Journal of Compater Assisted Tomography 18(4):533-538,July/August O 1994Raven Press. Ltd.. New York

MRI of Carotid Angiopathy after TherapeuticRadiation Tae Sub Chung, David M. Yousem, Frank J. Lexa, and Deborah A. Markiewicz

Objective: Our goal was to assesswhether significant secondaryatherosclerotic changesfrom radiationcan be detectedon SE MR of the neck. Materials and Methods: Pre- and postradiation MR scansof 16 patients with head and neck malignancieswere studied randomly, independently, and blindly by two readersto determine the frequency of narrowing of the carotid arterial lumen and obliteration of the carotid spacewithin the carotid sheath. Results: Interval narrowing of either the common, internal, or external carotid artery lumen was seenin 108of 192(56%)of vesselsevaluatedon postradiation MR scanscomparedwith preradiation studies.The differencesin the gradesof vesselluminal diameterwere statistically significant(p < 0.05for one readerand p < 0.0001for the other reader).Among the 16patients,3 patients had vesselswith a critical degreeofstenosis,newly appearingon postradiation scans.Seven of 16 patientshad diffuse obliterationof the planeswithin the carotid space. Conclusion:The incidence of acceleratedatherosclerosisfrom therapeutic radiation may be greater than expected in nonirradiated patients. Magnetic resonance scans are an effective, noninvasive method for this type of follow-up. Index Terms: Arteries. carotid-Radiation. treatment-Masnetic resonance imaging-Atherosclero sis-Angiopathy.

Instances of radiation-relatednarrowing of the cervical portion of the internal carotid arteries are infrequently reported in the imaging literature. In the experimental literature, however, histopathological changes in irradiated vessels are well described(1-4). Nonetheless,there are relatively few casereports that link strokesor transientischemic cerebrovascularepisodesto prior radiation of the head and neck (5-14). Thesefew casesuse clinical information and angiographyas the proof to suggest in that radiationinducesacceleratedatherosclerosis the cervical carotid branches. Successfulradiation therapy for head and neck cancershas increasedthe survival time of afflicted patients, but also puts them at risk for radiation-

From the Department of Diagnostic Radiology, YongDong SeveranceHospital,Seoul,SouthKorea (T. S. Chung);and Departmentsof Radiology (D. M. Yousem and F. J. Lexa) and Radiation Oncology (D. A. Markiewicz), Hospital of the University of Pennsylvania,Philadelphia,Pennsylvania,U.S.A. Address correspondenceand reprint requests to Dr. D. M. Yousem at Department of Radiology, Hospital of the University of Pennsylvania,3400SpruceSt., Philadelphia,PA 19104,U.S.A.

inducedcarotid atherosclerosis(13). The incidence and severity of radiationchangein the carotid vesselsof this populationhave been difficult to assess, in part becauseof the risks and invasivenessof carotid angiographyused to evaluate the arteries. Magnetic resonanceimaging has the advantageof noninvasivelyevaluatingthe carotid vesselsfor atheroscleroticvasculardisease.The absenceof signal from the vascularlumen on SE MR allows excellent contrastbetweenthe "black blood" and lesionsinvolving the vascularwatl (15-17).Moreover, on T1weighted scans, the fat in the neck has high signal intensity and allows easy evaluation of the vessel wall (15,18-21).Most studiesof carotidvesselsusing MR angiographyhave relied upon distortion of the arterial lumen and have not tried to visualize directly the atheroma or plaque in the wall (22,23). The aim of this study was to assesswhether significant arterial narrowing and secondary atherosclerotic changes occur after radiation therapy as determinedby MR findings.We comparedpre- and postradiation SE MR scansof the carotid walls and lumina to gaugeatheroscleroticchangeswithout resorting to angiography.

533

CHUNG ET AL. MATERIALS AND METHODS The pre- and postradiationMR scansof 16 consecutivepatientswho underwentradiation therapy for head and neck cancers were retrospectively reviewed. The postradiation MR scans were performed at least 6 months after the last radiation treatment,with the longestfollow-up interval being 30 months and the median interval 17.5 months. The posttreatmentscansin these 16 patientswere performed to serve as a baselinefor future surveillance. None of thesepatientshad suspectedresidual or recurrent tumor around the carotid vessels. Postradiation MR scans obtained 70 6549.99 60-64.99 50-59.99 -70 Gy, 56% of patients receiving 6569.99Gy, 59% of patients receiving 6f64.99 Gy, 3l% of patientsreceiving50-59.99Gy, and 55% of patientsreceiving 65 Gy) with the lower dose range (50 Gy; therefore, many patients who have received radiotherapy have a chance of developing radiation angiopathy. Since all 16 of our patientsreceived >40 Gy of radiation in our studies, we did not find a relationship between dosage and luminal narrowing. Anecdotally we had one patient with a difference in doses applied to the right and left carotid arteries.This one case demonstrateda greater degree of narrowing on the higher dose side. The latency between irradiation and subsequent neurologicalsymptomshas been studiedin one report with rangesfrom 6 months to >30 years (11). With time, luminal narrowing increases,but separating the effects of normal aging from radiation ef-

FlG. 2. A S3-year-oldman with right supraglotticcarcinoma.a: Bulky low intensitymass lesion is demonstrated at right supraglotticareawith near obstruction of the supraglottic air colu m n o n a x i a l T 1 - w e i g h t e dM R s c a n (800/20).Both walls of the carotid arteries (arrows) and the carotid space are well identifiedwithout thickening or obliteration.b: Six months after radiation (54 Gy) and total laryngectomy, there is diffuse thickening of both walls of the carotid arterieswith luminal narrowing(arrows).

J Comput AssistTomogr, Vol. 18, No.4, 1994

MR OF CAROTID ANGIOPATHY POST-XRT fects is nearly impossible.In our cases,three of the four patients who did not show narrowing of the carotid arteries on the postradiation study had follow-up MR scansjust 6 months to 1 year after radiation therapy. However, all 12patientswho had a follow-up MR scan over 1 year after radiation therapy showed significant luminal narrowing of carotid artery. These results suggestthe tendencytoward narrowing of postradiation arteries over time. Sevenof 16 of the patientsshowedcarotid space obliteration due to adventitial fibrosis with luminal narrowing on postradiation examination. Before attributing these changesto acceleratedatherosclerosis, one must eliminate the effects of postsurgical scarringor possibletumor recurrence.When combined with luminal narrowing and mural thickening, obliteration of the carotid space may be a suggestive sign of postradiationatherosclerosis. Although pathophysiologicchangesof postradiation atherosclerosisare well reported in the medical literature (1-4), their radiological significance has not been evaluatedexcept in small reports (5-14). Practical evaluation of the carotid artery has been difficult in asymptomatic patients due to the risks and invasivenessof carotid angiography.Magnetic resonance imaging is a noninvasive diagnostic method of evaluating arteries for atherosclerotic vascular disease.The absenceof signal from the vascularlumina allows excellent contrast between the black blood and lesionsinvolving the vascular wall. Moreover, lipid-containingtissues have a higher signal intensity comparedwith other tissue and are well seen against a dark flow background (15).Becausethis was a retrospectivestudy, we did not have the luxury of examining MR angiographic sequencesto evaluateluminal narrowing;however, we believethat an analysisof thin sectionaxial SE scansis capableof detectingthe extent of luminal narrowing we were examining in this study. The Tl-weighted axial images have been reported to correlate well with conventional angiography in 9l% of patients(16). Our 32 Tl-weighted axial MR images were of excellent quality for evaluating the vesselwalls. Moreover, the T1-weightedaxial MR images were very valuable for the evaluation of obliteration of the carotid space due to adventitial fibrosis. Magneticresonanceangiographywould be less useful in this regard. A trend toward an increasedrisk of cerebral vascular accidentsin patients after neck irradiation has been reported, although the risk is not statistically significant when compared with a control population (24). In our 16 cases, 57Vo(reader l) to 59% (reader 2) of the irradiated internal carotid arteries showed worsening stenotic changes.Three of 16 patients showed a new critical degree of internal carotid artery stenosis on postradiation SE MR scans.Theseresults suggestthat postradiationstenotic changes of the carotid arteries occur fre-

537

quently. Critical stenosisoccurs less often but can be a potential etiology of ischemic brain disease. The arterial stenosisand atherosclerotic changesin the carotid arteries are often not mentioned in reporting postradiationneck MR scans for patients with cancer. Even though the patients are clinically asymptomatic,long-termsurvivors of neck irradiation are at potential risk for acceleratedcarotid atherosclerotic diseaseand appropriate surveillance is recommended.Magnetic resonancescanningprovides a noninvasivemodality for this follow-up. REFERENCBS l. Lindsay S, Kohn HI, Dakin RL, Jew J. Aortic atherosclerosis in the dog after localized aortic x-irradiation. Circ Res 1962:10:51-60. 2. Lindsay S, EntenmanC, Ellis EE, Geraci CL. Aortic arteriosclerosisin the dog after localized aortic irradiation with electrons. Circ Res 1962;10:61-'7. 3. Fajardo LF, Berthrong M. Vascular lesions following radiation. Pathol Annu 1988;23:297-330. 4. FonkalsrudEW, SanchezM, ZerubavelR, MahoneyA. Serial changesin arterial structure following radiation therapy. SurgGynecolObstet 1977;145:395-400. 5. Benson EP. Radiation injury to large arteries: 3 further examples with prolonged asymptomatic intervals. Radiology 1973:.106:195J. 6. McCreadyRA, Hyde GL, Bivins BA, Mattingly SS, Griffin WO. Radiation-induced arterial injuries. Surgery 1983;93: 306-12. 7. Hayward RH. Arteriosclerosis induced by radiation. .Szrg CIin North Am 1972;52:35946. 8. Silverberg GD, Britt RH, Goffinet DR. Radiation-induced carotid artery disease.Cancer 1978;41: 130-7. 9. Painter MJ, Chutorian AM, Hilal SK. Cerebrovasculopathy following inadiation in childhood. N eurology 1975;25: 18994. 10. Wright TL, BresnanMJ. Radiation-inducedcerebrovascular diseasein children. Neurology 1976:,26:540-3. 11. Kearsley JH. Late cerebrovasculardisease after radiation therapy-report of two casesand a review of the literature. Austr alas Radiol 1983;27: I l-3. 12. Brant-Zawadzki M, Anderson M, DeArmond SJ, Conley FK, JahnkeRW. Radiation-inducedlarge intracranial vessel occlusivevasculopathy.AJR 1980;134:51-5. 13. HayashidaK, NishimuraT, Imakita S, Kumita SI, Uehara T, Hamada S. Embolic stroke following carotid radiation angiopathydemonstratedwith I-123 IMP brain SPECT. C/,n Nucl Med 1991;16:580-2. 14. Conomy JP, Kellermeyer RW. Delayed cerebrovascular consequencesof therapeuticradiation. A clinicopathological study of a stroke associatedwith radiation-related carotid arteriopathy. Cancer 1975 ;36:-17 02-8. 15. Herfkens RI, Higgins CB, Hricak H, et al. Nuclear magnetic resonance imaging of atherosclerotic disease. Radiology 1983;148: l6l-6. 16. WesbeyGE, HigginsCB, Amparo EG, Hale JD, Kaufman L, Pogany AC. Peripheral vascular disease: correlation of MR imagingand angiography.Radiology 1983;156:733-9. 17. Fishman MC, Naidich JB, Stein HL. Vascular magneticresonanceimaging. Radiol Clin North Am 1986;24:485-501. 18. Mohiaddin RH, Firmin DN, Underwood SR, et al. Chemical shift magnetic resonance imaging of human atheroma. Br Heart J 1989;62:81-9. 19. Vinitski S, Consigny PM, Shapiro MJ, Janes N, Smullens

J Comput AssistTomogr, Vol, 18, No.4, 1994

538

T, S. CHUNG ET AL.

SN, Rifkin M. Magnetic resonance chemical shifting and spectroscopyof atheroscleroticplaque. Invest Radiol l99l; 26:703-14. 20. Asdente M, Pavesi L, Oreste PL, Colombo A, Kuhn W, Tremoli E. Evaluation of atheroscleroticlesions using NMR microimaging.Atherosclerosis1990;80:243-53. 2 1 . Merickel MB, Carman CS, Brookeman JR, Mugler JP, Brown MF, Ayers CR. Identification and 3-D quantification of atherosclerosisusing magnetic resonanceimaging. Comput Biol Med 1988;18:89-102.

J Comput AssistTomogr, Vol. 18, No.4, 1994

22. Mohiaddin RH, Longmore DB. MRI studies of atherosclerotic vascular disease:structural evaluation and physiological measurements.Br Med Bull 1989;45:96U90. 23. Ross JS, Masaryk TJ, Modic MT, Harik SI, Wiznitzer M, SelmanWR. Magnetic resonanceangiographyof the extracranial carotid arteries and intracranial vessels: a review. N eur ology 1989;39: 1369J 6. 24. Elerding SC, FernandezRN, Grotta JC, Lindberg RD, Causey LC, McMurtrey MJ. Carotid artery disease following external cervical irradiation. Ann Surg 198I ; 194:609-15.