%ShunbngNo.of patients(%)5.042(33.6%)5.1—10.033(26.4%)10.1—15.014(11.2%)15.1—20.07(5.6%)20.1—25.06(4.8%)25.1—30.01

1(8.8%)30.1—40.05(4.0%)40.1—50.04(3.2%)>50.03(2.4%)Total125(100%)

mainly on metastatic liver tumor. In the present study, 125

patients with HCC with cirrhosis were assessed by intrahepatic-arterial @Tc-MAAin relation to lung shunting.

MATERIALSAND METHODS

PatientswithHCC(histology-provenor withraisedalpha-fetoproteinof over500pg/literandultrasoundevidenceof a livertumor)who underwenthepaticangiographyas an assessmentforresectabilityor selectiveinternal radiationtherapy were eligiblefor the study. Hepaticangiographywas performedthroughtheSeldingertechnique.Whenthe angiographiccatheterwas selectively placed in the hepatic artety, 111 MBq (3 mCi) of @“@TcMAA(AmershamPulmonateII, with 80%of the particles 10—60,m@in size and none greater than 150 sun) were injected into thehepatic artery following20 @gof angiotensin II (Ciba-Geigy,Basel, Switzerland).The catheter was then removed.

Thepatientwas thentransportedto the gammacamerasuiteandscintigraphicimagesof the lungs,liverandgastroduodenalregionswere takenwith an analog/digitalgammacamera (Philips,Hamburg,Germany).Thetotalcountrate wascomputedfromthedigitized image. Regions of interest (ROIs)were carefully drawnaroundthe organs(liverand lungs)and the percentageof activityshuntedinto the pulmonarycirculationwascalculatedas the ratioof lungcounts-to-totalcounts.

Tumor sizes were measured either by computerized tomographyor ultrasound.Thevascularityof the tumorson hepaticangiographywas independentlyassessedby an interventionalradiologist who did not know the results of the vascularity of thetumors by scintigraphic imaging. Vascularity on hepatic angiog

TABLE ILung-Shuntingas Detem*iedby @rc-MM

WdJiincreased use of intraarterialadministrationof chemotherapeuticand radioactiveparticulateagents,it is necessarytoassessagentdeliveiyin the lung.Methods:Technetium-99m-labeled macroaggregatedalbumin (@rc-MAA) deliveredthroughthe hepaticarterywasusedto determinethedegreeoflung shuntingin 125 patientswith hepatoceliularcarcinoma(HCC).Results:Thepercentageof lungshuntingvatiedamongpatientsandit rangedfromlessthan1%to67.2%,witha medianof8.1%.Thedegreeof shuntingdependedonthevascularityofthetumorsbutnotonthetumorsize.TheeffectofangiotensinIIon lungshuntingwastestedon six patientsandtherewas nosignificantdifferencefoundbetweenthosepatientswho werepre-treatedwfthangiotensinII and thosewho were not Onepatientwho underwenta liver resection,hada significantdecreasein lungshuntingfrom28.5%to lessthan 1%aftersurgary.Conclusion:The lackof effectof angiotensinII togetherwith the almostcompleteablationof lung shuntingby tumorresection suggested neoplastic blood vessels were responsiblefor the shunting.

KeyWords: hepatocellularcercinoma;lung shunting

J NucI Med 1994; 35:70-73

gional targeted treatment using chemoembolization,radioactive Lipiodol-'311and @°Ymicrospheres are increasingly used in treatment of inoperable hepatocellular carcinoma (HCC) (1). However, arteriovenous shunting isknown to occur in metastatic liver tumors (2). Large arteriovenous shunting makes regional chemotherapy less effective and increases systemic toxicity. In addition, therelatively low radiation tolerance of lungs has arousedmuch caution in treating HCC with s@@Ymicrospheres orLipiodol-'311. High lung shunting of radioactive isotopesmay induce radiationpneumonitis. Technetium-99m-MAAhas been employed to assess the degree of shunting intopulmonarycirculation(1,3—5).These reports concentrated

Received May 6, 19@3;rev@lonex@eptedSept 29, 1993.For correspondence and reprints cont@ Dr. W.T. Leun@ Dept. of @knk@al

Oncology,PIInCeOIWaIeSHosp@l,TheChineseUniversityotHongKong,Shatn,NewTerrltories,HongKong.

70 The Journal of Nuclear Medicine•Vol. 35 •No. 1 •January 1994

Measuring Lung Shunting in HepatocellularCarcinoma with Intrahepatic-ArterialTechnetium-99m Macroaggregated AlbuminWai-tong Leung, Wan-yee Lau, Stephen K.W. Ho, Michael Chan, Nancy W.Y. Leung, Joanna Lin,Con Metreweli, Philip J. Johnson and Arthur K.C. Li

Joint Hepatoma Clinic, Pthzce of Wales Hospita4 The Chinese University ofHong Kon& Hong Kong

by on August 13, 2020. For personal use only. jnm.snmjournals.org Downloaded from

I,

FIGURE 1.Technetium-MMscintlgramof a patient wWi 67.2%lungshunting.

FIGURE 2.Technetium-MMscintigramof a patient wfth 1% lungshunting.

Table 1 illustratesthe subdivision of 125patients accordingto the degree of lung shunting. Two representative scintigrams are shown in Figures 1 and 2 which have 67.2% and1%lung shunting respectively.

The size of the major tumor in each patient was measured and ranged from 2 cm to 26.2 cm with a median of12.5 cm. Plotting tumor sizes against lung shunting showedno correlation (Fig. 3, coefficient of linear regression r =0.0863).

The relationshipbetween the percentage of lung shunting and the tumor vascularity grading as assessed on hepatic angiographyis illustratedin Figure4. The lung shunting percentage of patients having vascular scores 1 and 2are compared with those having scores of 3 and 4 usingWilcoxon 2-sample test (p < 0.0001). This suggests thosewith more vascular lesions have larger lung shunting.

In six patients, lung shunting percentages were estimated by injecting @Fc-MAAthrough an implantableintra-arterial port-a-catheter with and without angiotensinII pretreatment. There was no difference in lung shuntingfor these patients (pairedStudent's t-test, p = 0.16).

One of the patients had a 15-cm right lobe tumor, avascular score of 4 on hepatic angiography and had undergone a righthepatic resection. The lung shunting droppedfrom 28.5% (pre-operative)to less than 1%after surgery.

raphy was graded by a scale from 0 (normal) to 4 (extremelyhypeivascular): Presence of mild tumor staining without an increase in the number of vessels was grade 1. Moderate tumorstaining and increased number of vessels were grade 2. Intensetumor staining and markedlyincreased numberof vessels whichwere also dilatedandtortuouswere grade3. Tumorshavingall thegrade 3 characteristicstogetherwith venous poolingwere classifled as grade 4. The degree of lung shuntingwas correlatedwiththe size and vascularity of the biggest tumorin the patient.

In six patientswho had implantedarterialport-a-catheters(Pharmacia,St. Paul,MN)inthegastroduodenalartetywhichledto the common hepatic artery, @Tc-MAAscans were repeatedwith and without angiotensin II by injection of the radioisotopethroughthesubcutaneousportsto assessthedegreeof changeoflung shunting. For one patient who had liver resection for HCCand placement of an arterial port-a-catheter, the @“@Tc-MAAscanwas repeatedaftersurgeiyandthe degreeof lungshuntingwascomparedwith that before the resection. Statisticalanalysis wasperformed using the Wilcoxon 2-sample test and paired Student'st-test.

RESULTS

From November 1990 to March 1993, 125 patients withHCC andcirrhosiswere entered into the study. The degreeof lung shunting was found to vaiy between patients andrangedfrom less than 1%to 672%, with a medianof 8.1%.

80r—0.0863

70

_@160

@50

t:

0 5 10 15 20 25 30

FIGURE 3. CorrelatIon between lungshuntingandtumorsize.

TumourSize(cm)

Lung Shunting in HepatocellularCarcinoma•Leung et al. 71

by on August 13, 2020. For personal use only. jnm.snmjournals.org Downloaded from

80n—I n=16 n —32 n—so n—18

70

60

50

40

30

20

I0

0

C.59.cC,,

HCC is usuallyvery vascular due to neovascularization.Capillarieswithin the tumor are expected to be embolizedby the @“Tc-MAAbecause of similarity in size and fewcan leak through if there is no arteriovenous shunting.However, arteriovenous shunting is well known in HCC.Therefore the leaked @“@Tc-MAAwill go through theseabnormal vascular channels and be detected in extrahepatic organs. The lungs, being a first-passorganand havingsimilar capillary structure, will hold up the leaked @‘@‘TcMAA. Little MAA will leak to other organs because thecapillary bed in the lungs is able to stop any further systemic leaking. Thus, measuringthe degree oflung shuntingcan indirectly reflect the degree of arterovenous shuntingin the liver tumor.

Other hepatic arterial perfusion scintigraphy using99mTc@MAAwas employed to assess the leakage of regional chemotherapy for liver metastases into the pulmo

nary circulation (2,6). The percentage of lung shunting wasfound to be between 6%and 26%(mean 12.3%)in a seriesof 20 patients (2) and between 0.4% and 32%(mean 6.2%)in another series of 67 patients (6). A much wider range(< 1%—67.2%)of pulmonary shunting was observed in our125 patients with primary liver cancer and cirrhosis. ThedegreeofshuntingappearstobehigherinpatientswithHCCwhen compared to metastatic liver cancer fromotherreports. Zeissman et al. (6) found that 49% of their patientswith metastatic tumors had a percentage equal to or largerthan 5%, whereas 66% of our patients with HCC had avalue equal to or largerthan 5%. Though this may not beentirely comparable to each other, it may suggest a trendtowards HCC having higher lung shunting.

Ourresults also showed that the degree of arteriovenousshunting to the lungs was influenced by the vascularity ofthe tumors but not by the tumor size. Thus, the patternofblood flow through the tumor seems to be the determiningfactor of pulmonary shunting. This may be due to thetumor itself which is vascular and rich in blood supply and

• 0000

0 000

SS

SS S

S@ I 2I tI 2

Vascularity3 40

FiGURE 4. Scatterdiagramplothnglungshuntingagainsttumorvascularity.

has more artenovenous shunts with resultant increases inlung shunting. The poor correlation between tumor sizeand lung shunting may be due to the fact that tumor sizecorrelates poorly with vascularity.

Angiotensin II was suggested to inducevasoconstrictionof normal blood vesselsbut not neoplasticblood vessels(7). It wasusedextensivelyin selectiveinternalradiationtherapy using @°Ymicrospheres (8) and for targetingcytotoxic-loaded microspheres into liver tumors (9). Comparison of the percentage of lung shunting with and withoutangiotensin II in our six patients showed no significantdifference. This finding is in agreement with that of Goldberg et al. (10) andsuggests thatthe vessels responsible forthe shunting are neoplastic in nature. The dramatic decrease in lung shunting after resection for HCC in onepatient concurred with Goldberg's findings, but more patients shouldbe studiedto prove the point.

In summary, arteriovenous shunting of @‘Tc-MAAtothe lungsafterinjectioninto the hepaticarterywas foundtovary somewhat and it was high in a large proportion ofpatientswith HCC. The degreeof shuntingwas related totumor vascularity rather than tumor size. Assessment oflung shunting by a @‘Tc-MAAscan before any regionaltargetedtreatmentis useful in excluding patients who maybe at high risk for excessivepulmonary irradiation afterinternalradiationtherapyor markedsystemic toxicity afterregional chemotherapy.

ACKNOWLEDGMENTS

The authorsthankMiss RebeccaYau for her assistanceintyping the manuscript and Mr. K.K. Man for his help in preparingthe figures.

REFERENCES

1. Lau WY, Li AKC. Therapeutic aspects of radioisotopes in hepatobiliarymalignancy.BrJSu,g 1992;79:711.

2. Zeissman HA, Thrall JH, Gyves JW, et al. Quantitative hepatic arterialperfusion scintigraphy and starch microspheres in cancer chemotherapy. INuci Med 1983;24:871—875.

72 TheJournalof NuclearMedicine•Vol.35•No.1 •Janualy1994

DISCUSSION

by on August 13, 2020. For personal use only. jnm.snmjournals.org Downloaded from

3. BlachardRJW,MorrowIM,SutherlandlB. Treatmentoflivertumorswithyttrium-90 micmspheres alone. I Can Assoc Radiol 1989;40:@6-210.

4. HoWe5, YipT-K, ShepherdFA, Ctal.HepatoceHularcarcinoma:pilottrialof treatmentwith @°Ymicrospheres.Radio1o@1989172857-860.

5. HerbaMi,IllescasFF,ThiriwellMP,etal.Hepaticmalignancies:improvedtreatment with intraarterial @°Y.Radiology 1989;169:311—314.

6. ZeissmanHA, ThrallJH, Yang PJ,et al. Hepaticarterialperfusionscintigraphy with Tc@ MAA Radiology 1984;152:167-lfl.

7. Sasaki Y, Imaoka S, Hasegawa Y, Ctal. Changes in distributiOnof hepaticblood flow induced by intra-arterialiafusion of angiotensinII in humanhepaticcancer. Cancer 1985-,55:311-316.

8. GrayBN, BurtonMA,KdlleherDK,AndersonJ, KiempP. Selectiveinternalradiation(SIR)therapyfor treatmentof livermetastases:measurement of response rate. I Stag Oncol 1989;42:192—1%.

9. Goldberg JA, Thomson JAK, Bradnam MS. et al. Angiotensin II as apotential method of targetingcytotoxic-loadedmicrospheres in patientswith colorectallivermetastases.Bri Cancer 1991;64:114-119.

10.GoldbergJA, BradnamMS, Kerr DJ, Ct al. Arteriovenousshuntingofmicrospberesin patientswith colorectallivermetastases:errors in assessment due to free pertechnetate, and the effect of angiotensm II. Nuci MedContmun1987;8:1033—1046.

LungShuntingin HepatocellularCarcinoma•Leunget al. 73

ERRATUMMIRD Pamphlet No. 14 “ADynamic Urinary Bladder Model for Radiation Dose Calculations,― appearingon pages 783—802of the May 1992 issue ofJNM contains results that are incorrect owing to an error in thecomputer code used in the calculations.

This error was discovered after publication of the pamphlet; the magnitude of the error introduced in thepublished results depends upon the radionucide as well as the specific model parameters; however, thepublished values are, on average, approximately 40% lower (ranging from less than 10% to greater than 60%lower). In addition, typographical errors were identified in the expressions involving the model description.

The pamphlet describes a dynamic urinary bladder model developed to provide physiologically realisticfeatures for bladder wall dose calculation, incorporates expanding bladder contents, and allows for variableurine entry rate, initial bladder contents volume, residual volume and first void time. Radiation dose estimatesare calculated for the bladder wall surface for 11 radiopharmaceuticals. Extensive tables and graphs arepresented for the dose to the bladder wall surface as a function of the variable parameters.

The MIRD Committee recognizes the importance of rectifying this situation. A revised Pamphlet No. 14,under preparation, will provide corrections and also take the opportunity to expand the list of radiopharmaceuticals presented. The availability and mode of distribution of this revision will be announced through theJournal. To assist the nuclear medicine community in the use of the dynamic bladder model, a computer codehas been installed at the Radiation Internal Dose Information Center, Oak Ridge Institute for Science andEducation, Oak Ridge, Tennessee 37831. At this time, interested individuals may obtain the corrected tablesfor any of the published radiopharmaceuticals by direct contact with Oak Ridge (Michael G. Stabin at615-576-3449).

The MIRb Committe sincerely regrets any inconvenience caused through errors in the publication.

by on August 13, 2020. For personal use only. jnm.snmjournals.org Downloaded from

1994;35:70-73.J Nucl Med.   Johnson and Arthur K.C. LiWai-tong Leung, Wan-yee Lau, Stephen K.W. Ho, Michael Chan, Nancy W.Y. Leung, Joanna Lin, Con Metreweli, Philip J.  Technetium-99m Macroaggregated AlbuminMeasuring Lung Shunting in Hepatocellular Carcinoma with Intrahepatic-Arterial

http://jnm.snmjournals.org/content/35/1/70This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 1994 SNMMI; all rights reserved.

by on August 13, 2020. For personal use only. jnm.snmjournals.org Downloaded from