Does tumor rupture during robot-assisted partial nephrectomy have an impact on mid-term tumor recurrences?

Partial nephrectomy (PN) is the treatment of choice for T1 renal cell carcinoma (RCC) because it provides comparable oncological safety while better preserving renal function, thus leading to a lower incidence of cardiovascular diseases.¹ Tumor enucleation is a safe procedure oncologically (perioperative, short-term, and long-term) when negative surgical margins are achieved by providing a microscopic layer of healthy kidney tissue on the surface of the tumor.^2,3,4

However, decreased distance between healthy parenchyma and the tumor pseudocapsule increases the risk of slitting into the tumor (positive surgical margin) or even rupturing the tumor during excision and tumor manipulation (tumor cell spillage). There is no clear definition of TR or so-called accidental slit into the tumor with consequent spillage of tumor cells into the operative field and abdominal cavity, the frequency of which has been underestimated and the clinical impact insufficiently investigated in the literature.^5,6,7 One simple inattentive move with sharp instrument by surgeon or assistant could disrupt already thin layer left on the surface of the tumor. Obviously, this would happen less frequently if more of the healthy tissue is left over the tumor capsule.

It has been known that a positive surgical margin in a low malignant tumor does not necessarily lead to recurrence of the disease but there is a higher chance of recurrence in tumors with higher malignant potential.⁸ On the other hand, a little is known if macroscopic spillage of the tumor cells occurs.^5,6,7 The purpose of this study was to investigate the rate of tumor recurrences and clinical impact of tumor rupture (TR) during robot-assisted partial nephrectomy (RAPN), what a surgeon should do in the case of this undesired event and how to avoid it. The rate of tumor recurrences was measured with radiological evidence of tumor in the locoregional region and abdominal cavity.

To the best of our knowledge, there have been only a few papers investigating the effect of tumor rupture or cyst rupture during robotic PN.^5,6,7 On the other hand, a positive surgical margin is much more widely researched and discussed. It seems that a positive surgical margin in cases of RCC (especially of low grade and size) is not associated with an increased risk of recurrence or decreased survival rates as opposed to transitional cell carcinomas or adrenocortical carcinomas.^12,13,14,15 In the context of surgical margin assessment, it is debatable if only TR of the bottom border of the tumor is relevant as rupture can occur far from healthy parenchyma interface. In that case, a surgeon could make a complimentary resection of the tumor bed, so minority of TRs result in a positive surgical margin. In addition, TR can occur when a surgeon or an assistant makes a macroscopic slit into the tumor or a tumor breaks because of manipulation during excision.

In our study, we observed 14 cases of intraoperative TR (14%), which is a high number, especially for something not usually reported in the literature. After a median of 39 months (IQR, 31–47 months), we recorded no cases of tumor recurrence. Interestingly, Khene et al. showed the same percentage of accidental surgical incision into the tumor (ASIT) as we did and concluded it as “common event that did not appear to compromise oncological outcome”.⁵ They observed 9% of recurrences in the ASIT group and 6% in the control group after median follow-up 36 months, while nearly 43% of their cases were high risk tumors (pT2–3A and/or Fuhrman Grade III–IV)⁵ as opposite to ours. Takagi et al. also showed high tumor grade along with pathological tumor upstaging from cT1 to pT3 to be risk factors for worse recurrence-free survival.¹⁶ In addition, Grossmann et al. presented a case report of peritoneal carcinomatosis of the cystic papillary renal cell carcinoma following intraoperative cyst rupture during partial nephrectomy.¹⁷ Apart from early recurrence, there is also a possibility of late recurrence (recurrence after 5 years) which occurs in around 3.5%; main predictive factors for it are higher pathological stage (≥ pT2) and age at surgery.^18,19

Among 14 cases of TR in our study, 86% were carcinomas, 7% were oncocytomas, and 7% were benign tumors, all of them were included because we did not want to solely investigate recurrences. One multicenter cohort study reported an 18.7% rate of intraoperative cystic renal masses rupture via an open or robot-assisted approach, which had no influence on tumor recurrences, including no cases of local or distal recurrences.⁶ Another group identified risk factors for cystic RCC rupture to be higher E (exophytic/endophytic) and N (nearness to collecting system or sinus) RENAL nephrometry scores, higher Bosniak category (specifically III), and surgeon's experience.⁷ Even though recurrence-free survival and cancer-free survival were worse if cystic RCC rupture occurred, it did not seem to influence overall survival.⁷

The only study that indeed investigated the impact of tumor rupture (in their paper called “effraction”) during RAPN showed the main determinants of accidental slit into the tumor to be size of the tumor and experience of the surgeon.⁵ According to our results, a high RENAL nephrometry score seems to be related to TR (P = 0.028). In a TR group, tumors tended to be larger (37 mm vs. 30 mm), but the result did not reach statistical significance (P = 0.2). With regard to surgeon experience, we observed a decrease in the number of TRs over time. In the first 20 cases, there were 5 (25%) TRs, but the percentage decreased to 11% in the following 80 procedures. However, this result did not reach statistical significance (P = 0.11). We suggest 3 reasons that could explain this: (1) with more experience, we started operating more difficult cases; (2) TRs also occur as a consequence of tumor manipulation by an assistant and are not solely dependent on the mistakes/experience of a surgeon; (3) due to a low number of cases, the results did not show the statistical significance. Tumor enucleation is more technically demanding, therefore it could be a risk factor for TR. It is an oncologically safe surgical technique whereby the surgeon leaves a microscopic layer of healthy kidney tissue on the surface of the tumor.^1,2,4 Generally, results regarding recurrences at the enucleation site differ in the literature (ranging from 0% to 8%, depending on the size of the tumor, pT stage, RENAL nephrometry score, follow-up duration). Benign tumors and lower pT stage RCCs did not recur after the follow-ups, whereas RCCs with higher pT stage did.^20,21,22 For example, in sporadic follow-up of RCCs of at least 4 years, there were no recurrences at the enucleation site.² According to Minervini et al. positive surgical margins, recurrence in the ipsilateral kidney (either at the enucleation site or elsewhere), and systemic recurrence were all found in 2.4% of cases, and < 1% of patients died due to metastatic RCC after the median follow-up of 61 months.² Similarly, Hu et al. observed positive surgical margins in 3.5% of cases and less than 1% of recurrences after a median follow-up of 2.7 years.²³ We performed tumor enucleation in most cases and observed a positive surgical margin in 1 case, which is comparable with the results in the literature.^2,23

We wanted to determine the influence of tumor type on the occurrence of tumor rupture. In our series, final pathology reports showed that most ruptured tumors were papillary RCCs, which is not surprising. Fragility is a typical feature of pRCC type I; this can be explained by its histology because its narrow papillae contain only microcapillaries without any binding and a tough pseudocapsule (specimens are described as a “minced meat” structure).¹ Some studies show the peritumoral pseudocapsule to be less developed (thinner, incomplete, or absent) in pRCCs compared with ccRCCs.^24,25 In addition, Hora et al. described 3 cases of spontaneous rupture of pRCCs or after minimal trauma due to extensive necrosis.²⁶ Moreover, pRCCs have been shown to have a substantial risk of renal tumor biopsy tract seeding (12.5%), indicating its malignant potential.²⁷ However, we did not observe tumor recurrence in any of the cases in the TR group.

We also wanted to determine the impact of tumor rupture on the possibility of complications during and after surgery. Pradere et al. showed that intraoperative cyst rupture during PN led to more postoperative complications⁶,which were not observed in our study. Our results showed that duration of the surgical procedure, duration of hospital stay, creatinine and hemoglobin levels (preoperatively and postoperatively) did not significantly differ between the TR and no-TR groups. Even though the estimated blood loss was higher in the TR group (50 vs. 20 mL, P = 0.13), the decrease in hemoglobin was not significantly different between the groups (25 vs. 22 g/L, P = 0.6). We observed 2 major complications (defined as Clavien-Dindo classification score 3 or more), but only in the no-TR group. On the other hand, WIT was significantly longer in the TR group, which could be explained in 3 ways: (1) tumor rupture with spillage of tumor tissue impairs visibility, resulting in more difficult tumor manipulation and further resection; (2) the surgeon decides to perform complementary resection of the tumor bed; (3) psychological stress experienced by the surgeon and decision making on how to proceed with the surgery. Interestingly, there was no case of tumor rupture within the no clamping group, which shows that bleeding during tumor resection alone with impaired visibility is not a sufficient reason for TR.

According to all these findings, we suggest that the surgeon should be careful to avoid TR when performing enucleation of kidney tumors. If pRCC is expected, we suggest enucleoresection instead of enucleation. The surgeon should always warn the assistant to be equally careful with any tumor manipulation (e.g., suction), especially if the tumor seems fragile. It is important that the surgeon stays focused and calm if TR occurs. Clear communication in the team is essential. The surgeon should assess the ability to control bleeding and extent of the spillage of the tumor cells, followed by the decision whether to convert to radical nephrectomy or even to open procedure for better visualization and control. If the surgeon decides to continue robot-assisted approach, sufficient irrigation of the surgical field and consequent suction are needed in order to remove spilled tumor cells. Moreover, a change in strategy (reduction of pneumoperitoneum pressure or switching to global ischemia) should also be considered. It is advisable to require the assistance of more experienced colleagues. After the procedure, patient documentation should be presented at the multidisciplinary team meetings in order to discuss potential adjuvant therapy or follow-up procedures and imaging. We believe that usage of three-dimensional models could make enucleations easier and decrease rates of surgical injury to the tumor.²⁸

There are a few limitations of our study. First, the median follow-up of 39 months is relatively short to observe local recurrences, even though in the study by Khene et al. they observed recurrences after nearly equal follow-up.⁵ Moreover, in the study by Takagi et al. median time from PN to recurrence was 19 months.¹⁶ Second, the definition of TR is questionable because there is no clear pathologic-surgical agreement on what TR is, therefore we used the one available in the literature.⁵ Third, due to the retrospective single-center design of the study, there is a possibility of biased interpretation of the results.

TR is a possible complication during RAPN, especially if tumor enucleation is performed on pRCCs with a higher RENAL nephrometry score, leading to prolonged WIT. We suggest proceeding with the resection of the tumor with a deeper resection plane and only eventually converting to radical nephrectomy or open PN, because it seems that TR has no mid-term risk of tumor recurrence or higher complication rate. The rate of long term effects of TR on tumor recurrences are still unknown.