A Multi-Institutional Matched-Pair Analysis of Robotic Partial Nephrectomy for Single vs Multiple Ipsilateral Renal Masses

Introduction

Approximately 81,000 new cases of kidney cancer are diagnosed annually in the United States and ∼14,890 people yearly will die from kidney cancer. ¹ A small group of patients diagnosed with kidney cancer have multiple renal masses in either one or both kidneys. These cases are often associated with hereditary syndromes. Historically, patients with multiple tumors in one kidney were frequently treated with radical nephrectomy. As partial nephrectomy (PN) technology and skill have evolved over time, patients with multiple ipsilateral renal masses are commonly offered PN (sometimes referred to as multiplex PN).

Nephron-sparing surgery is important in patients with multiple renal tumors, especially if they have a concomitant solitary kidney or hereditary syndrome. The primary goals are to prevent disease metastasis, while maximizing native renal function. ^2,3 Preservation of renal function has been shown to prevent chronic kidney disease, which is associated with elevated cardiovascular risk and increased mortality. ⁴

Prior studies have shown PN of multiple ipsilateral renal masses to have good oncologic and renal function outcomes, though some have shown higher complication rates as more tumors are resected. ⁵ In this multi-institutional study, we aim to compare renal function changes, complications, and warm ischemia time (WIT) of partial nephrectomy of a single renal mass (sPN) vs partial nephrectomy of multiple ipsilateral renal masses (mPN).

Materials and Methods

We retrospectively reviewed our multi-institutional PN database from August 2007 to March 2019. Patients with metastatic disease at presentation, completely off clamp procedures, solitary kidney procedures, and missing data observations (primarily nephrometry score) were excluded from analysis. We matched robotic sPN and mPN patients 3:1 using “nearest neighbor” propensity score matching based on age, Charlson comorbidity index (CCI), total tumor size, and nephrometry score. A caliper of 0.15 was used to limit the maximum allowable difference between matched observations.

This resulted in an incomplete 3:1 match as two observations were only able to match twice and one observation was only able to match once. Univariate analysis was performed using Wilcoxon rank-sum tests and chi-square tests for quantitative and qualitative comparisons, respectively, with Fisher's exact test being substituted appropriately for small sample qualitative comparisons. A multivariable linear regression model for clamp time and a multivariable logistic regression model for complication rate were used to determine potential significant associations with multiple tumor procedures while controlling for age, gender, CCI, and tumor size. All analyses were performed using R v4.1.0 with statistical significance set to p < 0.05.

Results

A total of 50 mPN and 146 sPN patients were matched. Table 1 compares pre- and postoperative surgical data of the two groups. In comparing mPN and sPN, mean total tumor size was 3.3 and 3.2 cm, respectively (p = 0.363). The mean nephrometry score in both groups was 7.3 and 7.2, respectively (p = 0.772). Estimated blood loss (EBL) was 137.6 and 117.8 mL, respectively (p = 0.184). The mPN group was found to have significantly longer operative time (174.6 vs 156.4 minutes, p = 0.008), significantly higher percentage of benign histology analysis (34% vs 19.2%, p = 0.036), and WIT (17.0 vs 15.3 minutes, p = 0.032).

There was no significant difference between the two groups in glomerular filtration rate (GFR) change (mPN −6.4% vs sPN −8.7%, p = 0.712). Complications (Clavien 2+) occurred in 10.2% of mPN and 11.3% of sPN patients (p = 0.837).

Our multivariable linear model estimating clamp time in minutes given in Table 2 predicts a nonstatistically significant difference of 1.4 minutes of additional WIT in the mPN group (p = 0.242). However, for every 1 mm increase in tumor size, there was a significant increase in WIT of 0.8 minutes (p = 0.017). Table 3 gives a multivariate logistic regression model estimating the odds of Clavien grade 2 or greater complications. There was no statistical difference in complication rates between groups in a multivariable model (odds ratio [OR] 1.00, p = 0.991). There was also no increased risk of complications based on age, gender, or CCI. There was, however, a significantly decreased risk of Clavien grade 2 or higher complications for every 1 mm increase in tumor size.

Discussion

In this multi-institutional matched-pair analysis comparing robotic sPN with mPN, we found no significant difference in percent GFR change, postoperative complications, or EBL. There was significantly longer total operating room time, and a higher percentage of benign pathology analysis in the mPN group. WIT was found to be significantly longer in the mPN group on univariate analysis, but this was not found to be significant in the multivariable linear model.

Historically, RN was the mainstay of surgical treatment for ipsilateral multifocal renal masses. ⁶ Multifocal PN has become an accepted alternative treatment modality for management of patients with multiple ipsilateral renal masses with acceptable oncologic outcomes and favorable renal functional outcomes. ⁷ As the advent of laparoscopy and robotic surgery has emerged, robotic PN has become increasingly utilized. ⁸ Yerram et al. compared trifecta outcomes in open and robotic PN in a total of 110 patients and found no differences between the two groups 16.3% vs 16.5%, respectively, and the robotic approach was found to have a significantly shorter length of stay. ⁹

Several previous studies have examined complications and WIT in patients undergoing mPN. Boris and colleagues reported on 10 patients undergoing robotic mPN. There were 24 total tumors removed, and of these 22 were proven malignant after surgery. Mean tumor size was 2.3 cm. The mean EBL was 360 mL. Mean operative time and WIT were 257 and 17.6 minutes, respectively. Mean WIT was 29.6 minutes when off-clamp PN were excluded. The mean change in GFR was −7.4%. ¹⁰ Laydner and colleagues performed nine robotic mPN surgeries (two or more ipsilateral tumors).

Mean operative time was 199 minutes. Mean tumor size was 2.2 cm. Mean nephrometry score was 6. Mean WIT was 21 minutes (in the three on-clamp patients). The mean GFR change was −4.0%. No complications were noted. ¹¹ The mean WIT and eGFR change was quite comparable with both studies in our mPN group at 17 minutes and −6.4%.

In 2019, Baiocco et al. published on their experience with multiplex partial nephrectomies (excision of three or more tumors) in solitary kidneys. They performed 78 multiplex PNs and 43 standard PNs. Each group had similar complication rates, renal functional outcomes (including rate of long-term dialysis), and oncologic outcomes. ¹² This is in line with our findings with regard to complications as we also found no significant difference in complication rates or renal functional outcomes between mPN and sPN.

In contrast to our findings, Maurice et al. described their experience with mPN (two or more ipsilateral tumors). They performed 59 robotic mPNs. They found a higher complication rate when more tumor excisions were performed, especially when three of more tumors were excised. Multivariable analysis showed the number of excisions to independently predict complications (OR 3.1, p = 0.041). ⁵

In comparing sPN and mPN in this study, there was no significant difference in WIT on multivariate analysis. In approaching mPN, various strategies to minimize WIT and achieve effective oncologic control have been described and utilized. Mercimek and colleagues reported on a case of a patient with five synchronous ipsilateral renal masses that were surgically removed. For the largest mass (2.7 cm), they clamped a segmental artery to perform PN. The remaining smaller masses were excised using an off-clamp technique.

This combined on- and off-clamp approaches can be useful for treatment of multiple tumors to minimize ischemia time. ⁷ We often employ a slight variation of this technique at one of our institutions where smaller tumors and even larger tumors are excised off-clamp until no longer feasible because of observation, after which point arterial clamping is performed. Metwalli also describes the use of an off-clamp tumor enucleation technique for mPN to minimize renal ischemia.

In addition, closure of Gerota's fascia at the end of the surgery is emphasized to prevent adhesions that would make a repeat surgery for recurrence more difficult. ² A different technique was described by Miyake and colleagues for mPN. They excised six tumors in three patients with solitary kidneys using robotic mPN. Their technique involves early artery unclamping after excision and inner layer suturing of the first tumor, followed by renorrhaphy. Then, the artery is reclamped for excision of the second tumor with early unclamping after the inner layer was sutured. Their trifecta outcomes were achieved. Mean WIT was 14.7 minutes.

One month after surgery, mean decrease in GFR was 31.8%, and no patients progressed to dialysis. ¹³ Another technique is en bloc resection of multiple tumors if they are in proximity to each other. The disadvantage of this technique is the removal of intervening normal parenchyma. Lastly, the most common technique, and one used primarily at our institutions, is excision and renorrhaphy of multiple tumors during the same arterial clamp time. Early unclamping was sometimes employed after the inner layer of suture was placed. This approach is ideal when there are three or fewer small tumors, and WIT can be minimized.

In mPN cases we are inherently more likely to be cognizant of clamp time, and more aggressively employ the aforementioned techniques: early unclamping, off-clamp resection of small tumors, and partially off-clamp resection of larger tumors. Increased mobilization of the kidney in mPN because of multiple mass locations may potentially play a role in ease of excision as well. All of these factors may in part explain the minimal difference in WIT between mPN and sPN.

Interestingly, we found that the mPN group had a significantly higher percentage of benign histology analysis in comparison with the sPN cohort (34% vs 19.2% benign histology analysis). Simhan et al. looked more specifically at the pathologic concordance rate between multifocal ipsilateral renal tumors after 76 mPNs. Pathologic concordance was observed in only 77.2% of malignant cases and 48.6% of benign cases. Only 11 (14%) of these patients underwent laparoscopic or robotic mPN. Mean decrease in GFR was 1.4%. Their overall benign histology rate was found to be 29.5%. ¹⁴ We believe that higher benign tumor rate in the mPN group is likely the result of more complex cysts being resected at the time of other tumor resection.

This study has several limitations. It is limited because of its retrospective nature. Despite propensity score matching, there is potential for possible bias in patient selection in the sPN group. There also may be some intangible effects that are hard to account for such as variation in anatomical factors that may add to intraoperative complexity. The involvement of multiple surgeons at multiple institutions and differing styles of PN with regard to WIT are likely to influence the results.

Despite this heterogeneity, we believe there is inherent value in this study as an effective tool in the armamentarium for urologists to counsel patients on the expected outcomes and potential complication rates of mPN. Future directions will focus on direct comparison of resection techniques of mPN and nuanced comparison of trifecta outcomes based on anatomical tumor location.

Conclusions

Robotic PN in our multi-institutional matched comparison of mPN and sPN showed no difference in complications, renal functional outcomes, or EBL. mPN was associated with increased operative time and WIT, though the WIT difference was not significant on multivariable analysis. Robotic mPN represents a safe and feasible approach to multifocal renal masses and complication rates can be readily compared with those associated with sPN. Robotic mPN should be considered and preferred over radical nephrectomy for patients with multiple renal masses given the higher rate of benignity, similar complication rates, and comparable outcomes when compared to sPN.