- Email: [email protected]
Radiation heart disease
Radiation Heart Disease Analysis of 16 Young (Aged 15 to 33 Years) Necropsy Patients Who Received Over 3,500Rads to the Heart
FRANK C. BROSIUS. III. M.D.* BRIJCE F. WALLER, M.D. WILLIAM C. ROBERTS, M.D. Rethesdo. &turyland
From the Pathology Branch, National Heart, Lung, and Blood Institute. National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Dr. William C. Roberts, Building 10A. Room SE-30, National Institutes of Health, Bethesda. MD 20205. Manuscript accepted September 11.1980. l Present address: Department of Medicine, University of Michigan Medical Center, Ann Arbor, MI 48104.
Certain clinical and necropsy findings are described in 16 young (aged 16 to 33 years) patients who received >3,500 rads to the heart Rve to 144 months before death. All 16 had some radiation-induced damage to the heart: 15 had thickened pericardia (Rve of whom had evidence of cardiac tamponadek eight had increased interstitial myocardial fibrosis, particularly in the right ventricle; 12 had Rbrous thickening of the mural endocardium and 13 of the valvular endocardium. Except for valvular thickening, the changes were more frequent in the right side of the heart than in the left, presumably because of higher radiation doses to the anterior surface of the heart. In six of the 16 study patients and in one of 10 control subjects, one or more major epicardial coronary arteries were narrowed from 76 to 100 percent in cross-sectional area by atherosclerotic plaque; one patient had a healed myocardial infarct at necropsy and one died suddenly. In 10 patients and in the 10 control subjects, the four major epicardial coronary arteries were examined quantitatively: 6 percent of the 469 five millimeter segments of coronary artery from the patients were narrowed from 76 to 100 percent (controls = 0.2 percent, p,,=,,“.O6)and 22 percent were narrowed from 51 to 75 percent (con= 12 percent). The proximal portion of the arteries in the patients had signiRcantly more narrowing than the distal portions. The arterial plaques in the patients were largely composed of fibrous tissue; the media were frequently replaced by Rbrous tissue, and the adventitia were often densely thickened by fibrous tissue. In five patients, there was focal thickening (with or without luminal narrowing) of the intramural coronary arteries. Thus, radiation to the heart may produce a wide spectrum of functional and anatomic changes but particularly damage to the pericardia and the underlying epicardial coronary arteries. In the early part of this century, the heart was considered a “radioresistant” organ [1,2]. With the introduction of megavoltage radiotherapy for treatment of neoplasms, however, it became apparent that the heart could, indeed, be damaged by high-dose radiation. The nature and extent of the damage, however, in both human subjects and in nonhuman experimental animals are still debated. Pericardial thickening with or without effusion or constriction has been confirmed from several centers [3-281. In several “case reports” clinical or necropsy evidence of myocardial ischemia and coronary arterial narrowing has been described in young persons who had received highdose mediastinal radiation [10.29-411. Increased myocardial fibrosis
March 1981 The American Journal of Medicine
Volume 70
519
RADIATION
HEART DISEASE-BROSIUS
ET AL.
has been described as a consequence of high-dose mediastinal irradiation [3,10,11,16,25,42-461. The effect of high-dose radiation on the mural and valvular endocardium has received virtually no attention. Furthermore, the extensiveness of the coronary disease has not been described previously in young necropsy patients who received megavoltage therapy to the mediastinum. In the report which follows, the effect of high-dose irradiation in 16 young necropsy patients is described. Particular attention is given to the mural and valvular endocardial damage and to the extent of the coronary narrowing that were found. PATIENTS
STUDIED
In 15 patients the apical portion of the heart was excluded from the direct radiation field. The interval from the end of the irradiation period to death ranged from five to 144 months [mean 55 months]. Ten patients received one or more chemotherapeutic agents, including six who received prednisone (for one to 12 six-week periods] and three who received doxorubicin. None had cushingoid features at any time. No patient had clinical evidence of chronic congestive cardiac failure and none had chest pain which could have been interpreted as evidence of myocardial ischemia. Six patients smoked cigarettes (10 to 20 daily). Electrocardiograms, recorded in 14 patients (all after irradiation), disclosed sinus rhythm in all, nonspecific S-T segment or T wave changes in nine, low (sum of QRS complexes in leads I, II and III 515 mv) voltage in five (four had pericardial effusions) and complete right bundle branch block in one. Five patients had large cardiac silhouettes on chest roentgenograms [Figure l], and five had clinical findings (distended neck veins, subcutaneous edema, paradoxical pulse and Kussmaul’s sign) of cardiac tamponade. Four of the five patients with clinical features of tamponade had pericardiocentesis with removal oflarge quantities of fluid [Figure 1). and two patients (Patients 9 and 12, Table I] subsequently had partial pericardiectomies because of recurring signs of cardiac tamponade. Blood hematocrit values about one month before death ranged from 20 to 48 percent [mean 33 percent): serum total protein, 3.5 to 7.5 g/d1 (mean 5.5 g/d]); and serum albumin, from 1.3 to 4.9 g/d1 (mean 2.8 g/dl). The highest total serum
AND FINDINGS
To be included in this study, patients had to be under 35 years of age at death, had to have had an autopsy, had to have received a total of more than 3,500 rads to the heart, and the heart specimens had to be examined by at least one of us. Sixteen patients fulfilled these criteria. Certain clinical and necropsy observations in them are summarized in Table I. The patients ranged in age from 15 to 33 years (mean 26 years) at death; 13 were men and three were women (Patients 6,10, and 13. Table I). Thirteen had Hodgkin’s disease; two (Patients 8 and 151, carcinoid; and one (Patient 10). histiocytic lymphoma. The number of rads administered to the anterior surface of the heart ranged from 3,987 to 8,769 [mean 5,592). and to the posterior surface, from 3,000 to 5,000 [mean 3,932). (The method for calculating the radiation dose is described in the Appendix.]
TABLE I
Clinical and Necropsy Findings in 16 Patients with Radiation Heart Disease
PaSent
A@ (Yr)
Aoe &I at Dx
1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16
15 19 24 24 25 25 26 26 27 29 29 29 30 30 33 33
12 16 23 20 22 23 19 24 20 27 17 21 25 27 20 24
Calculated Rad to Heart Ani Post
5,202 5,648 5,842 5,295 5,186 4,100 4,165 3,987 7,047 4,932 4,054 4,341 5,419 8,769 6,600 6,866
3,000 3,600 3,500 3,600 3,092 4,000 3,000 3,600 4,000 4,200 4,000 4,025 4,495 4,800 5,000 5,000
MO.from Rad to Death 34 35 22 42 39 11 78 5 53 20 144 84, 825 44 27,2§ 137 104
CT
TC’
0 0 + + 0 0 0 0 + 0
129 139 140 120 172 176 146 250 198 168 168
0 +
0 + 0 0
;;7 130 170 308
Hctt (%I
HW (9)
43 20 39 29 30 37 21 34
170 270 245 190 290 250 250 290 370 155 240 225 170
.. 31 27 27 35 31 45 48
230 350
TE p6 3oot 0 100 260 75 0 200 0 0 0 0 0 0 375 0 0
M
V
+ + + + 0 + + + 0 + + + + + + + + + + + 0 0 + + + + •t0 0 0 +++0+13
RV
IMF VS
+ 0 + 0 + + + 0 0 + ‘$’ + 0 0 0 o...o + 0 o...o 0 0 0 0 0 0 0 0
No. S-mm No. S-mm S TICA s >75% LV CAs XSA RV LV 0 0 + 0 0 ‘4’ 0 0 0 0 0 0
. so 54 44 55
.
50 44 42 37 43 ...
‘0 0 2 0 .. 8 0 0 10 12 ...
‘50
‘i
+ 0 + + 0 + 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0
0
5++
= no data; CA = coronary artery; CT = cardiac tamponade: Dx = diagnosis; ECG = electroNOTE: + = present; 0 = absent or none; . cardiogram: Hct = hematocrit; HW = heart weight; IMF = interstitial myocardial fibrosis; LV = left ventricle; M = mural; PE = pericardial effusion; RV = right ventricle; S = segments; TC = total serum cholesterol; TE = grossly thickened endocamium; TICA = thickened intramural coronary artery; V = valve; VS = ventricular septum: XSA = cross-sectional area. + Highest recorded value. t Value about one month before death. t Number of milliliters. 5 Received two courses of radiotherapy.
520
March 1981
The American
Journal of Medicine
Volume
70
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 1. Patient 14 (Table I). Chest radiogram after pericardiocentesis and injection of air into the pericardial sac and posterior half of the heart and lungs (left), and after they had been sectioned longitudinally (right). The pericardia are thickened primarily by fibrous tissue.
cholesterol values 115 patients) ranged from 120 to 308 mg/dl (mean 173 mg/dl); this value was >180 mg/dl in four patients. An additional 10 patients matched closely for age and sex to the study patients were used as control subjects. Certain characteristics of them are shown in Table II. None received therapeutic mediastinal irradiation, none ever had symptoms of cardiac dysfunction, all died from noncardiac conditions, and seven received one or more chemotherapeutic agents, including six who received prednisone (for one or more sixweek periods) and one who received doxorubicin. At necropsy, fluid was present in the pericardial sacs in 12 of the 16 study patients: in six, the amount was <50 ml: in the other six, 75 to 375 ml. The remaining four patients [Patients 9, 10, 12 and 13, Table I) had pericardial fibrous adhesions which obliterated the pericardial spaces, but two had had partial pericardiectomies. Four patients had fibrinous pericardial adhesions (Figure 2). In all but one patient the epicardial surface was thickened by fibrous tissue [Figure 2). The
hearts ranged in weight from 155 to 370 g (mean 246 g). None of the 16 patients had grossly visible or histologically discernible foci of myocardial coagulation necrosis. Only one patient [Patient 12, Table I, Figure 3) had a grossly visible scar on the ventricular wall at necropsy; in him it involved the posterior wall of the left ventricle, and the three major (right, left anterior descending and left circumflex) epicardial coronary arteries were severely narrowed. On histologic examination of at least one histologic section from the wall of each cardiac ventricle (mean 2.5 sections per ventricle], increased interstitial myocardial fibrosis was observed in the right ventricular free wall in eight patients (Figure 4), in the left ventricular free wall in three patients and in the ventricular septum in two patients. In one of the 10 control subjects, mild interstitial myocardial fibrosis was observed [Table II). Unequivocal damage to individual myocardial cells was not detected in any study patient or in any control subject by histologic study. The mural endocardium of one or more chambers was
Clinical and Necropsy Findings in 10 Control Subjects
TABLE II
Control Subjects
Age (vr)
bx
1 2 3 4 5 6 7 8 9 10
19 21 22 23 28 28 29 29 33 36
U’ AML ALL ALL CML HL Ch ET MM HS
Age at Dx (v) 19 18 20 19 21 28 28 28 30
.
($dt, Yi 200 178 154 147 2;2 182
.
Hct (%j
Alb (9Aflj
(;:,
ii 31 19 30 34 27 34
7.1 3.8 7.1 3.9 4.4 8.7 8.5 8.1
.
.
b:d 1.6 4.7 1.2 2.3 2.7 2.8 3.1
..
HW (g)
RV
370 280 500 430 220 360 200 270 280
0 0 0 0 0 0 0 0 0
..
+
IMF vs 0 0 0 0 0 0 0 0 0
LV
No. 5-mm Segmenlsof 4 Major CAs
0 0 0 p 0 0 0 .O 0
.
+
----
No. Narrowed >75%
63 22 58 62 42 37 28 46 48 34
0 0 0 0 1 0 0 0 0 0
= no data; Alb = albumin; AU = acute lymphocytic leukemia: AML = acute myelocytic leukemia; NOTE: + = present; 0 = absent or none; . BT = brain tumor; CA = coronary artery; Ch = choriocarcinoma; CML = chronic myelocytic leukemia; Dx = diagnosis; Hct = hamatocrit; HL = histiocytic lymphoma; HS = heat stroke: HW = heart weight; IMF = interstitial myocardial fibrosis; MM = malignant melanoma; TC = total serum cholesterol: U = unknown. * Idiopathic sudden death.
March 1981
The American Journal of Medicine
Volume 79
521
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 2. Fibrinous pericarditis. Patient 7 (Table I). Close-up of the epicardial surface of the right atrial appendage (RAA), right ventricle (RV) and pulmonary trunk (PT) demonstrating thickened epicardium and fibrinous deposits. b, Patient 12 (Table I). Photomicrograph of a portion of right ventricular surface showing thickened epicardium (E) and overlying fibrinous deposits (F). M = myocardium. Movat stain; magnification X 35, reduced by 16 percent.
thickened in 12 patients: right atrium in eight; right ventricle in 10 [Figure 5); left atrium in none and left ventricle in three. This thickening resulted from proliferation of fibrous tissue, containing some elastic fibers. One or more cardiac valves were focally thickened by fibrous tissue (often with substantial numbers of elastic fibers) in 13 patients; tricuspid in eight; pulmonic in two, mitral in 13 [Figure 6) and aortic in eight. The thickening in all patients was relatively mild and not sufficient
to produce valvular dysfunction. Neither of the two patients with the carcinoid syndrome had carcinoid heart disease
tigure 3-‘ Patient 12 (Table I). Transmural replacement fibrosis of the right ventricle (a) and of the left ventricle (b). The left ventricle is abnormally thinned. Movat stains; magnification each X 21, reduced by 32 percent.
522
March 1981 The American Journal of Medicine
1461. Six of the 16 study patients had one or more of their four major (right, left main, left anterior descending, left circumflex) epicardial (extramural) coronary arteries narrowed >75 percent in cross-sectional area by atherosclerotic plaque (Figures 7 and 8): only one of the four major arteries was so narrowed in one patient, two arteries in one patient: three arteries in three patients; and four arteries in one patient. Thus, of the 64 major coronary arteries in the 16 study patients, the lumens in 16 (25 percent] were narrowed from 76 to 100 percent in cross-sectional area, an average of one of four major coronary arteries per patient. Of the 40 major coronary arteries in the 10 control subjects, only one was narrowed from 75 to 100 percent in cross-sectional, area by atherosclerotic plaque. In 10 of the 16 patients and in all 10 control subjects, the four major epicardial coronary arteries were examined in detail as follows. Each was excised intact, roentgenographed and cut transversely (to its longitudinal axis) into 5-mm long segments. Each segment was labelled sequentially from either its aortic ostium or from its origin from the left main artery. The 5-mm segments were labeled, dehydrated [alcohol and xylene), embedded in paraffin, and two histologic sections were cut and stained from each paraffin block. The Movat stain was used on one histologic section, and all determinations of luminal narrowing were based on examination of the Movat-stained sections. The degrees of narrowing were based on histologic examination by two of us of each cross-section magnified 25 to 50 times. The judgment regarding the degree of luminal narrowing of each 5-mm segment was based on the degree of luminal obliteration within the luminal circle bordered by the internal elastic membrane. The circle was visually subdivided into four equal-sized quadrants, and the percent of cross-
Volume 70
RADIATION
HEART DISEASE-BROSIUS
ET AL.
Figure 4. Photomicrographs (a and b [boxed-portion of a]) of the right ventricular free wall in Patient 3 (Table I). The amount of interstitial fibrous tissue is increased. An intramural coronary artery (CA) is thickened and narrowed by fibrous tissue. Lower, close-up of a portion of right ventricular myocardium in Patient 6 (Table I) showing increased interstitial fibrous tissue. Hematoxylin and eosin stains, magnification X 11 (a), X 55 (b); and Movat stain; magnification X 256 (lower), reduced by 14 percent.
sectional area luminal narrowing in each 5-mm segment was determined as follows: 0 to 25,26to 50.51to 75 and 76 to 100. Reliability of such assessment in comparison to values obtained by video-planimetry has been shown to be very high [47]. Of the 469 five-mm long segments of major coronary artery examined in the 10 study patients, 28 (6percent) were narrowed from 76 to 100 percent in cross-sectional area by atherosclerotic plaque (controls = 0.2percent]: 103 (22 percent)
were narrowed from 51 to 75 percent (controls = 12 percent); 207 (44 percent) were narrowed from 26 to 50 percent (controls = 67 percept] and 131(28percent) were narrowed from 0 to 25 percent (contrdls = 21 percent) [Figure 9). The amount of narrowing in each of the four categories of narrowing was similar for each of the four major coronary arteries in the study patients [Figure 10); similarly, in the control subjects (Figure 10). A significantly (p <0.02) higher percent of 5-mm segments
Figure 5. Patient 7 (Table I). a, opened right ventricle showing thickened mural endocardium; the pulmonic valve cusps are also mildly thickened. b, photomicrograph of the thickened right ventricular endocardium and interstitial myocardial fibrosis. Movat stain; magnification X 55, reduced by 41 percent.
March 1981
The American
Journal of Medicine
Volume 70
523
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 6. Mitral valves in Patients 12 (a), 13 (b) and 3 (c) (Table I). Each valve, especially the anterior (A) leaflet, is focally thickened by fibrous tissue.
in the proximal halves of the right and left anterior descending coronary arteries in the study patients were narrowed >50 percent in cross-sectional area than in the distal halves of these arteries (Figure 11). In each of the 469 histologic sections of major coronary artery in the 10 study patients (plus 13 sections in Patient 16; total 48.21, and in each of the 438 sections of coronary artery in the 10 control subjects, the composition of intimal, medial and adventitial lesions was determined (Figure 12). Of the 482 sections of coronary artery examined in the 11 study patients, the intima1 lesions in all consisted mainly of fibrous tissue: in 339 sections (70 percent), entirely of fibrous tissue (controls = 365 sections [83 percent]]: in four sections (1 percent], of fibrous tissue plus calcific deposits (controls = 0); in 120 sections (25 percent), of fibrous tissue plus lipid deposits (controls = 68 sections [16 percent]); and in the remaining 19 sections (4 percent), of fibrous tissue, lipid and calcific deposits (controls = five sections [l percent]). Thus, 100 percent (482 sections) of the intimal lesions contained fibrous tissue, 29 percent (134 sections] contained lipid, usually both intracellular (“foam cells”) and extracellular (controls = 18 percent of sections contained lipid), and 5 percent (23 sections) contained calcific deposits (controls = 1 percent]. The internal elastic membrane was disrupted in 187 (39 percent) of 482 sections in the study patients and in 148 (34 percent) of the 438 sections in the control subjects. The media of the coronary arteries in the 482 sections in the 11 study patients was partially replaced by fibrous tissue in 364
Figure 7. Pa and medial and adventitial fibrosis. R = right; LM. = left main; LAD = left anterior descending; LC = left circ. = left obtuse marginal coronary arteries. The proximal portion of the left circumflex artery (LC-1) contains an organized thrombuz (T), just distal to which the artery is narrowed >95 percent in cross-sectional area by atherosclerotic plaque. The left oblique marginal artery is totally occluded as it tunnels into left ventricular myocardium. Movat stains; magnification X 27 (R, LM, LAD, LC-1 and LC-2); and X 35 (LOM), reduced by 38 percent.
524
March 1981
The American
Journal of Medicine
Volume
70
RADIATION
HEART
DISEASE-BROSIUS
ET AL.
Figure 8. Patient 8 (Table I). Major epicardial coronary arteries at sites of maximal luminal narrowing. R = right; LM = left main; LAD = left anterior descending; LC = left circumflex; and LD = left diagonal coronary arteries. Movat stains; magnification each X 20, reduced by 30 percent.
60I
-
50
-
m
Patients (469 Segments)
40
30
T
ns T
I
ns
20
10
o-25
26-50
Percent Luminal Narrowing Athersclerotic
51-75
in Cross-sectional
76-100
Area by
Plaque
Figure 9. Percent of 5 mm segments of all four major coronary arteries narrowed to various degrees in 10 patients with radiation heart disease and 10 control subjects.
March
1981
The American Journal of Medicine
Volume 70
525
RADIATION HEART DISEASE-BROSILJS
ET AL.
No. 5mm Segments PaIlemS LM = LAD= LC = R:
Controls
14 177 101 177
9 153 70 206
469
430
-TOWS =
0
PATIENTS
1 o-25
25-50
I 51-75
75-100
0.25
25-50
Percent Luminal Narrowng onCross-secttonal Area by Atlwrsclwot~c Plaque I
igure 10. Percent of 5 mm segments of the right (R), left main (LM), left anterior descending (LAD) and left circumflex (LC) coronary arteries narrowed to four categories of narrowing in 10 patients with radiation heart disease and in 10 control subjects.
sections (76 percent) (controls = 43 of 438 sections [lo percent]; p
526
March 1981 The American Journal of Medicine
COMMENTS Each of the 16 study patients described herein was young (aged 15 to 33 years [mean 26 years]], each had received >3,500 rads to the heart from five to 144 months before death for therapy of a malignant neoplasm, and the heart in each was studied in detail at necropsy. During life, five had clinical evidence of myocardial constriction from excessive pericardial fluid (tamponade), requiring pericardiocentesis in four and pericardiectomy in two: an additional patient, who during life never had evidence of cardiac disease, died suddenly and necropsy disclosed his death to be the result of severe coronary arterial narrowing. Thus, six of the 16 patients had clinical evidence of cardiac dysfunction. Although only six of our 16 patients had clinical evidence of cardiac dysfunction, at necropsy all had anatomic evidence of cardiac abnormality. All but one had fibrous thickening of the pericardium, two had fibrous pericardial adhesions, and eight had excessive pericardial fluid. The mural ‘or valvular endocardium or both also was thickened focally by fibrous tissue in 14 of ihe 16 patients, and the fibrous tissue located between myocardial cells (interstitial myocardial fibrosis) was
Volume 70
RADIATION
HEART
ET AL.
DISEASE-BRQSIUS
60
p <.005
Right
Left Anterior
Left Circumflex
Totals
Descending
L Fi
Coronary Artery
igure 11. Percent of 5 mm segments of the proximal and distal halves of the right, left anterior descending and left circumflex coronary arteries in the 10 study patients with radiation heart disease.
increased in amount in half the patients. The increase in fibrous tissue in all three layers of the ventricular wall (epicardium, myocardium and endocardium) was much more extensive and much more frequent in the right ventricle than in the left ventricle presumably because of higher radiation doses to the anterior surface of the heart. Although the increase in the fibrous tissue in the myocardial wall and mural endocardium caused no apparent clinical evidence of cardiac dysfunction in our patients, it is conceivable that the ventricular walls would have become more rigid or less compliant as the young patients aged. Likewise, the focal fibrous thickening of the cardiac valves in our patients produced no apparent valvular dysfunction, but had they lived for several decades longer, valvular dysfunction may have been the consequence. Indeed, in at least four patients mitral or aortic regurgitation or both have been reported to have developed after high-dose irradiation to the heart [9,10,22]. Although the increase in fibrous tissue in epicardium, myocardium and mural endocardium in our patients was far more extensive in the right ventricle than in the left ventricle, the fibrous thickening of the valves on the left side of the heart was decidedly
March
greater than that on the right-side. The fibrous thickening of both the mural and valvular endocardium, however, in our patients was relatively mild and could have been overlooked had not it been looked for specifically. Thickening of mural endocardium after high-dose irradiation has been described only once previously [16]. Although mild, it was unequivocal in our patients. The most important contribution of the present study is the detailed information regarding the epicardial coronary arteries. Damage to these vessels by high-dose irradiation has been reported previously [l&29-411, but the extent of, and the type of, damage to these arteries have not been described. It is now clear that high-dose irradiation can cause damage to the epicardial coronary arteries and allow intimal proliferation of mainly fibrous tissue to produce luminal narrowing. Of our 16 patients, 16 of the 64 major [right, left main, left anterior descending and left circumflex] epicardial coronary arteries were narrowed >75 percent in cross-sectional area, primarily by fibrous plaques. In contrast, of 10 control subjects of similar age and sex, only one of 40 major epicardial coronary arteries were similarly nar-
1961
The American Journal of Medicine
Volume
70
527
RADIATION HEART DISEASE-BROSIUS
I
n*
ET AL.
T
Fibrws
FbmurT*rwe
Icm,
-huma
gure 12. Morphology of coronary arterial lesions in radiation heart disease: percent of 5 mm segments of the four major epicardial coronary arteries showing various changes in the intima, media and adventitia in 11 study patients and 10 control subjects.
rowed. Of 469 five-milliliter segments of the four major epicardial coronary arteries in 10 study patients, 28 (6 percent) were narrowed >75 percent in cross-sectional area (controls = one of 438 sections [p = 0.06]) and another 22 percent were narrowed 51 to 75 percent (controls = 13 percent]. Thus, our study patients clearly had more coronary narrowing than did the control subjects, and none had pronounced recognized risk factors to premature atherosclerosis. The dominant component of the atherosclerotic plaques in both our study patients and in the control subjects was fibrous tissue: very little lipid was present in either. The study patients, however, had a striking loss of smooth muscle cells from the media (76 percent of sections versus 10 percent of sections in the controls]. Furthermore, adventitial fibrosis was noted in 49 percent of the coronary sections in the study patients and in only 3 percent of the sections in the control subjects. Thus, high dose radiation causes coronary luminal narrowing, particularly in the proximal portions of the major coronary arteries (possibly because they receive more rads than the distal portions], adventitial scarring and damage to the smooth muscle cells in the media.
528
March 1981
The American Journal of Medicine
The minor abnormalities in the intramural coronary arteries, mainly in the right ventricular wall, in the study patients do not appear sufficient to produce functional abnormality. Finally, a number of the 16 study patients were irradiated with techniques now considered unacceptable by most modern radiotherapists (viz., anterior-only ports, tumor doses >4,000 rads, 2 megavoltage radiation sources, multiple courses of radiotherapy). Indeed, many centers now utilize a “sub-carinal block’ to shield the heart from radiation, after a total tumor dose of 3,000 rads is reached [26]. Thus, it might be argued that the types of radiation damage herein demonstrated, will be diminished or eliminated through use of modern techniques. Although such a trend is to be hoped for, in four of the 16 patients radiation doses to both the anterior and posterior surfaces of the heart were <4,200 rad: radiation levels considered acceptable by modern therapists. Therefore, it still seems likely that at least in some young people who receive radiation to the heart, cured of their primary tumor, anatomic and functional cardiac abnormalities will ultimately develop due to the radiation therapy.
Volume 70
RADIATION HEART DISEASE-BROSIUS
ACKNOWLEDGMENT We are grateful to A. S. Lichter, M.D., National Cancer Institute, Bethesda, Maryland, for assistance in calculating radiation doses and interpreting radiotherapy techniques. APPENDIX The radiation doses to the anterior and posterior surfaces of the heart were determined as follows: (1) The posterior surface was considered to be at the mid-plane of the mediastinum (8 to 11 cm posterior to the chest surface) and, hence, to have absorbed the dose denoted in the patient’s records as the “midplane” or “tumor” dose. (2) The anterior surface was considered to be 3 cm posterior to the chest surface. $3)The fractional doses for both the posterior and anterior surfaces were deter-
ET AL.
mined by matching the “equivalent area” of each of the radiation ports with the distance from the radiation source on a standard fractional depth dose data table for a 6 megavoltage photon beam source (Radiation Oncology Branch, National Cancer Institute). (4) The a!solute radiation dose to the anterior surface thus equaled the fractional dose to the anterior heart surface divided by the fractional dose to the posterior heart surface times the absolute dose to the posterior heart surface [or “mid-plane dose”). (5) For nine patients who had been treated with a 2 megavoltage radiation source, the d,ose calculated for the anterior surface of the heart was multiplied [for anterior ports) or divided (for posterior ports) by a factor of 1.2 to compensate for the lower penetrance of x-rays from this source compared to those from a 6 megavoltage source. The doses thus calculated were considered to represent maximum possible dose.
REFERENCES 1. Desjardins
2.
3.
4. 5. 6.
7. 8. 9. 10. 11.
12.
13. 14.
15. 16. 17.
AU: Action of roentgen rays and radium on the heart and lungs. Am J Roentgen01 1932; 27: 153-176, 303-335,477-495. Leach JE: Effect of roentgen therapy on the heart: clinical study. Arch Int Med 1943; 72: 715-745. Fried JR, Goldberg H: Post-irradiation changes in the lungs and thorax. A clinical, roentgenological and pathological study, with emphasis on late and terminal stages. Am J Roentgen01 1940; 43: 877-895. Blumenfeld H, Thomas SF: Chronic massive pericardial effusion following roentgen therapy for carcinoma of the breast. Radiology 1945; 44: 335-340. Tricot R. Baillet J, Helmke G: La pericardite constrictive oost-radiotheraniaue 18 orooos d’un cas oersonnell. Arch kal Coeur 1954; 4j: 9i2-94& Gimlette TMD: Constrictive oericarditis. Br Heart 1 1959: 21: 9-E. Hurst DW: Radiation fibrosis of pericardium with cardiac tamponade. Can Med Assoc J 1959; 81: 377-380. Connolly DC, Buichell HB: Pericardltis: a 10 year survey. Am J Cardiol1961: 7: 7-13. Steinberg I: Effusive-constrictive radiation pericarditis. Two cases illustrating value of anglocardiography in diagnosis. Am J Cardiol1987; 19: 434-439. Cohn KE, Stewart JR, Fajardo LF, Hancock EW: Heart disease following radiation. Medicine (Baltimore] 1967; 46: 281-298. Stewart JR. Cohn KE. Fajardo LF, Hancock EW, Kaplan HS: Radiation-induced heart disease. Radiology -. 1967; 89: 302-310. Teng CY, Nemickas R, Tobin JR Jr, Szanto PB, Gunnar RM: Pericardial effusion follow& radiation to the chest. Report of a case. Dis Chest 1967; 527549-552. Muggia FM, Cassileth PA: Constrictive pericarditis following radiation therapy. Am J Med 1968; 44: 116-123. Roberts WC, Glancy DL, DeVita VT: Heart in malignant lymphoma (Hodgkin’s disease, lymphosarcoma, reticulum cell sarcoma a’nd mycosis fungoides). A study of 196 autopsy cases. Am J Cardiol1968; 2.2:85-107. Masland DS, Rotz CT Jr. Harris JH Jr: Postradiation pericarditis with chronic pericardial effusion. Ann Intern Med 1968; 69: 97-102. Fajardo LF, Stewart JR, Cohn KE: Morphology of radiationinduced heart disease. Arch Path01 1968; 86: 512-519. Morton DL, Kagan AR, Roberts WC, O’Brien KP, Holmes EC, Adkins PC: Pericardiectomy for radiation-induced pericarditis with effusion. Ann Thorac Surg 1969; 8: 195%208.
18.
19.
20.
21.
22.
23. 24.
25. 26. 27. 28. 29. 30. 31. 32.
Haas JM: Symptomatic constrictive pericarditis developing 45 years after radiation therapy to the mediastinum. A review of radiation pericarditis. Am Heart J 1969: 77: 8895. Pierce RH, Hafermann MI& Kagan AR: Changes in the transverse cardiac diameter following mediastinal irradiation for Hodgkin’s disease. Radiology 1969; 93: 619624. Lawson RAM, Ross WM, Gold RG. Blesovsky A, Barnsley WC: Post-radiation pericarditis. Report on four more cases with special reference to bronchogenic carcinoma. J Thorac Cardiovasc Surg 1972; 63: 841-847. Kagan AR, Hafermann M, Hamiton M. Pierce R, Morton D. johnson R: Etiology, diagnosis, and management of pericardial effusion after irradiation. Radio1Clin Biol197.2: 41: 171-182. Morton DL, Glancy DL, Joseph WL. Adkins PC: Management of patients with radiation-induced pericarditis with effusign: a note on the development of aortic regurgitation in two of them. Chest 1973: 64: 291-297. Martin RG, Rudkdeschel JC, Chang P, Byhardt R, Bouchard RJ, ‘Wiernik PH: Radiation-related pericarditis. Am J CardiollQ75; 35: 216-220. Byhardt R, Brace K. Ruckdeschel J, Chang P, Martin R, Wiernik PH: Dose and treatment factors in radiation related pericardial effusion associated with the mantle technique for Hodgkin’s disease. Cancer 1975; 35: 795802. Rudkdeschel JC. Chang P, Martin RG, et a].: Radiation-related pericardial effusions in patients with Hodgkin’s disease. Medicine [Baltimore] 1975.54: 245-259. Carmel RJ, Kaplan HS: Mantle irradiation in Hodgkin’s disease. Cancer 1976; 37: 2813-2825. Green B, Zornoza J, Ricks JP: Eccentric pericardial effuqion after radiation therapy of left breast carcinoma. Am J Roentgen01 1977; 128: 27-30. Schneider JS, Edwards JE: Irradiation-induced pericerditis. Chest 1979; 75: 560-564. Pearson HES: Coronary occlusion following thoracic radiotherapy, 2 cases. Proc R Sot Med 1957; 50: 516. Dollinger MR, Lavine DM, Foye LV Jr: Myocardial infarction following radiation (letter to the editor). Lancet 1965; 2: 246. Prentice RTW: Myocardial infarction following radiation [Letter to the Editor). Lancet 1965; 2: 388. Dollinger MR. Lavine DM, Foye LV Jr: Myocardial infarction due to post-irradiation fibrosis of the coronary arteries.
March 1981 The American Journal of Medicine
Volume 70
529
RADIATION HEART DISEASE-BROSIUS
33. 34. 35. 36. 37. 38. 39. 40.
530
ET AL
Case of successfully treated Hodgkin’s disease with lower esophageal involvement. JAMA 195: 316-319.1966. Huff H. Sanders EM: Coronarv-arterv occlusion after radiation. N Engl J Med 1972; 286: 780: Tracy GP. Brown DE, Johnson LW, Gottlieb AJ: Radiationinduced coronary artery disease. JAMA 1974; 228: 16601662. Rinerose. TL: Radiation-“induced” coronarvI arterv m, disease. ~~~ JiMA.1974; 230: 960-961. McRevnolds RA. Gold GL. Roberts WC: Coronarv heart disease after mediastinal irradiation for Hodgkin’s disease. Am J Med 1976; 60: 39-45. Ali MK, Khalil KG, Fuller LM, et al.: Radiation-related myocardial injury: management of two cases. Cancer 1976; 38: 1941-1946. Rodgers DL: Precocious myocardial infarction after radiation treatment for Hodgkin’s disease. Chest 1976; 70: 675-677. Murphy GK: Cancer and the coroner. JAMA 1977; 237: 786-788. Iqbal SM, Hanson EL, Gensini GG: Bypass graft for coronary artery stenosis following radiation therapy. Chest 1977; 71: 664-666.
March 1981
The American Journal of Medicine
41. Rasmussen
42. 43. 44. 45. 46. 47.
Volume 70
S, Dfissing M. Walbom-Jdrgensen S: Coronary heart disease-a possible risk in mega voltage therapy? Acta Med Stand 1978; 203: 237-239. Hartman FW, Bollinger A, Doub HP, Smith FJ: Heart lesions produced by deep x-rays. An experimental and clinical study. Bull Johns Hopkins Hosp 1927;41: 36-67. Fajardo LF, Stewart JR: Pathogenesis of radiation-induced myocardial fibrosis. Lab Invest 1973; 29: 244-257. DiMatteo 1, Vacheron A. Heulin A. et al.: Comolications cardiques de la radiotherapie thoracique. Arch Ma1 Coeur 1978; 71: 447-452. Bouvrain Y, Roujeau J, Rogard JM: Myocardite post-radiotherapique. Sem Hop Paris 1977:53: 433-436. Roberts WC, Sjoerdsma A: The cardiac disease associated with the carcinoid syndrome (carcinoid heart disease]. Am J Med 1964; 36: 5-34. Isner JM, Wu M, Virmani R, Jones AA, Roberts WC: Comparison of degrees of coronary arterial luminal narrowing determined by visual inspection of histologic sections under magnification among three independent observers and comparison to that obtained by video planimetry. An analysis of 559 five-mm segments of 61 coronary arteries from eleven patients. Lab Invest 1980; 68: 831-838.
FRANK C. BROSIUS. III. M.D.* BRIJCE F. WALLER, M.D. WILLIAM C. ROBERTS, M.D. Rethesdo. &turyland
From the Pathology Branch, National Heart, Lung, and Blood Institute. National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Dr. William C. Roberts, Building 10A. Room SE-30, National Institutes of Health, Bethesda. MD 20205. Manuscript accepted September 11.1980. l Present address: Department of Medicine, University of Michigan Medical Center, Ann Arbor, MI 48104.
Certain clinical and necropsy findings are described in 16 young (aged 16 to 33 years) patients who received >3,500 rads to the heart Rve to 144 months before death. All 16 had some radiation-induced damage to the heart: 15 had thickened pericardia (Rve of whom had evidence of cardiac tamponadek eight had increased interstitial myocardial fibrosis, particularly in the right ventricle; 12 had Rbrous thickening of the mural endocardium and 13 of the valvular endocardium. Except for valvular thickening, the changes were more frequent in the right side of the heart than in the left, presumably because of higher radiation doses to the anterior surface of the heart. In six of the 16 study patients and in one of 10 control subjects, one or more major epicardial coronary arteries were narrowed from 76 to 100 percent in cross-sectional area by atherosclerotic plaque; one patient had a healed myocardial infarct at necropsy and one died suddenly. In 10 patients and in the 10 control subjects, the four major epicardial coronary arteries were examined quantitatively: 6 percent of the 469 five millimeter segments of coronary artery from the patients were narrowed from 76 to 100 percent (controls = 0.2 percent, p,,=,,“.O6)and 22 percent were narrowed from 51 to 75 percent (con= 12 percent). The proximal portion of the arteries in the patients had signiRcantly more narrowing than the distal portions. The arterial plaques in the patients were largely composed of fibrous tissue; the media were frequently replaced by Rbrous tissue, and the adventitia were often densely thickened by fibrous tissue. In five patients, there was focal thickening (with or without luminal narrowing) of the intramural coronary arteries. Thus, radiation to the heart may produce a wide spectrum of functional and anatomic changes but particularly damage to the pericardia and the underlying epicardial coronary arteries. In the early part of this century, the heart was considered a “radioresistant” organ [1,2]. With the introduction of megavoltage radiotherapy for treatment of neoplasms, however, it became apparent that the heart could, indeed, be damaged by high-dose radiation. The nature and extent of the damage, however, in both human subjects and in nonhuman experimental animals are still debated. Pericardial thickening with or without effusion or constriction has been confirmed from several centers [3-281. In several “case reports” clinical or necropsy evidence of myocardial ischemia and coronary arterial narrowing has been described in young persons who had received highdose mediastinal radiation [10.29-411. Increased myocardial fibrosis
March 1981 The American Journal of Medicine
Volume 70
519
RADIATION
HEART DISEASE-BROSIUS
ET AL.
has been described as a consequence of high-dose mediastinal irradiation [3,10,11,16,25,42-461. The effect of high-dose radiation on the mural and valvular endocardium has received virtually no attention. Furthermore, the extensiveness of the coronary disease has not been described previously in young necropsy patients who received megavoltage therapy to the mediastinum. In the report which follows, the effect of high-dose irradiation in 16 young necropsy patients is described. Particular attention is given to the mural and valvular endocardial damage and to the extent of the coronary narrowing that were found. PATIENTS
STUDIED
In 15 patients the apical portion of the heart was excluded from the direct radiation field. The interval from the end of the irradiation period to death ranged from five to 144 months [mean 55 months]. Ten patients received one or more chemotherapeutic agents, including six who received prednisone (for one to 12 six-week periods] and three who received doxorubicin. None had cushingoid features at any time. No patient had clinical evidence of chronic congestive cardiac failure and none had chest pain which could have been interpreted as evidence of myocardial ischemia. Six patients smoked cigarettes (10 to 20 daily). Electrocardiograms, recorded in 14 patients (all after irradiation), disclosed sinus rhythm in all, nonspecific S-T segment or T wave changes in nine, low (sum of QRS complexes in leads I, II and III 515 mv) voltage in five (four had pericardial effusions) and complete right bundle branch block in one. Five patients had large cardiac silhouettes on chest roentgenograms [Figure l], and five had clinical findings (distended neck veins, subcutaneous edema, paradoxical pulse and Kussmaul’s sign) of cardiac tamponade. Four of the five patients with clinical features of tamponade had pericardiocentesis with removal oflarge quantities of fluid [Figure 1). and two patients (Patients 9 and 12, Table I] subsequently had partial pericardiectomies because of recurring signs of cardiac tamponade. Blood hematocrit values about one month before death ranged from 20 to 48 percent [mean 33 percent): serum total protein, 3.5 to 7.5 g/d1 (mean 5.5 g/d]); and serum albumin, from 1.3 to 4.9 g/d1 (mean 2.8 g/dl). The highest total serum
AND FINDINGS
To be included in this study, patients had to be under 35 years of age at death, had to have had an autopsy, had to have received a total of more than 3,500 rads to the heart, and the heart specimens had to be examined by at least one of us. Sixteen patients fulfilled these criteria. Certain clinical and necropsy observations in them are summarized in Table I. The patients ranged in age from 15 to 33 years (mean 26 years) at death; 13 were men and three were women (Patients 6,10, and 13. Table I). Thirteen had Hodgkin’s disease; two (Patients 8 and 151, carcinoid; and one (Patient 10). histiocytic lymphoma. The number of rads administered to the anterior surface of the heart ranged from 3,987 to 8,769 [mean 5,592). and to the posterior surface, from 3,000 to 5,000 [mean 3,932). (The method for calculating the radiation dose is described in the Appendix.]
TABLE I
Clinical and Necropsy Findings in 16 Patients with Radiation Heart Disease
PaSent
A@ (Yr)
Aoe &I at Dx
1 2 3 4 5 6 7 6 9 10 11 12 13 14 15 16
15 19 24 24 25 25 26 26 27 29 29 29 30 30 33 33
12 16 23 20 22 23 19 24 20 27 17 21 25 27 20 24
Calculated Rad to Heart Ani Post
5,202 5,648 5,842 5,295 5,186 4,100 4,165 3,987 7,047 4,932 4,054 4,341 5,419 8,769 6,600 6,866
3,000 3,600 3,500 3,600 3,092 4,000 3,000 3,600 4,000 4,200 4,000 4,025 4,495 4,800 5,000 5,000
MO.from Rad to Death 34 35 22 42 39 11 78 5 53 20 144 84, 825 44 27,2§ 137 104
CT
TC’
0 0 + + 0 0 0 0 + 0
129 139 140 120 172 176 146 250 198 168 168
0 +
0 + 0 0
;;7 130 170 308
Hctt (%I
HW (9)
43 20 39 29 30 37 21 34
170 270 245 190 290 250 250 290 370 155 240 225 170
.. 31 27 27 35 31 45 48
230 350
TE p6 3oot 0 100 260 75 0 200 0 0 0 0 0 0 375 0 0
M
V
+ + + + 0 + + + 0 + + + + + + + + + + + 0 0 + + + + •t0 0 0 +++0+13
RV
IMF VS
+ 0 + 0 + + + 0 0 + ‘$’ + 0 0 0 o...o + 0 o...o 0 0 0 0 0 0 0 0
No. S-mm No. S-mm S TICA s >75% LV CAs XSA RV LV 0 0 + 0 0 ‘4’ 0 0 0 0 0 0
. so 54 44 55
.
50 44 42 37 43 ...
‘0 0 2 0 .. 8 0 0 10 12 ...
‘50
‘i
+ 0 + + 0 + 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0
0
5++
= no data; CA = coronary artery; CT = cardiac tamponade: Dx = diagnosis; ECG = electroNOTE: + = present; 0 = absent or none; . cardiogram: Hct = hematocrit; HW = heart weight; IMF = interstitial myocardial fibrosis; LV = left ventricle; M = mural; PE = pericardial effusion; RV = right ventricle; S = segments; TC = total serum cholesterol; TE = grossly thickened endocamium; TICA = thickened intramural coronary artery; V = valve; VS = ventricular septum: XSA = cross-sectional area. + Highest recorded value. t Value about one month before death. t Number of milliliters. 5 Received two courses of radiotherapy.
520
March 1981
The American
Journal of Medicine
Volume
70
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 1. Patient 14 (Table I). Chest radiogram after pericardiocentesis and injection of air into the pericardial sac and posterior half of the heart and lungs (left), and after they had been sectioned longitudinally (right). The pericardia are thickened primarily by fibrous tissue.
cholesterol values 115 patients) ranged from 120 to 308 mg/dl (mean 173 mg/dl); this value was >180 mg/dl in four patients. An additional 10 patients matched closely for age and sex to the study patients were used as control subjects. Certain characteristics of them are shown in Table II. None received therapeutic mediastinal irradiation, none ever had symptoms of cardiac dysfunction, all died from noncardiac conditions, and seven received one or more chemotherapeutic agents, including six who received prednisone (for one or more sixweek periods) and one who received doxorubicin. At necropsy, fluid was present in the pericardial sacs in 12 of the 16 study patients: in six, the amount was <50 ml: in the other six, 75 to 375 ml. The remaining four patients [Patients 9, 10, 12 and 13, Table I) had pericardial fibrous adhesions which obliterated the pericardial spaces, but two had had partial pericardiectomies. Four patients had fibrinous pericardial adhesions (Figure 2). In all but one patient the epicardial surface was thickened by fibrous tissue [Figure 2). The
hearts ranged in weight from 155 to 370 g (mean 246 g). None of the 16 patients had grossly visible or histologically discernible foci of myocardial coagulation necrosis. Only one patient [Patient 12, Table I, Figure 3) had a grossly visible scar on the ventricular wall at necropsy; in him it involved the posterior wall of the left ventricle, and the three major (right, left anterior descending and left circumflex) epicardial coronary arteries were severely narrowed. On histologic examination of at least one histologic section from the wall of each cardiac ventricle (mean 2.5 sections per ventricle], increased interstitial myocardial fibrosis was observed in the right ventricular free wall in eight patients (Figure 4), in the left ventricular free wall in three patients and in the ventricular septum in two patients. In one of the 10 control subjects, mild interstitial myocardial fibrosis was observed [Table II). Unequivocal damage to individual myocardial cells was not detected in any study patient or in any control subject by histologic study. The mural endocardium of one or more chambers was
Clinical and Necropsy Findings in 10 Control Subjects
TABLE II
Control Subjects
Age (vr)
bx
1 2 3 4 5 6 7 8 9 10
19 21 22 23 28 28 29 29 33 36
U’ AML ALL ALL CML HL Ch ET MM HS
Age at Dx (v) 19 18 20 19 21 28 28 28 30
.
($dt, Yi 200 178 154 147 2;2 182
.
Hct (%j
Alb (9Aflj
(;:,
ii 31 19 30 34 27 34
7.1 3.8 7.1 3.9 4.4 8.7 8.5 8.1
.
.
b:d 1.6 4.7 1.2 2.3 2.7 2.8 3.1
..
HW (g)
RV
370 280 500 430 220 360 200 270 280
0 0 0 0 0 0 0 0 0
..
+
IMF vs 0 0 0 0 0 0 0 0 0
LV
No. 5-mm Segmenlsof 4 Major CAs
0 0 0 p 0 0 0 .O 0
.
+
----
No. Narrowed >75%
63 22 58 62 42 37 28 46 48 34
0 0 0 0 1 0 0 0 0 0
= no data; Alb = albumin; AU = acute lymphocytic leukemia: AML = acute myelocytic leukemia; NOTE: + = present; 0 = absent or none; . BT = brain tumor; CA = coronary artery; Ch = choriocarcinoma; CML = chronic myelocytic leukemia; Dx = diagnosis; Hct = hamatocrit; HL = histiocytic lymphoma; HS = heat stroke: HW = heart weight; IMF = interstitial myocardial fibrosis; MM = malignant melanoma; TC = total serum cholesterol: U = unknown. * Idiopathic sudden death.
March 1981
The American Journal of Medicine
Volume 79
521
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 2. Fibrinous pericarditis. Patient 7 (Table I). Close-up of the epicardial surface of the right atrial appendage (RAA), right ventricle (RV) and pulmonary trunk (PT) demonstrating thickened epicardium and fibrinous deposits. b, Patient 12 (Table I). Photomicrograph of a portion of right ventricular surface showing thickened epicardium (E) and overlying fibrinous deposits (F). M = myocardium. Movat stain; magnification X 35, reduced by 16 percent.
thickened in 12 patients: right atrium in eight; right ventricle in 10 [Figure 5); left atrium in none and left ventricle in three. This thickening resulted from proliferation of fibrous tissue, containing some elastic fibers. One or more cardiac valves were focally thickened by fibrous tissue (often with substantial numbers of elastic fibers) in 13 patients; tricuspid in eight; pulmonic in two, mitral in 13 [Figure 6) and aortic in eight. The thickening in all patients was relatively mild and not sufficient
to produce valvular dysfunction. Neither of the two patients with the carcinoid syndrome had carcinoid heart disease
tigure 3-‘ Patient 12 (Table I). Transmural replacement fibrosis of the right ventricle (a) and of the left ventricle (b). The left ventricle is abnormally thinned. Movat stains; magnification each X 21, reduced by 32 percent.
522
March 1981 The American Journal of Medicine
1461. Six of the 16 study patients had one or more of their four major (right, left main, left anterior descending, left circumflex) epicardial (extramural) coronary arteries narrowed >75 percent in cross-sectional area by atherosclerotic plaque (Figures 7 and 8): only one of the four major arteries was so narrowed in one patient, two arteries in one patient: three arteries in three patients; and four arteries in one patient. Thus, of the 64 major coronary arteries in the 16 study patients, the lumens in 16 (25 percent] were narrowed from 76 to 100 percent in cross-sectional area, an average of one of four major coronary arteries per patient. Of the 40 major coronary arteries in the 10 control subjects, only one was narrowed from 75 to 100 percent in cross-sectional, area by atherosclerotic plaque. In 10 of the 16 patients and in all 10 control subjects, the four major epicardial coronary arteries were examined in detail as follows. Each was excised intact, roentgenographed and cut transversely (to its longitudinal axis) into 5-mm long segments. Each segment was labelled sequentially from either its aortic ostium or from its origin from the left main artery. The 5-mm segments were labeled, dehydrated [alcohol and xylene), embedded in paraffin, and two histologic sections were cut and stained from each paraffin block. The Movat stain was used on one histologic section, and all determinations of luminal narrowing were based on examination of the Movat-stained sections. The degrees of narrowing were based on histologic examination by two of us of each cross-section magnified 25 to 50 times. The judgment regarding the degree of luminal narrowing of each 5-mm segment was based on the degree of luminal obliteration within the luminal circle bordered by the internal elastic membrane. The circle was visually subdivided into four equal-sized quadrants, and the percent of cross-
Volume 70
RADIATION
HEART DISEASE-BROSIUS
ET AL.
Figure 4. Photomicrographs (a and b [boxed-portion of a]) of the right ventricular free wall in Patient 3 (Table I). The amount of interstitial fibrous tissue is increased. An intramural coronary artery (CA) is thickened and narrowed by fibrous tissue. Lower, close-up of a portion of right ventricular myocardium in Patient 6 (Table I) showing increased interstitial fibrous tissue. Hematoxylin and eosin stains, magnification X 11 (a), X 55 (b); and Movat stain; magnification X 256 (lower), reduced by 14 percent.
sectional area luminal narrowing in each 5-mm segment was determined as follows: 0 to 25,26to 50.51to 75 and 76 to 100. Reliability of such assessment in comparison to values obtained by video-planimetry has been shown to be very high [47]. Of the 469 five-mm long segments of major coronary artery examined in the 10 study patients, 28 (6percent) were narrowed from 76 to 100 percent in cross-sectional area by atherosclerotic plaque (controls = 0.2percent]: 103 (22 percent)
were narrowed from 51 to 75 percent (controls = 12 percent); 207 (44 percent) were narrowed from 26 to 50 percent (controls = 67 percept] and 131(28percent) were narrowed from 0 to 25 percent (contrdls = 21 percent) [Figure 9). The amount of narrowing in each of the four categories of narrowing was similar for each of the four major coronary arteries in the study patients [Figure 10); similarly, in the control subjects (Figure 10). A significantly (p <0.02) higher percent of 5-mm segments
Figure 5. Patient 7 (Table I). a, opened right ventricle showing thickened mural endocardium; the pulmonic valve cusps are also mildly thickened. b, photomicrograph of the thickened right ventricular endocardium and interstitial myocardial fibrosis. Movat stain; magnification X 55, reduced by 41 percent.
March 1981
The American
Journal of Medicine
Volume 70
523
RADIATION HEART DISEASE-BROSIUS
ET AL.
Figure 6. Mitral valves in Patients 12 (a), 13 (b) and 3 (c) (Table I). Each valve, especially the anterior (A) leaflet, is focally thickened by fibrous tissue.
in the proximal halves of the right and left anterior descending coronary arteries in the study patients were narrowed >50 percent in cross-sectional area than in the distal halves of these arteries (Figure 11). In each of the 469 histologic sections of major coronary artery in the 10 study patients (plus 13 sections in Patient 16; total 48.21, and in each of the 438 sections of coronary artery in the 10 control subjects, the composition of intimal, medial and adventitial lesions was determined (Figure 12). Of the 482 sections of coronary artery examined in the 11 study patients, the intima1 lesions in all consisted mainly of fibrous tissue: in 339 sections (70 percent), entirely of fibrous tissue (controls = 365 sections [83 percent]]: in four sections (1 percent], of fibrous tissue plus calcific deposits (controls = 0); in 120 sections (25 percent), of fibrous tissue plus lipid deposits (controls = 68 sections [16 percent]); and in the remaining 19 sections (4 percent), of fibrous tissue, lipid and calcific deposits (controls = five sections [l percent]). Thus, 100 percent (482 sections) of the intimal lesions contained fibrous tissue, 29 percent (134 sections] contained lipid, usually both intracellular (“foam cells”) and extracellular (controls = 18 percent of sections contained lipid), and 5 percent (23 sections) contained calcific deposits (controls = 1 percent]. The internal elastic membrane was disrupted in 187 (39 percent) of 482 sections in the study patients and in 148 (34 percent) of the 438 sections in the control subjects. The media of the coronary arteries in the 482 sections in the 11 study patients was partially replaced by fibrous tissue in 364
Figure 7. Pa and medial and adventitial fibrosis. R = right; LM. = left main; LAD = left anterior descending; LC = left circ. = left obtuse marginal coronary arteries. The proximal portion of the left circumflex artery (LC-1) contains an organized thrombuz (T), just distal to which the artery is narrowed >95 percent in cross-sectional area by atherosclerotic plaque. The left oblique marginal artery is totally occluded as it tunnels into left ventricular myocardium. Movat stains; magnification X 27 (R, LM, LAD, LC-1 and LC-2); and X 35 (LOM), reduced by 38 percent.
524
March 1981
The American
Journal of Medicine
Volume
70
RADIATION
HEART
DISEASE-BROSIUS
ET AL.
Figure 8. Patient 8 (Table I). Major epicardial coronary arteries at sites of maximal luminal narrowing. R = right; LM = left main; LAD = left anterior descending; LC = left circumflex; and LD = left diagonal coronary arteries. Movat stains; magnification each X 20, reduced by 30 percent.
60I
-
50
-
m
Patients (469 Segments)
40
30
T
ns T
I
ns
20
10
o-25
26-50
Percent Luminal Narrowing Athersclerotic
51-75
in Cross-sectional
76-100
Area by
Plaque
Figure 9. Percent of 5 mm segments of all four major coronary arteries narrowed to various degrees in 10 patients with radiation heart disease and 10 control subjects.
March
1981
The American Journal of Medicine
Volume 70
525
RADIATION HEART DISEASE-BROSILJS
ET AL.
No. 5mm Segments PaIlemS LM = LAD= LC = R:
Controls
14 177 101 177
9 153 70 206
469
430
-TOWS =
0
PATIENTS
1 o-25
25-50
I 51-75
75-100
0.25
25-50
Percent Luminal Narrowng onCross-secttonal Area by Atlwrsclwot~c Plaque I
igure 10. Percent of 5 mm segments of the right (R), left main (LM), left anterior descending (LAD) and left circumflex (LC) coronary arteries narrowed to four categories of narrowing in 10 patients with radiation heart disease and in 10 control subjects.
sections (76 percent) (controls = 43 of 438 sections [lo percent]; p
526
March 1981 The American Journal of Medicine
COMMENTS Each of the 16 study patients described herein was young (aged 15 to 33 years [mean 26 years]], each had received >3,500 rads to the heart from five to 144 months before death for therapy of a malignant neoplasm, and the heart in each was studied in detail at necropsy. During life, five had clinical evidence of myocardial constriction from excessive pericardial fluid (tamponade), requiring pericardiocentesis in four and pericardiectomy in two: an additional patient, who during life never had evidence of cardiac disease, died suddenly and necropsy disclosed his death to be the result of severe coronary arterial narrowing. Thus, six of the 16 patients had clinical evidence of cardiac dysfunction. Although only six of our 16 patients had clinical evidence of cardiac dysfunction, at necropsy all had anatomic evidence of cardiac abnormality. All but one had fibrous thickening of the pericardium, two had fibrous pericardial adhesions, and eight had excessive pericardial fluid. The mural ‘or valvular endocardium or both also was thickened focally by fibrous tissue in 14 of ihe 16 patients, and the fibrous tissue located between myocardial cells (interstitial myocardial fibrosis) was
Volume 70
RADIATION
HEART
ET AL.
DISEASE-BRQSIUS
60
p <.005
Right
Left Anterior
Left Circumflex
Totals
Descending
L Fi
Coronary Artery
igure 11. Percent of 5 mm segments of the proximal and distal halves of the right, left anterior descending and left circumflex coronary arteries in the 10 study patients with radiation heart disease.
increased in amount in half the patients. The increase in fibrous tissue in all three layers of the ventricular wall (epicardium, myocardium and endocardium) was much more extensive and much more frequent in the right ventricle than in the left ventricle presumably because of higher radiation doses to the anterior surface of the heart. Although the increase in the fibrous tissue in the myocardial wall and mural endocardium caused no apparent clinical evidence of cardiac dysfunction in our patients, it is conceivable that the ventricular walls would have become more rigid or less compliant as the young patients aged. Likewise, the focal fibrous thickening of the cardiac valves in our patients produced no apparent valvular dysfunction, but had they lived for several decades longer, valvular dysfunction may have been the consequence. Indeed, in at least four patients mitral or aortic regurgitation or both have been reported to have developed after high-dose irradiation to the heart [9,10,22]. Although the increase in fibrous tissue in epicardium, myocardium and mural endocardium in our patients was far more extensive in the right ventricle than in the left ventricle, the fibrous thickening of the valves on the left side of the heart was decidedly
March
greater than that on the right-side. The fibrous thickening of both the mural and valvular endocardium, however, in our patients was relatively mild and could have been overlooked had not it been looked for specifically. Thickening of mural endocardium after high-dose irradiation has been described only once previously [16]. Although mild, it was unequivocal in our patients. The most important contribution of the present study is the detailed information regarding the epicardial coronary arteries. Damage to these vessels by high-dose irradiation has been reported previously [l&29-411, but the extent of, and the type of, damage to these arteries have not been described. It is now clear that high-dose irradiation can cause damage to the epicardial coronary arteries and allow intimal proliferation of mainly fibrous tissue to produce luminal narrowing. Of our 16 patients, 16 of the 64 major [right, left main, left anterior descending and left circumflex] epicardial coronary arteries were narrowed >75 percent in cross-sectional area, primarily by fibrous plaques. In contrast, of 10 control subjects of similar age and sex, only one of 40 major epicardial coronary arteries were similarly nar-
1961
The American Journal of Medicine
Volume
70
527
RADIATION HEART DISEASE-BROSIUS
I
n*
ET AL.
T
Fibrws
FbmurT*rwe
Icm,
-huma
gure 12. Morphology of coronary arterial lesions in radiation heart disease: percent of 5 mm segments of the four major epicardial coronary arteries showing various changes in the intima, media and adventitia in 11 study patients and 10 control subjects.
rowed. Of 469 five-milliliter segments of the four major epicardial coronary arteries in 10 study patients, 28 (6 percent) were narrowed >75 percent in cross-sectional area (controls = one of 438 sections [p = 0.06]) and another 22 percent were narrowed 51 to 75 percent (controls = 13 percent]. Thus, our study patients clearly had more coronary narrowing than did the control subjects, and none had pronounced recognized risk factors to premature atherosclerosis. The dominant component of the atherosclerotic plaques in both our study patients and in the control subjects was fibrous tissue: very little lipid was present in either. The study patients, however, had a striking loss of smooth muscle cells from the media (76 percent of sections versus 10 percent of sections in the controls]. Furthermore, adventitial fibrosis was noted in 49 percent of the coronary sections in the study patients and in only 3 percent of the sections in the control subjects. Thus, high dose radiation causes coronary luminal narrowing, particularly in the proximal portions of the major coronary arteries (possibly because they receive more rads than the distal portions], adventitial scarring and damage to the smooth muscle cells in the media.
528
March 1981
The American Journal of Medicine
The minor abnormalities in the intramural coronary arteries, mainly in the right ventricular wall, in the study patients do not appear sufficient to produce functional abnormality. Finally, a number of the 16 study patients were irradiated with techniques now considered unacceptable by most modern radiotherapists (viz., anterior-only ports, tumor doses >4,000 rads, 2 megavoltage radiation sources, multiple courses of radiotherapy). Indeed, many centers now utilize a “sub-carinal block’ to shield the heart from radiation, after a total tumor dose of 3,000 rads is reached [26]. Thus, it might be argued that the types of radiation damage herein demonstrated, will be diminished or eliminated through use of modern techniques. Although such a trend is to be hoped for, in four of the 16 patients radiation doses to both the anterior and posterior surfaces of the heart were <4,200 rad: radiation levels considered acceptable by modern therapists. Therefore, it still seems likely that at least in some young people who receive radiation to the heart, cured of their primary tumor, anatomic and functional cardiac abnormalities will ultimately develop due to the radiation therapy.
Volume 70
RADIATION HEART DISEASE-BROSIUS
ACKNOWLEDGMENT We are grateful to A. S. Lichter, M.D., National Cancer Institute, Bethesda, Maryland, for assistance in calculating radiation doses and interpreting radiotherapy techniques. APPENDIX The radiation doses to the anterior and posterior surfaces of the heart were determined as follows: (1) The posterior surface was considered to be at the mid-plane of the mediastinum (8 to 11 cm posterior to the chest surface) and, hence, to have absorbed the dose denoted in the patient’s records as the “midplane” or “tumor” dose. (2) The anterior surface was considered to be 3 cm posterior to the chest surface. $3)The fractional doses for both the posterior and anterior surfaces were deter-
ET AL.
mined by matching the “equivalent area” of each of the radiation ports with the distance from the radiation source on a standard fractional depth dose data table for a 6 megavoltage photon beam source (Radiation Oncology Branch, National Cancer Institute). (4) The a!solute radiation dose to the anterior surface thus equaled the fractional dose to the anterior heart surface divided by the fractional dose to the posterior heart surface times the absolute dose to the posterior heart surface [or “mid-plane dose”). (5) For nine patients who had been treated with a 2 megavoltage radiation source, the d,ose calculated for the anterior surface of the heart was multiplied [for anterior ports) or divided (for posterior ports) by a factor of 1.2 to compensate for the lower penetrance of x-rays from this source compared to those from a 6 megavoltage source. The doses thus calculated were considered to represent maximum possible dose.
REFERENCES 1. Desjardins
2.
3.
4. 5. 6.
7. 8. 9. 10. 11.
12.
13. 14.
15. 16. 17.
AU: Action of roentgen rays and radium on the heart and lungs. Am J Roentgen01 1932; 27: 153-176, 303-335,477-495. Leach JE: Effect of roentgen therapy on the heart: clinical study. Arch Int Med 1943; 72: 715-745. Fried JR, Goldberg H: Post-irradiation changes in the lungs and thorax. A clinical, roentgenological and pathological study, with emphasis on late and terminal stages. Am J Roentgen01 1940; 43: 877-895. Blumenfeld H, Thomas SF: Chronic massive pericardial effusion following roentgen therapy for carcinoma of the breast. Radiology 1945; 44: 335-340. Tricot R. Baillet J, Helmke G: La pericardite constrictive oost-radiotheraniaue 18 orooos d’un cas oersonnell. Arch kal Coeur 1954; 4j: 9i2-94& Gimlette TMD: Constrictive oericarditis. Br Heart 1 1959: 21: 9-E. Hurst DW: Radiation fibrosis of pericardium with cardiac tamponade. Can Med Assoc J 1959; 81: 377-380. Connolly DC, Buichell HB: Pericardltis: a 10 year survey. Am J Cardiol1961: 7: 7-13. Steinberg I: Effusive-constrictive radiation pericarditis. Two cases illustrating value of anglocardiography in diagnosis. Am J Cardiol1987; 19: 434-439. Cohn KE, Stewart JR, Fajardo LF, Hancock EW: Heart disease following radiation. Medicine (Baltimore] 1967; 46: 281-298. Stewart JR. Cohn KE. Fajardo LF, Hancock EW, Kaplan HS: Radiation-induced heart disease. Radiology -. 1967; 89: 302-310. Teng CY, Nemickas R, Tobin JR Jr, Szanto PB, Gunnar RM: Pericardial effusion follow& radiation to the chest. Report of a case. Dis Chest 1967; 527549-552. Muggia FM, Cassileth PA: Constrictive pericarditis following radiation therapy. Am J Med 1968; 44: 116-123. Roberts WC, Glancy DL, DeVita VT: Heart in malignant lymphoma (Hodgkin’s disease, lymphosarcoma, reticulum cell sarcoma a’nd mycosis fungoides). A study of 196 autopsy cases. Am J Cardiol1968; 2.2:85-107. Masland DS, Rotz CT Jr. Harris JH Jr: Postradiation pericarditis with chronic pericardial effusion. Ann Intern Med 1968; 69: 97-102. Fajardo LF, Stewart JR, Cohn KE: Morphology of radiationinduced heart disease. Arch Path01 1968; 86: 512-519. Morton DL, Kagan AR, Roberts WC, O’Brien KP, Holmes EC, Adkins PC: Pericardiectomy for radiation-induced pericarditis with effusion. Ann Thorac Surg 1969; 8: 195%208.
18.
19.
20.
21.
22.
23. 24.
25. 26. 27. 28. 29. 30. 31. 32.
Haas JM: Symptomatic constrictive pericarditis developing 45 years after radiation therapy to the mediastinum. A review of radiation pericarditis. Am Heart J 1969: 77: 8895. Pierce RH, Hafermann MI& Kagan AR: Changes in the transverse cardiac diameter following mediastinal irradiation for Hodgkin’s disease. Radiology 1969; 93: 619624. Lawson RAM, Ross WM, Gold RG. Blesovsky A, Barnsley WC: Post-radiation pericarditis. Report on four more cases with special reference to bronchogenic carcinoma. J Thorac Cardiovasc Surg 1972; 63: 841-847. Kagan AR, Hafermann M, Hamiton M. Pierce R, Morton D. johnson R: Etiology, diagnosis, and management of pericardial effusion after irradiation. Radio1Clin Biol197.2: 41: 171-182. Morton DL, Glancy DL, Joseph WL. Adkins PC: Management of patients with radiation-induced pericarditis with effusign: a note on the development of aortic regurgitation in two of them. Chest 1973: 64: 291-297. Martin RG, Rudkdeschel JC, Chang P, Byhardt R, Bouchard RJ, ‘Wiernik PH: Radiation-related pericarditis. Am J CardiollQ75; 35: 216-220. Byhardt R, Brace K. Ruckdeschel J, Chang P, Martin R, Wiernik PH: Dose and treatment factors in radiation related pericardial effusion associated with the mantle technique for Hodgkin’s disease. Cancer 1975; 35: 795802. Rudkdeschel JC. Chang P, Martin RG, et a].: Radiation-related pericardial effusions in patients with Hodgkin’s disease. Medicine [Baltimore] 1975.54: 245-259. Carmel RJ, Kaplan HS: Mantle irradiation in Hodgkin’s disease. Cancer 1976; 37: 2813-2825. Green B, Zornoza J, Ricks JP: Eccentric pericardial effuqion after radiation therapy of left breast carcinoma. Am J Roentgen01 1977; 128: 27-30. Schneider JS, Edwards JE: Irradiation-induced pericerditis. Chest 1979; 75: 560-564. Pearson HES: Coronary occlusion following thoracic radiotherapy, 2 cases. Proc R Sot Med 1957; 50: 516. Dollinger MR, Lavine DM, Foye LV Jr: Myocardial infarction following radiation (letter to the editor). Lancet 1965; 2: 246. Prentice RTW: Myocardial infarction following radiation [Letter to the Editor). Lancet 1965; 2: 388. Dollinger MR. Lavine DM, Foye LV Jr: Myocardial infarction due to post-irradiation fibrosis of the coronary arteries.
March 1981 The American Journal of Medicine
Volume 70
529
RADIATION HEART DISEASE-BROSIUS
33. 34. 35. 36. 37. 38. 39. 40.
530
ET AL
Case of successfully treated Hodgkin’s disease with lower esophageal involvement. JAMA 195: 316-319.1966. Huff H. Sanders EM: Coronarv-arterv occlusion after radiation. N Engl J Med 1972; 286: 780: Tracy GP. Brown DE, Johnson LW, Gottlieb AJ: Radiationinduced coronary artery disease. JAMA 1974; 228: 16601662. Rinerose. TL: Radiation-“induced” coronarvI arterv m, disease. ~~~ JiMA.1974; 230: 960-961. McRevnolds RA. Gold GL. Roberts WC: Coronarv heart disease after mediastinal irradiation for Hodgkin’s disease. Am J Med 1976; 60: 39-45. Ali MK, Khalil KG, Fuller LM, et al.: Radiation-related myocardial injury: management of two cases. Cancer 1976; 38: 1941-1946. Rodgers DL: Precocious myocardial infarction after radiation treatment for Hodgkin’s disease. Chest 1976; 70: 675-677. Murphy GK: Cancer and the coroner. JAMA 1977; 237: 786-788. Iqbal SM, Hanson EL, Gensini GG: Bypass graft for coronary artery stenosis following radiation therapy. Chest 1977; 71: 664-666.
March 1981
The American Journal of Medicine
41. Rasmussen
42. 43. 44. 45. 46. 47.
Volume 70
S, Dfissing M. Walbom-Jdrgensen S: Coronary heart disease-a possible risk in mega voltage therapy? Acta Med Stand 1978; 203: 237-239. Hartman FW, Bollinger A, Doub HP, Smith FJ: Heart lesions produced by deep x-rays. An experimental and clinical study. Bull Johns Hopkins Hosp 1927;41: 36-67. Fajardo LF, Stewart JR: Pathogenesis of radiation-induced myocardial fibrosis. Lab Invest 1973; 29: 244-257. DiMatteo 1, Vacheron A. Heulin A. et al.: Comolications cardiques de la radiotherapie thoracique. Arch Ma1 Coeur 1978; 71: 447-452. Bouvrain Y, Roujeau J, Rogard JM: Myocardite post-radiotherapique. Sem Hop Paris 1977:53: 433-436. Roberts WC, Sjoerdsma A: The cardiac disease associated with the carcinoid syndrome (carcinoid heart disease]. Am J Med 1964; 36: 5-34. Isner JM, Wu M, Virmani R, Jones AA, Roberts WC: Comparison of degrees of coronary arterial luminal narrowing determined by visual inspection of histologic sections under magnification among three independent observers and comparison to that obtained by video planimetry. An analysis of 559 five-mm segments of 61 coronary arteries from eleven patients. Lab Invest 1980; 68: 831-838.