Amsqraaalnnsmot©

BtawkcSfw

aimtne

FM

ls

©A

ROBOTIC-ASSISTED LAPAROSCOPIC PROSTATECTOMY:WHAT IS THE LEARNING CURVE?

S. DUKE HERRELL AND JOSEPH A. SMITH, JR

ABSTRACTlthough equipment for performance of robotic-assisted laparoscopic prostatectomy (RALP) is becomingore widely available, few surgeons have acquired the skill and experience necessary to master RALP. A

ignificant issue has been the requisite training and experience (ie, the “learning curve”) necessary for aualified surgeon to become an expert at RALP. The senior author (J.A.S.) has experience with �2500adical retropubic prostatectomy (RRP) and �350 RALP procedures. He assessed his learning curve inchieving expertise with RALP. Because there are no objective measures to define expertise with RALP,chievement of expert status depends on the surgeon’s comfort, experience, and results with alternativepproaches. Surgeons with extensive experience with open approaches may “set the bar” higher for theearning curve because of expectations. RALP results comparable to those obtained routinely with RRP wereot achieved until after �150 procedures. Surgeon comfort and confidence comparable to that with RRP didot occur until after 250 RALP procedures. Defining the learning curve for RALP as the point at which aurgeon could provide outcomes comparable to those obtainable with alternative surgical approacheseans that the point varies, depending on the experience and expertise of the surgeon. Surgeons whose soler dominant experience is with laparoscopic or robotic approaches may have a different perception ofhe learning curve compared with an experienced open surgeon. UROLOGY 66 (Suppl 5A): 105–107, 2005.

2005 Elsevier Inc.

cwflogef

tippnqOwrhco

oth nationally and internationally, experiencewith robotic-assisted laparoscopic prostatec-

omy (RALP) is expanding. Because of perceiveddvantages of this surgical approach comparedith alternative techniques, and influence by mar-eting pressures, hospitals increasingly are pur-hasing the da Vinci Surgical System (Intuitiveurgical, Sunnyvale, CA). The robotic instrumentsacilitate technical aspects of the surgery comparedith pure laparoscopic approaches.1,2

Nonetheless, regardless of preclinical trainingnd practice, there is a period during which thenexperience of the surgeon makes the operation

ore difficult or lengthy. This period is frequentlyermed the learning curve and often is defined by aumber of cases. There is no accepted standard forither a definition or measurement of the learning

rom the Department of Urologic Surgery, Vanderbilt Universityedical Center, Nashville, Tennessee, USAReprint requests: Joseph A. Smith, Jr, MD, Department of Uro-

ogic Surgery, A-1302 Medical Center North, Nashville, Tennes-

ree 37204. E-mail: joseph.smith@vanderbilt.edu

2005 ELSEVIER INC.LL RIGHTS RESERVED

urve. Typically, it is the self-declared point athich a surgeon states he or she has become com-

ortable performing the procedure. Thus, theearning curve could vary considerably, dependingn a number of surgeon-related factors. Some sur-eons may be slower to learn the procedure and oth-rs simply may require greater experience before theyeel comfortable with the operation.

Other factors that may be of overriding impor-ance are the experience and results achieved by anndividual surgeon using alternative surgical ap-roaches. A surgeon with limited open radicalrostatectomy experience or with results that mayot match those of some published series may beuicker to declare or perceive expertise with RALP.n the other hand, a highly experienced surgeonith outcomes comparable to those in contempo-

ary published series may set the performance barigher before either a perception or delivery ofomparable outcomes. We reviewed the progressf a surgeon highly experienced with radical ret-

opubic prostatectomy (RRP) in achieving ex-

0090-4295/05/$30.00doi:10.1016/j.urology.2005.06.084 105

pc

cf�tbasnssvcoc

wcwrraethrtRcRc

ptlTidbtc

weicasel

cstavriR

tctaeorc

PpRReTesgsfieali

fTp7ncm

otdacaafi

a

1

ertise with RALP and overcoming the learningurve.

PATIENTS AND METHODS

The senior author (J.A.S.) has performed �2500 RRP pro-edures. Published results have been comparable to thoserom other centers of excellence. A transfusion requirement of

1% has been achieved, and pathways have been constructedo limit postoperative length of stay and surgical-related mor-idity.3–5 In June 2003, RALP was introduced at our hospital,nd to date, �350 procedures have been performed by thisurgeon. During that same time frame, a nearly comparableumber of RRP procedures have been performed, and a pro-pective comparison has been made.6 Multiple outcome mea-ures have been recorded prospectively in an institutional re-iew board (IRB)–approved study, including blood loss,hange in serum hematocrit, transfusion requirement, lengthf stay, postoperative pain, quality of life, surgical margins,ontinence, and potency.

RESULTS

Blood loss, postoperative pain, and length of stayith RALP were minimal, and results with urinary

ontinence have been excellent.7 These resultsere demonstrated from the initiation of the expe-

ience, which implies that these factors are lesselated to the learning curve. There was a markednd steady decrease in operative times as experi-nce was gained.8,9 Surgical margin status and re-urn of erectile function are being analyzed andave improved with greater experience. Overall,esults comparable to those achieved with RRP byhe senior author were not achieved until �150ALP procedures. Further, self-perception of aomparable degree of comfort with RALP andRP was not recognized until �250 RALP pro-edures.

DISCUSSION

Ideally, the learning curve for surgery is com-leted during residency or fellowship training sohat an individual entering practice has achieved aevel of expertise comparable to other surgeons.his ideal is seldom met. Multiple studies show

mproved outcomes with various surgical proce-ures depending on surgeon volume, and this maye particularly true for radical prostatectomy. Al-hough the slope of the learning curve may de-rease, a true plateau may never be reached.For new surgical procedures or approaches withhich there may have been no or limited experi-

nce during training, the implications of the learn-ng curve become even more important. The con-ern, of course, is that a patient will have andverse outcome because of the inexperience of theurgeon. The dilemma is that the only way to gainxperience is to perform multiple procedures. Pure

aparoscopic radical prostatectomy is a techni- y

06

ally demanding procedure that requires uniqueurgical skills. Robotic assistance facilitates su-uring and other aspects of laparoscopic surgerynd helps flatten the learning curve. However,irtually every surgeon who has learned RALPecognizes and acknowledges the improvementn surgeon capability that comes with additionalALP experience.Surgeons frequently make statements about the

ime point at which they were “over the learningurve.” This is a self-declared perception of exper-ise that depends not only on self-perception, butlso on the definition of expertise. Highly experi-nced open surgeons may place a greater demandn themselves before declaring expertise. This canesult in an artifactual prolongation of the learningurve.Our experience seems to validate this concept.

atients at our hospital are managed with the sameerioperative care pathway, whether they undergoALP or RRP. Although some outcomes withALP were immediately equivalent to RRP, overallxperience was not comparable until �150 cases.his is true despite the participation of a highlyxperienced and fellowship-trained laparoscopicurgeon (S.D.H.) early in our series as table sur-eon. Further, the confidence level of the primaryurgeon in RALP did not approach his level of con-dence in RRP until �250 cases. The fact that oth-rs have declared the learning curve for RALP to bes few as 50 or even 25 cases may reflect more rapidearning or superior technical skills, but likely itmplies a different definition of expertise.

These observations have important implicationsor training and adoption of RALP by clinicians.he median number of radical prostatectomieserformed by urologists in the United States is onlycases per year. At this rate, most practitioners mayever overcome the learning curve for RALP. Clini-ians and hospitals must consider this sort of infor-ation when initiating robotic surgery programs.

CONCLUSION

The learning curve for RALP depends not onlyn the technical skills of the surgeon but also onhe self-perceived definition of expertise. This mayepend on the results a surgeon can attain withlternative, open approaches. It is likely that noomplete plateau of the learning curve exists forny surgical procedure. Standardized expectationsnd reporting of outcomes could help to better de-ne the true learning curve for RALP.

REFERENCES1. Menon M, and Tewari I: Robotic radical prostatectomy

nd the Vattikuti Urology Institute technique: an interim anal-

sis of results and technical points. Urology 61: 15–20, 2003.

UROLOGY 66 (Supplement 5A), November 2005

r1

rd

dp

iS

tm1

r9

srt

lr

U

2. Abbou CC, Honek A, Salomon L, et al: Laparoscopicadical prostatectomy with a remote controlled robot. J Urol65: 1964–1966, 2001.3. Koch MO, and Smith JA Jr: Blood loss during radical

etropubic prostatectomy: is preoperative autologous bloodonation indicated? J Urol 156: 1079–1080, 1996.4. Koch MO, Smith JA Jr, Hodge EM, et al: Prospective

evelopment of a cost-efficient program for radical retropubicrostatectomy. Urology 44: 311–318, 1994.5. Smith JA Jr: Techniques to decrease morbidity with rad-

cal prostatectomy. American Urological Association Updateeries 19: 273–280, 2000.

6. Smith JA Jr: Robotically assisted laparoscopic prostatec-

ROLOGY 66 (Supplement 5A), November 2005

omy: an assessment of its contemporary role in the surgicalanagement of localized prostate cancer. Am J Surg

88(suppl): 63S–67S, 2004.7. Herrell SD, and Smith JA Jr: Laparoscopic and robotic

adical prostatectomy: what are the real advantages? BJU Int5: 3–9, 2005.8. Webster TM, Herrell SD, Chang SS, et al: Robotic as-

isted laparoscopic radical prostatectomy versus retropubicadical prostatectomy: a prospective assessment of postopera-ive pain. J Urol (in press).

9. Farnham S, Herrell SD, Chang SS, et al: Comparison ofength of stay between radical retropubic and robotic assistedadical prostatectomy. Urology (in press).

107