Open Versus Robot-Assisted Partial Nephrectomy: Effect on Clinical Outcome

Introduction

T he incidence of small renal masses (SRMs) has increased with the widespread use of modern abdominal imaging techniques. Although the majority of patients still undergo radical nephrectomy (RN), nephron-sparing surgery (NSS) is recognized as the gold standard of management for localized SRMs ≤4 cm. ¹ Open partial nephrectomy (OPN) offers equivalent disease-specific survival and better overall survival and renal function than RN. ^{2 –4}

Minimally invasive approaches are a trend in contemporary surgery, and laparoscopic partial nephrectomy (LPN) is one of the options in NSS. LPN has emerged as a viable alternative to OPN, demonstrating comparable intermediate oncologic and functional outcomes with perioperative and convalescence benefits, but with the potential for increased ischemia times and complications. ^5,6 LPN is technically challenging, needing advanced skills for tumor excision and intracorporeal sutured reconstruction, which may limit its widespread acceptance. ⁵ Most studies of LPN have reported longer warm ischemia times than OPN. Indeed, the maximum tolerable warm ischemia time is still a matter of debate. ^{7 –9}

The introduction of robotic surgical systems has offered several benefits for renal surgeons. Articulating instruments and magnified three-dimensional vision facilitate precise tumor resection and renal reconstruction, while minimizing the warm ischemia time. ¹⁰ Therefore, robot-assisted partial nephrectomy (RPN) may bridge the gap between LPN and OPN by facilitating the technical challenge of LPN, and thus help surgeons achieve the standards of OPN while offering a minimally invasive approach. Recent reports on the outcomes of RPN compared with LPN have been promising. ¹¹ There are no studies that have compared RPN with OPN, however.

The objective of this study was to compare the outcomes of RPN and OPN at a single institution.

Patients and Methods

A retrospective review was performed to compare 69 and 234 consecutive patients who underwent RPN and OPN, respectively, between May 2003 and December 2010 by three surgeons at a single institution. This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital. Our hospital was opened in May 2003, and OPN has been performed since that time. The da Vinci robotic system was introduced in October 2007, and RPN was first performed in September 2008. The choice of surgical approach was decided after a full discussion between patients and surgeons. Included were 303 patients with a single sporadic tumor and a normal contralateral kidney. We excluded patients with single kidney, bilateral renal masses, or von Hippel-Lindau syndrome. Patients who choose RN or ablative radiation therapy also were excluded.

We evaluated the following clinical variables and perioperative outcomes between the RPN and OPN groups: Age, sex, body mass index (BMI), clinical tumor size, tumor laterality, tumor location, preoperative estimated glomerular filtration rate (eGFR), operative time, ischemic time, estimated blood loss (EBL), length of hospital stay, and intraoperative complications.

After OPN or RPN, specimens were evaluated for tumor size, histologic subtype, Fuhrman nuclear grade, and pathologic tumor-node-metastasis stage according to the 7th edition of the American Joint Committee on Cancer Cancer Staging Manual. ¹² Follow-up abdominal imaging studies, such as CT, were performed every 6 to 12 months after NSS to assess recurrence. Functional renal outcomes were assessed by comparison of preoperative and postoperative eGFR, which was calculated using the Modification of Diet in Renal Disease formula.

A three-arm technique was used during RPN, and two additional assistant ports were placed. Port location was tailored to the location of the mass and renal hilum. ¹³ Most of the hilar control procedures during RPN were performed with renal artery-only clamping. In the case of hilar tumors, the renal veins were also controlled with bulldog clamps at the discretion of the surgeon. Similar hilar control was performed in major OPN cases, although cold ischemia or no hilar clamping technique was used in some OPN cases, such as small or exophytic lesions.

Most renal reconstruction during RPN consisted of a two-layer repair. The first part of the repair involved intraparenchymal suturing to close the collecting system and achieve hemostasis with the use of continuous 3-0 polyglactin and Lapra-Ty^® clips. The second part of the repair involved parenchymal suturing with the use of 1-0 polyglactin and a sliding-knot technique. When the expected ischemia time was ≥30 minutes, early unclamping was performed after intraparenchymal suturing. ¹⁴ A supplementary tie over a Surgicel^® bolster was optional. A topical hemostatic agent (Tisseel^®; Baxter Corp., Deerfield, IL) was applied over the bolster and the base of the resection. Similar renal reconstruction was performed in major OPN cases.

All data analyses were performed using the Statistical Package for the Social Sciences^® 17.0 software (SPSS Inc, Chicago, IL). We tested the distribution of clinicopathologic parameters using chi-square and Student t tests. All P values were two-sided and a P<0.05 was considered significant.

Results

The preoperative characteristics of the 303 patients who underwent partial nephrectomies are summarized in Table 1. There were no significant differences between the two groups (RPN vs OPN) with respect to patient age (53.5 vs 54.4 y, P=0.609), sex (male proportion, 72.5% vs 70.1%, P=0.703), preoperative eGFR (83.3 vs 79.6 mL/min/1.73 m², P=0.146), tumor laterality (P=0.450), tumor location (P=0.170), and tumor size (2.37 vs 2.58 cm, P=0.256). The RPN group had a tendency to have a higher BMI compared with the OPN group (25.5 vs. 24.5 kg/m², P=0.012).

Table 2 shows the perioperative outcomes between the two groups. The mean operative time was longer in the RPN group (192 vs 143 min, P<0.001). The mean warm ischemia time was longer in the RPN cohort (22.99 vs 18.14 min, P<0.001). There were no significant differences between the two groups (RPN vs OPN) in terms of EBL (229 vs 217 mL, P=0.600) and transfusion rate (4.3% vs 4.7%, P=0.902). As expected, the length of hospitalization (6.2 vs 8.9 d, P<0.001) and use of postoperative analgesics (ketoprofen, 0.3 vs 0.9 ampules, P<0.001) were more favorable in the RPN cohort.

The histologic type, pathologic stage, and Fuhrman nuclear grade were not significantly different between the two groups (Table 3). The number of patients with positive surgical margins (PSMs) was zero for the RPN and six for the OPN groups (P=0.343). All of the patients with a positive final margin status initially elected surveillance over reoperation. Two patients had evidence of recurrent masses on follow-up imaging and underwent radiofrequency ablation (RFA).

Renal functional outcomes were likewise comparable between the two cohorts (Table 4). The creatinine levels were obtained at 1 month after surgery, and the postoperative eGFR and change in eGFR were calculated. There were no significant differences in postoperative eGFR (77.9 vs 74.3 mL/min/1.73 m², P=0.162) and change in eGFR (6.11 vs 5.25 mL/min/1.73 m², P=0.520). In addition, the renal functional outcomes between the RPN group and the warm ischemic OPN group were evaluated, except the cases of cold ischemia and no ischemia in OPN group, because cold ischemia protects renal ischemia. There were no statistically significant differences in terms of the renal functional outcomes.

The intraoperative complication rates were 4.27% (10/234) and 4.35% (3/69) in the RPN and OPN groups, respectively (P=0.999). Complications in the OPN group included nine pleural injuries and one ureteral injury. Complications in the RPN group included one colon, one spleen, and one renal vein injury. The primary repairs in each group were performed without further complications. One attempted RPN (1.4%) was converted to an OPN because of a narrow working space. Two operations in each group necessitated a RN. The reasons for conversion to RN were renal mass complexity and difficulty in hilar control, which was related to severe hemorrhage during attempted PN because of unclamping the unrecognized renal artery branch.

The overall postoperative complication rates were 15.4% (36/234) and 8.7% (6/69) in the OPN and RPN groups, respectively (P=0.158). Adverse events in the OPN group included 20 Clavien grade I complications (5 prolonged ileus, 9 wound problems, 1 urine leakage, and 5 minor problems), 5 Clavien grade II bleeding or hematuria necessitating blood transfusions, and 11 Clavien grade III complications (2 urine leakages necessitating ureteral stents, 9 prolonged bleeding or hematuria necessitating angioembolization). ¹⁵ Adverse events in the RPN group included three Clavien grade I complications (one prolonged ileus, one wound problem, and one minor problem) and three Clavien grade II bleeding or hematuria necessitating blood transfusions. ¹⁵ There were no Clavien grade III complications in the RPN group.

Discussion

There have been no prospective clinical trials that have reported the long-term oncologic outcomes of NSS. The large retrospective series of NSS for SRMs showed similar oncologic outcomes compared with RN, with cancer-specific survival rates at 5 and 10 years of 98.5% and 96.7%, respectively. ¹⁶ Comparing survival outcomes in our series may lack clinical significance because of the short follow-up period in the RPN group and the small number of events, such as low recurrence rate. Therefore, we evaluated the PSM rates instead of the cancer-specific survival rates because of the short-term follow-up in our series.

Our series showed that the PSM rates were 0% (0/69) for the RPN and 2.6% (6/234) for the OPN groups, respectively (Table 3). The more prevalent PSM rates in the OPN group might be related to selection bias. We preferred the easier cases in selecting the initial cases of RPN because we were gaining experience in this area. Two recurrences among six patients with PSM were reported and RFA was performed successfully.

Local recurrence after NSS was reported to be 0% to 5.9%. ⁷ Although PSM in renal-cell carcinoma does not always result in local recurrence and disease progression, surgeons should attempt to achieve clear margins during NSS. A 5.5% PSM rate was documented in the largest OPN series, which included 1390 cases. ¹⁷ Our series had an excellent PSM rate. Good optimal visualization of the tumor and tumor margin achieved by the intraoperative use of ultrasonography and an excellent robotic visualization system may have resulted in no PSMs in our RPN group. We anticipate future RPN cases with PSM, however, because we intent to deliberately expand the indications even in challenging cases.

Chronic kidney disease (CKD) is a graded and independent risk factor for cardiovascular events and death. ¹⁸ RN is an important risk factor for the development of CKD, and NSS should be considered in most patients with small renal tumors. ^18,19 In our RPN group, the mean warm ischemia time was documented to be longer than the OPN group. Mottrie and associates ²⁰ reported that the warm ischemia time (≤20 min) was optimized after the first 30 cases of RPN in the hands of a surgeon with extensive robotic experience. Similarly, the most recent 30 among 69 RPN cases in our series had a mean ischemia time of 19.5 minutes.

The recognition of the maximum tolerable ischemia time during NSS is essential to prevent ischemic renal injury during NSS. Controversy exists, however, regarding the safe maximum duration of ischemia during NSS. Thirty minutes has generally been considered the safe duration for warm ischemia. ²¹ Nevertheless, a recent prospective study showed that warm ischemia time of 25 minutes is a cutoff for irreversible renal damage. ²² Becker and colleagues ⁹ emphasized that hilar clamping should be performed within the minimum time (preferably with <20 min of warm ischemia), regardless of surgical approach.

The optimal warm ischemia time is difficult to achieve with LPN, even in expert hands. ⁵ Robotic instruments may be helpful to decrease the warm ischemia time during RPN because of the ability to operate freely, precisely, and rapidly compared with LPN. Moreover, the robotic reconstructive procedures can be performed safely, even among surgeons with less laparoscopic experience. ²³ Robotic instruments may bridge the gap between LPN and OPN, providing surgeons with the benefit of OPN and the merit of a minimally invasive approach.

The perioperative complication rates for OPN and LPN in a large series were documented to range from 4.1% to 38.6% and from 9% to 33%, respectively. ⁷ A recent RPN series showed perioperative results comparable to a LPN series. ²⁴ Our series also showed that intraoperative and postoperative outcomes in the RPN group were comparable to the OPN group.

There were more Clavien grade III complications, however, including urine leakage necessitating ureteral stents and prolonged bleeding necessitating angioembolization in the OPN group. This might be related to selection bias. We preferred the easier cases in selecting the initial cases of RPN because we were at the initial stage of the learning curve. In future, the number of cases of urine leakage and prolonged bleeding in the RPN group may increase, because we will encounter more difficult cases, such as hilar mass.

To our knowledge, our series is the first study to compare the clinical outcomes between OPN and RPN at a single institution. Because NSS is the gold standard of management for localized SRMs, it is reasonable to evaluate the safety and clinical outcomes of RPN as a new alternative surgery compared with OPN. Our RPN series was performed using no-hybrid technique. Interestingly, most robotic series have been performed using a hybrid procedure. ²⁴ The initial steps of the RPN procedure, such as colon mobilization, kidney mobilization, and exposure of the hilar structures were performed with standard laparoscopic transperitoneal dissection, and the remainder of the procedure was performed robotically.

There were several limitations to this study. The data were collected retrospectively and represented a single center experience. There was a possibility of selection bias interfering with accurate statistical analysis. Our study also had the possibility of type II error because of small sample size of the RPN group. We could not evaluate the efficacy and safety among the three groups (OPN, LPN, and RPN) because the number of LPN cases at our institution was too small. A larger sample size and longer follow-up are needed to confirm our findings and evaluate the oncologic outcomes in a clinically meaningful way.

Conclusions

The RPN group was favored in terms of length of hospitalization and postoperative analgesic use, and the OPN group was favored in terms of ischemic and operative time. Postoperative eGFR change, however, was not different between the two groups. RPN is a viable option as a NSS for SRMs considering perioperative parameters and postoperative renal function changes, in addition to the traditional benefits of a laparoscopic procedure. Long-term follow-up is needed to validate the efficacy of this procedure regarding long-term oncologic control and renal function.