% underwent RARP, 38.0% underwent ORP, and 0.9% underwent LRP. While the ascendancy of RARP has been widely discussed and previous reports have cited an 80% diffusion rate, this rate was based on unverified projections by the manufacturer of the da Vinci surgical system (Intuitive Surgical, Sunnyvale, CA, USA). The temporal trend of this adoption is an interesting academic question; unfortunately, longitudinal evaluation of this dissemination is not available, as the robot-assisted modifier was only recently introduced.

Moreover, this study provides significant evidence in the comparison of RARP and ORP. Multivariable analyses of propensity-matched cohorts adjusted for clustering demonstrate superior perioperative outcomes after RARP in virtually all of the examined outcomes. These contemporary results provide highly generalizable and clinically relevant data.

When individual postoperative complications were examined, cardiac, respiratory, and vascular complications were found to be less likely to occur in patients undergoing RARP than in patients undergoing ORP. These findings suggest that the minimally invasive nature of RARP might have a greater effect on medical than on surgical complications, such as wound and genitourinary complications. As such, it has been hypothesized that minimally invasive surgery is associated with attenuated immune and metabolic acute-phase responses [17], consistent with a reduction in tissue trauma. Such immunologic effects would likely affect the risk of cardiovascular and/or respiratory complications. Indeed, when subanalyses are performed according to CCI, the benefit of RARP is most significant in patients with more comorbidities (data not shown).

Two previous population-based comparisons (utilizing Surveillance Epidemiology and End Results–Medicare) assessed postoperative complications following MIRP compared with ORP. Hu et al. [8] performed a propensity score–adjusted analysis of 8837 men and found no difference in overall complications between the MIRP and ORP cohorts (OR: 0.95; 95% CI, 0.77–1.16). They did find, however, that patients undergoing MIRP had shorter length of stay and fewer respiratory and miscellaneous surgical complications and strictures. Conversely, men undergoing MIRP experienced more genitourinary complications, incontinence diagnoses, and erectile dysfunction diagnoses. Lowrance et al. demonstrated similar findings in a population of 4697 ORP patients and 1006 MIRP patients, noting that procedure type was not significantly associated with the likelihood of postoperative complications following adjustment for patient and tumor characteristics (OR: 0.93; 95% CI, 0.77–1.14) [18]. Based on these results, the authors concluded that ORP and MIRP outcomes were largely equivalent. These previous studies, based on similar datasets, had the same limitations: (1) including only patients >65 yr and (2) failing to account for two important confounders—the introduction and widespread dissemination of RARP, presumably supplanting LRP as the primary MIRP technique, and the learning curve effect that has occurred with the introduction of RARP over time [19].

Although a full discourse on the disparities in access to robotic surgery is outside the scope of this paper, there exist significant differences between patients undergoing RARP and ORP. Patients undergoing RARP were more often white and had fewer comorbidities. They were also more likely to undergo surgery at urban academic centers. These findings emphasize the need for meticulous adjustment for these differences when comparing perioperative outcomes. Specifically, based on previous reports, better surgical outcomes are expected at academic centers [20] and [21]. Recent data also demonstrate that patients without private insurance (Medicare, Medicaid, or self-pay) are more likely to experience intraoperative and postoperative complications, as well as pLOS [22].

One of the principal limitations of this population-based report is the lack of adjustment for tumor stage and other clinical/pathologic characteristics. Although propensity-based matching was performed, there may have been some unobserved differences between the groups for which we were unable to account. These differences include several patient determinants (body mass index, medication) and surgical determinants (extent of lymph node dissection). Moreover, cause of death is not recorded in the NIS. Mortality estimates are based on in-hospital rates. It is possible that the true mortality rate is underestimated, as some patients may have died at other institutions at which their mortality was not captured. Likewise, rates of rehospitalization and reintervention are not recorded. Another limitation of the NIS data is the lack of consistent surgeon identification; accordingly, we were unable to adjust for the effect of surgeon volume and/or the learning curve, two important determinants of operative outcomes. Lack of longitudinal data also provides a unique challenge to RP outcomes studies, as common morbidities (incontinence and erectile dysfunction) are beyond the scope of this study. Comparisons focusing on functional and oncologic outcomes are needed to further substantiate the findings of our report. Finally, we were not able to classify complications according to the Clavien classification or grade them using the Common Toxicity Criteria for Adverse Events [23] and [24].

5. Conclusions

In a broadly representative sample of patients undergoing RP in the United States, RARP has supplanted ORP as the most common surgical treatment of PCa. Superior perioperative outcomes were seen after RARP in virtually all of the examined outcomes, namely, blood transfusions, intraoperative and postoperative complications, and pLOS.

Author contributions: Quoc-Dien Trinh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Trinh, Sammon.

Acquisition of data: Trinh, Sammon, Sun, Schmitges, Bianchi, Hansen, Jeldres.

Analysis and interpretation of data: Trinh, Sammon, Sun.

Drafting of the manuscript: Trinh, Sammon, Sun, Ravi.

Critical revision of the manuscript for important intellectual content: Trinh, Sammon, Sun, Ravi, Ghani, Bianchi, Jeong, Shariat, Hansen, Schmitges, Jeldres, Menon, Karakiewicz.

Statistical analysis: Trinh, Sammon, Sun, Ravi.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: Shariat, Rogers, Peabody, Montorsi, Menon, Karakiewicz.

Other (specify): None.

Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Pierre I. Karakiewicz is partially supported by the University of Montreal Health Centre Urology Specialists, Fonds de la Recherche en Sante du Quebec, the University of Montreal Department of Surgery, and the University of Montreal Health Centre (CHUM) Foundation. Craig G. Rogers and James O. Peabody are speakers for Intuitive Surgical.

Funding/Support and role of the sponsor: None.

References

  • [1] A. Jemal, R. Siegel, J. Xu, E. Ward. Cancer statistics, 2010. CA: Cancer J Clin. 2010;60:277-300 Crossref.
  • [2] A. Bill-Axelson, L. Holmberg, M. Ruutu, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med. 2011;364:1708-1717 Crossref.
  • [3] J.C. Hu, Q. Wang, C.L. Pashos, S.R. Lipsitz, N.L. Keating. Utilization and outcomes of minimally invasive radical prostatectomy. J Clin Oncol. 2008;26:2278-2284 Crossref.
  • [4] J.A. Eastham. Robotic-assisted prostatectomy: is there truth in advertising?. Eur Urol. 2008;54:720-722 Abstract, Full-text, PDF, Crossref.
  • [5] M. Menon. Re: James A. Eastham, robotic-assisted prostatectomy: is there truth in advertising?. Eur Urol. 2009;55:e1 Abstract, Full-text, PDF, Crossref.
  • [6] V. Ficarra, G. Novara, W. Artibani, et al. Retropubic, laparoscopic, and robot-assisted radical prostatectomy: a systematic review and cumulative analysis of comparative studies. Eur Urol. 2009;55:1037-1063 Abstract, Full-text, PDF, Crossref.
  • [7] D.C. Kang, M.J. Hardee, S.F. Fesperman, T.L. Stoffs, P. Dahm. Low quality of evidence for robot-assisted laparoscopic prostatectomy: results of a systematic review of the published literature. Eur Urol. 2010;57:930-937 Abstract, Full-text, PDF, Crossref.
  • [8] J.C. Hu, X. Gu, S.R. Lipsitz, et al. Comparative effectiveness of minimally invasive vs open radical prostatectomy. JAMA. 2009;302:1557-1564 Crossref.
  • [9] R. de Vere White. Re: comparative effectiveness of minimally invasive vs open radical prostatectomy. Eur Urol. 2010;57:355-356 Abstract, Full-text, PDF, Crossref.
  • [10] W.T. Lowrance, E.B. Elkin, P.T. Scardino, J.A. Eastham. Re: comparative effectiveness of minimally invasive vs open radical prostatectomy. Eur Urol. 2010;57:538 Abstract, Full-text, PDF, Crossref.
  • [11] P.J. van Leeuwen. Re: Comparative effectiveness of minimally invasive vs open radical prostatectomy. Eur Urol. 2010;57:539-540 Abstract, Full-text, PDF, Crossref.
  • [12] M.E. Charlson, P. Pompei, K.L. Ales, C.R. MacKenzie. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383 Crossref.
  • [13] R.A. Deyo, D. Cherkin, M. Ciol. Adapting a clinical comorbidity index for use with ICD-9-CM administrative database. J Clin Epidemiol. 1992;45:613-619 Crossref.
  • [14] United States Census Bureau. 2000.
  • [15] R.B. D’Agostino. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Statist Med. 1998;17:2265-2281 Crossref.
  • [16] T.A. Stukel, E.S. Fisher, D.E. Wennberg, D.A. Alter, D.J. Gottlieb, M.J. Vermeulen. Analysis of observational studies in the presence of treatment selection bias: effects of invasive cardiac management on AMI survival using propensity score and instrumental variable methods. JAMA. 2007;297:278-285 Crossref.
  • [17] F.J. Vittimberga Jr., D.P. Foley, W.C. Meyers, M.P. Callery. Laparoscopic surgery and the systemic immune response. Ann Surg. 1998;227:326-334 Crossref.
  • [18] W.T. Lowrance, E.B. Elkin, L.M. Jacks, et al. Comparative effectiveness of prostate cancer surgical treatments: a population based analysis of postoperative outcomes. J Urol. 2010;183:1366-1372 Crossref.
  • [19] J. Schmitges, Q.-D. Trinh, F. Abdollah, et al. A population-based analysis of temporal perioperative complication rates after minimally invasive radical prostatectomy. Eur Urol. 2011;60:564-571 Abstract, Full-text, PDF, Crossref.
  • [20] B.A. Spencer, D.C. Miller, M.S. Litwin, et al. Variations in quality of care for men with early-stage prostate cancer. J Clin Oncol. 2008;26:3735-3742 Crossref.
  • [21] Q.D. Trinh, J. Schmitges, M. Sun, et al. Radical prostatectomy at academic versus nonacademic institutions: a population based analysis. J Urol. 2011;186:1849-1854 Crossref.
  • [22] Trinh QD, Schmitges J, Sun M, et al. Morbidity and mortality of radical prostatectomy differs by insurance status. Cancer. In press. doi:10.1002/cncr.26475.
  • [23] M.N. Simmons, I.S. Gill. Decreased complications of contemporary laparoscopic partial nephrectomy: use of a standardized reporting system. J Urol. 2007;177:2067-2073 discussion 2073 Crossref.
  • [24] L. Nogueira, D. Katz, R. Pinochet, et al. Critical evaluation of perioperative complications in laparoscopic partial nephrectomy. Urology. 2010;75:288-294 Crossref.

Footnotes

a Vattikuti Urology Institute, Henry Ford Health System, Detroit, MI, USA

b Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center, Montreal, Canada

c Department of Urology, University of Cambridge, Cambridge, UK

d Department of Urology, Universita Vita Salute San Raffaele, Milan, Italy

e Department of Urology, Weill Medical College of Cornell University, New York, NY, USA

f Martini-Clinic, Prostate Cancer Center Hamburg-Eppendorf, Hamburg, Germany

lowast Corresponding author. Vattikuti Urology Institute, Henry Ford Health System, 2799W. Grand Blvd., Detroit, MI 48202, USA. Tel. +1 313 405 2829; Fax: +1 514 227 5103.

1 These authors are equal contributing first authors.

z.star Please visit www.eu-acme.org/europeanurology to read and answer questions on-line. The EU-ACME credits will then be attributed automatically.