Experience With More Than 500 Minimally Invasive Hepatic Procedures: A Serious Note Of Caution

ORIGINAL ARTICLES

Experience With More Than 500 Minimally InvasiveHepatic Procedures

Joseph F. Buell, MD,* Mark T. Thomas, MD,† Steven Rudich, MD,† Michael Marvin, MD,*Ravi Nagubandi, MD,* Kadiyala V. Ravindra, MD,* Guy Brock, PhD,‡ and Kelly M. McMasters, MD*

Objective: To evaluate our experience with more than 500 mini-mally invasive hepatic procedures.Summary Background Data: Recent data have confirmed thesafety and efficacy of minimally invasive liver surgery. Despitethese reports, no programmatic approach to minimally invasive liversurgery has been proposed.Methods: We retrospectively reviewed all patients who underwenta minimally invasive procedure for the management of hepatictumors between January 2001 and April 2008. Patients were dividedinto 3 groups: laparoscopy with intraoperative ultrasound and biopsyonly, laparoscopic radiofrequency ablation (RFA), and minimallyinvasive resection. To compare the various forms of surgery, weanalyzed the incidence of complications, tumor recurrence, mortal-ity, and cost. Statistical analysis was performed using �2 analysis,Student t test, Kaplan-Meier survival analysis with the log-rank test,and multivariable Cox models.Results: A total of 590 minimally invasive hepatic procedures wereperformed during 489 operative interventions. The representative tumorhistologies were: hepatocellular carcinoma (HCC; N � 210), colorectalcarcinoma (N � 40), miscellaneous liver metastases (N � 42), biliarycancer (N � 20), and benign tumors (N � 176). Thirty-five patientsunderwent laparoscopic ultrasound and confirmatory biopsy alone;201 patients underwent 240 laparoscopic RFAs, and 253 patientsunderwent 306 minimally invasive resections. Conversion rates toopen surgery for the RFA and resection group were 2% overall. Onehundred ninety-nine (40.6%) patients were cirrhotic; 31 resections wereperformed in cirrhotic patients. Complication and mortality rates forRFA and resection were comparable (11% vs. 16%, and 1.5% vs.1.6%). However, complication rates (14% vs. 29%; P � 0.02) andmortality (0.3% vs. 9.7%; P � 0.006) rates were higher in thecirrhotic versus noncirrhotic resection group. Overall recurrencerates for RFA and resection groups were 24% and 23%, respec-tively. Local recurrence rates were higher in the RFA group (6.3%versus 1.5%; P � 0.06). Overall patient survival differed between

HCC patients receiving RFA alone and those receiving RFA andOLT (P � 0.0001). Overall survival for cancer patients receivingRFA versus resection differed significantly when unadjusted forother covariates (P � 0.01), and remained marginally significant ina multivariable model (P � 0.056).Conclusions: Minimally invasive hepatic surgery has become aviable alternative to open hepatic surgery. Our present data areequivalent or superior to those encountered in any large open series.Our experience with RFA confirms a low local recurrence rate andan excellent technique for bridging patients to transplantation. Mor-bidity and mortality rates for minimally invasive hepatic resectionsin cirrhotics, is similar to other reported open resection series. Thisseries confirmed excellent interim survival rates after laparoscopicHR and superiority over RFA in the treatment of cancer, withsignificantly lower local tumor recurrence rate.

(Ann Surg 2008;248: 475–486)

Minimally invasive surgical techniques have changed thelandscape of surgery worldwide. From the inception of

the laparoscopic cholecystectomy to robotic radical prosta-tectomy, minimally invasive surgery has changed the practiceof modern surgery.1–5 Often a minimally invasive approachto a standard open procedure has been received with skepti-cism and not infrequently considerable criticism. This skep-ticism was justified by the doubling of bile duct injuriesduring the initial era of laparoscopic cholecystectomy.6

Initial reports of laparoscopic hepatic surgery werelimited to small wedge resections performed by Gagner, andthe group at the University of Maryland in the early 1990s.7,8

During this same period, hepatic cryosurgery was largelyabandoned for what was hoped to be a safer form of thermalablation technology.9–11 The first large series of radiofre-quency ablation (RFA) for hepatic tumors was reported byCurley.11 This series was a combination of operative andpercutaneous procedures, but did confirm RFA was a safe andefficacious method for tumor ablation with impressive earlyterm results in terms of local disease control. Subsequentseries, by Siperstein and Poon, confirmed the laparoscopicapproach to cirrhotic ablations was also safe and effec-tive.12,13 However, subsequent studies reported greater localrecurrence rates for RFA of malignant liver tumors, leadingmany centers to adopt a more aggressive approach to resec-

From the *Departments of Surgery and Jewish Hospital TransplantationProgram, University of Louisville School of Medicine and †Departmentof Bioinformatics and Biostatistics, School of Public Health and Infor-mation Sciences, University of Louisville, Louisville, Kentucky; and‡Department of Surgery, University of Cincinnati, Cincinnati, Ohio.

Reprints: Joseph F. Buell, MD, FACS, Jewish Transplant Center, Division ofTransplantation, Department of Surgery, University of Louisville, Lou-isville, KY 40202. E-mail: [email protected].

Copyright © 2008 by Lippincott Williams & WilkinsISSN: 0003-4932/08/24803-0475DOI: 10.1097/SLA.0b013e318185e647

Annals of Surgery • Volume 248, Number 3, September 2008 475

tion when possible as opposed to RFA.14,15 Laparoscopicliver resection, if feasible, provides many of the advantagesof a less invasive procedure with complete tumor extirpationand assessment of resection margins.

After several years of sporadic reports and animalmodels, Cherqui was the first to report a large series ofsegmental resections.16–18 This led to a flood of small seriesconfirming safety of segmental and subsegmental resectionswith little addition to the surgical literature.19–27 O’Rourkepopularized the tunnel technique for the performance of alaparoscopic right hepatic lobectomy.28 This was followed bythe adaptation of laparoscopic hepatic surgery for pediatricand adult living-related donor hepatectomy by Cherqui andKoffron.29,30 Koffron et al also reported their experience with300 laparoscopic liver resections, the largest series reportedto date, with acceptable morbidity and no mortality; themajority of these patients had benign tumors.31 Much remainsto be learned regarding the proper application of minimallyinvasive techniques including robotics for patients with livertumors, however. Therefore, we reviewed our experiencewith more than 500 minimally invasive hepatic procedures.

PATIENTS AND METHODSWe retrospectively reviewed all patients who under-

went a minimally invasive procedure for the management ofhepatic tumors, between January 2001 and April 2008 at theUniversity of Cincinnati and the University of Louisville,Jewish Hospital; the study was approved by the InstitutionalReview Board at both institutions. To compare the variousforms of minimally invasive liver surgery, we analyzed theincidence of complications, hospital cost, local recurrencerates, and survival.

Statistical analysis was performed using various tech-niques including �2 analysis, Student t test, Kaplan-Meiersurvival analysis and log-rank tests, and Cox proportionalhazards models. Survival distributions for laparoscopic resec-tion, RFA, and other patient populations were estimated usingKaplan-Meier curves, and differences in survival distribu-tions between groups were determined using the log-ranktest. Three year survival rates along with 95% confidenceintervals were calculated based on the Kaplan-Meier esti-mates. Multivariable models for survival times were fit using theCox proportional hazards model, and significant effects deter-mined using Wald tests along with corresponding hazard ratios(HR) and 95% CIs. Statistically significant differences in fre-quency of complications, mortality, and tumor recurrence be-tween laparoscopic resection and RFA patients were determinedusing �2 analysis as the default and Fisher exact test whereindicated. �2 analysis was also used to compare differences infrequency of other dichotomous outcomes between patient pop-ulations. Differences in patient operating costs and other contin-uous variables were evaluated using Student t test.

Laparoscopic ResectionFor small liver tumors located at the periphery of the liver,

tumors in a cirrhotic liver, and lesions located in the caudate, apure laparoscopic approach is used when possible. For left lobeand anterior right lobe lesions, a supine position is selected. Incases of a major right lobe hepatectomy or a posterior lobe

lesion, a left lateral decubitus position is selected. In bothapproaches the patient is placed in a steep reverse Trendelenbergposition. A 30- or 45-degree laparoscope is placed at or adjacentto the umbilicus. Using the Hasson technique, the laparoscope isplaced followed by 2 to 3 additional working ports placed in theupper abdomen. After the abdomen is surveyed through thelaparoscope, the liver is then inspected using the intraoperativeultrasound. With the aid of the laparoscopic ultrasound a 2-cmmargin is marked on the surface of the liver with an ultrasonicdissector or argon beam coagulator.

Liver transection proceeds with mobilization followedby capsular fracture according to techniques previously pub-lished.32,33 Port access is obtained using balloon port trocars(Covidien, Norwalk, CT). Balloon port trocars are beneficialin that they can provide a mechanism to create upwardtraction on the abdominal wall, thereby improving workingspace for appropriate articulation of the endo-mechanicalstaplers. Capsular fracture can be achieved through a varietyof forms of electrosurgical energy and/or preablation. Oncethe capsule is incised 2 cm into the parenchyma, completedivision is carried out via the 60 mm- 2.5mm vascular loadendo-mechanical stapler (endoGIA, Covidien, Norwalk, CT).No vascular isolation or clamping is performed, unless vas-cular division is necessitated to achieve adequate surgicalmargins or lobar mobilization. Hemostasis is carried outthrough a combination of techniques including argon beamcoagulator, and/or topical hemostatic agents. Acute venousand arterial hemorrhage is managed through stapler reappli-cation, or a “quick stitch” as described by Koffron et al.31,32

During the pure laparoscopic approach for resection, thespecimen is placed in a 10 or 15 mm retrieval bag. In contrast,during hand-assisted resections specimens are removedthrough the protractor portion of the hand port.

Surgical drains are placed selectively and the majorityof patients are transferred to a regular or monitored floor bedwithout the need for an intensive care unit (ICU) stay.Patients are started on a diet the night of surgery or on the firstpostoperative day.

Hand-Assisted Laparoscopic ResectionThe conduct of a hand-assisted resection is similar to that

of the pure laparoscopic approach. As a principle, all majorhepatic resections are performed with the aid of a hand-assistdevice. During the initial phase of our program we used theLapDisc (Ethicon, Cincinnati, OH). With the advent of thesmaller profile locking Gelport (Applied Medical, Rancho SantaMargarita, CA) our group has converted to this hand-assistdevice. The hand-assist devices are uniformly placed in the rightupper quadrant. During the performance of left lobectomy andright anterior resections, the hand-assist device is placed in theright upper quadrant just above the level of the umbilicus. Thisallows the senior surgeon to stand on the patient’s right side andplace his right hand in the patient’s abdomen.

For performance of right lobectomy or right posteriorsegmentectomy, a more superior hand-assist port placementis necessary to allow the senior surgeon to reach the dome ofthe liver. The senior surgeon can then stand on the patient’sleft side and place his left hand in the patient’s abdomen. TheGelport also allows placement of instruments through the

Buell et al Annals of Surgery • Volume 248, Number 3, September 2008

© 2008 Lippincott Williams & Wilkins476

hand-assist device without loss of pneumoperitoneum. Clo-sure of the hand-assist site is followed by insertion of acatheter-based analgesia system.

Laparoscopic RFALaparoscopic RFA is performed in a similar fashion to

liver resection. The patient is positioned supine for the ap-proach to a left-sided or right anterior lesion. Similar to theapproach to right posterior resections, RFA for right sidedlesions is performed in a left lateral decubitus position.During both approaches, 2 ports are placed by Hasson tech-nique. Often the primary port is placed at the inferior marginof the umbilicus, so as to not disrupt a significant umbilicalvarix. The second port is often placed in the subcostal margin.Intraoperative ultrasound is performed through this secondport to coordinate percutaneous biopsy and RFA probe place-ment. All biopsies are performed through the assistance of a12-guage intravenous catheter sheath to avoid the potentialfor tumor seeding along the biopsy tract. All ablations havebeen performed using a 3.5- or 4.0-cm to 25-cm long mono-polar probe (Boston Scientific, Boston, MA) using theRF3000 generator. Two full cycles of therapy are performedaccording to an established treatment algorithm, confirming“roll off”, as measured through impedance.

Oncologic MonitoringPostoperatively, all cancer patients are followed in a

similar manner. Imaging via triphasic or triphasic abdominal,pelvic and chest computed tomography are performed atstandard intervals. These imaging intervals are 6 weeks, 3, 6,9, 12, 18, and 24 months. For individual cancers, the corre-sponding tumor markers are: colorectal cancer, carcinoem-bryonic antigen (CEA); HCC, alfa-fetoprotein (AFP); carci-noid tumors, and 5-hydroxyindoleacetic acid (5-HIAA)measured at the above intervals.

RESULTSStandard nomenclature was used to describe the type of

resection performed. All references to RFA are made usingCouniaud segment nomenclature. The study group comprised489 patients undergoing 590 minimally invasive procedures.Table 1 outlines the distribution of the 3 surgical groups withrespect to pathology and complications. The first subgroupcomprised 35 patients who underwent laparoscopic explorationand biopsy without a therapeutic procedure. In this group, 24patients had tissue diagnosis of cirrhosis. The majority of theseprocedures were performed to determine exclusion for therapeu-tic modalities, including resection, RFA, and/or transplantation.The second group comprised 253 patients who underwent 306

laparoscopic liver resections. In the resection group, 31 patientshad evidence of either grade IV fibrosis or cirrhosis. The re-maining group comprised 201 patients who underwent 249laparoscopic RFA procedures. In the RFA group, 144 patientshad tissue diagnosis of cirrhosis. The mean follow-up for thelaparoscopic resection group was 28 (median 26) and 30 (me-dian 25) months for the RFA group.

The majority of patients who underwent laparoscopicintervention were for malignancy; Table 2 outlines the principlehistology of tumors present in the 310 malignant patients and the179 benign tumors that underwent laparoscopic intervention.Malignant etiologies included HCC (N � 210), non-HCCprimary liver and biliary cancers (N � 20), noncolorectalmetastases (N � 42) and colorectal metastases (N � 40). Thebenign tumor group comprised hemangiomas (N � 46),cystic lesions (N � 40), adenomas (N � 25), regenerativenodules (N � 14), focal nodular hyperplasia (N � 21) andother solid masses (N � 31). Table 3 outlines the primary

TABLE 1. Operative Groups

No. Patients No. Procedures Cirrhotic Cancer Conversions to Open (%) LOS (days) Comps (%) Mortality (%)

IOUS 35 35 24 (66%) 21 (60%) 0 (0%) 2 2 (6%) 0 (0%)

Resections 253 306 31 (12%)* 106 (42%)* 6 (2.4%) 2.9 41 (16%) 4 (1.6%)

RFA 201 249 144 (72%) 185 (92%) 4 (2%) 2.4 23 (11%) 3 (1.5%)

Total 489 590 199 310 10 2.6 66 7

*P � 0.05 resection versus IOUS and RFA.

TABLE 2. Primary Histology of Patient

IOUS Laparoscopic Resection RFA Total

HCC 16 36 158 210

Non-HCC primary 3 12 5 20

NCR 2 27 13 42

CRC 0 31 9 40

Hemangioma 0 42 4 46

Cystic lesion 0 40 0 40

Adenoma 0 20 5 25

Regenerative nodule 12 0 2 14

FNH 0 19 2 21

Other solid masses 2 23 3 31

Total 35 253 201 489

TABLE 3. Primary Resection Performed

Cystic Benign Solid Cancer Total

Segment 11 29 32 72

Two or more segments 8 30 26 64

Left lateral lobe 5 17 20 42

Left lobe 6 5 13 24

Right lobe 9 17 7 33

Right triseg 0 2 3 5

Caudate 0 5 1 6

Central 1 4 2 7

Total 40 (16%) 107 (42%) 106 (42%) 253

Annals of Surgery • Volume 248, Number 3, September 2008 Experience With Minimally Invasive Hepatic Procedures

© 2008 Lippincott Williams & Wilkins 477

type of surgical resection performed for these lesions where-asreas the total resections performed: 112 segmentectomies,77 bi- or trisegmentectomies, 42 left lateral segmentectomies,24 left lobectomies, 33 right lobectomies, 5 right trisegmen-tectomies, 6 caudate lobectomies, and 7 central hepatic re-sections. Thirty-one cirrhotic resections were performed lapa-roscopically in this current series. These resections included18 segmentectomies, 6 bisegmentectomies, 4 left lateral seg-mentectomies, 2 left lobectomies, and 1 central hepatic re-section. No hemihepatectomies or extended resections wereattempted in this group.

Comparison of laparoscopic resections to modern co-horts of laparoscopic and open liver resections from North-western University revealed that laparoscopic resection inthis series compares favorably to those results. In general,laparoscopic resection was associated with reduced operativetime, reduced blood loss, complications, transfusion rate, andshorter lengths of hospital stay as is seen in Table 4. Nosignificant differences in bile leaks or local tumor recurrencewere observed. During the course of our resection series weexperienced 4 30-day mortalities. All deaths were in cancerpatients, with 3 in the cirrhotic group. In both our currentresectional series and the Northwestern series few operativeconversions were required. Table 5 chronicles tumor charac-teristics and resulting margins for the 106 patients whounderwent laparoscopic resection for: colorectal metastasesn � 31, noncolorectal metastases n � 27, and primary hepatic

tumors n � 48. No significant differences in tumor size wereencountered among tumors with differing histology, with theexception of cirrhotic versus noncirrhotic hepatocellular can-cers (3.1 vs. 6.4 cm; P � 0.001). Resection in cirrhotics wasreserved for smaller tumors with no significant difference inresulting margins when compared with the noncirrhotic hep-atocellular cancer group. (P � 0.09). During the course ofthis experience, 3 positive resection margins were left afterlaparoscopic resection. Two were during diagnostic resec-tions in patients with diffuse hepatic disease and the third wasin a mass-forming cholangiocarcinoma with ascending right-and left-ductal involvement. Two colorectal patients sufferedlocal marginal recurrences. Both patients were re-explored,the first was again resected for cure, whereas the second wasfound to have diffuse disease including a port site metastasis.The later patient is currently alive and undergoing systemicchemotherapy. As is seen in Table 5, recurrence rates variedamong the different histology groups (P � 0.02, Fisher exacttest), with the highest rate observed in the colorectal resectiongroup and the lowest in the hepatocellular cancer group (35%versus 6%).

In the RFA group there were 201 patients treated with themajority consisting of HCC patients. Table 6 outlines the tumorcharacteristics and surgical outcomes of these RFA patients. Inthis group the mean number of tumors per patient was 1.3 andthe mean tumor diameter was 3.2 cm. This group required 4conversions to an open procedure, 3 of which were in transplant

TABLE 4. Open Versus Laparoscopic Resection

NW OpenN � 100

NW LaparoscopicN � 300

Current LaparoscopicN � 253

Marseille OpenN � 100

Operative time 182 99 162 242

Blood loss 325 102 222 Unknown

Transfusion % 8* 1* 7*† 73

LOS 5.4 1.9 2.9 16

Bile leaks % 4‡ 3‡ 4‡ 7‡

Complications % 22§ 9§ 16§¶ 45

Local recurrence % 3 2 1 Unknown

Conversion N/A 1 2 N/A

*P � P �0.001 versus Marseille; †P � NS versus NW Open and Lap; ‡P � NS.§P � 0.001 versus Marseille; ¶P � NS versus NW open and P � 0.01 versus NW lap.

TABLE 5. Resection Results

No. Patients Tumor Size (cm) Positive Margin Margin (cm)

Recurrence

All Local Field Systemic

CRC 31 3.6 0 1.1 11 2 5 4

NCR 27 4.6 1* 1.0 8 0 4 4

Cholangio 8 5.5 2*† 0.6 2 0 2 0

HCC 36 4.6 0 0.7

Non-Cirrhotic‡ 16 6.4 0 1.1 2 0 2 0

Cirrhotic 20 3.1 0 0.5 0 0 0 0

Other 4 5.6 0 0.6 1 0 0 1

24 2 13 9

*Diagnostic resection, †ductal involvement of remaining R anterior and posterior ducts.‡P � 0.001 for tumor size; P � 0.09 for margin in cirrhotic versus non-cirrhotic.



candidates. Conversion proceeded to assure adequate pretrans-plant assessment of candidacy. The fourth patient was a carci-noid patient with prior thoracoplasty and impaired laparoscopicaccess. The overall complication rate for the RFA group was11%, but varied from 0% for the colorectal cancer patients to13.5% for HCC patients (P � NS). Three perioperative deathswere encountered in the RFA group, all in cirrhotic pa-tients who developed post procedure hepatorenal failure.The overall operative mortality for laparoscopic RFAgroup was 1.5% and the mean length of stay was 2.4 days.

Cirrhosis was, in general, associated with worse outcomein terms of morbidity and mortality. There were no significantdifferences in morbidity (Fig. 1) or mortality (Fig. 2) comparingcirrhotic to noncirrhotic patients who underwent RFA. How-ever, among patients who underwent laparoscopic resection,there was a statistically significant increase in both morbidity(29% versus 14%; P � 0.02) and mortality (9.7% versus0.3%, P � 0.006, Fisher exact test) among cirrhotic versusnoncirrhotic patients. Concurrently, there was a greater mor-tality rate among cirrhotic patients undergoing resection com-pared with RFA (P � 0.07, Fisher exact test).

As demonstrated by Fig. 3, overall tumor recurrencewas higher in the radiofrequency ablated tumor patients whencompared with the cohort of laparoscopic resection patients.A total recurrence rate of 24% was observed in the RFAgroup, Whereaswhereas in the resected group, the total re-currence rate was 23%. (P � NS). Local recurrence was alsoobserved to be statistically higher in the RFA group (6%versus 1.5%; P � 0.06, Fisher exact test).

In a small sample financial analysis we examined theimpact of a laparoscopic approach on Current ProceduralTerminology (CPT) based professional fees (Table 7) andDiagnosis-Related Group (DRG) related cost/payment ratios(Table 8). In traditional open hepatic surgery several CPTcodes are used, including 47100, 47120, 47122, 47125, and47130; for laparoscopic hepatic resections, CPT code 47379with a technical modifier is used since individualized CPTcodes do not yet exist for these procedures. In the hospitalbased DRG system codes 191 and 192 are most often used,whereas, with the change in DRG codes in September 2007,our laparoscopic practice now uses the new DRG codes 405,406, and 407. This represents an effort by the DRG system to

TABLE 6. RFA Results

N Patients N Tumors Mean Tumor Size (cm) Conversion (%) Operative Mortality (%) Comps (%) LOS (days)

Benign 16 1.3 2.4 0 (0%) 0 (0%) 1 (6%) 2.4

CRC 9 1.5 1.5 0 (0%) 0 (0%) 0 (0%) 3.2

NCR 13 2 2.1 1 (7.7%) 0 (0%) 1 (8%) 2.5

HCC 155 1.3 3.2* 3 (1.9%) 3 (1.9%) 21 (13.5%) 2.4

Overall 201 1.3 3.2 1.9% 1.5% 11% 2.4

*P � 0.4 versus NCR; 0.06 versus CRC.

FIGURE 1. Complications in cirrhotics.



FIGURE 3. Tumor recurrence.

FIGURE 2. Mortality in cirrhotics.



better compensate for minor and major complications in-volved in the care of liver patients. Cost data from 34 openand 29 laparoscopic resections performed between 2004 and2006 were analyzed. The open group was composed of 35%lobar resections and the laparoscopic group 45%. Actual costdata confirmed significant cost savings for a laparoscopicapproach to lobar ($21,131 vs. $36,821; P � 0.01) andsegmental resections ($18,139 vs. $23,471; P � 0.02). Thedifference in operating room cost was significantly lowerfor laparoscopic lobar resections ($12,131 vs. $21,025 P �0.001) but not segmental resections ($13,179 vs. $15,336;P � 0.18). Intragroup differences in total cost between majorand minor resections was observed in open resections($36,821 vs. $23,471; P � 0.001) but not the laparoscopicapproach ($21,131 vs. $18,139; P � 0.5). Because of thechange in DRG coding we elected to examine the last 5 openresections (2 major and 3 minor) and the last 10 laparoscopic

resections (4 major and 6 minor), the DRG charges weresignificantly higher for open lobar resections compared withlaparoscopic ($82,797 vs. $54,625; P � 0.0005). Actual costdata demonstrated a smaller nonsignificant cost savings forlobar resection ($25,457 vs. $23,691; P � 0.2).

DISCUSSIONSince the adoption of laparoscopic cholecystectomy in

the early 1990s, there has been a gradual and steady evolutionof the application of minimally invasive surgical techniquesto more complex surgical procedures. This revolution hasbeen driven directly by 2 principles: 1) financial incentivesfrom shortened hospital stays and 2) demand by educatedconsumers of health care for less invasive procedures thatallow quicker return to normal function.

The initial skepticism of minimally invasive approacheswas proven prudent by the early era of laparoscopic cholecys-tectomy. During this era, the incidence of bile duct injuriesdoubled with the laparoscopic approach.6 The rate of thiscomplication eventually decreased with routine intraoper-ative cholangiograms, improved training, and credential-ing procedures.34,35 Skepticism eventually turned into en-thusiasm, as referral rates increased in favor for the lessinvasive laparoscopic approach to the traditional openNissen fundoplication, splenectomy, and adrenalectomy.Despite the shift in surgeons’ perspectives regarding lapa-roscopic surgery, a significant and often vocal resistancesurfaced against the application of laparoscopic techniquesin the resection of colon cancer. Continued advances insurgical technology and increased surgeon experience hastranslated into successful application of laparoscopic ap-proaches for resection of colon cancer, now confirmed inrandomized clinical trials.36 –39 Subsequently, many cen-

FIGURE 4. Treatment algorithm.

TABLE 7. CPT Charges

Open CollectionRate %

LaparoscopicCollection Rate %

MIS/Open ChargeRatio

Wedge 16% 18% 1.12

Lobe 26% 34% 1.31

TABLE 8. DRG Cost

Wedge Lobe

Lap/open cost 0.94 0.81

Lap/open payment 1.07 1.0

Cost-to-payment ratio 1.14 1.23

Lap-to-open LOS ratio 0.28



ters have accepted and now advocate the laparoscopicapproach to colectomy. Unfortunately, a similar resistanceto laparoscopic liver resection has arisen, particularly withrespect to malignancy. Our experience with minimallyinvasive liver surgery and all its therapeutic modalities hasled us to propose an algorithm for treatment of both benignand malignant tumors of the liver. Fig. 4 outlines ourcurrent treatment algorithm.

Several distinct needs and events have led to the de-velopment of a minimally invasive liver program at ourcenter. The first need was to accurately stage cirrhotic pa-tients being evaluated for liver transplantation. Laparoscopicexploration can identify advanced peritoneal or multifocaldisease that is underappreciated by standard preoperativeimaging. The impetus for this approach arose in the setting ofa busy liver transplant, hepatobiliary and laparoscopic donornephrectomy program. We found that appropriate stagingeliminated unnecessary transplant evaluations, minimizingexpense and often improving cold ischemic times for liverallografts. Laparoscopy eliminated the majority of aborted livertransplants for advanced cancer, minimizing the need to proceedto a secondary recipient. The reliability of this approach isconfirmed by a number of terminated transplant evaluations withlow incidence of complications and short hospital stays.40

This initial comfort with intraoperative ultrasonogra-phy and percutaneous tumor biopsy evolved into laparoscopicRFA. We initially used RFA for the management of unre-sectable primary and metastatic liver cancer. Eventually weadopted pretransplant ablation, to prevent tumor growthwhereas the patient awaited a suitable organ for transplanta-tion. Several techniques have been described for bridgingcirrhotic HCC cancer patients to transplantation, includingchemoembolization, RFA and most recently yttrium-90 mi-crospheres.41–43 The concept of pretransplantation ablationfor stabilization and down-staging has been a long acceptedtherapeutic modality. This concept was even employed in theoriginal article by Mazzefaro and colleagues defining theMilan Criteria for liver transplantation.44 In reviewing ex-plantation specimens, it is clear that there is a circumferentialhemmorhagic zone surrounding the treated tumor. This zoneis felt to prohibit tumor seeding during the transplant hepa-tectomy. Initial reports from the Northwestern group ob-served almost universal necrosis of explanted tumors afterpretransplant yttrium-90 therapy.45 Unfortunately, subse-quent reports noted only about 80% tumor necrosis, which isconsistent with other forms of ablation therapy.41–43 In ourexperience, RFA has been most useful in bridging hepatocel-lular cancer patients to transplantation. Fig. 5 represents aKaplan-Meier anlaysis of HCC patients, comparing RFAalone versus RFA as a bridge to transplantation. A significantsurvival benefit was observed in using RFA as a bridge fortransplantation (P � 0.001, log-rank test). Three year sur-vival rates were also different between the 2 groups (RFAalone: 0.508, 95% CI (0.401, 0.643); RFA with OLT: 0.96,95% CI (0.905, 1.0)). This effect remained significant afteradjustment for gender, age, cirrhosis, and size of tumor in amultivariable Cox model (P � 0.0004, Wald test). Theestimated hazard ratio from the multivariable Cox model for

HCC patients receiving RFA alone versus those who receivedRFA plus OLT was 13.4, with a 95% CI of (3.2, 56.6).Unfortunately, the majority of the RFA alone patients werenot candidates for transplantation due to continued substanceabuse, medical comorbidities, or inadequate social support.

RFA as a therapeutic modality has had mixed results interms of local recurrence rates, ranging from 6% reported byPoon in cirrhotic HCC patients, to a 40% recurrence ratereported by Harrison et al in pretransplant cirrhotic HCCpatients.13–14 Early data from Curley et al reported lowrecurrence rates and excellent short term survival.11 Unfor-tunately, long-term results from MD Anderson now show ahigher local recurrence rate of 25% in RFA-treated colorectalcancer liver metastases.46,47 Several authors have demon-strated percutaneous RFA is associated with a greater localrecurrence rate when compared with operative RFA pa-tients.14,46–48 The higher recurrence rates in percutaneouslytreated patients may be the result of inadequate tumor visu-alization or inability to assess multifocal disease. An “oveneffect” has been postulated to explain lower local recurrencerates in cirrhotic HCC patients. Thermal ablation of tumor ina surrounding fibrous or cirrhotic liver serves to insulate thetumor ablation, augmenting treatment temperatures, and po-tentially therapeutic efficacy. Our current series is weightedtoward cirrhotic HCC; consequently, we believe our serieshas a low local recurrence similar to that reported by Poon.13

Despite this similarity, the log-rank test for differences inoverall survival between cirrhotic and noncirrhotic patientsundergoing RFA alone was nonsignificant (see Fig. 6). The3-year survival rate for cirrhotic patients was 0.692, with a95% CI of 0.611, 0.784, whereas for noncirrhotic patients the3-year survival rate was 0.724, with a 95% CI of 0.610,0.858. This leads one to speculate that other technical factorsmay contribute to lower recurrence rates as well.

FIGURE 5. Kaplan-Meier curves for HCC patients who re-ceived RFA alone versus RFA as a bridge to transplantation.



Clinical analysis of RFA, from our group and other largegroups, has led to the adoption of our current treatment algo-rithm outlined in Fig. 4. In a large series by Poon and Siperstein,higher local recurrence rates were associated with treatment oftumors larger than 5 cm and tumors in proximity to largevascular structures.12,47,49 Higher recurrence rates are postulatedto arise from inadequate tumor ablation in large lesions orinadequate treatment temperatures produced by the “heat sink”effect of adjacent large blood vessels.

Laparoscopic resection was the final procedure adoptedby our group after initial experience with laparoscopy for otherless technically demanding procedures. Our experience withlaparoscopic resection began with small peripheral tumors, andprogressed to major lobar resections. Based on our experiencewith hand-assisted donor nephrectomy, liberal use of a hand-assist device made transition from small peripheral tumors tomajor lobectomy safe and rapid. Initially, our group was con-servative and approached �10% of liver resections with alaparoscopic approach.50 Our current practice is to approach70% of liver tumors initially through a laparoscopic approach.The most controversial and challenging resections in this latergroup have been the caudate, central, and cirrhotic resection.The caudate resection is technically challenging due to thenature of access to this lobe. The central resection evolved fromthe performance of pedicle transection during major lobectomy.Hand-assist mobilization of the hilar plate allowed safe preco-agulation and resection of segment IV. One must be cautioned,that there is a definite learning curve to be overcome during thetransition from open to minimally invasive hepatic resection; wepropose that surgeons experienced in open liver resection cansafely adopt laparoscopic liver resection by such a graduatedprogram of increasingly complex procedures.

Unlike some European groups, American surgeonshave been more hesitant to approach cirrhotic tumor resec-tions.18,51,52 Our large experience with cirrhotic RFA and

major laparoscopic resection has increased our comfort andwillingness to resect cirrhotic HCC. A novel indication forcirrhotic resection has been to assess potential transplantcandidates with questionable vascular invasion. Peripheralcirrhotic tumors can be resected and pathologic examinationof resected tumors can clarify the presence or absence ofmajor vascular invasion. In the absence of major vascularinvasion post resection patients can be safely referred fortransplantation. This approach is delineated in our currenttreatment algorithm (Fig. 4). Nonetheless, the increased mor-bidity and mortality rates associated with laparoscopic he-

FIGURE 6. Kaplan-Meier curves for cirrhotic and noncirrhoticpatients who underwent RFA therapy.

FIGURE 7. Kaplan-Meier curves for patients receiving resec-tion, cirrhotic versus noncirrhotic.

FIGURE 8. Kaplan-Meier curves for cancer patients receivingRFA versus resection treatment.



patic resection in cirrhotic patients as demonstrated in thiscurrent study indicates that such procedures should be ap-proached with caution. The Kaplan-Meier survival analysisdemonstrates a significantly higher survival for noncirrhoticpatients undergoing curative resection. (Fig. 7). The laparo-scopic approach does not necessarily translate into increasedsafety for patients with decompensated liver function.

Our experience with laparoscopic liver resection con-firmed prior reports that there are several distinct advantagesto a minimally invasive approach, including shorter operativetimes, lower blood loss, low transfusion rates, and shorterhospital stay. Resection for benign disease appears to be safe,with almost no mortality and minimal morbidity incurredwith all observed deaths in cancer patients and/or cirrhoticcancer patients. However, only patients with truly symptom-atic benign tumors should be considered for resection, and thesubstantial potential risks of major liver resection must beunderstood by the patient in the context of the natural historyof the particular benign liver tumor. A major liver resection,even if performed using minimally invasive techniques, stillcarries substantial potential risks that must be clearly ex-plained. Even with the success and relative safety of laparo-scopic resection for benign disease demonstrated in thisstudy, we do not believe that indications for resection shouldbe liberalized because a minimally invasive approach isavailable. In our clinical algorithm, symptomatic lesions arecandidates for laparoscopic resection or RFA for unresectablepatients. However, laparoscopic resection or RFA for grow-ing, symptomatic benign tumors and large adenomas shouldbe considered a reasonable treatment option, especially inlight of the possibilities of spontaneous rupture and a smallrisk of malignant degeneration of large adenomas.

In the setting of laparoscopic resection for malignancy,our data indicate that aside from safety and efficacy, theintermediate long-term survival data are encouraging. Amongcancer patients, we observed that laparoscopic resection hadsignificantly improved overall survival compared with RFAas a primary treatment modality (Fig. 8, P � 0.01, log-ranktest). Significant differences also existed in 3-year survivalrates between the 2 groups (RFA: 0.672, 95% CI (0.600,0.753); resection: 0.840, 95% CI (0.747, 0.945)). Thesegroups were similar in terms of age (58.9 versus 54; P � NS),and number of lesions (1.3 versus 1.2; P � NS), but signif-icant differences exist in the RFA group in terms of a malepredominance (67% versus 36%; P � 0.001), cirrhotic pre-dominance (72% versus 12%; P � 0.001) and smaller lesions(3.1 versus 5.3 cm; P � 0.001). However, the effect of RFAversus resection still remained marginally significant afteradjustment for all of the above covariates (P � 0.056, Waldtest). The estimated hazard ratio of RFA versus resectionpatients from the multivariable model was 2.4, with a 95% CIof (0.98, 5.87). Despite a higher incidence of cirrhosis in theRFA patients both the cirrhotic resection and RFA patientshad a mean MELD of 10 and a median MELD of 9. Thesedata, along with the observation in many studies of a greaterlocal recurrence rate for RFA compared with resection53,54

would indicate that surgical resection should be advocatedover RFA whenever possible. Like many other centers that

were early and enthusiastic adopters of RFA because it sparesnormal liver parenchyma, can be performed safely, and wasinitially reported to be associated with low rates of localrecurrence, we have adopted a more aggressive approach toresection, especially when it can be done with the advantagesof a minimally invasive approach.

CONCLUSIONSWe have demonstrated that minimally invasive ap-

proaches to the treatment of liver tumors can be performedsafely and effectively. However, laparoscopic hepatic surgery isa highly technically demanding procedure that requires signifi-cant cross-training and/or partnership in the specialties of trans-plantation, laparoscopy, and surgical oncology, and is bestadopted by a program of graduated procedural complexity.55

REFERENCES1. Flowers JL, Lefor AT, Steers J, et al. Laparoscopic splenectomy in

patients with hematologic diseases. Ann Surg. 1996;224:19–28.2. Gill IS, Zippe CD. Laparoscopic radical prostatectomy: technique. Urol

Clin North Am. 2001;28:423–436.3. Jamieson GG, Watson DI, Britten-Jones R, et al. Laparoscopic Nissen

fundoplication. Ann Surg. 1994;220:137–145.4. Peters JH, Ellison EC, Innes JT, et al. Safety and efficacy of laparoscopic

cholecystectomy. A prospective analysis of 100 initial patients. AnnSurg. 1991;213z 3–12.

5. Zucker KA, Bailey RW, Gadacz TR, et al. Laparoscopic guided chole-cystectomy. Am J Surg. 1991;161:36–42; discussion 42–44.

6. Fletcher DR, Hobbs MS, Tan P, et al. Complications of cholecystectomy:risks of the laparoscopic approach and protective effects of operativecholangiography: a population-based study. Ann Surg. 1999;229:449–457.

7. Gagner M, Rogula T, Selzer D. Laparoscopic liver resection: benefitsand controversies. Surg Clin North Am. 2004;84:451–462.

8. Lefor AT, Flowers JL. Laparoscopic wedge biopsy of the liver. J AmColl Surg. 1994;178:307–308.

9. Pearson AS, Izzo F, Fleming RYD, et al. Intraoperative radiofrequencyablation or cryoablation for hepatic malignancies. Am J Surg. 1999;178:592–599.

10. Steele G. Jr. Cryoablation in hepatic surgery. Semin Liver Dis. 1994;14:120–5.

11. Curley SA, Izzo F, Ellis LM, et al. Radiofrequency ablation of hepatocel-lular cancer in 110 patients with cirrhosis. Ann Surg. 2000;232:381–391.

12. Berber E, Rogers S, Siperstein A. Predictors of survival after laparo-scopic radiofrequency thermal ablation of hepatocellular cancer: a pro-spective study. Surg Endosc. 2005;19:710–714.

13. Poon RT. Current role of laparoscopic surgery for liver malignancies.Surg Technol Int. 2007;16:73–81.

14. Harrison LE, Koneru B, Baramipour P, et al. Locoregional recurrencesare frequent after radiofrequency ablation for hepatocellular carcinoma.J Am Coll Surg. 2003;197:759–764.

15. Jiang HC, Liu LX, Piao DX, et al. Clinical short-term results of radiofre-quency ablation in liver cancers. World J Gastroenterol. 2003;9:192.

16. Haberstroh J, Ahrens M, Munzar T, et al. Effects of the Pringlemaneuver on hemodynamics during laparoscopic liver resection in thepig. Eur Surg Res. 1996;28:8–13.

17. Kurian MS, Gagner M, Murakami Y, et al. Hand-assisted laparoscopicdonor hepatectomy for living related transplantation in the porcinemodel. Surg Laparosc Endosc Percutan Tech. 2002;12:232–237.

18. Cherqui D, Laurent A, Tayar C, et al. Laparoscopic liver resections: afeasibility study in 30 patients. Ann Surg. 2000;232:753–762.

19. DeMatteo RP, Fong Y, Jarnagin WR, et al. Recent advances in hepaticresection. Semin Surg Oncol. 2000;19:200–207.

20. Cuschieri A. Laparoscopic hand-assisted hepatic surgery. SeminLaparosc Surg. 2001;8:104 –113.

21. Chen HY, Ker CG, Juan CC, et al. Laparoscopic subsegmentectomy forhepatocellular carcinoma with cirrhosis: a case report. Kaohsiung J MedSci. 2000;16:582–586.



22. Gigot JF, Hubert C, Banice R, et al. Laparoscopic management ofbenign liver diseases: where are we? HPB (Oxford). 2004;6:197–212.

23. O’Rourke N, Fielding G. Laparoscopic right hepatectomy: surgicaltechnique. J Gastrointest Surg. 2004;8:213–216.

24. Dagher I, Proske JM, Carloni A, et al. Laparoscopic liver resection:results for 70 patients. Surg Endosc. 2007;21:619–624.

25. Cherqui D. Laparoscopic liver resection. Br J Surg. 2003;90:644–646.26. Ardito F, Tayar C, Laurent A, et al. Laparoscopic liver resection for

benign disease. Arch Surg. 2007;142:1188–1193; discussion 1193.27. Descottes B, Glineur D, Lachachi F, et al. Laparoscopic liver resection

of benign liver tumors. Surg Endosc. 2003;17:23–30.28. Gigot JF, Glineur D, Santiago Azagra J, et al. Laparoscopic liver

resection for malignant liver tumors: preliminary results of a multicenterEuropean study. Ann Sur. 2002;236:90–97.

29. O’Rourke N, Shaw I, Nathanson L, et al. Laparoscopic resection ofhepatic colorectal metastases. HPB (Oxford). 2004;6:230–235.

30. Cherqui D, Soubrane O, Husson E, et al. Laparoscopic living donorhepatectomy for liver transplantation in children. Lancet. 2002;359:392–396.

31. Koffron AJ, Kung R, Baker T, et al. Laparoscopic-assisted right lobedonor hepatectomy. Am J Transplant. 2006;6:2522–2525.

32. Koffron AJ, Auffenberg G, Kung R, et al. Evaluation of 300 minimallyinvasive liver resections at a single institution. Ann Surg. 2007;246:385–394.

33. Buell JF, Koffron AJ, Thomas MJ, et al. Laparoscopic liver resection.J Am Coll Surg. 2005;200:472–480.

34. Hunter JG. Avoidance of bile duct injury during laparoscopic cholecys-tectomy. Am J Surg. 1991;162:71–76.

35. Connor S, Garden OJ. Bile duct injury in the era of laparoscopiccholecystectomy. Br J Surg. 2006;93:158–168.

36. Falk PM, Beart RW Jr, Wexner SD, et al. Laparoscopic colectomy: acritical appraisal. Dis Colon Rectum. 1993;36:28–34.

37. Musser DJ, Boorse RC, Madera F, et al. Laparoscopic colectomy: atwhat cost? Surg Laparosc Endosc. 1994;4:1–5.

38. Jayne DG, Guillou PJ, Thorpe H, et al. Randomized trial of laparo-scopic-assisted resection of colorectal carcinoma: 3-year results of theUK MRC CLASICC Trial Group. J Clin Oncol. 2007;25:3061–3068.

39. Bonjer HJ, Hop WC, Nelson H, et al. Laparoscopically assisted vsopen colectomy for colon cancer: a meta-analysis. Arch Surg. 2007;142:298 –303.

40. Kim RD, Nazarey P, Katz E, et al. Laparoscopic staging and tumorablation for hepatocellular carcinoma in Child C cirrhotics evaluated fororthotopic liver transplantation. Surg Endosc. 2004;18:39–44.

41. Maluf DG, Stravitz RT, Williams B, et al. Multimodality therapy and livertransplantation in patients with cirrhosis and hepatocellular carcinoma: 6years, single-center experience. Transplant Proc. 2007;39:153–159.

42. Lu DS, Yu NC, Raman SS, et al. Percutaneous radiofrequencyablation of hepatocellular carcinoma as a bridge to liver transplan-tation. Hepatology. 2005;41:1130 –1137.

43. Kulik LM, Carr BI, Mulcahy MF, et al. Safety and efficacy of 90Yradiotherapy for hepatocellular carcinoma with and without portal veinthrombosis. Hepatology. 2008;47:71–81.

44. Mazzaferro V, Battiston C, Perrone S, et al. Liver transplantation for thetreatment of small hepatocellular carcinomas in patients with cirrhosis.N Engl J Med. 1996;334:693–699.

45. Sato K, Lewandowski RJ, Bui JT, et al. Treatment of unresectableprimary and metastatic liver cancer with yttrium-90 microspheres(TheraSphere): assessment of hepatic arterial embolization. CardiovascIntervent Radiol. 2006;29:522–529.

46. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomesfollowing hepatic resection, radiofrequency ablation, and combinedresection/ablation for colorectal liver metastases. Ann Surg. 2004;239:818–825; discussion 825–827.

47. Aloia TA, Vauthey JN, Loyer EM, et al. Solitary colorectal liver metastasis:resection determines outcome. Arch Surg. 2006;141:460–466.

48. Ong ES, Alassas AM, Litwan A, et al. Percutaneous versus open/laparoscopic radiofrequency ablation: Is there a difference. HPB Surg.2008;10:77.

49. Ng KK, Poon RT, Lo CM, et al. Analysis of recurrence pattern and itsinfluence on survival outcome after radiofrequency ablation of hepato-cellular carcinoma. J Gastrointest Surg. 2008;12:183–191.

50. Buell JF, Thomas MJ, Doty TC, et al. An initial experience andevolution of laparoscopic hepatic resectional surgery. Surgery, 2004;136:804–811.

51. Montorsi M, Santambrogio R, Bianchi P, et al. Survival and recurrencesafter hepatic resection or radiofrequency for hepatocellular carcinoma incirrhotic patients: a multivariate analysis. J Gastrointest Surg. 2005;9:62–67; discussion 67–68.

52. Dagher I, Lainas P, Carloni A, et al. Laparoscopic liver resection forhepatocellular carcinoma. Surg Endosc. 2008;22:372–378.

53. White RR, Avital I, Sofocleous CT, et al. Rates and patterns ofrecurrence for percutaneous radiofrequency ablation and open wedgeresection for solitary colorectal liver metastasis. J Gastrointest Surg.2007;11:256–263.

54. Guglielmi A, Ruzzenente A, Valdegamberi A, et al. Radiofrequencyablation versus surgical resection for the treatment of hepatocellularcarcinoma in cirrhosis. J Gastrointest Surg. 2008;12:192–198.

55. Koffron AJ, Kung RD, Auffenberg GB, et al. Laparoscopic liver surgeryfor everyone: the hybrid method. Surgery. 2007;142:463–468.

DiscussionsDR. RONALD W. BUSUTTIL (LOS ANGELES, CALIFORNIA): This

is one of the largest series using a laparoscopic approach tohepatic lesions, and I believe the technique is gaining tractionand results in a relatively low rate of complications and mortalitywith a conversion rate to the classic approach of about 2%.

What this paper does is to provide a detailed descriptionof the institutionally employed algorithm for a laparoscopicapproach, which should be of benefit to those who wish toembark on this methodology. However, there are some cave-ats that I think must be addressed.

First, only half of the patients in this series underwentactual laparoscopic resection, whereas the remainder receiveddiagnostic ultrasound and radiofrequency ablation. Sixty per-cent of the resections were for benign or cystic lesions,whereas the other 40% were for malignancy. Of those withcancer, only 25% had a major (greater than 3 segmental)resection. What was the specific complication in this high-risk group? How do these results compare with an openapproach when you conduct a case-matched analysis for stageof tumor and degree of hepatic impairment? I think only withthis analysis would we be able to make a valid comparisonbetween the laparoscopic approach and the open approach.

Secondly, you operated on many patients with benignconditions, as I already alluded to. Has the laparoscopicapproach changed your indication for resection of benignlesions? Generally speaking, unless a patient is symptomaticor has the possibility of harboring malignancy or parasiticdisease, we do not operate.

Third, you posed an argument for a cost benefit fromthe laparoscopic approach in terms of less pain medication,reduced complications, et cetera. You also demonstratedimproved reimbursement, which is of obvious interest to usall. How has this been achieved? Were both groups compa-rable, or was this solely a result of modifier codes, and do youthink that the higher reimbursement issue was a geographicphenomenon?



Fourth, you allude to the concept that the laparoscopicapproach may result in superior oncologic outcomes. I wouldlike to see some data to support that.

Finally, you report a higher complication and mortalityrate among the cirrhotic patients undergoing these proce-dures, which is obviously intuitive and we have all experi-enced this. However, there is no discussion of the stratifica-tion of these patients, for example with the Barcelona stagingcriteria, to better evaluate these patients.

DR. CHRISTOPH BROELSCH (DUESSELDORF, GERMANY):When we talk about central resections, I do not think we aretalking about the same procedures as we do result in surgery,because there is a meticulous dissection with the reconnection ofbile ducts. So I think what you allude to is a central resection.

The second question is, what drives your decision tochoose a laparoscopic versus open approach? Is this theDRG? Or when you discuss the case with your oncologist isyour mind set on utilizing a laparoscopic approach until youhave to convert? Or do you consider a decision, this is one wedo open or is this is a case where I will approach laparoscopi-cally because I talked to the patient and the patient wants it.What is your decision algorithm?

DR. JEAN C. EMOND (NEW YORK, NEW YORK): In theinterest of truly minimizing patient invasion, I am curiouswhy you subject so many patients to laparoscopic RFA orbiopsy. In our practice almost all of that work is donepercutaneously and the patients can go home the same day.Are these selected patients? Are they particularly compli-cated, or is this just a tactical choice you make to get the workinto surgical hands as opposed to the radiologist’s?

DR. HENRI BISMUTH (VILLEJUIF, FRANCE): My question ison benign tumor, and mainly on indication, for you said thatthe mean size of the tumor was around 2 cm. Usually for themost frequent type of benign tumor of liver, that is FocalNodular Hyperplasia or angioma, this kind of small tumor isasymptomatic. Why do you need resection for these patients?

DR. JOSEPH F. BUELL (LOUISVILLE, KENTUCKY): First, Iwould like to discuss the point raised by Dr. Emond. Recentlypublished data demonstrated that percutaneous RFA has ahigher local recurrence rate. This has been our concern and wehad preliminary data that suggested this. Hence, a laparoscopicforay is probably beneficial to rule out patients with multiplelesions not seen by CT. This avoids further evaluation andprovides a cost savings for patients going on to transplantation.The majority of the patients that underwent laparoscopic RFArather than resection were high risk cirrhotic patients withadvanced MELDs and/or advanced Child’s class.

Regarding Dr. Broelsch’s statements, I completelyagree that our central resection is probably much differentthan your central resection. These are areas of inferior seg-

ment 4 and 5 resections with potential involvement of seg-ment 8. They divide the hilar plate.

Several branches of the hilar plate are usually clippedand divided to lift and elevate, but it is a dissimilar operationas you describe.

The decision to proceed with a laparoscopic versusopen approach is taken very seriously with the referringdoctors and the patients. We must ensure that they are wellsuited to undergo a laparoscopic procedure.

If there is any inconclusiveness, they would be con-verted. We do not approach any patient that has potentialvascular invasion, requiring a portal vein reconstruction, orcaval resection, with a laparoscopic approach. Any patientwith more than what we perceive to be 4 lesions, we wouldapproach with an open surgical procedure.

Professor Bismuth, we have seen smaller lesions thatwere clinically significant. We proceed with surgical resec-tion of benign lesions for those patients that present withsymptoms and symptoms only. Several patients within thisclassification were initially thought to have malignant lesionsand were found later to have benign lesions, so that doesadmix the data and decrease the size of the lesions evaluated.

Professor Busuttil, we only proceed with those patientsthat were symptomatic, and we are not changing the criteriafor benign resection.

As far as complications, these fall into 2 major classes:Medical complications and surgical complications. Approxi-mately 30% of our patients have medical complicationsbecause of their advanced age, whether it is Olgilve’s syn-drome or other odd complications that add to our complica-tion rates. Those surgical complications that we encounter areodd items such as pulmonary embolism, a bile duct bilestenosis or bile collection of approximately 4%. These are theareas where we see complications sporadically among alltypes of resection whether they be segmental or lobar.

When we looked at the cost data, we looked specificallyat lobar resections and segmental resections. It appears thatthe cost of the actual operation and devices used are probablymore expensive in the segmental resections, but once wepurchase the devices, performing the major lobar resectionsbecomes cost saving.

You suggested a stage-to-stage comparison, and I be-lieve that is absolutely warranted. A prospective randomizedstudy in a multi institutional fashion proceeding with a properoncological outcome is what should be done. We are utilizingthis now, in fact, in a phase 1/phase 2 type of series, butagain, prospective randomized data must be reported. This isalso salient with cirrhotic patients.

We must determine the MELD criteria. As a group weare proceeding to evaluate patients with genomic and pro-teomic fingerprinting, and I believe this will probably be themost beneficial reason and methodology for these patients.



Experience With More Than 500 Minimally Invasive Hepatic Procedures: A Serious Note Of Caution

Documents

Transcript of Experience With More Than 500 Minimally Invasive Hepatic Procedures: A Serious Note Of Caution