- Email: [email protected]
Primary sarcoma of the heart
PRIMARY I~ZPORT
OF A
ARLIE
SARCOMA
CASE WITH R,. BARNES, AND
OF THE
HEART
ELECTROCARDIOGRAPHIC STUDIES”~
M.D.,
DONALD
M.
ALBERT ROCHESTER,
SNELL,
AND PATHOLOQICAL
C. BEAVER, M.D.
M.D.,
MINN.
NVASION of the heart by a neoplastic process, whether primary or secondary, is relatively rare. Lymburner, in a review of 8,550 postmortem examinations at The Mayo Clinic, found 4 cases of primary tumor of the heart, and in 52 casesthe heart was the site of metastatic invasion by neoplasms. Adami stated that the relative immunity of t,he heart to neoplastic involvement “is probably to be attributed to t,he fact that the heart, above all organs, is constantly in a state of great efficiency, well-nourished, well-innervated, and functionally always active. ” The rarity of cardiac neoplasms is attested by the fact that Lymburner found only 226 cases of primary tumors of t,he heart recorded in the literature, to which he added 4 cases. The first reports of cases are attributed, by Beck and Thatcher, t,o Zollicofferus (1685) and Theophy Boneti (1700). According to Perlstein, the first authentic modern cases are those of Albers (1835) and Bodenheimer (1865). In recent years, a number of writers have reported additional cases and have reviewed the literature; among t,he more important reviews of this type are those of Perlstein, Beck and Thatcher, Meroz, Uehlinger, Karrenstein, Goldstein, Bradley and Maxwell, Diebold, Pommer, and Matras. The consensus of opinion of these writers is that the majority of primary cardiac tumors are of mesenchymal origin, and may be classified as spindle-cell, round-cell or mixed-cell sarcomas, the first-named being the type most commonly encountered; a few rhabdomyosarcomas have also been noted. The primary focus of these tumors is variable. Beck and Thatcher cited Link’s series of cases, and noted the relatively high incidence of primary involvement of the auricles; the ventricles, valves? and intra-auricular septum are considerably less common as primary sites. The symptoms of these tumors are, as one might expect, as variable as is their distribution. Numerous analyses of the symptoms have been made, but no constant clinical syndrome can be said to have been established. As Meroz has pointed out, there can be no characteristic symptoms if the tumor does not interfere with the cardiac mechanism.
I
*Submitted for publication tFrom the Sections on Medicine of the Mayo Clinic.
November Cardiology
23, and 480
1933. Pathologic
Anatomy
and
the
Division
of
BARNES
ET
AT,. :
PRIMARY
SARC‘OII
A OF
TTEAR’I
4X1
He went on to state that if the valve orifices are involved, stenosis, regurgitation, or both, may result, and the pulmonary and general If the pericirculation may be involved singly or in combination. cardial surface is involved, an effusion, usually hemorrhagic, results; involvement of the cavities may produce multiple embolie phenomena, and involvement of the conduction system may produce arrhythmia of various types. Meroz’ conception of the symptoms of cardiac tumors is based chiefly on hypothetical grounds. Nomicki, Bradley and Maxwell, Matras, and others, from a study of case records, have outlined the symptoms most commonly recorded in t,he literature. among these may be mentioned : (1) progressive circulatory failure, for which no obvious anatomical basis can be clemonst,rated; (2) progressively increasing valvular stenosis, with murmurs which change on movement or from day to day; (3) signs of mediastinal tumor, sometimes associated with gross changes in the appearance of the heart on roentgenological examination ; and (4) recurrent hemorrhagic pericardial or pleural effusions. So far as we are able to judge from available reports in the literature, the diagnosis of a primary cardiac neoplasm has not been made during life. Lymburner reviewed 4 cases in which secondary tumors of the heart had been recognized during life! including the case reported by Willius and Amberg in 1921. The case reported here is of particular interest because the coexistence of signs of pericarditis, heart-block, and metastasis to the muscles of the shoulder girdle permitted a tentative clinical diagnosis of malignant involvement of the heart. Curiously enough, the two principal diagnostic points in this ease, heart-block and metastasis to the skeletal system, appear to be particularly uncommon, Bradley and Maxwell stated that heart-block has been noted in two cases, and Goldstein cited the case of Armstrong and Monckeberg, in which this condition developed in an infant with a cardiac tumor. In Beck and Thatcher’s case a metast,atie deposit was noted in the right deltoid muscle. In Perlstein’s series of thirty cases, metastasis was noted in the lungs, liver, pancreas, and mediastinal lymph nodes, and in the suprarenal glands, but not in the skeletal system. The diagnosis in this case is therefore no contradiction of the often-quoted statement that diagnosis of a cardiac tumor is “either impossible or a matter of good fortune.” CASE
REPORT
A woman, a housekeeper, aged sixty-two years, was admitted to St. Mary’s Haspita1 December 10, 1932, because of pain in the thorax and dyspnea on exertion. These symptoms had been present for about two months and had been accompanied by definite loss of weight and strength. For about one month she had suffered from night sweats, and was known to have had a daily rise in temperature to 100” or 10Z” F. The past history was entirely irrelevant to the complaint at the time of admission.
t’llysiwl cxaminatitril ptirnt was tlecidedly
ill.
gnr~z ~~vitle~llcc~ ot’ Tlw lw~l~etxtnw
wnsi~lwxlll(~ was I IL!”
F.,
10~s of \\-(light, and the ~)ulw
and rate
thus 120
brats cwh minute. Cyanosis or peril)hcrnl ,drm:t could not 1w made ant, Ijut tlrtinite orthopnea was present. There wwc no oral foci. ~~rnmin:rtion of the Inngs gavel csxcntially negative wsults. Tl1clT wc1’c no signific;lnt nl~d~m~inal findings, c~scel)t some tenderness in the ulqwr rwd of 2 lbrt~ious lalmrotomy scar. Examination of the pelvis gave essentially urgatire reanlts. excel)t for t5?dwc*c of ~~~cwol~:~usal illIt ,\‘:,s volutionarp change. The principal physical findings conwrned the heart. ~alrulnr murmurs wuld not lw not enlarged, the tones wcrc c,lear, and drtinitt, heard. Loud friction sounds WC w heard :I I1 over tlw l)rwnrdinm, varying somt’Results of urinalysis and stwlogi~2il what on chnngc in tlicx patient’s condition. ‘I’1161 ~~l11(,t,iitr:ttioll of hen~oglol~it~ was 11 .4 jpl. ; test for syphilis ww negative. c-rpthrocytes numl~cwd .+,:tNl,(N:ll and Irll<~~w~tc~s (i,Y!llll in c:lch (.ulli(. millitn(~tcr of w~12~led :I c,:ilGtird tulwrculous l,wwss in I,lood. A roent,genogram of the tllor:ls 111C upper lobe of the left 11111g. TWI ~111lur~w of tlic l~lood did not revwl growtl~ in forty-tight llours. An agglutination twt for .4/cctligcl1c.c nhorlu.~ w:w ncgaiirc. csng~lectrocardiogr:1Illli(, cxnmination dis~~losc~d :I rate of 106, sinus tarhyc~ardia, I I and gerated P-waw in derivation TTI and s111rred QHS ~~nml~lcsrs in derivations IIT. -4 diagnosis of subacute, fibrinous periwrditis was made, hut no explanation UBS advanced at the time as ‘LO its etiology. Hecausc of its probable infectious nature, hoxever, a cour~c! of sodium cncodylatr was begun, the dosage consisting of 7% grains (0.48 gm.) giren intrnwnouslg t-wiw daily. Pivcl days later the pcricardial friction sounds were altered; in fact, with the patient in the sitting position they were almost inwudihlc. Thaw wow 110 signs of pericardial effusion, and she xas cl&i&e17 better. T)ewmIwr :!:I tlu 1 ;wrjca,rdial friction sounds had conllJlt+r’ly disalllwared, and the tcmperaturc rtwminrd lwi~w 100” P. for t\w~nty-four llours. .4(1ministration of sodium carodylat~~ was diswntinnrd at this time*. On this date thus pain in the tllOl%S was entirely relir\ cd, I)nt on 1)ecemher 26 it re:clqw:~rcd, and the temperature Iwgnn to l~eccimt~ elel-;ltcAd in the afternoon to al)out IW” F. AllOtllPI ~~OII~SC of sodium cwodylutc W;IS lwyun, l,u1- within threw days a ditfuse, arsenical type of dermatitis dereloped, which con~l~cll~~d ns to al)andon this form of treatment. The dermatitis cxleared rapidly nndrr twa tmrnt, and yawed tlrc ltatient 110 wrious inconvenience. On January 3, 1933, it, WIS nntiwd that tht~ ~~ardi:lc~ tones were mu~11 IMS distinct, wlwcially in the region of the n.lwx. Thc~ were no definite signs of lberi~~ardial (Bfiusion, and roentgenograms of 111~ thorax did not reveal signifirant widening of An c,lectror;lrdiogr:l:ll m;~dr on this date> rewaled :I rate of the cardiac shadow. 116, left ventricular prepondrr;~nc.~~, dil)hasir T-wv:i\-e in derivations I and II, notrlld I’-nave in derivation II, and notcllrtl QXS coml,lexes in derivation TIT. (‘ultnre of the blood did not revc.al grow\-th in forty-eight hours. On the night of January .? acut(s pain drvelol~ed in the region of thr right shoulder. On the following morning the joint was painful to touch, nucl there was :I moderate amount of induration along the lwstrrior lwrder of the deltoid musrlr. r,0f :d treatmwt with hot packs, and full dosw of salic$ntrs given during the next few days and l)y January !I the joint was swollen, failed to have any rffwt on this wndition, hot, and indurated. 4 tt,nt:ltirc diagnosis of metastatic alxwcss of the right deltoid muscle was made, and after a further lwriod of ohservatiw of three dn,ys, the muscle was incised under yrocnine aneatlrcsix. Th
differentiated that a definite conclusion could nut he rv:~~hrtl as to their I~roInrI~lc During the next four or source. The cultures did not reveal growth of organisms. The pain in the thorax five da.ys there was little c,hange in the lmti;nt ‘s condition. ,und shoulder wntinued, and the perivardial friction sounrls ecbutinued to Iw rlrfinitely ,tudible
at times
and ahsent
at others.
On the morning of January Iuml~odorsal region, and both
1X consideral,le edem a wts pwnt in the left arm, tlus lower extremities. The ~~ardiac rhythm, lvhich Ijreviand gross irregularity was noted. An ously bad been regular, was grratiy altered, and this garc evidence of an auricular rat{% electrocardiogram was made at onw, auricnlo~elltri~ular disswiot.iou. of 116 and :L rentricuiar rate of SO, with complete On the following day the cardiac rhythm lwcnmt~ regular, lmt the c,lcctro(,arcliogralil revealed :I prolonged :Inriculo~entriclll~lr conduction time of O.:lL seconds. A diagnosis of tumor of the heart, with inwI~;ement of the c~onduction systeni and Iwrieardiurrr. Rr~~:ruw of the and with metastasis to tlw Cght deltoid muscle was suggested. uncertainty with regard to the I~rcvious l)iopsy. anot1ic.r YJK~~~III~~ was t:tken from the same reqioii. A somewhat more satisfactory l)iopsy was obtained from tire muscle, which on examination re\ealed a highly undifferentiated, infiltrating, do’generating cellular tumor, prolnrlrly :i sarcoma. This dt~rrloIm~mt appeared to &al)lish definitely the diagnosis of sarcomatom involrement of the heart. Owing to the patient’s greatly weakened condition the search for a primary source wax necessarily Kepcated roentgenologiwl txuminavurtailrd, and the results were entirely uegative. tions of the thorax and a series of cultures of the t,lood gart’ no significant results. E lwtrowrdiograms were made .January 41 , 22, 2.5, and 27, all of them giving evidence of eompletr auriculoventricular dissociation. The Js~ril~hrral edema, which had twtw noted previously, inercased slowly. The patient lrw:cme more and inor<* She took stuljorous, and roused only to eomI)lain of lmiu in the ut?eeted sboultler. almost no food, and appeared to fail very rapidly. A course of radium therapy was given over the right shoulder in the I~OJW of relieving the pain and IwhaI)s of effecting slight temporary imTrovemcnt in her general condition. She was transferred to her home by mhnlnnrr J:muar~ 31, when the treatrrirnt with radium had Her course there was marked 1)s Irrogressiye failure and in I ,ccn completed. creasing edema and orthopnea, as well as l)y other signs of circulatory failure. Death o,aeurred Februa,rv I 14 . E:~amin;ttiun shortly In~fore death gave no further .clinieal CYidence of sareomatosis, except some Intiler questional~le nodules in the region of the left deltoid muscle. Nr~,ap’s!/.-i\:ecroI)sy \~a:, I~~foru~rd Hi-62 hours after death. The h~vly was extrtwiely em:rci:rted, and the lower c~strrmities were mark~vllp t~lrm:~to~~s. JlultiI)l<, I)urpurie~ llate1Ie.s were present on the skin of the f;iw aud forehead. The skin t,f the deltoid mus~~lc, was thickwed and reve;rl(d uiult,iI)le the right arni, ovc~rlyiiip 111 hot11 pleural lirm, tised, smdll, shotlike, wl~~hamx~~~s nod&%, Iby I’alI):ition. cavities there was clear, dark :unl~cr fluid, .?O(I CL. in t!ic right cavity and l,W1 V.I.. in the left. The pericardial cavity was olrliterated, Imrtiallp 11p tihrous and fibrinous adhesions between the l)arietal and risceral layers, t)ut also 1~7 the inter, I)osit,ion of soft, grayish white, nroplastic tissue. The nonplastic tissue appeared md to lw entirely confined within to arise from both parirtal and riscernl layers, the sw j the external surfaw of thtx parietal Iwrieardium was smooth, glistening. and alrparently free from tumorous inrolvrinmt. Oil m:~nm I seI)aration of the pericardial surfaces, the internal Iswictol surface an11 the visceral s~ufaw disclosed multiple grayish white, rel:~tirrl~soft, slightly elewtetl, plaquelike masses of tissue. These were so numerous that diffuse involvement was :rpproximate~l. IJut where disercte, these areas wre from :! mm. to .Z mm. in irrcguhrr diameter (Fig. 1).
484
THE
AMERICAN
HEART
JOURNAL
The heart and adherent pericardium weighed 3% gm. The wall of the. right auricle was diffusely thickened and replaced by relatively soft, friable, grayish white tissue, which was contiguous Tvith similar appearing tissue of the epicardium. ,Llthough the aurieular wall appeared to be diffusely involved, there were some areas which were distinctly nodular. These were especially prominent above the posterior cusp of the tricuspid valve, where the endocardium was eroded by tumor. Here the tumor extended somewhat into the auricular chamber, as slightly elevated and roughened nodules; the two largest of these were each I cm. in diameter. Elsewhere the endocardium of the auricle was smooth and apparently not extensively invaded; where considerable tumor tissue existed in the wall, however, there was an evenly rounded bulging of the corresponding portion of the wall toward the aurieular chamber. This was especially the ca.se in the region of the interauricular septum, below the foramen oval?. The thickness of the wall of the right auricle was usually
0.5 to 1.5 cm.,
but
at the
scptmn
it
wa,s 0.5 cm.
thick.
The
triruspid
Fig. I.-Neoplastic involvement of the pericardium over the posterior portion The pericardium has been reflected from the epicardium. the left ventricle. Fig. 2.-.-Involvement of the right auricle and right ventricle by the neoplasm. tumor of the auricle is particularly prominent above the tricuspid valves. Most tensive involvement of the right ventricle nppears on the interventricular wall; incision appears through the center of this arca.
Of The exan
valves were normal. The wall of the right ventricle was involved in a way similar to that of the right auricle, but the involvement was less extensive. There was diffuse, grayish white, neoplastic infiltration of the ventricular muscle below the posterior papillary muscle, extending downward to the apex, and anteriorly to the base of the anterior papillary muscle; the tumor extended toward the left in the interventrieular septal portion of the right ventricle, as far as the papillary mu&e Xost, of the neoplastic r*han$e observed in the right vent.ricle was in of the conus. this portion. There was no tumor obsrrvahlr in the anterior wall (except near the apex) in the region of the eouus arteriosus, in the base beneath the tricuspid valves, At the apex, the wall of the nor in the extreme left portion of the septal region. right ventricle was 1.5 cm. thick; elsewhere the thickness averaged 0.8 cm. NoThe where in the right ventricle did the tumor extend through the endocardium.
wall was invaded by neopulmonary valves were normal. (Fig. t’.) The left aurit’ular plasm only in the region of the interauricfiulas septum and the auricular appendage. The wall of the appendage was 1 cm. thick. Elsewhcrc, except in the interauricular septum, the wall of the left auricle was 1 to 2 mm. thick. The mitral valves were normal. The left ventricle was normal; its muscle was nl~proximately 1 cm. thirk. The aortic valves were normal, The large vessels, to and from the heart, mere normal, without neoplasm surrounding them, except in their intrapericardial portion. Both lungs revealed numerous very small, mostly subpleural, metastatie nodules, the largest of which was 1.5 cm. in diameter. The thymus gland was replaced by adipose tissue; there were no metastatic nodules. About the arch of the aorta, and the bifurcation of the’ trachea, some lymph nodes were enlarged. In one region the nodes were conglomerate and formed a firm, grayish white mass 3 cm. in diameter. Lymph nodes elsewhere were somewhat enlarged, hut on gross inspection disclosed
F:g.
I”:g.
3.
.I.
polymorphism of the tumor, as it involves the epicardium of the right Large cells with abundant cytoplasm, others with tendency auricle, may be seen. to spindle formation, especially in the background, may be noted (hematoxylin and eosin, X365). Fig. I.-Mediastinal lymph node. Cells with abundant reticulated cytoplasm, and Fig.
small
3.-The
cytoplasmic
vacuoles
(Mallory’s
phosphotungstic
acid
hematoxylin,
X187.5).
The thyroid gland was atrophic, but otherwise revealed no no evidence of tumor. lesions. The cervical nodes were not remarkable in any way. The esophagus, stomach a.nd intestinal tract were normal. The pancreas was normal, except for one firm pinkish white nodule 0.5 cm. in diameter in t,he body of the gland. There was one metastatie nodule 1 cm. in diameter in the liver. The spleen was normal. The left suprarenal gland was enlarged and appeared to be diffusely involved by an infiltrative type of neoplastie process without loss of identifiable units of the gland. The right suprarenal gland was similar, but revealed only a small fragment of tumor. There were no metastatic growths in the kidneys. The left kidney was atrophic, and weighed 30 gm., apparently representing old, healed pyelonephritis. The right kidney ~a.9 correspondingly enlarged, and weighed 200 gm.; it was essen-
1.
I:
Fig. .i.-A, Mediastinal Ivmgh node. Giant wll with abunclant c~toylaslll anSl multiple nuclei. Faint long.itudinal fihrillw and rows of evenly placed granules I(‘.Sentbling embryonic ,sarcous elements or striations ma)- also lie seen faintly in the, VYtOPkISm
(Msllory’s
phosphotungstic
acid
metastasis to deltoid muscle. Polymorphous wincllc forma ma\- tw notrcl Orl‘nvitox~-lin
hematox>-lin, chaydcter
nnA wsin.
X 1451)). of cell i
X 251))
R, with
Biopsy many
revr:tlr:l o\:i\l
111
phosphotungstic-acid hematoxylin rtain, was wrcaled as wry fine, scarcely visible longitudinal fibrillae, with more dwl~ly stained granules in dotlikr lines. These suggested concentration points of cmhryounl sarvous elements, the striations of rwhryoniv muscle ~11s ( Fig. 5.1 ). Nuclei of the tumor ccllx were usually laxgc!. hyperchromatic structures, often with finely diridtvl chromatin. This gave SOUP nwlei tht! appearance of :I meshwork of c.llrom:ltin. Sonle wlls had two or three large nuclei. Mitotic: figures were fairly numerous. Cells were often undergoing retrogressive changes, and in thew thv nuclei werr in a state of karyorrhexis. In several portions of the tumor necrosis was a.dvanced. This was especially the case in the epicardium. Lymphocytes and polymorphonuclear nentrophilic leueocytes were l.ollert,ed in aggregations, particularly in the cpi~ardium and pericardium, and at
the periphery of necrotic portions of the tumor. Vessels of the epicardium and myocardium were often surrounded I)p tumor cells, and somr of them contained cells of similar appearance which packed their lumina. Metastatic involvement was in the right. deltoid muscle (two proved to be present, by microscopic s&ions, specimens for biolsy ; Fig. .SB), lungs, liver, suprarenal glands and lymph nodes In all of these the morphological c3,aractcr(mediastitml, aortic and mesenterie). istirs of the cells were either identical with or similar to those observed in the heart and pericardium. The extreme degrees of polymorphism which were disclosed by the tumor in the pericardium and my-ocardium were eshilbited less frequently by the marked variations in size, shape and staining of the metastatic growths; however, The metastatic growths in the deltoid muscle, mediastinal lymph cells did occur. nodes, and lungs, as in portions of the cardiac tumor, revealed extreme degrees of necrosis; however, there mere always areas of preserved cells near the periphery of the growth. In the aortir and mesenteric lymph nodes, and in the suprarenal glands, the tumor cells were evidently young and well preserved, and offered excellent 01~ portunity for detailed study of individual cells. In all situations about tlic tumors, the blood vessels appeared to be invaded, and in sections elsewhere, such as in the preparations of the leiomyomas of t,he uterus, tumor cells were often seen packing the lumina of the vessels. The nodule described in the Itody of the l~ancreas proved to be an adenoma composed of cells derived from islands of Langcrhans. The anatomical diagnosis was rhabdomyosarcoma, primary in the heart, with metastatic involvement of the l)erir:urdium, right deltoid muscle, liver, lungs, supra renal glands, and lyml)Ii IKK~FS. COMMENT
Elect~ora~~ioyra2,hy.-The rlect,rocardiographic findings in this case, in conjunction wit,h the specimen of muscle removed for biopsy, furThe sudden appearance of nished the key to the correct diagnosis. complet,e auriculoventricular dissociation when previous elect’rocartliograms had not. indicated any disturbance Of conduction, suggested that, the malignanl- process mbicb hat1 invaded the deltoid muscle hacl also invaded t,be bundle Of His. There was no way Of knowing, during life, that. the malignant process was primary in the heart. The electrocarcliogram in the ease reported by Willius and Amberg, mentioned in the early part of this paper, revealed incomplete bundlebranch block of a type indicating involvement Of the right bnndlebranch (new berminology ). This electrocardiographic change is entirely consistent wit,11 the fact that the only portion of the right ventricle that was not involved by the tumor was the right anterior port,ion of the conus. The RS-T changes present in the tracing could ea.sily be explained by the concluetion defect in the right. buncllcbranch. T~loycl published the electrocardiogram in a case in which the P-R interval was 0.28 second. At necropsy, a tumor was found in the region of the auriculoventricular node. The electrocardiogram in a case reported by Houck and Bennett was essentially negative except for slight changes in the S-T interval, possibly caused by treatment with digitalis. The tumor in this case involved the left auricle.
488
THE
AMERICAN
HEART
JOITRNAI,
In the electrocardiogram published by Siegel and Young, the T-waves were inverted in all leads. and the RS-T contours were convex in all leads. There was no evidence of defects of conduction in the tracing. One of us (Barnes) has seen an almost exact replica of this tracing develop a few weeks after occlusion of the anterior descending branch of the left coronary artery complicated b;v extensive pericarditis. It is interesting to note that Siegel and Young found that the tumor involved an area of the left ventricle. corresponding to that involved by infarction when the descending branch of the left coronary artery is occluded; but they also recorded invasion of the epicardium of the posterior surface of the heart by the tumor. This might correspond to the epicardial in\-olvement observed in the case of coronary occlusion just, c.ited as haviil, m been encountered by one of us. The epicardial involvement also might account for the close similarity of the tracings in the two cases. It is our belief that a tumor invading the left ventricle, that does not involre either bundle-branch if locallized t.o the ant.erior apical or the posterior basal portion of that ventricle, would produce an elcctrocar~7iogral11 which is similar to that obtained in the stage of fibrosis subsequent to acut,e myocardial infarction. Elevation of the level of take-off’ of the RS-T segment would not be anticipatetl from tumorous invasion of the left ventricle, for the acute cellular and vascular reaction which follows acitte myocarTJikemisr, a changing 11%T segment, as dial infarction is lacking. seen in acute myocarclial iufawtion, would not. be c~spect.ed in a tumor of the left veiit,ricle, for thcsr RS-9’ changes in acute infarction are associated with a rapidly changing cellular reaction. This point is lvell illustrated in the case reportecl by Siegel and Young. On the basis of published tracings and on t,heoretical grounds, it is to be anticipated that most cJf the electrocardiographic changes that will be observed in cases of tumor of the heart will result from neoplastic invasion of the ventricles. It is probable that, invasion of the conducting systems mill account for a majority of abnormal electrocardiograms. Because of the much more frequent, invasion by t,umors of the right ventricle than of the left, right. bundle-branch defects should exceed those of the left bundle-bra.nch. Invasion of the muscle of the left ventricle without involvement of the bundle-branches or of the pericardium should produce t,racings closely similar to those obtained in the fibrotic stage following acute myocardial infarction. either alone or in conjunction with inInvasion of the epicardium, volvement of the ventricles, may produce or modify changes in t,he waves.“’ 7 If a patient, known to have or to have hacl a neoplasm, more or less suddenly exhibits these electrocardiographic changes in the absence of any other pathological process in the heart to explain them, then they constitute important presumptive evidence that the malignant process has invaded the heart.
BARNES
ET
AI,.
:
l’RIMARY
SARC’OXA
OP
IIE~AK’I
4x9
Pathology.-As we have stated previously, primary sarcoma of the heart is a pathological rarity. Perlstein, in 1918, by careful selection, admitted thirty cases of this condition from a compilation of the literature, and added an additional case. Lymburner recently (1933) reviewed the literature exhaustively, and was able to collect fiftyseven cases. Primary sarcomas of the heart have been designated as spindle-celled, giant cell, lymphosarcoma, round-celled sarcoma, mixed-cell sarcoma, fibrosarcoma or myxosarcoma, in agreement w&h the histological details presented by t,he cellular constituents of the tumors. Apparently the diagnosis of rhabdomyosareoma has been made only once, this in a case reported by Bradley and Maxwell? in 1928. Primary rhabdomyoma of the heart has been less often diagnosed, according to reports in the literature. Of this type of tumor, Lymburner could find only forty-seven examples, and added another case. Primary cardiac rhabdomyomas are considered to be benign tumors, in the sensethat in none of the examples referred to in t,he literature have metastatic growths be.en noted, although they have been multiple in the heart. Rhabdomyomas predominate in the early years of life. Of the 48 cases collected by Lymburner, 20 occurred within the first year of life, and only 5 between the twent,ieth and forty-fifth years. Rhabdomyomas of the heart have a further distinctive characteristic in being associated frequently with functional or anatomical developmental defects, epilepsy and tuberous sclerosis of the brain being the two most common of many such disturbances which have been found. It is believed that rhabdomyomas are an expression of abnorma.1 tissue differentiation, or represent secondary developments from embryonic tissue rests. They present a characteristic microscopic picture, in which the large, irregularly processed and vacnolated, so-called “spider cells, ” and long spindIe cells with delicate IongitudinaI fibrillae and cross st,riae form the essential differential points. The cells frequently contain excess glycogen. The relat,ionship which the typical rhabdomyoma of the heart might bear to primary sarcoma of the heart is difficult or impossible t.o ascertain accurately, since only once has the differentiation in sarcoma been considered sufficient to admit the diagnosis, primary rhabdomyosarcoma. From a more or less superficial survey of the field, it would appear that the two t,ypes of t,umor are distinct, and with possible exceptions, it seems improbable that primary sarcoma of the heart arises from malignant, transformations of a rhabdomyoma. Transitional stages apparentjIg do not exist, for as Lymburner remarked: “All reported cases (of rhabdomyoma) seemed to possess a similar microscopic appearance, regardless of the age of the patient, and evidently had obtained their final development in embryonic life.” Primary sarcomas and rhabdomyomas of the heart are further distin-
390
THJS
AMERIt’AN
HEART
JOUKNAI,
guished in that they appear in different age groups. Although cardiac rhabdomyomas predominate among young people, cardiac sarcomas predominate among middle-aged and older people; only two Cases have been recorded in which the patients were less than twenty years of age. Also, primary sarcomas are riot, associated with developmfmtal defects, whereas rhabdomyomas usually are. In reviewing cases reported in t.hr literature. we have found that t$here was one which strikingly resembled t,he case reported here. The case was described by Bradley and Jlaxwell. Their patient was a man, aged sixty-t.wo years; his clinical history was similar to that recorded for our patient. The gross and microscopic appearances of the tumor in rach case were similar, although not identical. The histological differentiation of the tumor described by Bradley and Maxwell was as a polymorphous cell sarcoma with many long spindle cells revealing fibrillae and acidophilic cytoplasm. Sections of t,he tumor were snbmitted by Bradley and Maxwell to Mallory. Ewing and Kroders. 811 concurred in the diagnosis of rhabdomyosarcoma. We have concluded,* as Bradley and Maxwell did in their case, that the diagnosis of the tumor in our case should be rhabdomyosarcoma, primary in the heart. In proof of this we must necessarily first offer evidence that the tumor was of primary cardiac origin. As evidence me may cite t,he clinical history. which gave every indication that the pa,tient ‘s illness from beginnin, 0’ to cncl wax car&w. Further proof of the cardiac origin is offered by the macroscopic findings at necropsy. had occurred, scarcely an! Relatively speaking, although metastasis tumor was found outside the hearts and pericardium. The pericardium and epicardium were involved throughout, and the right auricle and right ventricle were extensively infiltrated. The metastatic areas were small and nowhere gave thr impression of primary sarcoma. (lollversely, the cardiac tumor had no resemblance to metastatic tumors of the heart. many examples of which we have seen. Microscopic examination gives aclditional evidence of the primary cardiac nat.itre of the tumor, by way of its histologica, differentiation. The cells of the tumor were ext,reniely lmIyniorpIious ; there \vere many oval or long spindle forms ; the larger cells had abundant acitlophilic cytoplasm : in a, few cells faint, fibrillae and striations were observed. Kecatwe of the extreme degree of malignancy, the cells were usually t,otalIy undifferentiated. However, there were some. as indicated, which revealed sufficient detail to allow identification of the exceedingly suggestive fibrillae and st,riations. These could ha,rdIy be expect,ed to be so clearly defined as in the slow]) growing rhabdomyomas. because of the high degree of malignancy of the tllmor. The sumniat~ion of t,his evidence, me believe, warrants the conclusion, as previonslp stated. *Dr. tumor
H. and
E. Robertson have concurrerl
and in
DI-. the
AL C. diagnosis
Broders of
ilk+\-c reviewml rhsbdon?;\‘osnr~om;l.
the
sections
of
this
UARNES
ET
AL.
:
PRIMARY
SAR(‘OILIA
OF
lIEAWl
197
SUMMARY
A case is presented in which a diagnosis of tumor of the heart was made during life. The electrocardiographic abnormalities that have been reported in cases of cardiac tumor are reviewed; in our case, the electrocardiographic changes observed were important in est,ablishing the diagnosis. Careful study of the pathological changes in our caw. we believe, permits the conclusion that, the t,umor was a rhabdomposarcoma, primary in the heart. REFERENCES
The l’rinviplps of Pathology, Phila~lvlphia. 1. Adami, J. G., and Nichols, A. G.: 1908-1909, p. 158, Lea B Fehiger. Herzblock, bedingt durch primlren 2. Armstrong, H., and Monckeberg, J. G.: Herztumor, hei cinem 5-jahrigen Kincl, Drutsczhrs Arch. f. klin. 31~11. 162: 144, 1911. 3. Barnes, A. H., and Mann, .F. (1.: ~:le~tr~~ar~liog~aphi~ Changcbs Following Ligation of the Coronary Arteries of t,he Dog, Ax HEART J. 7: 4iT, 19X. 4. Beck, Ccl1 Sarcoma of the Heart. Arch. C. S., and Thatcher, H. 1;1.: Spintlle ht. Med 36: 830 1925 5. Bradley, E’B., and Maxwell, E. S.: Primary Neoplasnrs of the Heart; Report of an Unusual Case, J. A. M. A. 91: 1X!!, 192X. 6. Diebold, Otto : t’ber das Primdre Herxsarkom. Ztschr. f. Krrislaufforsch. 22: 785, ‘1930. 7. Fowlcr, W. M., Rathe, H. W., ant1 Smith, I’. M.: ‘l’hc ~:lrc*troc~arrlit)g~aI)hi(, Changes Following the Ligation of the Small Branrhen of t,hc (!oronar! Artcries, AM. HEART J. 8: 370. 19X:. 8. Goldstein, H. I.: Tunlors of the Heart, New York M. .J. 115: 97, 1.58, 19%. !I. Houck, B. H., anI1 Bennett,, G. A.: l’olvpoid Fihroma of the Left Aurivlrt (HoCalletl ‘Cartliar Myxm~a) (kusing ‘:; Bxll-\~alvv Action, >{M. HRAKT .1. 5: 787, 1930. IO. Karrenstcin: I:in E’all van P’ibroc~l:lstol,l~~~lll ~1~s Herzen:, un11 k-asuistisches zur Frage tlrr Herzgesrhwiilstc, besontic.rs IIV* Mpxomc~, Virvhow ‘s Arch. f. path. Anat. u. Physiol. 194: l?i, 1908. 11. Lloyd, P. C.: Heart Block Due to Primary I,yI:lpha”gio-en(lothrliolla oC Atrio-ventricular Node, Bull. Johns Hopkins Hosp. 44: 149, 1929. 1”. Lymburner, R. M.: Tumors of the Heart; Histopathologic autl Clinicaal Rtutly, Canad. M. A. J. (In press). 13 . Matras, A.: Ein Primares farkom des Hcrzens, Ztschr. f. Kreislaufiorscll. 19’ . “.) O33 192i. 14. Meroz, E.: A Clinical Study of Three Casts of Primary Tumor of the Heart. lnternat Clin 4’ * 231, 1917. 15. Perlstein, .I.: sarcoma of thr Heart. Ani. J. hr. Se. 156: ?I-&, 1918. 1 Ii. Pomnier, G.: %ur Kc~untnis ~Icr Priitliircv Hc~zg~~sc*hwiilstc~, %tsvhr. t’. Kireislaufforscthung. 23: 65. 1941. Anna M.: 1 i. &Siegel, M. I,., &,I Young, ~:l~ctrucarclioKraphi~ Findings in TUIIIOI.:. of the Heart: With a Report of a Casv, AM. HURT J. 8: 682, 1933. tiber einer Fall van diffusem Rhabdomyom des Hcrzens, 18. Uehlinger, Erwin: Virrhow’s Arch. f. path. Anat. u. Physiol. 258: il9, 1925. 19. Willius, I!. A., and Amberg, Samuel: Two Casvs of Secondary Tumor of the Heart in Children in One of Which thv Diagnosis Was Matlv During T,ifc, M. Clin. North linierica 13: 1X17. 19::O.
OF A
ARLIE
SARCOMA
CASE WITH R,. BARNES, AND
OF THE
HEART
ELECTROCARDIOGRAPHIC STUDIES”~
M.D.,
DONALD
M.
ALBERT ROCHESTER,
SNELL,
AND PATHOLOQICAL
C. BEAVER, M.D.
M.D.,
MINN.
NVASION of the heart by a neoplastic process, whether primary or secondary, is relatively rare. Lymburner, in a review of 8,550 postmortem examinations at The Mayo Clinic, found 4 cases of primary tumor of the heart, and in 52 casesthe heart was the site of metastatic invasion by neoplasms. Adami stated that the relative immunity of t,he heart to neoplastic involvement “is probably to be attributed to t,he fact that the heart, above all organs, is constantly in a state of great efficiency, well-nourished, well-innervated, and functionally always active. ” The rarity of cardiac neoplasms is attested by the fact that Lymburner found only 226 cases of primary tumors of t,he heart recorded in the literature, to which he added 4 cases. The first reports of cases are attributed, by Beck and Thatcher, t,o Zollicofferus (1685) and Theophy Boneti (1700). According to Perlstein, the first authentic modern cases are those of Albers (1835) and Bodenheimer (1865). In recent years, a number of writers have reported additional cases and have reviewed the literature; among t,he more important reviews of this type are those of Perlstein, Beck and Thatcher, Meroz, Uehlinger, Karrenstein, Goldstein, Bradley and Maxwell, Diebold, Pommer, and Matras. The consensus of opinion of these writers is that the majority of primary cardiac tumors are of mesenchymal origin, and may be classified as spindle-cell, round-cell or mixed-cell sarcomas, the first-named being the type most commonly encountered; a few rhabdomyosarcomas have also been noted. The primary focus of these tumors is variable. Beck and Thatcher cited Link’s series of cases, and noted the relatively high incidence of primary involvement of the auricles; the ventricles, valves? and intra-auricular septum are considerably less common as primary sites. The symptoms of these tumors are, as one might expect, as variable as is their distribution. Numerous analyses of the symptoms have been made, but no constant clinical syndrome can be said to have been established. As Meroz has pointed out, there can be no characteristic symptoms if the tumor does not interfere with the cardiac mechanism.
I
*Submitted for publication tFrom the Sections on Medicine of the Mayo Clinic.
November Cardiology
23, and 480
1933. Pathologic
Anatomy
and
the
Division
of
BARNES
ET
AT,. :
PRIMARY
SARC‘OII
A OF
TTEAR’I
4X1
He went on to state that if the valve orifices are involved, stenosis, regurgitation, or both, may result, and the pulmonary and general If the pericirculation may be involved singly or in combination. cardial surface is involved, an effusion, usually hemorrhagic, results; involvement of the cavities may produce multiple embolie phenomena, and involvement of the conduction system may produce arrhythmia of various types. Meroz’ conception of the symptoms of cardiac tumors is based chiefly on hypothetical grounds. Nomicki, Bradley and Maxwell, Matras, and others, from a study of case records, have outlined the symptoms most commonly recorded in t,he literature. among these may be mentioned : (1) progressive circulatory failure, for which no obvious anatomical basis can be clemonst,rated; (2) progressively increasing valvular stenosis, with murmurs which change on movement or from day to day; (3) signs of mediastinal tumor, sometimes associated with gross changes in the appearance of the heart on roentgenological examination ; and (4) recurrent hemorrhagic pericardial or pleural effusions. So far as we are able to judge from available reports in the literature, the diagnosis of a primary cardiac neoplasm has not been made during life. Lymburner reviewed 4 cases in which secondary tumors of the heart had been recognized during life! including the case reported by Willius and Amberg in 1921. The case reported here is of particular interest because the coexistence of signs of pericarditis, heart-block, and metastasis to the muscles of the shoulder girdle permitted a tentative clinical diagnosis of malignant involvement of the heart. Curiously enough, the two principal diagnostic points in this ease, heart-block and metastasis to the skeletal system, appear to be particularly uncommon, Bradley and Maxwell stated that heart-block has been noted in two cases, and Goldstein cited the case of Armstrong and Monckeberg, in which this condition developed in an infant with a cardiac tumor. In Beck and Thatcher’s case a metast,atie deposit was noted in the right deltoid muscle. In Perlstein’s series of thirty cases, metastasis was noted in the lungs, liver, pancreas, and mediastinal lymph nodes, and in the suprarenal glands, but not in the skeletal system. The diagnosis in this case is therefore no contradiction of the often-quoted statement that diagnosis of a cardiac tumor is “either impossible or a matter of good fortune.” CASE
REPORT
A woman, a housekeeper, aged sixty-two years, was admitted to St. Mary’s Haspita1 December 10, 1932, because of pain in the thorax and dyspnea on exertion. These symptoms had been present for about two months and had been accompanied by definite loss of weight and strength. For about one month she had suffered from night sweats, and was known to have had a daily rise in temperature to 100” or 10Z” F. The past history was entirely irrelevant to the complaint at the time of admission.
t’llysiwl cxaminatitril ptirnt was tlecidedly
ill.
gnr~z ~~vitle~llcc~ ot’ Tlw lw~l~etxtnw
wnsi~lwxlll(~ was I IL!”
F.,
10~s of \\-(light, and the ~)ulw
and rate
thus 120
brats cwh minute. Cyanosis or peril)hcrnl ,drm:t could not 1w made ant, Ijut tlrtinite orthopnea was present. There wwc no oral foci. ~~rnmin:rtion of the Inngs gavel csxcntially negative wsults. Tl1clT wc1’c no signific;lnt nl~d~m~inal findings, c~scel)t some tenderness in the ulqwr rwd of 2 lbrt~ious lalmrotomy scar. Examination of the pelvis gave essentially urgatire reanlts. excel)t for t5?dwc*c of ~~~cwol~:~usal illIt ,\‘:,s volutionarp change. The principal physical findings conwrned the heart. ~alrulnr murmurs wuld not lw not enlarged, the tones wcrc c,lear, and drtinitt, heard. Loud friction sounds WC w heard :I I1 over tlw l)rwnrdinm, varying somt’Results of urinalysis and stwlogi~2il what on chnngc in tlicx patient’s condition. ‘I’1161 ~~l11(,t,iitr:ttioll of hen~oglol~it~ was 11 .4 jpl. ; test for syphilis ww negative. c-rpthrocytes numl~cwd .+,:tNl,(N:ll and Irll<~~w~tc~s (i,Y!llll in c:lch (.ulli(. millitn(~tcr of w~12~led :I c,:ilGtird tulwrculous l,wwss in I,lood. A roent,genogram of the tllor:ls 111C upper lobe of the left 11111g. TWI ~111lur~w of tlic l~lood did not revwl growtl~ in forty-tight llours. An agglutination twt for .4/cctligcl1c.c nhorlu.~ w:w ncgaiirc. csng~lectrocardiogr:1Illli(, cxnmination dis~~losc~d :I rate of 106, sinus tarhyc~ardia, I I and gerated P-waw in derivation TTI and s111rred QHS ~~nml~lcsrs in derivations IIT. -4 diagnosis of subacute, fibrinous periwrditis was made, hut no explanation UBS advanced at the time as ‘LO its etiology. Hecausc of its probable infectious nature, hoxever, a cour~c! of sodium cncodylatr was begun, the dosage consisting of 7% grains (0.48 gm.) giren intrnwnouslg t-wiw daily. Pivcl days later the pcricardial friction sounds were altered; in fact, with the patient in the sitting position they were almost inwudihlc. Thaw wow 110 signs of pericardial effusion, and she xas cl&i&e17 better. T)ewmIwr :!:I tlu 1 ;wrjca,rdial friction sounds had conllJlt+r’ly disalllwared, and the tcmperaturc rtwminrd lwi~w 100” P. for t\w~nty-four llours. .4(1ministration of sodium carodylat~~ was diswntinnrd at this time*. On this date thus pain in the tllOl%S was entirely relir\ cd, I)nt on 1)ecemher 26 it re:clqw:~rcd, and the temperature Iwgnn to l~eccimt~ elel-;ltcAd in the afternoon to al)out IW” F. AllOtllPI ~~OII~SC of sodium cwodylutc W;IS lwyun, l,u1- within threw days a ditfuse, arsenical type of dermatitis dereloped, which con~l~cll~~d ns to al)andon this form of treatment. The dermatitis cxleared rapidly nndrr twa tmrnt, and yawed tlrc ltatient 110 wrious inconvenience. On January 3, 1933, it, WIS nntiwd that tht~ ~~ardi:lc~ tones were mu~11 IMS distinct, wlwcially in the region of the n.lwx. Thc~ were no definite signs of lberi~~ardial (Bfiusion, and roentgenograms of 111~ thorax did not reveal signifirant widening of An c,lectror;lrdiogr:l:ll m;~dr on this date> rewaled :I rate of the cardiac shadow. 116, left ventricular prepondrr;~nc.~~, dil)hasir T-wv:i\-e in derivations I and II, notrlld I’-nave in derivation II, and notcllrtl QXS coml,lexes in derivation TIT. (‘ultnre of the blood did not revc.al grow\-th in forty-eight hours. On the night of January .? acut(s pain drvelol~ed in the region of thr right shoulder. On the following morning the joint was painful to touch, nucl there was :I moderate amount of induration along the lwstrrior lwrder of the deltoid musrlr. r,0f :d treatmwt with hot packs, and full dosw of salic$ntrs given during the next few days and l)y January !I the joint was swollen, failed to have any rffwt on this wndition, hot, and indurated. 4 tt,nt:ltirc diagnosis of metastatic alxwcss of the right deltoid muscle was made, and after a further lwriod of ohservatiw of three dn,ys, the muscle was incised under yrocnine aneatlrcsix. Th
differentiated that a definite conclusion could nut he rv:~~hrtl as to their I~roInrI~lc During the next four or source. The cultures did not reveal growth of organisms. The pain in the thorax five da.ys there was little c,hange in the lmti;nt ‘s condition. ,und shoulder wntinued, and the perivardial friction sounrls ecbutinued to Iw rlrfinitely ,tudible
at times
and ahsent
at others.
On the morning of January Iuml~odorsal region, and both
1X consideral,le edem a wts pwnt in the left arm, tlus lower extremities. The ~~ardiac rhythm, lvhich Ijreviand gross irregularity was noted. An ously bad been regular, was grratiy altered, and this garc evidence of an auricular rat{% electrocardiogram was made at onw, auricnlo~elltri~ular disswiot.iou. of 116 and :L rentricuiar rate of SO, with complete On the following day the cardiac rhythm lwcnmt~ regular, lmt the c,lcctro(,arcliogralil revealed :I prolonged :Inriculo~entriclll~lr conduction time of O.:lL seconds. A diagnosis of tumor of the heart, with inwI~;ement of the c~onduction systeni and Iwrieardiurrr. Rr~~:ruw of the and with metastasis to tlw Cght deltoid muscle was suggested. uncertainty with regard to the I~rcvious l)iopsy. anot1ic.r YJK~~~III~~ was t:tken from the same reqioii. A somewhat more satisfactory l)iopsy was obtained from tire muscle, which on examination re\ealed a highly undifferentiated, infiltrating, do’generating cellular tumor, prolnrlrly :i sarcoma. This dt~rrloIm~mt appeared to &al)lish definitely the diagnosis of sarcomatom involrement of the heart. Owing to the patient’s greatly weakened condition the search for a primary source wax necessarily Kepcated roentgenologiwl txuminavurtailrd, and the results were entirely uegative. tions of the thorax and a series of cultures of the t,lood gart’ no significant results. E lwtrowrdiograms were made .January 41 , 22, 2.5, and 27, all of them giving evidence of eompletr auriculoventricular dissociation. The Js~ril~hrral edema, which had twtw noted previously, inercased slowly. The patient lrw:cme more and inor<* She took stuljorous, and roused only to eomI)lain of lmiu in the ut?eeted sboultler. almost no food, and appeared to fail very rapidly. A course of radium therapy was given over the right shoulder in the I~OJW of relieving the pain and IwhaI)s of effecting slight temporary imTrovemcnt in her general condition. She was transferred to her home by mhnlnnrr J:muar~ 31, when the treatrrirnt with radium had Her course there was marked 1)s Irrogressiye failure and in I ,ccn completed. creasing edema and orthopnea, as well as l)y other signs of circulatory failure. Death o,aeurred Februa,rv I 14 . E:~amin;ttiun shortly In~fore death gave no further .clinieal CYidence of sareomatosis, except some Intiler questional~le nodules in the region of the left deltoid muscle. Nr~,ap’s!/.-i\:ecroI)sy \~a:, I~~foru~rd Hi-62 hours after death. The h~vly was extrtwiely em:rci:rted, and the lower c~strrmities were mark~vllp t~lrm:~to~~s. JlultiI)l<, I)urpurie~ llate1Ie.s were present on the skin of the f;iw aud forehead. The skin t,f the deltoid mus~~lc, was thickwed and reve;rl(d uiult,iI)le the right arni, ovc~rlyiiip 111 hot11 pleural lirm, tised, smdll, shotlike, wl~~hamx~~~s nod&%, Iby I’alI):ition. cavities there was clear, dark :unl~cr fluid, .?O(I CL. in t!ic right cavity and l,W1 V.I.. in the left. The pericardial cavity was olrliterated, Imrtiallp 11p tihrous and fibrinous adhesions between the l)arietal and risceral layers, t)ut also 1~7 the inter, I)osit,ion of soft, grayish white, nroplastic tissue. The nonplastic tissue appeared md to lw entirely confined within to arise from both parirtal and riscernl layers, the sw j the external surfaw of thtx parietal Iwrieardium was smooth, glistening. and alrparently free from tumorous inrolvrinmt. Oil m:~nm I seI)aration of the pericardial surfaces, the internal Iswictol surface an11 the visceral s~ufaw disclosed multiple grayish white, rel:~tirrl~soft, slightly elewtetl, plaquelike masses of tissue. These were so numerous that diffuse involvement was :rpproximate~l. IJut where disercte, these areas wre from :! mm. to .Z mm. in irrcguhrr diameter (Fig. 1).
484
THE
AMERICAN
HEART
JOURNAL
The heart and adherent pericardium weighed 3% gm. The wall of the. right auricle was diffusely thickened and replaced by relatively soft, friable, grayish white tissue, which was contiguous Tvith similar appearing tissue of the epicardium. ,Llthough the aurieular wall appeared to be diffusely involved, there were some areas which were distinctly nodular. These were especially prominent above the posterior cusp of the tricuspid valve, where the endocardium was eroded by tumor. Here the tumor extended somewhat into the auricular chamber, as slightly elevated and roughened nodules; the two largest of these were each I cm. in diameter. Elsewhere the endocardium of the auricle was smooth and apparently not extensively invaded; where considerable tumor tissue existed in the wall, however, there was an evenly rounded bulging of the corresponding portion of the wall toward the aurieular chamber. This was especially the ca.se in the region of the interauricular septum, below the foramen oval?. The thickness of the wall of the right auricle was usually
0.5 to 1.5 cm.,
but
at the
scptmn
it
wa,s 0.5 cm.
thick.
The
triruspid
Fig. I.-Neoplastic involvement of the pericardium over the posterior portion The pericardium has been reflected from the epicardium. the left ventricle. Fig. 2.-.-Involvement of the right auricle and right ventricle by the neoplasm. tumor of the auricle is particularly prominent above the tricuspid valves. Most tensive involvement of the right ventricle nppears on the interventricular wall; incision appears through the center of this arca.
Of The exan
valves were normal. The wall of the right ventricle was involved in a way similar to that of the right auricle, but the involvement was less extensive. There was diffuse, grayish white, neoplastic infiltration of the ventricular muscle below the posterior papillary muscle, extending downward to the apex, and anteriorly to the base of the anterior papillary muscle; the tumor extended toward the left in the interventrieular septal portion of the right ventricle, as far as the papillary mu&e Xost, of the neoplastic r*han$e observed in the right vent.ricle was in of the conus. this portion. There was no tumor obsrrvahlr in the anterior wall (except near the apex) in the region of the eouus arteriosus, in the base beneath the tricuspid valves, At the apex, the wall of the nor in the extreme left portion of the septal region. right ventricle was 1.5 cm. thick; elsewhere the thickness averaged 0.8 cm. NoThe where in the right ventricle did the tumor extend through the endocardium.
wall was invaded by neopulmonary valves were normal. (Fig. t’.) The left aurit’ular plasm only in the region of the interauricfiulas septum and the auricular appendage. The wall of the appendage was 1 cm. thick. Elsewhcrc, except in the interauricular septum, the wall of the left auricle was 1 to 2 mm. thick. The mitral valves were normal. The left ventricle was normal; its muscle was nl~proximately 1 cm. thirk. The aortic valves were normal, The large vessels, to and from the heart, mere normal, without neoplasm surrounding them, except in their intrapericardial portion. Both lungs revealed numerous very small, mostly subpleural, metastatie nodules, the largest of which was 1.5 cm. in diameter. The thymus gland was replaced by adipose tissue; there were no metastatic nodules. About the arch of the aorta, and the bifurcation of the’ trachea, some lymph nodes were enlarged. In one region the nodes were conglomerate and formed a firm, grayish white mass 3 cm. in diameter. Lymph nodes elsewhere were somewhat enlarged, hut on gross inspection disclosed
F:g.
I”:g.
3.
.I.
polymorphism of the tumor, as it involves the epicardium of the right Large cells with abundant cytoplasm, others with tendency auricle, may be seen. to spindle formation, especially in the background, may be noted (hematoxylin and eosin, X365). Fig. I.-Mediastinal lymph node. Cells with abundant reticulated cytoplasm, and Fig.
small
3.-The
cytoplasmic
vacuoles
(Mallory’s
phosphotungstic
acid
hematoxylin,
X187.5).
The thyroid gland was atrophic, but otherwise revealed no no evidence of tumor. lesions. The cervical nodes were not remarkable in any way. The esophagus, stomach a.nd intestinal tract were normal. The pancreas was normal, except for one firm pinkish white nodule 0.5 cm. in diameter in t,he body of the gland. There was one metastatie nodule 1 cm. in diameter in the liver. The spleen was normal. The left suprarenal gland was enlarged and appeared to be diffusely involved by an infiltrative type of neoplastie process without loss of identifiable units of the gland. The right suprarenal gland was similar, but revealed only a small fragment of tumor. There were no metastatic growths in the kidneys. The left kidney was atrophic, and weighed 30 gm., apparently representing old, healed pyelonephritis. The right kidney ~a.9 correspondingly enlarged, and weighed 200 gm.; it was essen-
1.
I:
Fig. .i.-A, Mediastinal Ivmgh node. Giant wll with abunclant c~toylaslll anSl multiple nuclei. Faint long.itudinal fihrillw and rows of evenly placed granules I(‘.Sentbling embryonic ,sarcous elements or striations ma)- also lie seen faintly in the, VYtOPkISm
(Msllory’s
phosphotungstic
acid
metastasis to deltoid muscle. Polymorphous wincllc forma ma\- tw notrcl Orl‘nvitox~-lin
hematox>-lin, chaydcter
nnA wsin.
X 1451)). of cell i
X 251))
R, with
Biopsy many
revr:tlr:l o\:i\l
111
phosphotungstic-acid hematoxylin rtain, was wrcaled as wry fine, scarcely visible longitudinal fibrillae, with more dwl~ly stained granules in dotlikr lines. These suggested concentration points of cmhryounl sarvous elements, the striations of rwhryoniv muscle ~11s ( Fig. 5.1 ). Nuclei of the tumor ccllx were usually laxgc!. hyperchromatic structures, often with finely diridtvl chromatin. This gave SOUP nwlei tht! appearance of :I meshwork of c.llrom:ltin. Sonle wlls had two or three large nuclei. Mitotic: figures were fairly numerous. Cells were often undergoing retrogressive changes, and in thew thv nuclei werr in a state of karyorrhexis. In several portions of the tumor necrosis was a.dvanced. This was especially the case in the epicardium. Lymphocytes and polymorphonuclear nentrophilic leueocytes were l.ollert,ed in aggregations, particularly in the cpi~ardium and pericardium, and at
the periphery of necrotic portions of the tumor. Vessels of the epicardium and myocardium were often surrounded I)p tumor cells, and somr of them contained cells of similar appearance which packed their lumina. Metastatic involvement was in the right. deltoid muscle (two proved to be present, by microscopic s&ions, specimens for biolsy ; Fig. .SB), lungs, liver, suprarenal glands and lymph nodes In all of these the morphological c3,aractcr(mediastitml, aortic and mesenterie). istirs of the cells were either identical with or similar to those observed in the heart and pericardium. The extreme degrees of polymorphism which were disclosed by the tumor in the pericardium and my-ocardium were eshilbited less frequently by the marked variations in size, shape and staining of the metastatic growths; however, The metastatic growths in the deltoid muscle, mediastinal lymph cells did occur. nodes, and lungs, as in portions of the cardiac tumor, revealed extreme degrees of necrosis; however, there mere always areas of preserved cells near the periphery of the growth. In the aortir and mesenteric lymph nodes, and in the suprarenal glands, the tumor cells were evidently young and well preserved, and offered excellent 01~ portunity for detailed study of individual cells. In all situations about tlic tumors, the blood vessels appeared to be invaded, and in sections elsewhere, such as in the preparations of the leiomyomas of t,he uterus, tumor cells were often seen packing the lumina of the vessels. The nodule described in the Itody of the l~ancreas proved to be an adenoma composed of cells derived from islands of Langcrhans. The anatomical diagnosis was rhabdomyosarcoma, primary in the heart, with metastatic involvement of the l)erir:urdium, right deltoid muscle, liver, lungs, supra renal glands, and lyml)Ii IKK~FS. COMMENT
Elect~ora~~ioyra2,hy.-The rlect,rocardiographic findings in this case, in conjunction wit,h the specimen of muscle removed for biopsy, furThe sudden appearance of nished the key to the correct diagnosis. complet,e auriculoventricular dissociation when previous elect’rocartliograms had not. indicated any disturbance Of conduction, suggested that, the malignanl- process mbicb hat1 invaded the deltoid muscle hacl also invaded t,be bundle Of His. There was no way Of knowing, during life, that. the malignant process was primary in the heart. The electrocarcliogram in the ease reported by Willius and Amberg, mentioned in the early part of this paper, revealed incomplete bundlebranch block of a type indicating involvement Of the right bnndlebranch (new berminology ). This electrocardiographic change is entirely consistent wit,11 the fact that the only portion of the right ventricle that was not involved by the tumor was the right anterior port,ion of the conus. The RS-T changes present in the tracing could ea.sily be explained by the concluetion defect in the right. buncllcbranch. T~loycl published the electrocardiogram in a case in which the P-R interval was 0.28 second. At necropsy, a tumor was found in the region of the auriculoventricular node. The electrocardiogram in a case reported by Houck and Bennett was essentially negative except for slight changes in the S-T interval, possibly caused by treatment with digitalis. The tumor in this case involved the left auricle.
488
THE
AMERICAN
HEART
JOITRNAI,
In the electrocardiogram published by Siegel and Young, the T-waves were inverted in all leads. and the RS-T contours were convex in all leads. There was no evidence of defects of conduction in the tracing. One of us (Barnes) has seen an almost exact replica of this tracing develop a few weeks after occlusion of the anterior descending branch of the left coronary artery complicated b;v extensive pericarditis. It is interesting to note that Siegel and Young found that the tumor involved an area of the left ventricle. corresponding to that involved by infarction when the descending branch of the left coronary artery is occluded; but they also recorded invasion of the epicardium of the posterior surface of the heart by the tumor. This might correspond to the epicardial in\-olvement observed in the case of coronary occlusion just, c.ited as haviil, m been encountered by one of us. The epicardial involvement also might account for the close similarity of the tracings in the two cases. It is our belief that a tumor invading the left ventricle, that does not involre either bundle-branch if locallized t.o the ant.erior apical or the posterior basal portion of that ventricle, would produce an elcctrocar~7iogral11 which is similar to that obtained in the stage of fibrosis subsequent to acut,e myocardial infarction. Elevation of the level of take-off’ of the RS-T segment would not be anticipatetl from tumorous invasion of the left ventricle, for the acute cellular and vascular reaction which follows acitte myocarTJikemisr, a changing 11%T segment, as dial infarction is lacking. seen in acute myocarclial iufawtion, would not. be c~spect.ed in a tumor of the left veiit,ricle, for thcsr RS-9’ changes in acute infarction are associated with a rapidly changing cellular reaction. This point is lvell illustrated in the case reportecl by Siegel and Young. On the basis of published tracings and on t,heoretical grounds, it is to be anticipated that most cJf the electrocardiographic changes that will be observed in cases of tumor of the heart will result from neoplastic invasion of the ventricles. It is probable that, invasion of the conducting systems mill account for a majority of abnormal electrocardiograms. Because of the much more frequent, invasion by t,umors of the right ventricle than of the left, right. bundle-branch defects should exceed those of the left bundle-bra.nch. Invasion of the muscle of the left ventricle without involvement of the bundle-branches or of the pericardium should produce t,racings closely similar to those obtained in the fibrotic stage following acute myocardial infarction. either alone or in conjunction with inInvasion of the epicardium, volvement of the ventricles, may produce or modify changes in t,he waves.“’ 7 If a patient, known to have or to have hacl a neoplasm, more or less suddenly exhibits these electrocardiographic changes in the absence of any other pathological process in the heart to explain them, then they constitute important presumptive evidence that the malignant process has invaded the heart.
BARNES
ET
AI,.
:
l’RIMARY
SARC’OXA
OP
IIE~AK’I
4x9
Pathology.-As we have stated previously, primary sarcoma of the heart is a pathological rarity. Perlstein, in 1918, by careful selection, admitted thirty cases of this condition from a compilation of the literature, and added an additional case. Lymburner recently (1933) reviewed the literature exhaustively, and was able to collect fiftyseven cases. Primary sarcomas of the heart have been designated as spindle-celled, giant cell, lymphosarcoma, round-celled sarcoma, mixed-cell sarcoma, fibrosarcoma or myxosarcoma, in agreement w&h the histological details presented by t,he cellular constituents of the tumors. Apparently the diagnosis of rhabdomyosareoma has been made only once, this in a case reported by Bradley and Maxwell? in 1928. Primary rhabdomyoma of the heart has been less often diagnosed, according to reports in the literature. Of this type of tumor, Lymburner could find only forty-seven examples, and added another case. Primary cardiac rhabdomyomas are considered to be benign tumors, in the sensethat in none of the examples referred to in t,he literature have metastatic growths be.en noted, although they have been multiple in the heart. Rhabdomyomas predominate in the early years of life. Of the 48 cases collected by Lymburner, 20 occurred within the first year of life, and only 5 between the twent,ieth and forty-fifth years. Rhabdomyomas of the heart have a further distinctive characteristic in being associated frequently with functional or anatomical developmental defects, epilepsy and tuberous sclerosis of the brain being the two most common of many such disturbances which have been found. It is believed that rhabdomyomas are an expression of abnorma.1 tissue differentiation, or represent secondary developments from embryonic tissue rests. They present a characteristic microscopic picture, in which the large, irregularly processed and vacnolated, so-called “spider cells, ” and long spindIe cells with delicate IongitudinaI fibrillae and cross st,riae form the essential differential points. The cells frequently contain excess glycogen. The relat,ionship which the typical rhabdomyoma of the heart might bear to primary sarcoma of the heart is difficult or impossible t.o ascertain accurately, since only once has the differentiation in sarcoma been considered sufficient to admit the diagnosis, primary rhabdomyosarcoma. From a more or less superficial survey of the field, it would appear that the two t,ypes of t,umor are distinct, and with possible exceptions, it seems improbable that primary sarcoma of the heart arises from malignant, transformations of a rhabdomyoma. Transitional stages apparentjIg do not exist, for as Lymburner remarked: “All reported cases (of rhabdomyoma) seemed to possess a similar microscopic appearance, regardless of the age of the patient, and evidently had obtained their final development in embryonic life.” Primary sarcomas and rhabdomyomas of the heart are further distin-
390
THJS
AMERIt’AN
HEART
JOUKNAI,
guished in that they appear in different age groups. Although cardiac rhabdomyomas predominate among young people, cardiac sarcomas predominate among middle-aged and older people; only two Cases have been recorded in which the patients were less than twenty years of age. Also, primary sarcomas are riot, associated with developmfmtal defects, whereas rhabdomyomas usually are. In reviewing cases reported in t.hr literature. we have found that t$here was one which strikingly resembled t,he case reported here. The case was described by Bradley and Jlaxwell. Their patient was a man, aged sixty-t.wo years; his clinical history was similar to that recorded for our patient. The gross and microscopic appearances of the tumor in rach case were similar, although not identical. The histological differentiation of the tumor described by Bradley and Maxwell was as a polymorphous cell sarcoma with many long spindle cells revealing fibrillae and acidophilic cytoplasm. Sections of t,he tumor were snbmitted by Bradley and Maxwell to Mallory. Ewing and Kroders. 811 concurred in the diagnosis of rhabdomyosarcoma. We have concluded,* as Bradley and Maxwell did in their case, that the diagnosis of the tumor in our case should be rhabdomyosarcoma, primary in the heart. In proof of this we must necessarily first offer evidence that the tumor was of primary cardiac origin. As evidence me may cite t,he clinical history. which gave every indication that the pa,tient ‘s illness from beginnin, 0’ to cncl wax car&w. Further proof of the cardiac origin is offered by the macroscopic findings at necropsy. had occurred, scarcely an! Relatively speaking, although metastasis tumor was found outside the hearts and pericardium. The pericardium and epicardium were involved throughout, and the right auricle and right ventricle were extensively infiltrated. The metastatic areas were small and nowhere gave thr impression of primary sarcoma. (lollversely, the cardiac tumor had no resemblance to metastatic tumors of the heart. many examples of which we have seen. Microscopic examination gives aclditional evidence of the primary cardiac nat.itre of the tumor, by way of its histologica, differentiation. The cells of the tumor were ext,reniely lmIyniorpIious ; there \vere many oval or long spindle forms ; the larger cells had abundant acitlophilic cytoplasm : in a, few cells faint, fibrillae and striations were observed. Kecatwe of the extreme degree of malignancy, the cells were usually t,otalIy undifferentiated. However, there were some. as indicated, which revealed sufficient detail to allow identification of the exceedingly suggestive fibrillae and st,riations. These could ha,rdIy be expect,ed to be so clearly defined as in the slow]) growing rhabdomyomas. because of the high degree of malignancy of the tllmor. The sumniat~ion of t,his evidence, me believe, warrants the conclusion, as previonslp stated. *Dr. tumor
H. and
E. Robertson have concurrerl
and in
DI-. the
AL C. diagnosis
Broders of
ilk+\-c reviewml rhsbdon?;\‘osnr~om;l.
the
sections
of
this
UARNES
ET
AL.
:
PRIMARY
SAR(‘OILIA
OF
lIEAWl
197
SUMMARY
A case is presented in which a diagnosis of tumor of the heart was made during life. The electrocardiographic abnormalities that have been reported in cases of cardiac tumor are reviewed; in our case, the electrocardiographic changes observed were important in est,ablishing the diagnosis. Careful study of the pathological changes in our caw. we believe, permits the conclusion that, the t,umor was a rhabdomposarcoma, primary in the heart. REFERENCES
The l’rinviplps of Pathology, Phila~lvlphia. 1. Adami, J. G., and Nichols, A. G.: 1908-1909, p. 158, Lea B Fehiger. Herzblock, bedingt durch primlren 2. Armstrong, H., and Monckeberg, J. G.: Herztumor, hei cinem 5-jahrigen Kincl, Drutsczhrs Arch. f. klin. 31~11. 162: 144, 1911. 3. Barnes, A. H., and Mann, .F. (1.: ~:le~tr~~ar~liog~aphi~ Changcbs Following Ligation of the Coronary Arteries of t,he Dog, Ax HEART J. 7: 4iT, 19X. 4. Beck, Ccl1 Sarcoma of the Heart. Arch. C. S., and Thatcher, H. 1;1.: Spintlle ht. Med 36: 830 1925 5. Bradley, E’B., and Maxwell, E. S.: Primary Neoplasnrs of the Heart; Report of an Unusual Case, J. A. M. A. 91: 1X!!, 192X. 6. Diebold, Otto : t’ber das Primdre Herxsarkom. Ztschr. f. Krrislaufforsch. 22: 785, ‘1930. 7. Fowlcr, W. M., Rathe, H. W., ant1 Smith, I’. M.: ‘l’hc ~:lrc*troc~arrlit)g~aI)hi(, Changes Following the Ligation of the Small Branrhen of t,hc (!oronar! Artcries, AM. HEART J. 8: 370. 19X:. 8. Goldstein, H. I.: Tunlors of the Heart, New York M. .J. 115: 97, 1.58, 19%. !I. Houck, B. H., anI1 Bennett,, G. A.: l’olvpoid Fihroma of the Left Aurivlrt (HoCalletl ‘Cartliar Myxm~a) (kusing ‘:; Bxll-\~alvv Action, >{M. HRAKT .1. 5: 787, 1930. IO. Karrenstcin: I:in E’all van P’ibroc~l:lstol,l~~~lll ~1~s Herzen:, un11 k-asuistisches zur Frage tlrr Herzgesrhwiilstc, besontic.rs IIV* Mpxomc~, Virvhow ‘s Arch. f. path. Anat. u. Physiol. 194: l?i, 1908. 11. Lloyd, P. C.: Heart Block Due to Primary I,yI:lpha”gio-en(lothrliolla oC Atrio-ventricular Node, Bull. Johns Hopkins Hosp. 44: 149, 1929. 1”. Lymburner, R. M.: Tumors of the Heart; Histopathologic autl Clinicaal Rtutly, Canad. M. A. J. (In press). 13 . Matras, A.: Ein Primares farkom des Hcrzens, Ztschr. f. Kreislaufiorscll. 19’ . “.) O33 192i. 14. Meroz, E.: A Clinical Study of Three Casts of Primary Tumor of the Heart. lnternat Clin 4’ * 231, 1917. 15. Perlstein, .I.: sarcoma of thr Heart. Ani. J. hr. Se. 156: ?I-&, 1918. 1 Ii. Pomnier, G.: %ur Kc~untnis ~Icr Priitliircv Hc~zg~~sc*hwiilstc~, %tsvhr. t’. Kireislaufforscthung. 23: 65. 1941. Anna M.: 1 i. &Siegel, M. I,., &,I Young, ~:l~ctrucarclioKraphi~ Findings in TUIIIOI.:. of the Heart: With a Report of a Casv, AM. HURT J. 8: 682, 1933. tiber einer Fall van diffusem Rhabdomyom des Hcrzens, 18. Uehlinger, Erwin: Virrhow’s Arch. f. path. Anat. u. Physiol. 258: il9, 1925. 19. Willius, I!. A., and Amberg, Samuel: Two Casvs of Secondary Tumor of the Heart in Children in One of Which thv Diagnosis Was Matlv During T,ifc, M. Clin. North linierica 13: 1X17. 19::O.