Combination of repeated single-session percutaneous ethanol injection and transarterial chemoembolis

PO Box 2345, Beijing 100023, ChinaWorld Gastroenterol 2006 June 21: 12(23): 3707-3715orld Joumal of Gastroenterology ISSN 1007-9327Tjgwwignet.com@2006 The wJG Press. All rights reserved.CⅣ RCACERCombination of repeated single-session percutaneous ethanolinjection and transarterial chemoembolisation compared torepeated single-session percutaneous ethanol injection itpatients with non-resectable hepatocellular carcinomaArne Dettmer, Timm D Kirchhoff, Michael Gebel, Lars Zender, Nisar P Malek, Bernhard Panning, Ajay Chavan,Herbert Rosenthal, Stefan Kubicka, Susanne Krusche, Sonja Merkesdal, Michael Galanski, Michael P MannJoerg s BleckArne Dettmer, Michael Gebel, Lars Zender, Nisar P Malek, RESULTS: The 1-and 3-year survival of all patientsStefan Kubicka, Susanne Krusche, Michael P Manns, Joerg was 73% and 47%. In the subgroup analyses, theS Bleck, Department of Gastroenterology, Hepatology and combination of TACE and PEI(1)was associated withEndocrinology, Hannover Medical School, GermarTimm D Kirchhoff, Herbert Rosenthal, Michael Galanski, and 43%)compared to PEI treatment alone(2)(1-,3,ermanyBernhard Panning Department of Anaesthesiology, hannoveafter initial stratification to TACE(3) yielded comparableMedical School, Germanyresults(1-,3-5-year survival: 91%, 40%, and 30%)Ajay Chavan, Department of Diagnostic and Interventional while PEI after stratification to best supportive care(4)Radiology, Klinikum Oldenburg, Germanywas associated with decreased survival (1-,3-5-yearSonja Merkesdal, Division of Rheumatology, Hannover Medical survival: 50%, 23%, 12%). Apart from the chosenSchool, Germanytreatment modalities, predictors for better survival wereCo-first-authors: Timm d Kirchhofftumor number(n< 5),tumor size(< 5 cm), no ascitesCorrespondence to: Timm D Kirchhoff, MD, Hannover Medi- before PEI, and stable serum cholinesterase after PEI (Pal School, Department of Diagnostic Radiology OE 8220, Car<0.05). The mortality within 2 wk after PEI was 2.8%euberg-Str. 1, D-30625 HannoverGermany. kirchhoff. timm @mh-hannover de(n 3). There were 24(8.9%)major complicationsTelephone:+49511-32342lFax:+49-511-5323885after PEI including segmental liver infarction, focal liverReceived:2005-11-29Accepted: 2006-01-09necrosis, and liver abscess. All complications could bemanaged non-surgicallyCONCLUSION: Repeated single-session PEI is effectiveAbstractin patients with advanced HCc at an acceptable andmanageable complication rate. Patients stratified to aAIM: To evaluate the treatment effect of percutaneous combination of TACE and PEI can expect longer survivalethanol injection(PEI) for patients with advanced, than those stratified to repeated PEI alone. Furthermorenon-resectable HCC compared with combination of patients with large or multiple tumors in good clinicaltransarterial chemoembolisation(TACE)and repeated status may also profit from a combination of TAce andsingle-session PEI, repeated single-session PEI alone, reconsideration for secondary PEIrepeated TACE alone, or best supportive carec 2006 The WJG Press. All rights reservedMETHODS: All patients who received PEI treatmentduring the study period were included and stratified to Key words: HCC; Single-session PEI; TACE; Survivalone of the following treatment modalities according to Predictionphysical status and tumor extent: combination of TACEand repeated single-session PEI, repeated single-session Dettmer A, Kirchhoff TD, Gebel M, Zender L, Malek NPPEI alone, repeated TACE alone, or best supportive Panning B, Chavan A, Rosenthal H, Kubicka S, Krusche $carePrognosticvalueofclinicalparametersincludingMerkesdalS,GalanskiM,MannsMp,BleckJs.combinAtionOkuda-classification, presence of portal vein thrombosis, of repeated single-session percutaneous ethanol injectionpresence of ascites, number of tumors, maximum tumor and trardiameter, and serum cholinesterase(CHE), as well as single-sl中国煤化工 -njection in patientsChild-Pugh stage, a-fetoprotein(AFP), fever, incidence of with nCNMH GarcinomaWorldcomplications were assessed and compared between the Gastroergroups. Survival was determined using Kaplan-Meier andmultivariate regression analyses.http://www.wjgnet.com/1007-9327/12/3707.asp3708 ISSN 1007-9327 CN 14-1219/R World J Gastroenterol June 21, 2006 Volume 12 Number 23INTRODUCTIONLimitedThe prognosis of untreated HCC remains poor. The3-year survival ranges from 21% to 28% and declinesto about 8% for advanced tumors". The only therapiesTACE-PEIthat currently offer potential cure are surgical resectionand liver transplantation. Despite improved surgicalstatusTACE(3)(date not shown)techniques, only about 20% of HCC are amenable toresection, mainly due to advanced tumor growth andReducedthe underlying cirrhosis 2. Systemic chemotherapy hasPEI alonedinicalbest supportive fi supportive careficant survival benefit in comparisonstatuscare(4best supportive care. Hence, locoregional modalities liketransarterial chemoembolization ( TACE, percutaneousure 1 Stratification algorithm for the treatment of advanced unresectable HCCethanol injection(PED), laser induced thermo therapyITT), and radiofrequency ablation(RFA) are employedidely in non-resectable HCC in an attempt to prolong treatment from July 1999 until July 2003 in our institutionrred mocwith large and disseminated HCC in good physical status, were included into the present study.whereas repeated PEI has shown efficacy in cases of smalltumors Both modalities have proven their effectiveneStratificationin terms of tumor response and survival, even moreIn an interdisciplinary gastroenterological-radiologicalconference, patients were stratified to one of the followingn combinetreatment modalities according to physical status andPEI has gained importance in small HCC due to its tumor extent( Figure 1): combination of TACE andsafety and ease of operation. The method is both effectiveand economically sound. The survival data of patients alone, repeated TACE alone, or best supportive care. Afterafter surgical resection with 5-year survival rates between thorough explanation of the procedures, informed consent32%-59%5,18221. The amount of ethanol applicable dtwas obtained from each patient prior to therapy. The dataone session was initially limited. With the introductionof the TACE- PEI combination group(1)and of theof the single-session high-dose technique under generalPEI alone(2) group were analyzed. In addition, patientanaesthesia by Livraghi and co-workers, the indicationswith extensive disease(large or multiple tumors)werefor PEI have been extended. The injection of higher reconsidered and in some cases restratified to secondaryquantities of ethanol during one session under mechanicalPEI treatment(Figure 1). These patients were evaluated intwo separate subgroups: patients switched secondarilyventilation offers the advantage of treating larger tumors PEI after being initially stratified to repeated TACE (3)and more lesions at a time, requiring fewer interventions. after being initially stratified to best supportive care( orFurthermore, tumors in difficult accessible areas(eliver segments, 1 or 2)can be better accessed in deep Exclusion criteriainspiration under mechanical ventilationAccording to physical status and tumor extentContraindications against regional treatment werepatients with non-resectable HCC were stratified in an extrahepatic tumor manifestation, sepsis, hepaticinterdisciplinary conference to repeated single-session PEL, encephalopathia, elevated serum ammonia >47 umol/L,repeated TACE sessions, a combination of TACE and platelet counts <50000/uL, prothrombin activity <50%repeated single-session PEl or to best supportive careserum creatinine >120 umol/L, left ventricular ejectionAim of this prospective study was to evaluate all patientsfraction 50%, heart attack within the previous 12 mo,of our institution who received PEI treatment from 1999 heart insufficiency >NYHA stage II. The inclusion into theto 2003. Using multivariate regression analyses of preliver transplantation program was not a contraindicationtreatment variables, survival predictors were identified. InFurther contraindications against PEI were significantsubgroup analyses, patients stratified to repeated PEI alone perihepatic ascites or inadequate parameters for generalwere compared to those stratified to a combination of anaesthesia.TACE and PEI. Furthermore, long-term results of thoseFurther contraindications against TACE were portalpatients who were reevaluated and switched secondarily vein thrombosis, impossible catheterization of theto PEI treatment after being initially stratified to repeated hepatic artery, cholinesterase(CHE)<2 mU/mL(21C),TACE and of those patients who received PEI after being inadequate functional, hematological or biochemicalinitially stratified to best supportive care forr advancefor cherdisease were analysed)OutoPtYHa中国煤化工ion included physicalMATERIALS AND METHODSC MH Gf clinical parametersPatientstoxicitIes,events and survival and the evaluation ofOne hundred and one patients(age range 35-88 years, tumor markers as well as hematological and biochemical85% male, mean observation time 26+ 17 mo)with non- parameters. To classify cardiac function and to detectresectable histologically-confirmed HCC who received PEI right heart hypertrophism and pulmonary hypertensionww海敝Dettmer A et al Combination therapy of non-resectable HCC3709echocaralography, cardiac scintigraphy, and ECG were viation or frequencies [n( %)Overall survival of all patients and survival of thesubgroups were determined by Kaplan-Meier analyPE/ treatmentKaplan-Meier curves were statistically compared by theSingle-session PEI therapy was performed during log-rank-test. In order to quantify the influence of thegeneral anaesthesia under realtime ultrasound guidance. treatment strategies on the survival outcome, a multivariateBefore each session the patients underwent a complete logistic regression analysis (enter method) was carriedabdominal ultrasound examination. After thorough skin out. The clinical parameters that differed between thedisinfection, absolute sterile ethanol (96%)was injected four treatment groups at baseline(Okuda-classificationtranscutaneously into the tumor using a 0.7 mm needle presence of portal vein thrombosis, presence of ascites,(Angiomed, Karlsruhe, Germany)and a biopsy-transducer number of tumors, maximum tumor diameter, and CHE)(3.5 MHz, Toshiba, Neuss, Germany). Starting at the most were entered into the regression model as well as Childdistal part of the tumor, ethanol was continously injected Pugh stage, a-fetoprotein(AFP), fever, incidence ofwhile carefully retracting the needle until the tumor complications, and the type of treatment administeredechogenicity became homogenous. After complete ethandThe model rendered the relevance of the variables testeddministrationshort-term follow-up ultrasound control competitively(R-square-value)and identified significantexaminations were performed. Intravenous infusions of predictors for survantibiotics and of fluid replacement were given before andSubsequently, Kaplan-Meier curves were computedfor two days after PEIfor all patients stratified separately for every significantSix weeks after therapy, doppler ultrasound control predictor of the regression analysis in order to demonstratexaminations were performed. In case of detectable the influence of the predictors on survival probabilitiesremaining tumor vascularisation(indicating remaining vitalA P-value <0.05 was considered significant, andumor tissue), tumor growth or new lesions, further PEIP-value <0.001 was considered highly significant.treatment was initiated. In case of uncertain vascularisationcontrast-enhanced MRI scan were performed. In case of RESULTSno detectable tumor vascularization, patients were followed Disease severity at outsetup every three months.The patient set included a high proportion of patients withadvanced stages of liver cirrhosis and tumor extent. FortyTACE treatmentone percent of the 101 patients presented with cirrhosisTACE was performed under local anaesthesia. Before of Child-Pugh stage B, 15% with stage C. Okuda-scoreseach TACE session, a biphasic CT scan of the abdomen II and III were present in 62%. Portal vein thrombosiswas performed to show tumor extent, and to check for was found in 18% and ascites requiring intensified diureticpossible extrahepatic disease. After aortography and therapy in 42% of the patients. In only 14% uninodulara selectiventericoportography to verify vascular tumors were detected, in 70% up to 5 tumor lesions and inanatomy and the patency of the portal vein, a selective 15% multiple tumors. The mean maximum tumor size washepaticography was performed usingard diagnostic 53(+ 27) mm in diameter. High mean AFP levels of 1188catheter or, if necessary, a coaxial system. According ng/mL(+ 6802)were seen, and the mean CHE levels wereto tumor vascularisation and distribution a mixture of reduced to 3.3 mU/mL(+ 1.6). The hepatitis B status wascisplatin(50 mg/m2), doxorubicin(50 mg/m), 450 to 900 positive in 27% of the patients, and hepatitis C status inmg degradable starch microspheres(Spherex, Pharmacia, 22%Erlangen, Germany) and 5 to 30 mL iodized oil ( LipiodolThe baseline characteristics of the four subgroupsUltra-Fluide, Guerbet, Sulzbach/Ts, Germany) were differed due to the selection criteria employed. Aadministered under fluoroscopic control. In case of stasis comparison of the frequencies and mean values of clinicalor reflux, the injection was stopped until the arterial variables is shown in Table 1.flow resumed. After complete administration, a control Significant deviations of the subgroups were shownhepaticography was performed to document the arterialfor the Okuda score, the frequencies of portal veinperfusion and the reduction of tumor vascularisation. thrombosis and ascites, the number of tumor lesions, theIntraarterial analgesia(Dolantin, Pethidin, Aventis, Bad maximum size of tumor diameter and the liver parametersSoden/Ts, Germany)was sufficient to control local pain CHE, bilirubin, and AST.in all cases. Intravenous infusions of antibiotics and fluidreplacement were given before therapy and for two days Treatmentafter TACE. accordiOverall, 268 PEI-interventions were performed in 101patients(2.7+ 1.7 sessions per patient). The mean injectedStatistical analysiscthanolher session. The meanData analysis was performed employing the Statistical hosp中国煤化工43 patients0.5%Package for the Social Sciences Version 12.0 for Windows intensCNMHGtO reduce pre-treat(SPSS Inc, Chicago IL, USA). For comparison of distri- ment5 w Ma Juaualcation algonithmbutions the Wilcoxon-test was used, for metric variables patients received combination treatment of TACE and re-the chi-square test and the Fisher's exact test for nominal peated PEI (1)and 34 patients received PEI alone (24). Tenariables.Results were presented as mean t standard de- patients primarily stratified to repeated TACE treatment3710 ISSN 1007-9327 CN 14-1219/R World J Gastroenterol June 21, 2006 Volume 12 Number 23Table 1 Baseline patient characteristics of the four evaluated treatment subgroupsPEl alonePEI after best TACE-PEI Sec PEL, after p-valuesupp. care(2)n=34(4)n=20(1)n37(3)n"10Child-Pugh18(49%)14(4119⑤51%)4(40%Okuda2(10%)16(46%3(33‰)3(16%)3(8%)Gender33(89%)Portal vein thromb yes9(26%)1(10%)No of tumors4(12%)123453(15%)4(11%)1(5%)9(24%)0(0%)17(46%)1(10%)0(0%Tumor sizeNo of Pel2301.6)3(28)26(12)33(23)75(12)77(16)78(18)FP235(515)(476)28(497tBilirubin36(36)0008mU/mL28(16)37(14)226(7218(41)Protein几L几几几几15(6)33(24)35③31)U/212153(172)55(14)60(25)63018)85(18)120(56)124(62)NS(35)72(17NSNS: Not significant, n(%),Mean(standard deviation). TACE-PEl combiase, AP: Alkaline phosphatase, G-GT: Gammaeasons for discontinuation of TACE in the patientscontinued secondarily on repeated PEI sessions wereadverse events (n= 6), efficacy failure/progressive disease(n= 3)or achievement of preconditions for PEI treatmentdue to tumor regression(n= 1). In those patients primarilystratifed to best supportive care secondary PEI treatmentwas performed because of improvement of general healthstatus(n=9), of patients request(n=7), of reduction ofascites(n=2), or as a bridge to liver transplantation(n= 2)atients (n= 101)CensoredSeven of the 101 patients received liver transplantsduring follow-up(7%). Two of these patients had been2243648stratified to combination treatment of TACE and PEI (1)Months after start of treatmentfour patients to repeated PEI alone(2), and one patientsecondarily to repeated PEI after primary stratificationFigure 2 Kaplan-Meier curve showing the survival probabilities of all 101 patients. TACE ( 3)Survivalnly,were switched after TACE discontinuation to sec- TheYH中国煤化工 Ites of the01 patientsondary PEI treatment (3). Another 20 patients who were wetCNMH Gilities of 720 afrer 1ered treatable by PEI and received subsequently repeated year, 43% after 3 years, and 32% after 5 years(Figure 2)PEI sessions in order to decrease tumor load in individualA comparison of the Kaplan-Meier survival curvestherapy settingsfor the four subgroups is shown in Figure 3. Survivalw忘教掘Dettmer A et a/ Combination therapy of non-resectable HCc37111.0口 Secondary PEI/ TACE0.8TACE-PEI combTACE-PEl comb.sec. PEI/b supp. care口 tumor<=5cmPEI alonetumor 5 cm00。12240.0L364860728420Months after start of treatmentMonths after start of treatmentPEl after best supportive care. The highest survival probabilities can be probabilities of the subgroup with smaller tumors are significantly higher(logrankin the combination treatment group initally stratified to TACE followed by P=0.02)repeated single session PElTable 2 Multivariate regression model of predictorsongterm survival of three years and more after start of therapyP-alue odd's ratios ConfidenceAFP400 ng/mL<38℃CHE-level at Change <1 mU/mL 0.052410.1No of tumor0011113322302lesions+ no ascites← censoredChild-PughchildaNo ascites12247-303020406080No thrombosisNSMonths after start of treatmentOkurNo complic5 Comparison of the Kaplan-Meier curves for the parameter ascites priormplicationstification(ascites vs no ascites). Significantly higher survival probabilitiesoup without ascites(logrank: P<0.001)Maximumamor sizeTACE-PEl comb( 0.0512-14treatmentps other groupsbetween the four subgroups at the outset of the study,P-values, Odd's ratios and confidence intervalls are shown for all significant pre- treatment ascites, the number of tumor lesions,themaximum tumor size, and CHE were significant predictorsin the multivariate enter-model Table 2). Additionally,the type of treatment administered proved to haveindependent predictive value in the moprobabilities differed between the subgroups. Significant The probability to survive longer than three years wasdifferences were detected by log rank-tests comparing 6-fold higher in patients with stable CHE levels undersecondary PEI after best supportive care (4)to therapy and in patients with tumors of less than 5 cm incombination treatment of TACE and PEI (1)(P 0.001) maximum diameter, 11-fold higher in patients with fiveand to PEI treatment alone (2)(P <0.05). There was a tumor lesions or less and 12-fold higher in patients withouttrend towards better survival in the combination group(1) ascites prior to therapy. Treatment with TACE and PEIcompared to PEI alone(2)(P=0. 1)combination (1)led to a 4-fold higher survival probabilityThe 1-, 3-, and 5-year survival probabilities of the compared to the other subgroups(2-4)independently ofsubgroups were 90%, 52%0, 43% for the combination the clinical parameters tested in the modelgroup(1);65%, 50%, 37% for the PEI alone group(2)91%, 40%, 30% for the secondary PEI group after TACe Stratification of predictors in Kanlan-Meier analysesdiscontinuation(3); and 50%0, 23%, 12% for the secondary Survi中国煤化工 for all significantPEI group after best supportive care stratification (4)paranCNMHaccorr-,-(Figures 4-7). ThePredictors of three-year-survivaldifferences of the stratified subgroups were significant inFrom all clinical baseline characteristics that differed all parameters shown (logrank-test: P<0.05). The survival3712 ISSN 1007-9327 CN 14-1219/R World J Gastroenterol June 21, 2006 Volume 12 Number 23bleedings into liver or abdomen, 3 thromboses of theportal vein, 1 thrombosis of the superior mesentericvein, I pancreatitis, 1 cholecystitis, and 1 cholangitis).Allcomplications were managed non-surgically, and one liverabscess needed local drainage. There were no instances ofneedle-track seeding0The mortality within 2 wk after PEI was 2.8%(3口 no ChE decreasepatients). Two patients died with the signs of an acuteno Che deccens.right heart decompensation(one patient sufferingfrom pre-existing alcoholic congestive heart diseaseCHE decreaseOHE decrease-cens20406080previously unknown right ventricular hypertrophy). TheMonths after start of treatmentthird patient developed a hypertensive crisis followed by anapoplectic insult.evels after PEI (decrease >1 mU/mL vs no decrease). The subgroupE levels can expect significantly better survival probabilities (logrank: DISCUSSIONIn patients with HCC, survival is influenced by theetiology and characteristics of the tumors as well as thedegree of cirrhosis and the remaining hepatic functionSince the degree of cirrhosis is rarely affectable in mostcases, improval of survival can be expected mainlythrough effective tumor treatment. To date, there is still nostandard treatment for advanced, unresectable HCC. thelack of efficient systemic chemotherapy against HCC andthe non-resectability in most cases led to the developmentof various locoregional treatment modalities. Increasedanti-tumor efficacy might be achieved through acombination of different modalities carefully tailored to>tumorseach patient+>5 tumors - censoredPEi was introduced in 1983 and has become a020406080standard treatment for small HCC The rationale forMonths after start of treatmentPEI is complete tumor necrosis resulting from cellularFigure 7 Comparison of the Kaplan-Meier curves for the parameter number oftein denaturation, and chemical occlusiontumor les(5 or less tumors vs more than 5 tumors). The survival probabilities of tumor-feeding vessels. The method is best applicableof the group with 5 or less tumors are significantly higher(logrank: P<0.05)in those hccs that are surrounded by firm livercirrhosis reducing the washout of ethanol, Ultrasoundguidance allows for realtime monitoring of ethanolwas improved significantly in patients with stable CHE- injection and distribution. Although regular PEI can belevels under therapy as a parameter of the liver function, performed under local anaesthesia, general anaesthesiain patients without ascites prior to therapy, in patients and endotracheal intubation have greatly facilitated thewith a maximum tumor diameter of less than 5 cm, and in implementation of the single-session technique in termspatients with five or less tumor lesionsof reduced pain and patient movement, thus extendingc range oof treatable tumors. The major draw-TACE complicationsof PEI is a comparatively high recurrence rate of morea total of 67 TACE procedures were performedhan 30%/4. Plausible reasons are inhomogeneous ethanol(10.4%)procedures complications were recordeddistribution within the tumor and the limited effect onwere two cases of reversible leukopenia, oneextracapsular tumor spread. About 30% of small HCCersible pancytopenia, two cases of dissection of the (<3 cm) have already developed invasive growth leadiepatic artery, one case of reversible liver failure, and one to microscopic intrahepatic metastases not detectable byInguinal hematomaeven in small tumore, effect of PEl may be incompletecurrent imaging. ThePE complicationsIt is postulated that while the visible tumors are ablatedSixty-nine(25. 7%)complications occurred after the 268 by PEl, microscopic metastases are effectively destroyedPEI sessions performed. Forty-five of these( 16.8%)were by TACE. Furthermore, TACE is supposed to break upconsidered minor [40 ascites(29 discrete, 7 intermediateIntra中国煤化工 a fibrous wall around4 severe), 3 pleural effusion, 1 oedema of the lower leg]. theus leading to a moreIn 24 interventions(8.9%)major complications wereCN MH Gol during subsequentbserved (5 segmental liver infarctions, 2 focal liver PEI". Hence, the combination of TACE and repeatednecroses,4 liver abscesses, 1 pulmonary embolism, 5 PEl, introduced in 1991, has shown superior resultsww亦要据Dettmer A et a/ Combination therapy of non-resectable HCC3713compared to PEI or TACE alone -4.This concepttreatment of Hcc can be considered effective for ouralso adopted by our institution as described previouslypatients with advanced disease.After careful interdisciplinary evaluation patThe 10 patients with large(7 cm) or multiple(n> 5)non-resectable HCC were treated by a combination of tumors who were switched from repeated TACE treatmentTACE and repeated PEI, if possible. Those patients not to secondary PEI after re-evaluation(Figure 1),wereamenaable to combination treatment were stratifiedindeed amenable to secondary PEI after receiving initialeither modality alone or to best supportive care according TACE as stated by other investigators. Interestingly, 1to their individual situation. Furthermore, patients with 3, and 5-year survival of this group (91%o, 40%o, and 30%70)large or multiple tumors were reevaluated under therapy was comparable to those patients with limited diseaseand secondarily switched to PEI when considered possible who were primarily stratified to combination treatment(90%/, 52%/, and 43%)(Figure 3). From our experience itIn the present prospective study there was a compara- may be advisable to reconsider patients for PEI even aftertively high proportion of patients with decreased liver repeated TACE treatment, especially if TAce therapysynthesis Child-Pugh B(41%)and C( 15%)and portal cannot be continuedvein thromboses(18%). Only 14% of the patients hadThe 1-, 3, and 5-year-survival of the 20 patients withsingle tumors while 15% had multiple disease and the severely impaired liver function and advanced tumors, whomean tumor size was 53 mm(Table 1). Because both liver were switched to repeated PEI after initially being stratifiedfunction and tumor extent affected therapy decisions, the to best supportive care, was comparatively poor(50%,present study was not a controlled trial but a prospective 23%%, and 12%). However, it was superior to a large set oftherapy evaluation based on a stratification scheme patients treated by best supportive care only or systemicaccording to individual patient characteristics. The results therapy from our institution(1- and 3-year-survival: 32%were therefore not unexpected: patients stratified to PEIand 2%). Large tumors can technically be treated bytherapy alone had worse pre-treatment conditions in terms PEI as a salvage therapy, however, at the increasing risk ofof clinical status and liver function compared to those systemic, toxic side-effects and other severe complicationsstratified to the combination of TACE and PEI (Table 1). as previously suggested". From our experience we wouldOverall observed 1- and 3-year-survival of all 101 not generally recommend single-session PEI as a salvagepatients was 72% and 47%. Kaplan-Meier analyses revealed therapy in patients with both disseminated disease anda 1, 3, and 5 year survival probability of 90%, 52%0, and severely impaired liver function who cannot receive TACE43%after initial stratification to TACE followed by Pel as initial treatment. Still, there may be patients in thisand of 65%, 50%, and 37% after PEI alone(Figure 3). group who might profit. It seems as if sufficient clinicalSuperior survival of combining TACE and PEI compared status and/or the effect of initial TACE before PEI isto either modality alone has been proven by several needed to enhance anti-tumoral activity and survival.investigators. Bartolozzi et al detected superior recurrence-In the multithe parametersfree survival of patients after combination of TACE and tumor size and number, no ascites prior to therapy, and100%and 72% aftcr 1 and 3 years), compared to stable levels of CHE under therapy were associatedrepeated TACE alone. Allgaier et al found a survival with better survival. The type of treatment administeredbenefit of patients stratified to combination of TACE and seemed to have independent predictive value, however,PEI(us Pel alone) due to a lower recurrence rate Koda the significance of this result may be reduced due to theet al showed superior survival after 1, 3, and 5 years(100%, limited patient numbers(Table 2). Although screening80%, and 40%)for patients after combination treatment(us programs improved the early detection of HCC, largePEI alone), and Kamada ef al recorded 1-, 3, and 5-year non-resectable tumors are still frequently encounteredsurvival rates of 90%%, 65%, and 50% after combination and remain a major therapeutic challenge. As expected,treatment214tumor size did affect survival. This might be explained byOur results are hence in line with the above data; the fact that small tumors tend to have lower gradings thanowever, there are differences in terms of patient criteria. large tumors. Histological studies revealed that tumors upThe former studies did apply comparatively strict inclusion to 3 cm diameter are mostly well differentiated while largecriteria in terms of tumor number and size. It should be tumors are in general less differentiated.Furthermorenoted that the patients consecutively included into the there is a greater risk in larger HCC of peripheral satellitepresent study were in more advanced stages of both tumor tumors that escape pre-treatment staging. The functionalextent and liver cirrhosis. Being a liver transplantation status of the liver has been known to be a major predictorcenter in Northern Germany, advanced stages of tumor in PEI treatment as it is reflected in the significance ofextent and cirrhosis were included in an attempt to pre-treatment ascites and CHE-levels under therapy inprolong survival of those patients awaiting possible liver the present study. Both TACE and PEI might furthertransplantation: patients with multinodular disease up impair the remaining liver function. The regeneration ofto 5 tumors with less than 5 cm diameter or singular functionchuma is possibly delayed as indicatedtumors up to 7 cm in diameter were primarily stratified to by decr中国煤化工 after treatment(datacombination treatment. In this setting, the present survival not shedata were higher than expected. The overall survival ofCNMHGlow-risk procedureour patients after combination therapy is comparable. severe complications, including death have beeno reported survival data of tumor-free patients with reported. In the present study on repeated singleliver cirrhosis. Consequently, the present combination session PEI in advanced tumors the overall complicatiowww.wjgnet.com3714 ISSN 1007-9327 CN 14-1219/R World J Gastroenterol June 21, 2006 Volume 12 Number 23rate was 25.7%(n=69)including 8.9%(n= 24)majorl1998;10:907-909complications, however, all could be managed non11 Allgaier HP, Deibert P, Olschewski M, Spamer C, Blum Usurgically. We consider these instances acceptable becauseGerok Blum HE. Survival benefit of patients with nopeagle session PEI represents a more aggressive approachable hepatocellular carcinoma treated by a combination ofransarterial chemoembolization and percutaneous ethanolthan regular PEI. Giorgio et al stated a mortality of 1.8%ection--a single-center analysis including 132 patients. Int Iafter single-session PEI. There were also three deathsCaneer1998;79:601-60514 days after PEIamong our patients.12 Kamada K, Kitamoto M, Aikata H, Kawakami Y, Kono hnamura M, Nakanishi T, Hayama K. Combination of tran-Two patients died with the signs of right ventricularcatheter arterial chemoembolizationlatin-lipiodoldecompensation, one patient with a hypertensive crisissuspension and percutaneous ethanol injection for treatmentfollowed by an apoplectic insult. Although there was noof advanced small hepatocellular carcinoma. Am J Surg 2002direct link to PEI, a connection seems possible, as other184:284290authors stated cases of pulmonary hypertension after PEI 13 Bartolozzi C, Lencioni R, Caramella D, Vignali C, Cioni Rwere possibly due to shunting 2. We did not observe anMazzeo S, Carrai M, Maltinti G, Capria A, Conte PF. Treat-hent of large HCC: transcatheter arterial chemoembolizationinstance of needle track seeding, as described by othersIn conclusion, in our set of patients with advancedanscatheter arterial chemoembolization. Radiology 1995: 197non-resectable HCC the combination of TaCe and 14 Koda M, Murawaki Y, Mitsuda A, Oyama K, Okamoto K,812-818repeated single-session PEI is effective and shows superiorIdobe Y, Suou T, Kawasaki H. Combination therapy withresults compared to repeated single-session PEl alonetranscatheter arterial chemoembolization andEven for patients with large(>7 cm) or multiple (n>5)ethanol injection compared with percutaneous ethanol injtumors initially stratified to tepeated TACE treatmenttion alone for patients with small hepatocellular carcinoma: aalone pei treatment is effective in terms of survival atrandomized control study. Cancer 2001; 92: 1516-1524an acceptable complication rate. The predictors including15 Livraghi T, Giorgio A, Marin G, Salmi A, de Sio L, Bolondi L,Pompili M, Brunello F, Lazzaroni S, Torzilli G, Hepatocellularlimited tumor size and number, no pre-treatment ascites,rcinoma and cirrhosis in 746 patients: long-term results ofand stable levels of CHE under therapy are independentlyercutaneous ethanol injection. Radiology 1995; 197: 101-108associated with better survival. Our experience further 16 Llovet JM, Real Ml, Montana X, Planas R, Coll S, Aponte JAsuggests that in advanced disease carefully tailoredAyuso C, Sala M, Muchart J, Sola R, Rodes J, Bruix J. Arterialcombination therapy is more advantageous than singlembolisation or chemoembolization versus symptomatic treat-ment in patients with unresectable hepatocellular carcinoma: atherapy. The optimal strategy against advanced HCCrandomised controlled trial. Lancet 2002: 359: 1734-1739however, has still to be determined by future trials17 Lo CM, Ngan H, Tso wK, Liu CL, Lam CM, Poon RT, Fan ST,Wong J. Randomized controlled trial of transarterial lipiodolhemoembolization for unresectable hepatocellular carcinomaREFERENCESHepatology 2002; 35: 1164-117118 Yamamoto J, Okada S, Shimada K, Okusaka T, Yamasaki s1 Llovet JM, Bustamante J, Castells A, Vilana R, Ayuso Mdel C,Ueno H, Kosuge T Treatment strategy for small hepatocellularSala M, Bru C, Rodes ], Bruix J. Natural history of untreatedcarcinoma: comparison of long-term results after percutaneousnonsurgical hepatocellular carcinoma: rationale for the designhanol injection therapy and surgical resection. Hepatologyand evaluation of therapeutic trials. Hepatology 1999; 29: 62-672001:34:707-7132 Sturm Jw, Keese MA, Bonninghoff RG, Wustner M, Post S. 19 Pompili M, Rapaccini GL, Covino M, Pignataro G, CaturelliLocally ablative therapies of hepatocellular carcinoma. Okolo-E, Siena DA, Villani MR, Cedrone A, Gasbarrini G Prognosticgie2001;2 Suppl5:35-45ctors for survival in patient3 Simonetti RG, Liberati A, Angiolini C, Pagliaro L. Treatmentsmall hepatocellular carcinoma after percutaneous ethanol inction therapy. Cancer 2001; 92: 126-135omized controlled trials, Ann Onco! 1997: 8: 117-13620 Ueno S, Tanabe G, Nuruki K, Oketani M, Komorizono.Y,4 Head HW, Dodd GD 3rd. Thermal ablation for hepatocellularHokotate H Fukukura y Baba Y imamura Y, Aikou t Picarcinoma. Gastroenterology 2004; 127: $167-178nosis of hepatocellular carcinoma associated with Child class b5 Omata M, Tateishi R, Yoshida H, Shiina S Treatment of hepand C cirrhosis in relation to treatment: a multivariate analysistocellular carcinoma by percutaneous tumor ablation methodsof 411 patients at a single center. J Hepatobiliary Pancreat SurgEthanol injection therapy and radiofrequency ablation Gastro-2002:9469-477enterology 2004; 127: $159-21 Ebara M, Okabe S, Kita K, Sugiura N, Fukuda H, Yoshikawa6 Thomas MB, Zhu AX Hepatocellular carcinoma: the need forM, Kondo F, Saisho H. Percutaneous ethanol injection forsmall hepatocellular carcinoma: therapeutic efficacy based on7 Kurokohchi K, Watanabe S, Masaki T, Hosomi N, Miyauchi Y,20-year observation. Hepatol 2005; 43: 458-464Himoto T, Kimura Y, NakaiS, Deguchi A, Yoneyama H, Yosh-22 Shiina S, Teratani T, Obi S, Hamamura K, Koike Y, Omataida S, Kuriyama S Comparison between combination therapyM. Nonsurgical treatment of hepatocellular carcinoma: fromof percutaneous ethanol injection and radiofrequency ablationpercutaneous ethanol injection therapy and percutaneousand radiofrequency ablation alone for patients with hepatocelmicrowave coagulation therapy to radiofrequency ablation.alar carcinoma. World Gastroenterol 2005: 11: 1426-1432Oncology 2002: 62 Suppl 1: 64-688 Acunas B, Rozanes I Hepatocellular carcinoma: treatment 23 Livraghi T, Benedini V, Lazzaroni S, Meloni F, Torzilli G, Vet-th transcatheter arterial chemoembolization. eur radioltori C. Long term results of single session percutaneous etha199932:8689ol injection in patients with large hepatocellular carcinomaYH, Wu JC, Lee PC, Chang FY, Lee SD. Sur-ival benefit of cirrhotic patients with hepatocellular carci- 24中国煤化工 hoff T, Wohlberedt tnoma treated by percutaneous ethanol injection as a salvageScand gastroenterol 2002: 37: 350-355CNMHG10 Kirchhoff T, Chavan A, Galanski M. Transarterial chem-analysis of 389 patients. Br/ Cancer 2005: 92: 1862-1868embolization and percutaneous ethanol injection therapy 25 Ebara M, Ohto M, Sugiura N, Kita K, Yoshikawa M, Okudain patients with hepatocellular carcinoma. Eur/ GastroentenK, Kondo F, Kondo Y Percutaneous ethanol injection for thew数掘Dettmer A et a/ Combination therapy of non-resectable HCC3715treatment of small hepatocellular carcinoma. Study of 95 pa-programme of cirrhotic patientstients.J Gastroenterol Hepatol 1990; 5: 616-626ment of hepatocellular carcinoma: a cost effectiveness analysis26 Castroagudin JF, Delgado M, Martinez SM, Abdulkader L,Gut2001;48:251-259Potel J, Tome S, Otero E, Varo E Prospective histopathological 33 Jonas33 Jonas S, Bechstein wO, Steinmuller T, Herrmann M, Radkeanalysis of hepatocellular carcinoma treated with percutane-C, Berg T, Settmacher U, Neuhaus P, Vascular invasion andous ethanol injection in patients on the waiting list for liverhistopathologic grading determine outcome after liver trans-transplantation. Transplant Proc 2005; 37: 1477-1479plantation for hepatocellular carcinoma in cirrhosis. Hepatology27 Higuchi T, Kikuchi M, Okazaki M. Hepatocellular carcinoma2001;33:1080-1086after transcatheter hepatic arterial embolization. A histopatho- 34 Nanashima A, Tanaka K, Yamaguchi H, Shibasaki s, Morinologic study of 84 resected cases. Cancer 1994; 73: 2259-2267S, Yoshinaga M, Sawai T, Nakagoe T, Ayabe H. Fibrosis and28 Dimitrakopoulou-Strauss A, Strauss LG, Gutzler F, Irn-nflammatory activity in noncancerous tissue and mitotic in-gartinger G, Kontaxakis G, Kim DK, Oberdorfer E, van Kaickdex of cancer tissue in patients with hepatocellular carcinomaG. Pharmacokinetic imaging of 11C ethanol with PET in eightrelationship to clinicopathological factors and prognosis afterpatients with hepatocellular carcinomas who were scheduledhepatic resection. Dig Dis Sci 2003: 48: 1517-1522for treatment with percutaneous ethanol injection. Radiology35 Yamamoto J, Okada S, Shimada K, Okusaka T, Yamasaki S,1999:211:681-68Ueno H, Kosuge T Treatment strategy for small hepatocellular29 Tanaka K, Okazaki H, Nakamura S, Endo O, Inoue S, Taka-carcinoma: comparison of long-term results after percutaneousmura Y, Sugiyama M, Ohaki Y Hepatocellular carcinomaethanol injection therapy and surgical resection. Hepatologytreatment with a combination therapy of transcatheter arterial2001;34:707-713embolization and percutaneous ethanol injection. Radiology 36 Vilana R, Bruix J, Bru C, Ayuso C, Sole M, Rodes J. Tumor191:179:71371size determines the efficacy of percutaneous ethanol injection30 Pessione F, Ramond M], Peters L, Pham BN, Batel P, Rueff B,for the treatment of small hepatocellular carcinoma. HepatologyValla DC. Five-year survival predictive factors in patients with1992:16:353357excessive alcohol intake and cirrhosis. Effect of alcoholic hepa- 37 Tapani E, Soiva M, Lavonen J, Ristkari S, Vehmas T Compli-titis,smoking and abstinence. Liver Int 2003: 23: 45-53cations following high-dose percutaneous ethanol injectiongio A, Tarantino L, de Stefano G, Perrotta A, Aloisio V,nto hepatic tumors. Acta Radiol 1996; 37: 655-659del Viscovo L, Alaia A, Lettieri G Ultrasound-guided per- 38 Meloni F, Lazzaroni S, Livraghi T. Percutaneous ethanolcutaneous ethanol injection under general anesthesia for theinjection: single session treatment. Eur J Ultrasound 2001; 13treatment of hepatocellular carcinoma on cirrhosis: long-tern07115results in 268 patients. Eur ULtrasound 2000: 12: 145-15439 Lencioni R, Cioni D, Crocetti L, Bartolozzi C. Percutaneous32 Bolondi L, Sofia S, Siringo S, Gaiani S, Casali A, Zironi G, Pi-ablation of hepatocellular carcinoma: state-of-the-art. Liverscaglia F, Gramantieri L, Zanetti M, Sherman M. SurveillandTranspl 2004; 10: 591-S97S-Editor Wang J L- Editor Zhu Lh E- Editor Ma Wh中国煤化工CNMHGWww