Laparoscopic Completion Nephrectomy for Local Surgical Bed Recurrence After Partial Nephrectomy: An Analysis of Procedural Complexity and Feasibility

Introduction

Local recurrence represents a growing clinical concern after partial nephrectomy (PN) given increased utilization, particularly for lesions of greater anatomic complexity. ¹ Despite reports citing feasibility of percutaneous ablation and repeat PN in this setting, emphasis placed on the management of metachronous lesions arising away from the original nephrectomy bed limits generalizability of these studies to cases of true locoregional recurrence within the ipsilateral kidney. ^{2 –7} As such, repeat extirpation in the form of completion nephrectomy remains widely utilized in the absence of severe renal dysfunction or familial tumor syndromes.

Historically, an open approach has been advocated for repeat ipsilateral renal surgery over concerns related to dense adhesions, visceral injury, and intraoperative hemorrhage. ³ However, as experience with minimally invasive techniques has matured, several groups have explored the feasibility of laparoscopic and robotic approaches. Although ability to safely perform laparoscopic nephrectomy in a previously operated field has been shown for benign indication, the role of minimally invasive surgery for true local recurrence of malignancy is less defined. ⁸ Most reports investigate repeat partial rather than completion nephrectomy, with surgery being performed primarily for de novo lesions distant from the original nephrectomy bed, often in populations with hereditary kidney cancer syndromes. ^6,7

As the biologic aggressiveness of local recurrences within the nephrectomy bed likely exceeds that of metachronous multifocal tumors, we contend that architectural derangements in the ipsilateral kidney as well as technical considerations during surgery differ between these two disease states. ^3,4,9 In this respect, studies that have described the role of minimally invasive techniques in a previously operated kidney may not accurately capture the complexity involved with management of true locoregional recurrence. In this study, we explore perioperative outcomes related to the use of laparoscopic completion nephrectomy for the management of true locoregional recurrence after initial PN for renal cell carcinoma (RCC).

Patients and Methods

After receiving Institutional Review Board approval, data on patients undergoing elective minimally invasive or open PN for RCC between 2006 and 2016 at a tertiary center were collected from a prospectively maintained institutional database. Patients with solitary kidney, benign pathology, or familial RCC were not included. Demographic information, surgical approach, tumor pathology, R.E.N.A.L. nephrometry score, follow-up duration, disease status, and time and site of recurrence were evaluated. Local recurrence was considered if imaging demonstrated a new lesion(s) within or immediately adjacent to the prior PN bed. Definitive diagnosis of recurrence was assigned only after tissue confirmation; metachronous lesions in the ipsilateral kidney away from the original PN bed or in the contralateral kidney were not considered recurrence. New lesions in distant organs were considered metastatic.

Individuals with locoregional recurrence after PN comprised the primary study group. Management of local recurrence was retrospectively reviewed. For cases treated with extirpative therapy, information was collected on patient demographics (age, gender, body mass index, Eastern Cooperative Oncology Group performance status, and Charlson comorbidity index), preoperative clinical factors (history of abdominal surgery apart from previous PN and presence of concomitant metastatic relapse), surgical approach (laparoscopic vs open), intraoperative variables (operative time, operative blood loss, and complications), and postoperative parameters (complications, admission to intensive care unit [ICU], and hospital length of stay).

Postoperative complications were classified based on Clavien–Dindo criteria. Information regarding size and stage of locoregional recurrence was based on pathology derived from the extirpative specimen; size was measured as the sum of lesions if multiple foci were present.

The primary study objective was to assess feasibility of effective laparoscopic completion nephrectomy by characterizing adverse perioperative events, including need for open conversion. The secondary objective was to identify variables associated with occurrence of these complications. Clinical and pathologic features studied were summarized with medians and interquartile ranges or frequency counts and percentages for continuous and categorical variables, respectively. Univariable comparisons between patients who underwent effective laparoscopy vs those requiring open conversion were performed using Mann–Whitney U, chi-square test, and Fisher's exact test. Associations between perioperative clinical variables and risk of open conversion and adverse events were evaluated using chi-square test and Fisher's exact test. Statistical analyses were performed using SAS v9.4 (SAS Institute, Cary, NC). All tests were two-sided and p-values <0.05 were considered statistically significant.

Results

A total of 1259 patients underwent PN for pathology-confirmed RCC during the study period, of which 1142 (90.7%) and 117 (9.3%) were performed via minimally invasive and open techniques, respectively. Localized tumor recurrence (with or without concomitant metastatic relapse) was observed in 45 (3.6%) patients. Percutaneous ablation was performed in five cases, whereas completion nephrectomy was performed in 40 cases, of which 37 were attempted laparoscopically. The three patients not considered in the analysis had undergone an open completion nephrectomy as their index PN had also been performed via an open approach. All patients undergoing laparoscopic completion nephrectomy had undergone initial minimally invasive PN via an intraperitoneal approach. Four patients for whom perioperative information was missing were excluded from analyses, leaving 33 patients in the final study cohort.

Characteristics of the incident tumor at time of initial PN as well as clinical features at time of laparoscopic completion nephrectomy are delineated in Table 1. No major complications were sustained during the original PN, and there were no occurrences of urine leak or hemorrhage. Open conversion during completion nephrectomy was performed in 12 patients (36.4%), with the most common reasons being failure to progress (8) and intraoperative hemorrhage (3); one case was converted due to unexpected invasion of the adjacent colon. Of note, concomitant metastasectomy and/or lymph node dissection was performed in only two cases, both of which were completed via a laparoscopic approach not requiring open conversion. One case involved the resection of satellite metastatic nodules in the omentum, whereas the other involved regional lymph node dissection.

Pathologic and perioperative outcomes related to completion nephrectomy are presented in Table 2, stratified by whether open conversion was necessary. Median blood loss (725 mL vs 175 mL, p < 0.001) and operative time (280 minutes vs 160 minutes, p < 0.001) were significantly higher in individuals requiring open conversion. Overall, 16 (48.5%) patients experienced a postoperative complication within 30 days of surgery, 9 of whom (27.3%) had a major event (Clavien grade ≥3). Nature of major complications is delineated in Table 3. Patients requiring open conversion were at significantly higher risk of major postoperative complications (58% vs 9.5%, p = 0.005) and had a longer median hospital length of stay (4.5 days vs 2 days, p = 0.026). Incidence of ICU admission was 58.3% in the open conversion cohort compared to 14.3% in patients without open conversion, although the difference was not statistically significant.

Patients requiring intraoperative conversion were more likely to have clinical evidence of synchronous metastatic relapse (p = 0.016) or higher nephrometry score of their incident tumor at time of initial PN (p < 0.001, Table 4). No differences in body mass index, performance status, Charlson comorbidity index, incidence of previous abdominal surgery (aside from initial PN), size, stage or margin status of incident tumor, or size of recurrent lesion were observed between open conversion and no conversion groups.

Discussion

Although minimally invasive techniques are widely used for treatment of RCC in the primary setting, perioperative outcomes have not been clearly studied in cases of repeat renal surgery for true locoregional recurrence after PN. We demonstrate that laparoscopic completion nephrectomy, while feasible, is associated with a relatively high incidence of conversion to open surgery. Indeed, the 36% open conversion rate observed in this series is in stark contrast to the 4.3% and 2.9% open conversion rates previously cited by the authors of the current study during index laparoscopic partial and radical nephrectomy procedures. ¹⁰ Interestingly, risk of open conversion was concentrated primarily in patients having previously undergone PN for lesions with high anatomic complexity, as defined by R.E.N.A.L. nephrometry score, or who had evidence of synchronous metastatic relapse.

The clinical relevance of these findings is underscored by significantly higher rates of postoperative morbidity in patients requiring conversion to open surgery; indeed, risks of major postoperative complications and ICU admission were increased in this setting as was postoperative hospital stay. In addition, two perioperative deaths occurred in the conversion group.

Several factors may explain the relationship observed between higher R.E.N.A.L. score of the incident tumor and risk for intraoperative open conversion during attempted laparoscopic salvage nephrectomy. Bruner et al. demonstrated a significantly higher incidence of urine leak after PN for anatomically complex tumors, with this risk increasing ∼90% for each unit increase in R.E.N.A.L score. ¹¹ As these findings support the notion that there is more frequent entry into the collecting system during treatment of higher complexity tumors, subclinical urinary extravasation, even when not resulting in overt symptomatology or postprocedural complications, could conceivably augment the degree of postoperative inflammation and fibrosis within the nephrectomy bed as well as around the remnant renal unit. ^11,12

Moreover, higher R.E.N.A.L score has been shown in several independent series to be associated with prolonged operative time related to more extensive kidney dissection, complex kidney reconstruction, and longer warm ischemia duration. ^13,14 More prominent tissue inflammation and fibrosis expected under these circumstances would likely further complicate any future attempt at ipsilateral kidney surgery.

A higher rate of open conversion was also observed in the setting of synchronous metastatic relapse, a correlation which mirrors the technical complexity involved with cytoreductive nephrectomy for metastatic RCC. In a contemporary series of 116 cases of cytoreductive nephrectomy attempted laparoscopically, open conversion was reported in 14%. In addition, major intraoperative complications approached 12%, irrespective of surgical approach, and exceeded 15% in the postoperative setting. ¹⁵ A population-based study using state-level inpatient data similarly demonstrated significantly higher risk for major complication as well as in-hospital mortality after nephrectomy for metastatic vs nonmetastatic disease. ¹⁶ These findings attest to a more aggressive phenotype of renal tumors with simultaneous metastatic involvement, likely inducing a heavier burden of neovascularity as well as a perirenal desmoplastic response that can obscure natural tissue planes. In this regard, synchronous locoregional and metastatic relapse after nephron-sparing surgery would be expected to only further increase the complexity already inherent in repeat ipsilateral renal surgery.

The clinical relevance of our findings extends beyond the high incidence of open conversion in patients undergoing laparoscopic salvage nephrectomy, but relates more so to the substantial morbidity encountered during a minimally invasive attempt. We observed the overall rate of major postoperative complications among all-comers to approach 30%, with the majority experienced by individuals requiring open conversion. These statistics far exceed rates cited in contemporary series for minimally invasive (5.8%) and open (13%) radical nephrectomy performed in the setting of surgically naive kidneys, and thus speak to the challenging nature of repeat renal surgery that transcends surgical approach. ¹⁷ Indeed, several series evaluating the feasibility of repeat PN, primarily in patient groups with solitary kidney or hereditary syndromes, have also revealed a high complication burden for both open and robotic approaches. ^3,5,6,18 The perioperative complication rate among patients undergoing repeat PN at the National Cancer Institute surpassed 50%, with the vast majority of postoperative events being classified as major in nature. Serving as a further testament to the complexity of repeat intervention is their reported median estimated blood loss of 2400 mL and median operative time of 8.5 hours. ⁵ Recent series using a robotic platform in similar patient groups report more favorable outcomes for repeat renal surgery compared to the traditional open approach. ^6,7 A follow-up study from the group at the National Cancer Institute using robotic repeat PN contextualized their nearly 60% rate of postoperative complications by indicating that major complications were observed in only 11.5% of patients; open conversion, however, was still necessary in 16% of patients. ⁶ As open conversion during minimally invasive nephrectomy has been shown to be associated with a 2.5 increased odds of 30-day mortality, our results encourage careful patient selection when considering a minimally invasive approach to completion nephrectomy. ¹⁹

It is also important to consider that the technical aspects to surgically managing de novo metachronous lesions distant from the original nephrectomy bed likely differ from those involved with treatment of recurrent lesions described in our report. In this regard, the original site of resection, around which scarring is expected to be most prominent, is generally not accessed, whereas it would need to be when treating true locoregional recurrence within the PN bed. Furthermore, renal hilar dissection was often omitted in contemporary series investigating repeat ipsilateral surgery, with substantial proportions of patients undergoing off-clamp tumor excision in an effort to not only maximize preservation of renal function but also purposefully minimize complication risk. ^{3,5 –7}

Dissection of the renal hilum has been identified as among the more complex aspects of repeat kidney surgery given dense adhesions which complicate mobilization of adjacent viscera as well as limit ability to skeletonize the renal vessels. In a review of 25 cases of repeat PN performed in solitary kidneys, the most common intraoperative complication was significant vascular injury, occurring in 25% of patients, all at the time of hilar dissection. ⁵ All patients undergoing salvage nephrectomy in our study required hilar dissection to enable access to and division of the main vessels. In our experience, dense fibrosis in the region of the hilum resulting from prior dissection, as has also been described by Liu and colleagues, often obscures clear visualization of the hilar vasculature as well as precludes the safe, en masse transection of the hilar region with a vascular stapler; this is related primarily to the concern that incorporation of extraneous perivascular fibrotic tissue would interfere with adequate coaptation of the stapler device. ⁵ As such, to ensure adequate vascular control, we have found that some level of hilar dissection is necessary to safely accommodate the vascular stapler.

In this regard, the complexity inherent in performing completion nephrectomy may surpass that of repeat ipsilateral surgery for de novo lesions and thus may not be accurately captured in the established literature. Given the high risk for open conversion and ensuing morbidity in this context, surgeons attempting a laparoscopic approach to completion nephrectomy should approach such cases with special care and preparation—ensuring open conversion tools is immediately accessible, having in place an institutional rapid transfusion protocol, and considering stand-by availability of vascular surgery services. Moreover, considering the technical complexity demonstrated for such cases, they should be performed by skilled laparoscopic surgeons at high-volume centers.

This study has several limitations beyond its retrospective design. Due to the lack of a comparison group undergoing open ipsilateral renal surgery for locoregional recurrence, it is difficult to discern the true impact of intraoperative open conversion during an initial laparoscopic approach and whether complication rates would have differed had these procedures been initially performed with an open approach. Nevertheless, conversion to open surgery has been identified in previous series as an independent risk factor for 30-day mortality. ¹⁹ In addition, the study is limited by the relative infrequency of completion nephrectomy for surgical bed recurrence. As such, multivariable analysis was not performed, introducing possibility for unidentified confounders which explain the relationship with variables associated with conversion. However, given the relative novelty of this data set and the overall infrequency with which local recurrence after PN is encountered, we believe our study findings establish important factors to consider during surgical planning before salvage nephrectomy. Finally, all salvage procedures were performed using laparoscopy, and thus the results may not be applicable to a robotic platform, although the current role of robotic surgery for total nephrectomy remains controversial given significantly higher procedure-related costs. ^20,21

In conclusion, laparoscopic completion nephrectomy for true locoregional recurrence is a complex and technically demanding procedure associated with a high rate of open conversion and significant postoperative morbidity, particularly in cases where laparoscopy is aborted. Although laparoscopy is feasible in this setting, careful patient selection may optimize surgical outcomes, as high nephrometry score of the incident tumor and the concomitant presence of metastatic relapse were associated with risk of open conversion. These features, when present, should help guide procedure selection as well as aid in the counseling of patients regarding surgical approach.