Impact of perinephric fat volume and the Mayo Adhesive Probability score on time to clamping in robot-assisted partial nephrectomy

The aim of this study is to evaluate the association of perinephric fat volume (PNFV) and the Mayo Adhesive Probability (MAP) score with time to clamping (TTC) in robot-assisted partial nephrectomy (RAPN). The study subjects consisted of 73 tumors in 72 patients who underwent transperitoneal RAPN at a single cancer center between February 2020 and July 2022. Clinical characteristics including R.E.N.A.L. nephrometry score, MAP score and PNFV were evaluated in a multivariate analysis in relation to TTC, which was classified into two groups based on median TTC. PNFV and MAP score were analyzed separately. PNFVs were measured by SYNAPSE VINCENT® by a single expert urologist. Median TTC was 67 (range: 36–119) min. Spearman’s rank correlation analysis indicated that a significant correlation was observed between PNFV and MAP score with a value of 0.81 (p < 0.0001). Univariate analysis revealed that R.E.N.A.L. nephrometry score ≥ 7 (p = 0.036), posterior tumor location (p = 0.033), MAP score ≥ 3 (p = 0.02) and PNFV ≥ 250 ml (p = 0.02) were significant factors for prolonged TTC. In a multivariate analysis including PNFV (analysis 1), R.E.N.A.L. nephrometry score ≥ 7 (OR 3.54, p = 0.018) and PNFV ≥ 250cm3 (OR 3.94, p = 0.010) were independent factors for prolonged TTC. Similarly for MAP score (analysis 2), R.E.N.A.L. nephrometry score ≥ 7 (OR 3.54, p = 0.018) and MAP score ≥ 3 (OR 3.94, p = 0.010) were independent factors for prolonged TTC. Both MAP score and PNFV may have a significant impact on TTC.

Many factors contribute to the difficulty of PN, including tumor size and location, and its relationship to blood vessels and urinary collecting system. The R.E.N.A.L. nephrometry score consists of five elements of salient renal tumor anatomy and is widely used as an objective reproducible tool [4]. This scoring system has been reported to predict perioperative outcomes [5,6].
In contrast, the difficulty of intraoperative manipulation in RAPN depends on tumor complexity as well as the amount of perinephric fat conditions, which reflect patient-specific factors. The Mayo Adhesive Probability (MAP) score is a quantitative indicator of perinephric fat (PNF) information that is also used [6,7]. MAP score is simple and feasible for preoperative evaluation [8]. While the MAP score evaluates fat characteristics, the actual perinephric fat volume (PNFV) evaluation is also important in PN.
Robot-assisted PN (RAPN) is increasingly becoming the approach of choice among patients undergoing PN. RAPN overcomes some of the technical difficulties pertaining to the laparoscopic technique [9,10], with the exception of highly skilled surgeons [11]. Even in the era of RAPN, there is a growing need to evaluate the time to dissection of the perirenal fat tissue. Particularly in transperitoneal RAPN, the time to dissection of the perirenal fat tissue is almost equivalent to the time to clamping (TTC) of the renal artery. Evaluating TTC is important to estimate the impact of perirenal fat on surgical outcomes in RAPN.
In the present study, we aim to evaluate the impact of PNFV and MAP score on TTC in RAPN by analyzing patient and tumor characteristics.

Patients
This study protocol was approved by the Institutional Review Board of our institution. Between February 2020 and July 2022, we performed a retrospective analysis of a total of 73 tumors in 72 patients who underwent RAPN via transperitoneal approach for localized clinical T1a-3a renal tumors at the National Cancer Center Hospital East, Chiba, Japan.

Surgical procedure
All RAPN procedures were performed by three expert surgeons using the da Vinci Xi Surgical System (Intuitive Surgical, Sunnyvale, CA, USA) via the transperitoneal approach. Four robotic arms and basically one 12-mm assistant port were placed in a flank position at approximately 60°. The following robotic instruments were used: monopolar curved scissors, fenestrated bipolar forceps, ProGrasp forceps, and 0° scope. Surgical procedures were basically performed according to the method reported by Kaouk [12]. The key steps to renal artery clamping were as follows: (1) bowel mobilization, (2) hilum dissection, (3) tumor exposure, identification and demarcation, and (4) renal artery clamping. All renal artery clamping was performed before the start of renal parenchymal resection. The kidney was mobilized within Gerota's fascia. The renal artery was clamped with a bulldog clamp under warm ischemia. The renal vein was also clamped when tumors were in close contact with the main branches of the renal vein. Renorrhaphy was performed in two layers. In particular, the inner suture was placed using a 3/0 barbed suture in a running fashion.

Data collection
The clinical, surgical, and pathological characteristics were evaluated, including anatomical features of RCC and perinephric fat status. Preoperative computed tomography (CT) images were reviewed by a single expert urologist (N.Y.). Anatomical features of RCC was assessed using the R.E.N.A.L. nephrometry score which consists of (R)adius (tumor size as maximal diameter), (E)xophytic/endophytic properties of the tumor, (N)earness of tumor deepest portion to the collecting system or sinus, (A)nterior (a)/posterior (p) descriptor and the (L)ocation relative to the polar line [4]. MAP score was evaluated as follows: posterior perirenal fat thickness was measured from the renal capsule to the posterior abdominal wall by selecting the CT slice of interest at the level of renal hilum. Perinephric fat stranding was also defined as follows: absent (0 points), moderate (2 points), severe (3 points), as previously described [7].
Evaluation of PNFV was carried out using the following methods. The region of interest (ROI) for PNF area on the affected side was defined as the fatty area within the anatomical structures, including the lateroconal fascia, fusion fascia, psoas muscle, lumbar quadrate muscle and diaphragm [13]. Based on the ROI, 3D images for PNF were reconstructed from the DICOM data of preoperative CT, and the volume of PNF was quantitatively calculated using the SYN-APSE VINCENT ® system (Fuji Medical Systems, Tokyo, Japan) (Fig. 1). PNFV was calculated by subtracting renal volume and tumor volume from PNF area volume.
The cutoff values of age, body mass index (BMI), R.E.N.A.L. nephrometry score, MAP score and PNFV were set at 70-year-old, 25 kg/m 2 , score of 7, score of 3250 cm 3 , with the highest value of "sensitivity − (1 − specificity)" in the receiver operating characteristics (ROC) analysis using prolonged TTC. TTC was defined as the time from console start to renal artery clamping.

Statistical analysis
Differences in the distribution of variables between groups were evaluated using a Chi-square test or Fisher's exact probability test. Spearman's rank correlation, Pearson's correlation or Kendall's rank correlation analysis were used to assess the association between parameters. Uni-and multivariate logistic regression analyses were conducted to evaluate preoperative parameters associated with prolonged TTC. The prolonged TTC group was defined by dividing the patients into two groups based on the median TTC, which was the same criteria as that previously reported in laparoscopic surgery [14]. A reduced multivariate model was developed using the backward method, in which the variable with the highest p value was eliminated from each iteration of the multivariate analysis. Two-tailed p < 0.05 was regarded as significant. Statistical analyses were performed using JMP software version 13 (SAS Institute, Cary, NC, USA). Table 1 summarizes the clinicopathological characteristics of our study cohort. In this series, the median age, BMI, TTC and PNFV were 77 (range: 34-87), 23.7 kg/m 2 (range: 14.5-35.5), 67 min (range: 36-119) and 249.7 cm 3 (range: 25.9-751.1), respectively. There were 46 (63.0%) and 36 (49.3%) patients with nephrometry score ≥ 7 and MAP score ≥ 3, respectively. Two patients (3.0%) with pathological T3a tumors were included.

Results
The association between the PNFV and MAP score are shown in Fig. 2. Spearman's rank correlation analysis indicated that a significantly strong correlation was observed   Table 2 shows the outcomes of logistic regression analyses that assessed the impact of several parameters, including the PNFV or MAP score, which were analyzed separately, on TTC. Univariate analysis revealed that R.E.N.A.L. nephrometry score ≥ 7 (p = 0.036), tumor posterior location (p = 0.033), MAP score ≥ 3 (p = 0.02) and PNFV ≥ 250 cm 3 (p = 0.02) were significant factors for prolonged TTC. In multivariate analysis including PNFV (analysis 1), R.E.N.A.L. nephrometry score ≥ 7 (OR 3.54, p = 0.018) and PNFV ≥ 250cm 3 (OR 3.94, p = 0.010) were independent factors for prolonged TTC. Similarly in terms of MAP score (analysis 2), R.E.N.A.L. nephrometry score ≥ 7 (OR 3.54, p = 0.018) and MAP score ≥ 3 (OR 3.94, p = 0.010) were independent factors for prolonged TTC.
As shown in Table 3, no significant association was found between PNFV and perioperative outcomes, including operative time, warm ischemic time, estimated blood loss (EBL), positive  surgical margin rate and complications of Clavien classification ≥ 3. Similarly, no significant association was found between MAP score and the aforementioned outcomes. According to the multivariate analysis, our cohort was divided into three groups: no risk factor (n = 17), one risk factor (n = 39), two risk factors (n = 17) for both analyses 1 and 2. The median TTC was significantly increased among the three groups: 56 min, 70 min and 80.5 min (p = 0.0024 in analysis 1), and 56 min, 70 min and 80 min (p = 0.0026 in analysis 2).

Discussion
In the present study, we clarified the effect of both patientspecific factors and tumor-specific factors on TTC. Unlike radical nephrectomy, partial nephrectomy is characterized by the fact that the extent of the dissection operation varies with the tumor location and influences tumor identification, and that the fat status influences tumor identification. The current study is significant in that it provides data on the degree of difficulty of the procedure for surgeons, which is useful for preoperative evaluation.
Various fat-related parameters have been reported as factors affecting surgery. Obesity has been reported to be a risk factor of laparoscopic surgery. Several investigators have reported the association between surgical difficulty and obesity [15][16][17]. From the perspective of fat volume, visceral fat volume measured by CT is related to prolonged pneumoperitoneum time of laparoscopic radical nephrectomy [14]. In addition, visceral fat area also affects operative time and is reported to have a higher impact than BMI in laparoscopic RN [18]. In terms of robot-assisted urological surgery, high BMI is associated with longer operative time and blood loss in upper urinary tract and renal surgery [19]. Other studies demonstrated that obesity, particularly increased visceral fat, was associated with surgical difficulty and increasing the risk of postoperative complications in PN [20,21].
When we look at the perirenal fat environment, PNF is a frustrating surgical variable encountered during PN, limiting renal mobilization and renal tumor identification. In assessing fat status, it is important to evaluate not only its partial characteristics, but also total volume. We evaluated PNFV using the VINCENT system. In this method, a manual process was used to set the ROI for the entire perinephric fat area, but the automated algorithm recognizes kidney and tumor regions by maximizing an evaluation measure consisting of the sum of the target likelihood values for each Hounsfield unit value from voxels around the long axis, the edge strength of the boundary, and the fit to the ellipsoid model [14,22]. This system has been widely adopted as a method to measure the volume of a specific target and has been reported to calculate accurate volumes with high reproducibility and inter-observer agreement [22,23].
Adherent perinephric fat (APF) is also one of the major factors complicating partial nephrectomy in obese patients resulting in difficult dissection, higher blood loss and conversion to open surgery or radical nephrectomy [24]. Other studies have indicated that APF is associated with adverse perioperative outcomes including longer operative time [25] and greater EBL [26]. The MAP score, which is based on the posterior PNF thickness and stranding, was found to be significantly related to prolongation of the dissection phase time during RAPN [27] or operative time and EBL [28]. This point has been reported in laparoscopic surgery as well [29]. The MAP score is the most widely used model to assess APF. Posterior perinephric fat thickness has been reported to be significantly related to APF and complications in RAPN [7]. It should be noted that while posterior perinephric fat thickness can be measured quickly, the scoring of fat stranding is subjectively mediated and may introduce bias due to inter-observer variability [30]. In addition, the pathogenesis of APF may correlate with inflammation, and cancer-related inflammation has been implicated in the development and progression of RCC [31]. In contrast, it is interesting to note that histological analysis revealed that larger adipocytes, no inflammatory infiltrate, and no difference in fibrosis in APF was observed with MAP score of ≥ 2 [32]. Further analysis of APF in this regard is warranted. In our multivariable analysis, R.E.N.A.L. nephrometry score was also found to be an independent factor for prolonged TTC. R.E.N.A.L. nephrometry score is well known to be related to the surgical margin, WIT and complications [5]. Although our study evaluated TTC, it was thought that the identification and marking of the tumor before clamping can influence TTC, especially in the case of the endophytic type as a tumor-specific factor. Furthermore, there was no difference in operation time and the incidence of postoperative complications regardless of fat volume or APF status in robot-assisted partial nephrectomy. The reason for the former is that many Asian patients were not large in body size and there was a only small group of patients with MAP score of 5. Consequently, there was no difference in perinephric fat manipulation. The reason for the latter is that the number of complications in our series was lower to begin with. Several previous studies have also reported that MAP scores do not correlate with complication rates [25,33].
Several limitations exist in the present study. First, this study of a relatively small patient cohort has possible selection bias due to the retrospective nature. Gender and tumor location could also be independent factors as the sample size increases. Second, TTC is only one aspect of surgical outcomes. It should be noted that this endpoint is limited and not representative of perioperative outcomes. Third, the RAPNs were performed by multiple surgeons during the study period. However, all RAPNs were performed by surgeons highly experienced in the da Vinci Xi system. Therefore, our conclusions may not be directly applicable to situations relevant to all surgeons and institutions. Finally, the perinephric fat assessment was obtained only from Japanese patients. Therefore, they should not be applied to other races until our results are validated in a global cohort, since there are clear differences in fat composition and body size by race.

Conclusions
PNFV and the MAP score can predict prolonged TTC. In addition, The MAP score could be an alternative to PNFV in the assessment of TTC. The no association between complications and perinephric fat status was observed, suggesting that this was probably due to the predominantly thin patient cohort.