Robotic partial nephrectomy without renal hilar occlusion

Authors


Jihad Kaouk, The Cleveland Clinic Foundation – Glickman Urological Institute, Laparoscopic/Minimally Invasive Surgery, 9500 Euclid Avenue/A100, Cleveland, OH, USA.
e-mail: kaoukj@ccf.org

Abstract

Study Type – Therapy (case series)
Level of Evidence 4

OBJECTIVE

To evaluate operative outcomes among patients undergoing robotic partial nephrectomy (RPN) without renal hilar clamping.

PATIENTS AND METHODS

This was a prospective observational study of patients undergoing RPN under perfused conditions (pRPN). Patients with solitary, radiographically enhancing renal cortical lesions gave consent for pRPN. Salient demographic data, including age, body mass index (BMI) and preoperative tumour size were obtained. Operative data, including mean operative time, estimated blood loss (EBL), and the presence of any complications, were collected. Renal function was evaluated before and after RPN. Remote adverse events were noted. The pRPN group was then retrospectively compared to a contemporary group of patients who had RPN with renal hilar occlusion. Endpoints for comparison included operative time, warm ischaemia time, EBL, length of hospitalization, and the rate of adverse events.

RESULTS

Between February 2008 and December 2008, eight had underwent pRPN; the mean age was 59.3 years, mean BMI 28.7 kg/m2, mean operative time 167 min, mean EBL 569 mL and mean hospitalization 3.75 days. Pathology showed renal cell carcinoma in five patients and oncocytoma in three; the mean tumour size was 2.4 cm. Final pathological margins were negative in all patients. Adverse events included one transfusion and one deep venous thrombosis. When compared to the contemporary group who had RPN with hilar clamping, the operative time was shorter (P = 0.035) and EBL greater (P = 0.018) in the pRPN group. There was no significant difference between the groups in transfusion rate, and no significant difference in renal function before and after surgery either group.

CONCLUSIONS

For selected small renal cortical masses, RPN is safe without renal hilar occlusion. The EBL was higher during pRPN but with no significant difference in the rate of transfusion.

Abbreviations
(p)(R)(L)PN

(perfused) (robotic) (laparoscopic) partial nephrectomy

RN

radical nephrectomy

BMI

body mass index

EBL

estimated blood loss

DVT

deep venous thrombosis

NSS

nephron-sparing surgery.

INTRODUCTION

Nephron-sparing surgery (NSS) is assuming an increasingly prominent role in the management of small, radiographically enhancing renal masses [1]. Partial nephrectomy (PN) offers equivalent disease-specific survival and better overall survival and renal function than radical nephrectomy (RN) [2,3]. Laparoscopic PN (LPN) similarly offers comparable disease-specific survival but with considerably less morbidity and improved convalescence [4]. However, LPN is a technically challenging operation that requires advanced laparoscopic skills and, in most cases, temporary renal arterial occlusion that allows tumour resection and renal reconstruction in a relatively bloodless field [5]. Unfortunately, occluding the renal artery is associated with an uncommon but very real risk of vascular injury and, given the inability to use effective in-situ renal hypothermia during laparoscopy, places the kidney at risk of renal injury after ischaemia [5–7]. For selected peripheral small renal masses, renal arterial occlusion can be avoided during open surgery given the rapidity with which directed haemostasis and renorrhaphy is possible [8]. Duplicating such a technique during LPN is challenging given the complexity of renal reconstruction with this approach. Use of the robotic operating platform during pelvic and abdominal surgery has allowed surgeons to perform complex reconstructive procedures with more precision, dexterity and rapidity [9]. In an attempt to further refine the surgical technique and improve patient outcomes during LPN, we present our experience with robotic PN (RPN) without renal hilar control in the management of selected superficial renal tumours.

PATIENTS AND METHODS

Following Institutional Review Board approval, a prospective observational study was conducted to evaluate the surgical and perioperative outcomes after RPN without renal arterial occlusion for the treatment of small, radiographically enhancing renal masses. Selection criteria included patients with solitary, enhancing cortical renal masses that were <4 cm based on preoperative imaging (Fig. 1). Patients with endophytic lesions, centrally located or hilar lesions, previous ipsilateral renal surgery and/or ablation, and those with solitary renal units were excluded from the study. All patients had a preoperative history taken, physical examination and CT of the kidneys with three-dimensional reconstruction. Appropriate candidates were accrued and consented for RPN without renal arterial occlusion (perfused, pRPN). Baseline demographic data including age, body mass index (BMI), preoperative serum creatinine level, and the presence of comorbidities, if any, were obtained.

Figure 1.

CT image showing a 2.3 × 2.2 × 2.4 cm enhancing renal mass selected for left pRPN; pathology showed a clear cell RCC with negative margins.

During RPN patients were placed in the lateral decubitus position with the operative side facing up. Access to the peritoneal cavity was achieved using a Veress needle. The abdomen was insufflated with CO2 to a maximum pressure of 15 mmHg. Four 12-mm trocars were placed as depicted in Fig. 2a,b. The base of the robot was positioned perpendicular to the patient at the level of the 12th rib (Fig. 3). The robotic camera was docked through the ‘middle’ upper quadrant port. Robotic arms 1 and 3 were docked with a ‘port-in-port’ configuration through the ‘upper’ and ‘lateral’ ports (Fig. 2a). The second robotic arm was not used. The remaining 12-mm port was used by the bedside assistant to provide suction/irrigation and introduce and retrieve sutures during renal reconstruction.

Figure 2.

(a,b) Robotic port configuration during RPN. A port-in-port configuration was used. The bedside assistant port was placed medially to maximize the range of motion. An additional port was placed inferiorly should hilar control with a Satinsky clamp be necessary.

Figure 3.

Schematic representation of the room configuration during RPN.

The colon was reflected from the lower pole to the upper pole of the kidney. The ureter and gonadal vein were identified and preserved. Given the peripheral nature of the lesions in this group, ureteric catheterization was not routine. The renal hilum was identified but not dissected should emergent arterial occlusion be required. Gerota’s fascia was dissected over the kidney and the lesion identified. Intraoperative renal ultrasonography was used to confirm the nature, size, and depth of the lesion. The Harmonic scalpel (Ethicon Endo-surgery, Cincinnati, OH, USA) was used to excise the mass in question with a margin of normal renal parenchyma. All lesions were sent to pathology for frozen-section analysis. For renal reconstruction we used a 2–0 polyglactin suture placed through the capsule of the kidney and sequentially through the parenchyma of the operative bed (parenchymal sutures). Pre-placed Hem-O-Lok clips® (Weck Closure Systems, Research Triangle Park, NC, USA) were used to secure the entry and exit sites of the suture at the renal capsule. Additional 0 polyglactin sutures were placed through the renal capsule and tied over a Surgicel bolster (capsular sutures) (Fig. 4). A topical haemostatic agent (VitagelTM, Orthovita, Malvern, PA, USA) was sprayed under the bolster and at the base of the bed of resection. A Jackson-Pratt drain was placed in all patients.

Figure 4.

The capsular sutures with pre-placed clips are sutured over a Surgicel bolster.

Operative data included operative time, estimated blood loss (EBL), the presence of any intraoperative complications, and the number of sutures used during reconstruction. Postoperative data included the length of hospitalization, mean visual analogue pain scale score at discharge, and the presence of any adverse events. Pathology results were reviewed for tumour type, tumour size and margin status. Patients were evaluated at 1 month after surgery with a nuclear renal functional scan (MAG3) to ensure adequate renal perfusion. Serum creatinine levels were obtained at 1 and 3 months after surgery. Follow-up imaging was otherwise dictated individually based on the patient’s pathology.

The pRPN group was then compared retrospectively with a contemporary group of patients who had RPN with renal hilar occlusion. Endpoints for comparison included age, BMI, pathological tumour size, operative time, tumour size, EBL, number of sutures required during reconstruction, length of hospitalization, and the rate of adverse events.

Descriptive analyses were used to describe the characteristics of the patient samples, expressed as the mean (sd) percentages and frequencies. Independent sample t-tests were used to compare pathological tumour size, age, BMI, operative time, EBL, number of sutures used and length of hospitalization. The Mann–Whitney U-test was used to evaluate warm ischaemia time, and chi-square tests to evaluate adverse events. Serum creatinine values before and after surgery were compared with a repeated measures t-test; for all tests, statistical significance was indicated at P ≤ 0.05 a priori.

RESULTS

Between February 2008 and December 2008, eight patients with radiographically enhancing renal cortical masses were treated with pRPN; the details of this group are given in Table 1. All patients had pRPN on the left (five) or right (three) sides as previously described, with no intraoperative complications or conversions. Immediately after surgery one patient required a blood transfusion for symptomatic anaemia, one developed a deep venous thrombosis (DVT) and was placed on anticoagulation, and one returned to the emergency room for mild nausea but did not require re-admission.

Table 1.  The demographic data and comparative outcomes between patients undergoing pRPN and RPN with renal hilar occlusion
Mean (range) variablepRPNRPN + hilar occlusionP
N patients  8 20 
Age, years 59.3 (38–78) 60.2 (37–79)0.850
BMI, kg/m2 28.7 (21.1–37.2) 29.8 (22.7–44.3)0.610
Serum creatinine, mg/dL  0.87 
Operative time, min167 (118–215)197 (140–270)0.035
EBL, mL569 (250–2000)220 (50–600)0.018
Pathological tumour size, cm  2.38 (1.1–3.5)  2.72 (1.4–4.7)0.341
Warm ischaemia time, min  0 23.8<0.001
N parenchymal sutures  2.5  2.5 
N capsular sutures  3.25  3.4 
Hospitalization, days  3.75 (3–5)  4.050.631
Adverse events:   
DVT, DVT/pulmonary embolism  1  10.864
Transfusion  1  3 
Embolization  1 

Pathology showed stage T1a RCC in five patients and oncocytoma in three. All pathological margins were negative; the mean pathological tumour size was 2.39 cm (Table 1).

Nuclear renal scintigraphy (MAG-3) was completed after surgery in seven of the eight patients with no evidence of renal functional impairment or other abnormalities. There was no significant difference in serum creatinine values before and after pRPN (P = 0.059). At a mean follow-up of 15 months, no adverse events were reported. Repeat CT was negative for recurrence in the five patients with pathological evidence of RCC.

When compared to the contemporary matched group of patients who had RPN with renal hilar occlusion, there was no significant difference in age, BMI, pathological tumour size, number of parenchymal sutures used, number of capsular sutures used, or the length of hospitalization (Table 1). The operative time was significantly longer in the RPN clamped group, and EBL was significantly greater in the pRPN group. Adverse events including the need for blood transfusion were similar between the groups. One patient in the clamped group required re-admission for bleeding and underwent angio-embolization. The clamped group had a significant difference in serum creatinine level before and after surgery (0.866 vs 1.02 mg/dL, P = 0.021).

DISCUSSION

The widespread use of advanced imaging has led to a marked increase in the number of incidentally discovered small renal masses [10]. This stage migration has allowed urologists to investigate less invasive treatments, including laparoscopic NSS. Previously NSS was reserved primarily for patients at risk of end-stage renal disease after RN [11]. Evidence-based studies have subsequently shown equivalent disease-specific outcomes between PN and RN, with improved renal functional outcomes in the former [2,3].

NSS is a technically demanding procedure that requires the operating surgeon to efficiently excise the tumour and reconstruct the kidney. For open surgery these critical manoeuvres are completed rapidly and decisively. The kidney can be compressed to decrease copious bleeding and the field readily evacuated by an assistant while swift, directed haemostasis is achieved. For most small renal masses, renal arterial occlusion is not required during open NSS [8]. With more complex lesions, transient hilar clamping can be used with in situ renal hypothermia that limits post-ischaemia perfusion injury and allows the operating surgeon to meticulously reconstruct the kidney without the duress of a 30-min warm ischaemia threshold [12].

Although laparoscopic NSS offers superior convalescence and decreased pain than open NSS, the procedure is considerably more challenging. Directed haemostasis is difficult to achieve after tumour excision, as bleeding obscures the operative field [13]. As such, the vast majority of renal lesions require hilar occlusion during tumour excision and renorrhaphy. Thus far, outcomes with in situ renal hypothermia during LPN have been mixed [14,15]. The kidney is therefore placed at considerable risk of reperfusion injury if warm ischaemia is carried beyond 30 min.

A recent study by Lane et al.[16] addressed the renal functional outcomes after PN. Predictors of a lower postoperative GFR included a lower preoperative GFR, solitary kidney, older age, gender, tumour size and longer ischaemic interval. Of these predictors, only surgical ischaemia was identified as a modifiable risk factor. The authors advocated limiting warm ischaemia during open and laparoscopic NSS [16]. However, even in experienced hands, the mean warm ischaemia time during LPN is ≈30 min [17].

When the robotic operating platform was introduced and approved by the USA Food and Drug Administration in 2000, the landscape of laparoscopic surgery was profoundly changed. Combining the advantages of laparoscopy with a considerably shorter learning curve, robotics was quickly embraced and is currently used for many extirpative and reconstructive urological procedures [18,19]. The robotic system is ideally suited for reconstructive procedures, given its superior optics, its capacity to filter tremor, and its ability to operate freely and precisely in a confined field. Moreover, the articulating ability of the robot allows the operating surgeon to perform complex reconstruction with rapidity and at angles that would be otherwise difficult to reach with axial laparoscopic operating instruments.

RPN has been well described [20–22] and the outcomes have been generally favourable, with excellent oncological and renal functional results. Warm ischaemia times during RPN approach 30 min [20–22]. Comparative studies with laparoscopy are preliminary but show significantly better outcomes in patients treated with RPN [23]. There are no published data on RPN without renal hilar occlusion.

The primary goal of the present study was to show that for selected small renal cortical masses, RPN might be safe and feasible without occluding the renal hilum. In general, we were able to maintain a clear operative field with the judicious use of the suction/irrigator and apply directed haemostasis with monopolar cautery or suture ligation. This approach allowed us to identify and target active bleeding that can often be associated with delayed haemorrhage and is sometimes not appreciated under ischaemic conditions. Moreover, meticulous directed haemostasis might prevent renal injury associated with bulkier, compressive parenchymal sutures.

Although reconstructive times were not compared to our 500 previous LPNs we feel that the robot allows us to perform renorrhaphy with more precision and rapidity. When compared to demographically similar patients with corticomedullary tumours who had RPN with hilar occlusion, adverse events were similar and the operative time significantly shorter in the pRPN group. The mean EBL was higher in pRPN group but this was probably due to an isolated outlier (EBL 2000 mL). Transfusions were required in three patients in the clamped group and one in the pRPN group. One patient in the clamped group required additional intervention (Clavien Class IIIb complication) in the form of angio-embolization.

The primary limitation of our study is the few patients who had pRPN, but we were very selective in choosing patients for pRPN and plan to expand our indications based on our favourable outcomes. Second, the comparative results between the pRPN and clamped groups were not intended to prove that one technique was better than the other, but rather to identify a subgroup of tumours that might be adequately and safely approached without renal hilar clamping.

In conclusion, based on our experience, pRPN is an appropriate and oncologically sound treatment for selected, small enhancing renal cortical masses. Operative outcomes and adverse events, including the need for a blood transfusion, are comparable to that in contemporary, matched patients who had RPN with renal hilar clamping. The avoidance of warm ischaemia with the unclamped approach might translate into improved long-term renal function. A prospective, randomized, comparative study is needed to confirm these findings.

CONFLICT OF INTEREST

Jihad H. Kaouk is a Paid Speaker for Intuitive Surgical, Inc.

Ancillary