Journal of Clinical Pathology, 1978, 31, 984-989
AtheroembolismW. F. KEALY
From the Department of Histopathology, Kingston Hospital, Kingston upon Thames, Surrey, UK
SUMMARY A review of the literature has shown that atheroembolism as a cause of clinically evidentdisease is an entity that is little documented. Sections of tissues from necropsies on patients over theage of 60 years from 1970 to 1977 inclusive were reviewed and examined for evidence of athero-embolism, and an incidence of 0 79% was found. In every case of embolism the aorta showedadvanced atheroma, sometimes with aneurysm formation. The ESR in some instances was increased,sometimes markedly so. Some of the problems of antemortem diagnosis are discussed. To emphasisethe possible clinical importance of the condition, a rare instance of spinal cord infarction due toatheroembolism is described.
Atheroembolism from an ulcerating atheromatousaorta is generally accepted as one of the noncardiacsources of peripheral emboli, yet surprisingly littledocumentation of this entity exists (Handler, 1956Wagner and Martin, 1972). Embolisation mayoccur spontaneously, or as a consequence of surgicalmanipulation of a greatly diseased artery, or followaortography (Gore and Collins, 1960; Schipper etal., 1975). Flory (1945), in a definitive account,stimulated interest in the condition. The true in-cidence of atheroembolism is not known but hasbeen variously described as frequent (Handler, 1956),rare (Sayre and Campbell, 1959), infrequent (Goreand Collins, 1960), and more common than isgenerally believed (Schipper et al., 1975). In themajority of instances, atheroembolism is of littleclinical significance since patients with this diseasealready suffer from advanced atherosclerosis.Nevertheless there are several described instances ofclinical importance (Schornagel, 1961).
Material and methods
Sections of blocks taken from necropsies on allpatients over 60 years between 1970 and 1977inclusive were reviewed and examined for evidenceof atheroembolism. One or more sections werepresent of the following tissues: kidney (99-6% ofcases), liver (92%), heart (47 %), spleen (29 %),pancreas and gastrointestinal tract (18% each),thyroid and bone (16% each), and brain andprostate (11 % each). Atheroembolism was identified
Received for publication 21 February 1978
in sections stained with haematoxylin and eosin bythe presence of biconvex cholesterol clefts in thelumina of small arteries (Fig. 1), occasionallyaccompanied by foamy histiocytes. In doubtfulcases, downblocks were taken and stains for elasticwere done on some sections. Frozen sections offormalin-fixed tissues from one necropsy wereexamined using polarised light, and a modifiedLiebermann-Burchardt sterol reaction was appliedto some of these. The sizes of those arteries contain-ing emboli were noted by measuring their diametersto the outer margins of the muscle coats using aneyepiece graticule.The case notes and necropsy records of those
patients showing atheroembolism were scrutinised.To stress the possible clinical importance of this
condition, a case of atheroembolism resulting ininfarction of the spinal cord is described.
Results
Excluding those of the lung, sections from 2126necropsies were examined. Examination of frozensections using polarised light showed doubly refrac-tile cholesterol crystals inside the lumina of someblood vessels, and these gave a positive Liebermann-Burchardt reaction (colourless to copper-red toviolet) characteristic of cholesterol.
Lesions of varying ages were seen, from the mostrecent without any vascular reaction to older lesionswith multinucleated foreign body giant cells,endarteritic thickening, and vascular fibrosis.
Evidence of atheroembolism was found in 16patients, nine men and seven women with anaverage age of 76-7 years. All had been hypertensive,
984
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from
Atheroembolism
Fig. 1 Section of kidneywith cholesterol cleftsinside lumen of smallartery. Haematoxylin andeosin x 110
Table Details ofpatients with evidence ofatheroembolism
Case Sex Age ESR Aortic atheroma Site of emboli Cause of death
I F 81 - + + + Kidney HydronephrosisCarcinoma of cervix
2 M 75 - + + + Kidney, ileum Pulmonary embolismCerebral infarct
3 M 61 - + + + + Kidney, brain Cerebral + myocardial infarct4 F 85 90 + + + Kidney Myocardial infarct5 M 75 34 ++++A Spleen Ruptured aortic aneurysm
Chronic bronchitis + emphysema6 F 75 60 + + + + Kidney, spleen Bronchopneumonia
Cerebral infarct7 M 79 103 ++++A Kidney Ruptured aortic aneurysm8 F 77 - +++ + A Kidney Ruptured aortic aneurysm9 M 78 6 + + +A Kidney Cerebral and myocardial infarct10 M 70 35 ++ +A Spleen Ischaemic heart disease11 M 72 - + + + + Kidney, spleen Perforation of colon
Cerebral infarct12 F 86 67 + + + + Spleen, kidney, liver, Ischaemic heart disease
colon, ileum Ileal perforation13 F 75 - + + + +A Kidney Cerebral infarct14 M 75 116 + + + + Spleen, kidney, pancreas, Carcinoma of prostate
prostate, liver, ileum Diverticulitis of colon15 M 76 82 + ++ +A Kidney Chronic bronchitis + emphysema16 F 88 - + + + + Spleen Cerebral infarct
+ + + marked; + + + + severe; A = aneurysm.
and the Table shows the organs involved in eachcase. The aorta in all instances showed marked toadvanced atherosclerosis. Most showed calcificationand ulceration, some with aneurysm formation, andcases 14 to 16 had extensive soft porridge-like
atheroma. The ESR in the final admission, some-times considerably raised, is also shown in somecases.The sizes of the arteries plugged with athero-
matous material varied from 150 IL to 1100 tu in
985
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from
986
diameter with an average size of 464 ,u. When thekidneys were involved they were, in the majority ofinstances, grossly scarred and contracted. Micro-scopically, all showed marked ischaemic changes,and the emboli were lodged in the inter- and intra-lobular arteries. There was no evidence of infarctionin any kidney. Because of the marked arterio-sclerotic changes associated with hypertension andischaemia, the effects of the emboli were impossibleto assess. Infarction of the spleen was present in twocases, one of which showed atheromatous pluggingof an artery at the apex of the infarct (Fig. 2).Infarction of the brain was present in case 3, andatheromatous plugging of a small meningeal arterycould be seen. When the small and large bowel wereinvolved, the emboli were lodged in the small vesselsof the submucosa. No evidence of infarction of thebowel wall was seen, but in case 12 (Table) ulcerationand perforation of the ileum was present. Section ofthe pancreas in case 14 showed focal areas offibrosis. No definitive ischaemic changes were seenin sections of liver (Fig. 3) or prostate. No evidenceof embolism was seen in sections of heart or bone.Sections of thyroid often showed areas of haemor-rhage and cholesterol cleft formation but an intra-vascular location of the clefts could not bedemonstrated.
W. F. Kealy
Case history
A 76-year-old diabetic man complained of the suddenonset of weakness in both legs seven days beforeadmission to hospital. The weakness was morepronounced in the left leg and lasted for about halfan hour. On the day before admission the weaknessreturned, accompanied by numbness and pins andneedles in both legs and feet, and he had difficulty inpassing water.
Physical examination showed a regular pulse anda blood pressure of 160/90 mmHg. The fundi werenormal apart from some colloid bodies. The eyemovements were full without nystagmus and thepupils were normal and reacted to light. There wasno facial weakness. The upper limbs showed noabnormality apart from diminished tendon reflexes.Abdominal reflexes were present and equal. Thelower limbs showed severe weakness of flexion ofthe hips, knees, and ankles, more pronounced on theleft side. The knee reflexes were absent and the anklejerks were diminished. There was no left plantarresponse and the right was weak extensor. Vibrationsense was absent at the ankles but present at theknees, and light touch was felt on both legs thoughpain sensation was diminished below the knees. Thenext day the patient died suddenly.
Fig. 2 Section of spleenwith atheroembolic
\s4S \ \ >:.<:t ^ plugging of small arteryand splenic infarct (bottomleft). Haematoxylin andeosin x 110
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from
Atheroembolism
Fig. 3 Section of liverwith atheroembolicplugging ofportal arterywith giant cell reaction.Haematoxylin and eosinx 225
RELEVANT NECROPSY FINDINGS
Apart from massive pulmonary embolism andmarked aortic atheroma, pathology was limited tothe spinal cord, which showed intense congestion atthe T9 segment downwards for about 6 cm, and a
small amount of blood was present in the meninges.
MicroscopySections from blocks taken from the whole length ofthe spinal cord showed scattered areas of demyelina-tion at the T12 and LI segments, most extensive inthe left lateral and posterior columns and to a lesserextent in the right lateral and posterior columns(Fig. 4). Sections stained with haematoxylin andeosin showed these areas to be due to infarction,occurring about a week previously, and to includethe left anterior horn. The posterior spinal arterieswere occluded by platelet thrombi and cholesterolclefts with varying degrees of organisation (Fig. 5).
Discussion
Multiple emboli of atheromatous material in smallarteries have been recognised only recently as
significant lesions (Sayre and Campbell, 1959). Theclinical significance depends upon the organ involvedand the sizes of the channels occluded although, inthe absence of infarction, this is difficult to assess as
all patients already show an advanced degree ofatherosclerosis. The aorta is the principal source ofthese emboli but they may originate in any majorbranch, for example, the carotid arteries (Gore andCollins, 1960). That the atheromatous materialfound plugging small arteries is in fact embolic innature has been well dealt with by Flory (1945), who,in an examination of 267 cases with advanced aorticatherosclerosis, did not see any evidence of embolismin 63 cases where ulceration of the aorta was notpresent but found nine instances of embolism out of204 cases where plaque erosion was present.Thurlbeck and Castleman (1957) claimed thatatheroembolism is never found in patients whoseaorta is smooth. Flory (1945) questioned whetherthese lesions might represent a localised form ofatherosclerosis but stated that the crystals were neverfound inside the intima of moderate-sized arteriesbut were always present in the lumina of the vessels.He supported his point by the injection of athero-matous material into the ear veins of rabbits,producing similar lesions in the small arteries of thelungs.The antemortem diagnosis of atheromatous
embolism is naturally very difficult. A clinical pictureof acute or chronic renal failure may occur whereatheromatous material dislodged from a severelyatherosclerotic aorta obstructs the renal arteries
987
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from
W. F. Kealy
Fig. 4 Section of spinalcord TJ2 segment, showingextensive infarction.Loyez x 10
4
S* ..:..;;. .. ..:.*'.:.. .:.'
:S * :
..
46.:*..
*a,*'V
Fig. 5 Section of spinalcord showing athero-embolic plugging ofsmallmeningeal artery.Haematoxylin and eosinx 110
4C
pibA. e 3
988
AS le,
*q
.xWs
'. '..
-0 ..,
,:...; i:%:.
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from
A theroembolism 989
(Kassirer, 1969). In the present series, those patientswith renal emboli showed rapidly developing renalfailure in their final admissions. Occasionally,evidence of atheroembolism may be seen in a renalbiopsy (Heptinstall, 1974), and emboli may some-times be seen by ophthalmologists in the smallretinal arteries. Also amputation specimens of lowerlimbs may show evidence of embolism in the smallarteries of the toes (Schipper et al., 1975). Thepatient (case 12, Table) who, at necropsy, hadatheroembolism to the ileum with ulceration andperforation presented with nausea, vomiting, diar-rhoea, and the passage of blood per rectum. Case 3presented with evidence of cerebral infarction buttwo cases (11 and 12) with splenic infarction had nosymptoms referable to that part.The presence of a raised ESR in some patients is
interesting but may be confusing. Atheroembolismmay be associated with a low-grade fever, eosino-philia, and a high ESR, and these may suggestcollagen disease, polyarteritis, vasculitis, or subacutebacterial endocarditis (Zak and Elias, 1949; Schipperet al., 1975). Case 14 presented with anaemia,rouleaux formation, high ESR, and increased IgGwith an M band in the beta-gamma region, raisingthe possibility of multiple myeloma, which wassubsequently discounted.
Cerebral and myocardial infarction, acute pan-creatitis, massive infarction of the gall bladder, andulcers of the gastric, duodenal, and jejunal mucosaehave sometimes been attributed to atheroembolism(Handler, 1956; Carnevaleand Delany, 1973; Schipperet al., 1975), and the case of infarction of thespinal cord described previously is a very rareoccurrence (Garland et al., 1966).
Sixteen instances of atheroembolism occurringout of 2126 cases examined (079%) is probably afalsely low incidence since it is dependent upon thenumber, size, and site of blocks taken for micro-scopical examination, and, bearing in mind thatatherosclerosis of the aorta is extremely common,instances of atheroembolism are probably morefrequent.
I thank Dr M. Powell, Kingston Hospital, for hisadvice and Mr J. Spicer for technical assistance.
References
Carnevale, N. J., and Delany, H. M. (1973). Cholesterolembolization to the cecum with bowel infarction.Archives of Surgery, 106, 94-96.
Flory, C. M. (1945). Arterial occlusions produced byemboli from eroded aortic atheromatous plaques.American Journal ofPathology, 21, 549-558.
Garland, H., Greenberg, J., and Harriman, D. G. F.(1966). Infarction of the spinal cord. Brain, 89, 645-662.
Gore, I., and Collins, D. P. (1960). Spontaneous athero-matous embolization. American Journal of ClinicalPathology, 33, 416-426.
Handler, F. P. (1956). Clinical and pathologic significanceof atheromatous embolization, with emphasis on anetiology of renal hypertension. American Journal ofMedicine, 20, 366-373.
Heptinstall, R. H. (1974). Pathology of the Kidney, 2ndedition, Vol. 1, pp. 230-231. Little Brown, Boston.
Kassirer, J. P. (1969). Atheroembolic renal disease. NewEngland Journal of Medicine, 280, 812-818.
Sayre, G. P., and Campbell, D. C. (1959). Multipleperipheral emboli in atherosclerosis of the aorta.Archives ofInternal Medicine, 103, 799-806.
Schipper, H., Gordon, M., and Berris, B. (1975). Athero-matous embolic disease. Canadian Medical AssociationJournal, 113, 640-644.
Schornagel, H. E. (1961). Emboli of cholesterol crystals.Journal ofPathology and Bacteriology, 81, 119-122.
Thurlbeck, W. M., and Castleman, B. (1957). Athero-matous emboli to the kidneys after aortic surgery. NewEngland Journal of Medicine, 257, 442-447.
Wagner, R. B., and Martin, A. S. (1972). Peripheralatheroembolism. Confirmation of a clinical concept,with a case report and review of the literature. Surgery,73, 353-359.
Zak, F. G., and Elias, K. (1949). Embolization withmaterial from atheromata. American Journal ofMedicalScience, 218, 510-515.
Requests for reprints to Dr W. F. Kealy, Department ofPathology, 37 Coombe Road, Kingston upon Thames,Surrey KT2 7BD.
copyright. on A
ugust 20, 2021 by guest. Protected by
http://jcp.bmj.com
/J C
lin Pathol: first published as 10.1136/jcp.31.10.984 on 1 O
ctober 1978. Dow
nloaded from