Comparison of Robotic and LaparoscopicUltrasound Probes for Robotic Partial Nephrectomy

Bartosz F. Kaczmarek, MD, Shyam Sukumar, MD, Ramesh K. Kumar, MD, Nolan Desa,Kristen Jost, Mireya Diaz, PhD, Mani Menon, MD, and Craig G. Rogers, MD

Abstract

Objectives: To evaluate and compare perioperative outcomes of robotic partial nephrectomy (RPN) using ro-botic and laparoscopic ultrasound probe for tumor identification.Materials and Methods: Data from 75 consecutive RPN procedures using a laparoscopic ultrasound probe( January 2009- November 2010) and 75 consecutive RPN procedures using a robotic ultrasound probe(November 2010- November 2011) were collected. Perioperative outcomes of the two groups were retrospec-tively analyzed.Results: A total of 72 patients underwent 75 consecutive RPN using the laparoscopic ultrasound probe followedby 73 patients who underwent 75 consecutive RPNs using the robotic ultrasound probe. Characteristics weresimilar between groups, and tumors had a similar complexity (mean nephrometry score 6.6 vs. 6.8, p = 0.534),mean operating room time (234 vs. 218 min, p = 0.095), mean console time (173 vs. 156 min, p = 0.071), mean bloodloss (171 mL vs. 164 mL, p = 0.79), and positive tumor margin rates (1.2% vs. 2.2%, p = 1) did not achieve sig-nificance. All patients are free of cancer recurrence after a mean follow up of 25.7 months in the laparoscopicprobe group and of 10.2 months in the robotic probe group.Conclusions: Robotic ultrasound probes for tumor identification during RPN had comparable perioperativeoutcomes and surgical margin rates as a laparoscopic ultrasound probe, but with the advantage of surgeonautonomy.

Introduction

Partial nephrectomy is a standard treatment for smallrenal masses,1,2 offering comparable oncologic results to

radical nephrectomy with the advantage of renal functionpreservation.3–5 Robotic partial nephrectomy (RPN) hasemerged as a viable option for minimally invasive nephronsparing surgery.6–8

Tumor identification during partial nephrectomy that isnecessary for oncologic control can be facilitated by intra-operative ultrasonography.9–13 RPN using a laparoscopicultrasound probe requires the probe to be controlled by thebedside assistant. A robotic ultrasound probe enables thesurgeon, rather than the assistant, to control the probe. Weevaluate our experience with RPN using a robotic ultrasoundprobe under surgeon control and compare the outcomes toRPN using a laparoscopic ultrasound probe.

Materials and Methods

RPN was performed using a laparoscopic ultrasound probefor tumor identification by an experienced robotic kidney

surgeon in 75 consecutive surgeries in 72 patients betweenJanuary 2010 and November 2010. From November 2010 tillNovember 2011, we performed another set of 75 consecutivesurgeries on 73 patients with the utilization of a roboticultrasound probe.

Our RPN technique and subsequent modifications havebeen previously described.6,14 After hilar dissection, kidneymobilization and opening of Gerota’s fascia to expose therenal capsule in the region of the tumor intraoperative ultra-sound was used for tumor identification and delineating ofmargins of resection. Regardless of the probe type used, thetransducer was placed circumferentially along the edge of thetumor to identify the junction between tumor and normalparenchyma. The ultrasound images were displayed as apicture-on-picture image on the console screen using theTilePro feature15 of the da Vinci surgical system (IntuitiveSurgical, Sunnyvale, CA). The renal capsule was scored withcautery approximately a probe-width distance away from thetumor edge to provide an adequate margin of normal pa-renchyma. The renal hilum was then clamped (or in selectedcases, left unclamped), and tumor excision was carried outalong the scored line.

Vattikuti Urology Institute, Henry Ford Hospital, Detroit, Michigan.

JOURNAL OF ENDOUROLOGYVolume 27, Number 9, September 2013ª Mary Ann Liebert, Inc.Pp. 1137–1140DOI: 10.1089/end.2012.0528

1137

In the first group, we used a four-directional laparoscopicprobe introduced through the assistant port, enabling four-directional movement of the probe by the bedside assistant(Fig. 1A). In the second group, an intraoperative ultrasoundwas performed with a robotic ultrasound probe. Technicalcharacteristics of both probes are presented in Table 1. AllRPN procedures were performed by an experienced surgeonand a bedside assistant with more than 100 cases of experi-ence. The robotic probe was introduced through the primaryassistant port, either by dropping it down a 12 mm trocar orby passing it with a microfrance grasper through a 15 mmtrocar, which facilitates faster needles exchange, laparoscopicsponge deployment and enables specimen bag placement andtumor extraction. The probe consists of a transducer with agrooved ridge that fits the robotic grasping instrument and aflexible cable connecting the transducer to the externally lo-cated ultrasound console (Fig. 1B). The surgeon can grasp thenotch on the back of the transducer, enabling the ultrasoundto be controlled directly by the robotic surgeon.

Patient demographics and perioperative outcomes fromboth groups were analyzed. Tumor characteristics as well asR.E.N.A.L. nephrometry scores were assessed based on pre-operative CT images. Postoperative complications using theClavien scale were recorded and evaluated.16 Statistical sig-nificance of comparison between the samples was assessedvia t-test and chi-square test for continuous and categoricalvalues, respectively. Fisher exact test was used for variableswith two categories. Analyses were performed by a statisti-cian using SAS 9.2 for Windows (Carey, NC).

Results

There were 75 consecutive RPN procedures performed in72 patients using the laparoscopic ultrasound probe, and 75consecutive RPN procedures were performed in 73 patientsusing the robotic ultrasound probe. Three patients in thelaparoscopic group and 2 patients in the robotic group un-derwent a second RPN procedure for contralateral tumors.The two patient groups had similar patient characteristics(Table 2). Both groups had similar tumor characteristics, in-cluding R.E.N.A.L. nephrometry scores (6.6 vs. 6.8), exceptthe robotic probe group had a slightly higher endophyticcomponent (42.8 vs. 55.3%, p = 0.004). Perioperative outcomesare shown in Table 2. The robotic ultrasound probe group hadsimilar outcomes to the laparoscopic probe, including oper-ating time (234 vs. 218 min), mean console time (173 vs.

156 min), mean warm ischemia time (WIT) (19.4 vs. 19.0 min),mean estimated blood loss (171 vs. 164 mL), and positivesurgical margins (1.2% vs. 2.2%). There was no statisticallysignificant difference in measured variables between groupswhen controlling for tumor size and complexity in univariateor multivariate analysis. One patient with a positive margin inthe laparoscopic probe group underwent a completion ne-phrectomy with final pathology negative for cancer.

Functional outcomes noted by mean epidermal growthfactor receptor (eGFR) preoperative results compared withthe laparoscopic group were 75.3 versus 78.6 followed bymean eGFR at 1 month postoperatively 70.4 versus 72.3. Aftera mean follow up of 25.7 and 10.3, there was no evidence ofradiographic recurrence.

Discussion

Intraoperative ultrasonography is useful during partialneprectomy for tumor identification to facilitate completetumor removal. Gilbert et al. reported the use of in-traoperative ultrasonography to identify renal cell carcinomain patients with non-palpable tumors.9 Assimos et al. reportedthe use of intraoperative ultrasonography in partial nephrec-tomy for the identification of tumor margins for deep in-traparenchymal lesions.10 Several other studies have describedthe use of intraoperative ultrasound during PN for tumoridentification to help achieve negative margins and identifypotential satellite lesions.11–13

The use of intraoperative ultrasonography has also foundapplication in RPN, where the surgeon can take advantage ofthe ultrasound image superimposed as a picture-on-picturedisplay on the console screen using the TilePro feature.15,17

However, RPN using a laparoscopic ultrasound probe ren-ders the surgeon dependent on the assistant to control theprobe. The robotic ultrasound probe enables the precision of arobotic instrument as well as direct surgeon control. Yakoubiet al. described the use of the robotic ultrasound probe inlaboratory conditions, reporting advantages of direct surgeoncontrol and less instrument clashing in the external operativefield.18 Our group previously described the feasibility of RPNusing a robotic ultrasound probe in 22 patients.19 We nowreport our updated experience in 75 patients and compareoutcomes with RPN patients in whom a laparoscopic probewas used. Our study demonstrated that the robotic ultra-sound probe achieved comparable perioperative outcomes

FIG. 1. (A) A four-directional laparoscopic ultrasoundprobe controlled by the assistant during RPN. (B) Roboticultrasound probe with a flexible cable connecting the trans-ducer to the externally located ultrasound console.

Table 1. Technical Characteristics of Probes

Laparoscopic probe Robotic probe

Probe model Hitachi-AlokaUST-9144-LAP

Hitachi-AlokaUST-5550R

Probe frequency 5–12 MHz 5–13 MHzArray type Convex LinearProbe width 47.5mm 38mmProbe articulation

range4 directional Full range of

a roboticinstrument

Probe shaft Rigid shaft No shaft, flexiblecable

Probe purchasecost

*$30000 *$23000

Data received from manufacturer representative.

1138 KACZMAREK ET AL.

and surgical margins rates to RPN with a laparoscopic ul-trasound probe but with the benefit of surgeon autonomy.

Both groups had tumors of similar complexity as defined byR.E.N.A.L. nephrometry scores,20 although the robotic ultra-sound group had a higher tumor endophytic percentage. Sunand colleagues reported that percent endophytic component isthe single element of the nephrometry score system for whichindependent practicing urologists would recommend in-traoperative ultrasound,21 suggesting its unique importance.There was a trend toward a shorter mean console time in therobotic group by approximately 17 minutes ( p = 0.07). A largerstudy is warranted to further examine this trend.

Although adequate mobilization of the kidney can allow alaparoscopic ultrasound probe to achieve sufficient identifi-cation of tumors in challenging locations during RPN, thisrequires more dissection by the surgeon and less autonomyover probe control. Examples of challenging angles for alaparoscopic ultrasound include the far edges of the tumor, inwhich the probe and/or the kidney should be turned to placethe probe flat on the kidney surface parallel to the tumor edge(Fig. 2). It can be especially challenging with tumors in aposterolateral or upper pole location. We had an experiencedassistant for all cases. The surgeon independence offered bythe robotic ultrasound probe could be more important whenan experienced assistant is not available. However, we feel

that a skilled bedside assistant is still important for other stepsof the surgery. We found that the use of a 15 mm assistanttrocar allows the robotic probe to be passed conveniently with alaparoscopic grasper. We have routinely utilized a 15 mm as-sistant port, as it facilitates faster needle exchanges, laparoscopic

Table 2. Patient Population and Perioperative Data

Laparoscopic probe (n = 75) Robotic probe (n = 75) p-value

Mean patient age, range 60.8 [39–81] 59.1 [32–79] p = 0.370No. of males 54 (72%) 47 (62.67%) p = 0.206Mean BMI 31.35 [21.1–49] 30.1 [17–49] p = 0.245Mean ASA 2.6 [1–4] 2.7 [2–4] p = 0.376Mean CCI 2.4 [0–8] 2.4 [0–7] p = 0.376Mean tumor size [cm] 2.74 [1–7] 3.15 [1–8] p = 0.091No. of tumors > 4cm 8 (10.7%) 17 (22.7%) p = 0.128No. of tumor side left 45 (60%) 38 (50.7%) p = 0.127No. of multifocal tumors 5 (6.9%) 7 (9.6%) p = 0.764No. of tumors removed 82 89No. of solitary kidney 3 (4%) 3 (4%)Mean tumor endophytic% 42.8% 55.3% p 50.004No. tumors > 50% endophytic 20 (26.7%) 29 (38.7%) p = 0.163Mean nephrometry score 6.6 [4–11] 6.76 [4–11] p = 0.543Mean OR time [min] 234 [92–513] 218 [82–437] p = 0.095Mean console time [min] 173 [60–312] 156 [46–270] p = 0.071Mean EBL [mL] 171 [25–1000] 164 [25–800] p = 0.791Mean WIT [min] 19.4 [0–37] 19 [0–35] p = 0.767Mean LOS [days] 2.53 [1–9] 2.69 [1–9] p = 0.382Conversions to radical 1 (1.33%)a 1 (1.33%)a

Retroperitoneal surgeries 3 (4%) 1 (1.33%)Complications Clavien 1 & 2 3 (4%) 3 (4%)Complications Clavien 3 & 4 3 (4%)b 3 (4%)c

Final path positive marginsd 1 (1.22%) 2 (2.25%) p = 1Tumor recurrence 0 0Mean followup [months] 25.7 [16.6–38.1] 10.3 [3.9–16.3] p<0.001Mean eGFR preop 75.32 [32.4–119.6] 78.62 [40.63–148.1] p = 0.299Mean eGFR 1 month postop 70.41 [7.68–159.83] 72.27 [38.67–162.97] p = 0.633

aTumor nephrometry score of 11.dNumber per tumor resected.bARF requiring dialysis, delayed splenic bleed requiring exploration, pseudoaneurysm requiring embolization.cICU admission for elevated blood pressure, splenic bleed requiring repair, pseudoaneurysm requiring embolization.ASA = American Society of Anesthesiologists; BMI = body mass index; CCI = Charlson comorbidity index; RPN = robotic partial nephrectomy;

WIT = warm ischemia time; EBL = estimated blood loss; eGFR = epidermal growth factor receptor; OR = operating room; LOS = length of stay.Significant differences between the cohorts are highlighted in bold.

FIG. 2. (A) Laparoscopic ultrasound probe: challengingangles for ultrasound include the near and far edges of thetumor, in which the angulation of the probe may make itdifficult to place the probe flat on the kidney surface parallelto the tumor edge without extra mobilization of the kidney.(B) Robotic ultrasound probe: the probe angle can be ad-justed with the robotic instrument.

ROBOTIC ULTRASOUND PROBE IN ROBOTIC PARTIAL NEPHRECTOMY 1139

sponge placement, specimen bag deployment, and tumor ex-traction. However, a 15 mm trocar is not required to use therobotic probe, and those who prefer a 12 mm assistant port canhave the robotic probe introduced in a drop-in fashion.

The robotic ultrasound probe lacks a hinging/articulationmechanism that is intrinsic to the laparoscopic probe, whichmay contribute to its lower cost. In the course of the study, wesubjectively observed that the robotic probe resulted in lessprobe slippage and eliminated the need of probe grabbing forreadjustment. This could make it less susceptible to me-chanical issues with prolonged use. Future studies are nee-ded to evaluate whether this would translate into financialsavings. We did not do a formal cost analysis to assesswhether it would be beneficial for centers already perform-ing intraoperative ultrasound with a laparoscopic probe topurchase a robotic ultrasound probe. However, the majorityof cost for an intraoperative ultrasound is for the ultrasoundmachine, not the probe. If centers doing RPN have alreadymade the purchase of an ultrasound machine for an in-traoperative ultrasound that is compatible with a roboticultrasound probe, then the additional cost for a roboticultrasound probe could be justified given the increasedsurgeon independence.

Conclusion

Use of a robotic ultrasound probe for tumor identificationduring RPN yielded comparable perioperative outcomesand surgical margin rates as RPN performed with a laparo-scopic ultrasound probe, but with the advantage of surgeoncontrol of probe movements and independence from theassistant.

Disclosure Statement

No competing financial interests exist.

References

1. Fergany AF, Hafez KS, Novick AC. Long-term results ofnephron sparing surgery for localized renal cell carcinoma:10-year followup. J Urol 2000;163:442–445.

2. Campbell SC, Novick AC, Belldegrun A, et al. Guideline formanagement of the clinical T1 renal mass. J Urol 2009;182:1271–1279.

3. Huang WC, Levey AS, Serio AM, et al. Chronic kidney diseaseafter nephrectomy in patients with renal cortical tumours: Aretrospective cohort study. Lancet Oncol 2006;7:735–740.

4. Sun M, Trinh QD, Bianchi M, et al. A non-cancer-relatedsurvival benefit is associated with partial nephrectomy. EurUrol 2012;61:725–731.

5. Van Poppel H, Da Pozzo L, Albrecht W, et al. A prospective,randomised EORTC intergroup phase 3 study comparingthe oncologic outcome of elective nephron-sparing surgeryand radical nephrectomy for low-stage renal cell carcinoma.Eur Urol 2011;59:543–552.

6. Sukumar S, Rogers CG. Robot-assisted partial nephrectomy.J Endourol 2011;25:151–157.

7. Gettman MT, Blute ML, Chow GK, et al. Robotic-assistedlaparoscopic partial nephrectomy: technique and initialclinical experience with DaVinci robotic system. Urology2004;64:914–918.

8. Benway BM, Bhayani SB, Rogers CG, et al. Robot-assistedpartial nephrectomy: An international experience. Eur Urol2010;57:815–820.

9. Gilbert BR, Russo P, Zirinsky K, et al. Intraoperativesonography: Application in renal cell carcinoma. J Urol1988;139:582–584.

10. Assimos DG, Boyce H, Woodruff RD, et al. Intraoperativerenal ultrasonography: A useful adjunct to partial nephrec-tomy. J Urol 1991;146:1218–1220.

11. Polascik TJ, Meng MV, Epstein JI, et al. Intraoperative sono-graphy for the evaluation and management of renal tumors:Experience with 100 patients. J Urol 1995;154:1676–1680.

12. Matin SF, Gill IS. Laparoscopic ultrasonography. J Endourol2001;15:87–92.

13. Fazio LM, Downey D, Nguan CY, et al. Intraoperativelaparoscopic renal ultrasonography: Use in advanced laparo-scopic renal surgery. Urology 2006;68:723–727.

14. Patel MN, Bhandari M, Menon M, et al. Robotic-assistedpartial nephrectomy. BJU Int 2009;103:1296–1311.

15. Rogers CG, Laungani R, Bhandari A, et al. Maximizingconsole surgeon independence during robot-assisted renalsurgery by using the Fourth Arm and TilePro. J Endourol2009;23:115–121.

16. Dindo D, Demartines N, Clavien PA. Classification of sur-gical complications: A new proposal with evaluation in acohort of 6336 patients and results of a survey. Ann Surg2004;240:205–213.

17. Bhayani SB, Snow DC. Novel dynamic information inte-gration during da Vinci robotic partial nephrectomy andradical nephrectomy. J Robotic Surg 2008;2:67–69.

18. Yakoubi R, Autorino R, Laydner H, et al. Initial laboratoryexperience with a novel ultrasound probe for standard andsingle-port robotic kidney surgery: Increasing console sur-geon autonomy and minimizing instrument clashing. Int JMed Robot 2012;8:201–205.

19. Kaczmarek BF, Sukumar S, Petros F, et al. Robotic ultrasoundprobe for tumor identification in roboitc partial nephrectomy:Initial series and outcomes. Int J Urol 2013;20:172–176.

20. Canter D, Kutikov A, Manley B, et al. Utility of theR.E.N.A.L. nephrometry scoring system in objectifyingtreatment decision-making of the enhancing renal mass.Urology 2011;78:1089–1094.

21. Sun MR, Wagner AA, San Francisco IF, et al. Need for in-traoperative ultrasound and surgical recommendation forpartial nephrectomy: Correlation with tumor imaging featuresand urologist practice patterns. Ultrasound Q 2012;28:21–27.

Address correspondence to:Craig G. Rogers, MD

Vattikuti Urology InstituteHenry Ford Health System2799 W. Grand Boulevard

Detroit, MI 48202

E-mail: crogers2@hfhs.org

Abbreviations UsedASA¼American Society of AnesthesiologistsCCI¼Charlson comorbidity indexEBL¼ estimated blood loss

eGFR¼ epidermal growth factor receptorLOS¼ length of stayOR¼ operating room

RPN¼ robotic partial nephrectomyWIT¼warm ischemia time

1140 KACZMAREK ET AL.