Paper # 008 Versión en Español Versión en Español

Purkinje cell hamartoma (Histiocytoid cardiomyopathy). Study of a case in an 18 months infant.

Marcial Garcia-Rojo, Carlos Gamallo, Felipe Moreno

[Title] [Introduction] [Materials & Methods] [Results] [Picture 1] [Discussion] [Bibliography]

Other findings


[Definition] [Clinical] [Gross] [Microscopy] [Techniques] [Other findings]

LITERATURE REVIEW. Anatomic Pathology


Interpretation of the process


I) Abnormal development of the cardiac conduction system:

The lesions have been described as Nodular aggregations of Purkinje cells or Purkinje cells tumor(5).

The possibility must be considered of an affection primarily or mainly concerned with the myocardial fibers of the conduction system(5). The foamy cells exhibit a structure which is very similar to that of immature Purkinje cells(16). Sudanophilic lipids can also be demonstrated in the cells of the normal specific cardiac tissue(16). The distribution of archnocytes(subendocardial along the ventricular and septal wall) exactly corresponds to the location of the bundle of His and its branches, and of peripheral Purkinje cells(16). The lack of involvement of both cardiac nodes is well in the line with the evidence of a separate ontogenetic derivation of the nodes on the one hand and the bundles on the other hand(16). The activity of acetylcholinesterase is present normally in specific cardiac tissue and is lacking in working myocardium. Positive reaction for this enzyme activity is found in at least some of their typical foamy cells(16). The primary fault appears to be failure of Purkinje cells to mature and differentiate into longitudinally arrayed fibers(23). Despite this, ECG were essentially normal the first few months of life, and that could be explained by a late-stage "dedifferentiation" of Purkinje cells, which seems improbable. Or it could be that the intrinsic cellular abnormality which failed to permit normal postnatal development into longitudinally disposed Purkinje fibers also caused these cells to enlarge somewhat, the consequences of which may be loss of conducting ability by them(23). The presence of xanthomatous fibers, both in the bundle of His and in the common myocardial fibrocells, might warrant consideration of a DIFFUSE lesion of the specific myocardium, i.e., involving the bundle of His to the outermost periphery of Purkinje network(5). This interpretation is strengthened by the fact that the lesions are focal and uniform, i.e., without grading toward the normal limitrophe myocardial fibers, from which they are clearly separated(5).

The occasional epicardial or valvular location of lesions, however, are not consistent with the view that the conduction system proper is abnormal(8). Whether the cell of origin of this lesion is a Purkinje cell or a nonspecific myocardial cell cannot be established at this time(8).

The distribution of the lipid-filled cells, primarily found in the subendocardial connective tissue adjacent to the A-V valves and extending towards the A-V node( an EMBRYOLOGICALLY peculiar region) arouses the possibility that this region may have distinct derangements(20). The possibility of a CONGENITAL defect involving the DIFFERENTIATION of certain groups of cardiac muscle cells cannot be excluded, although it is difficult to explain all of the features (particularly the hypertrophy) on the basis of such a defect.

The occurrence of endocardial fibroelastosis in one case(4) suggest the presence of myocardial functional defect, at least at a time when the cells are morphologically abnormal, since increased transmural pressure is a postulated factor in endocardial fibroelastosis(4).


II) Hamartoma de células de Purkinje:

In some authors' opinion, it is a myocardial hamartoma.

The lesions represent an abnormality of cardiac development of a HAMARTOMATOUS nature. The features of a hamartoma that are illustrated by this lesion include location of the cellular aggregates in parts of the heart in which the cell type may be found normally or heterotypically, a tumorlike growth pattern that is relatively proportionate to the surrounding normal tissue with no apparent mitotic figures, the prevalence in infants suggesting that the lesion may have been present at birth, an the association of developmental abnormalities in other organs(8). The constituent large, rounded cells with few myofibrils suggest a primitive myocyte morphology. Similar considerations have led to the classification of cardiac rhabdomyomas as hamartomatous malformations(8).


III) Cardiomyopathy

The clinical and pathological alterations described have benn frequently described as a CARDIOMYOPATHY, the etiology of which is unclear(3,13). Besides mitochondrial cardiomyopathy there was involvement of skeletal muscle(22)In this case the diffuse hyperplasia of mitochondria was demonstrable with the same intensity in all regions of the heart, and all heart muscle cells showed the usual elongated form. The authors separated this case from the cases of oncocytic or so-called histiocytic transformation of heart muscle cells(22), in which there is a focal transformation of heart muscle cells(22).

The "transitional cells" have been offered as further supportive evidence of a cardiomyopathic disease process( 27,3,4,36).

The localized collections of the abnormal cells, and the apparent successful cure of the condition by surgical excision, argue against cardiomyopathies as they are currently understood(8).

On the basis of the lack of postoperative recurrence of ventricular tachycardia and the normalization of ventricular function after the cardiac rate and rhythm were restored to normal, it would seem reasonable to assume that this lesion is not a progressive cardiomyopathy(9).

The possibility that we are dealing with a TOXIC form of cardiomyopathy should also be considered(14). A retrospective survey revealed no toxic exposure(2,14). Exposure to toxic agents could only be documented in one patient(20).

In one case there was some narrowing of the outflow region in left ventricle, but in typical HYPERTROPHIC CARDIOMYOPATHY the hyperplasia of partially abnormal mitochondria with extreme reduction of myofibrills seen here has never been observed(22). Adult patients with hypertrophic cardiomiopathy have mitochondrial proliferation and myofibrillar loss, but it is not associated with cellular enlargement or spherical cell shape and histiocyte-like appearance(3). Degenerated cardiac muscle cells showing mitochondrial proliferation in adult patients do not form clusters or nodules(3).

Las alteraciones clínicas y anatomopatológicas descritas en esta lesión se interpretan con frecuencia como una miocardiopatía de etiología desconocida(3,13). Para algunos autores, la lesión cardiaca es una miocardiopatía oncocítica y el agente es un virus, como la rubéola. El efecto viral tan especial en estas células (hiperplasia de mitocondrias) podría estar relacionado con el nivel materno de inmunidad materna(12).

Las células de aspecto intermedio apoyarían que se tratase de una miocardiopatía(3,4,27,36).

Algunos autores han considerado la posibilidad de una forma tóxica de miocardiopatía(14), pero un estudio retrospectivo demostró la ausencia de exposición a tóxicos(2,14), excepto en un paciente(20).

En un paciente, además de una "miocardiopatía mitocondrial" difusa en todo el corazón, incluso en células de forma normal, se identificó afectación del músculo esquelético(22).

Argumentos en contra de una miocardiopatía:

  1. La distribución localizada de acúmulos de células anormales(8,9).
  2. El tratamiento exitoso mediante extirpación quirúrgica de la lesión(8,9).


En un caso se describió un estrechamiento del tracto de salida del ventrículo izquierdo, pero en la miocardiopatía hipertrófica típica no se ha descrito la hiperplasia de mitocondrias parcialmente anómalas con reducción llamativa de miofibrillar(22), excepto en algunos pacientes adultos, aunque en estos últimos no hay un aumento del tamaño celular con formas esféricas de las células y aspecto histiocitario y las células musculares anómalas no se disponen en grupos o nódulos(3).


IV) Metabolism error

Evidence favouring an INBORN ERROR OF METABOLISM: Biochemical studies of cardiac muscle showed a marked defect of reducible cytochrome b, suggesting that the disease may be due to an INBORN ERROR OF MITOCHONDRIAL ELECTRON TRANSPORT(15). Similar morphologic changes have been described in skeletal muscle of many patients with "mitochondrial myopathies. In these disorders an increasing number of specific biochemical errors, including several distinct defects of the respiratory chain components, have been documented(15). A secondary impairment of lipid oxidation (decrease of reducible cytochrome b and cytochrome cc1) could explain the abnormal accumulation of lipid droplets in the hearts of patient with histiocytoid cardiomyopathy(15). Histiocytoid cardiomyopathy represents the first biochemically documented disorder of cardiac mitochondria(15).

This case with deficiency of myocardial cytochrome B(15) does not appear to be this clinicopathologic entity since the baby's illness began with failure to thrive and led to a fatal cardiac arrest without reported arrhythmia(8).

It has been considered a LIPIDOSIS that affects the heart, part of a generalized disorder (11). A congenital storage disorder confined to one small part of the heart(11). The changes observed are not suggestive of an inborn error of metabolism involving the heart(3). Other authors think it is not a lipidoses(2).

The ultraestructural studies indicate that accumulation of extraorganelle or nonmembrane lipid is NOT a significant feature of abnormal cells(4).

A storage disease is practically excluded by the histological and fine structural investigation(22).

Both myocardial and connective tissue cells are affected in this disorder(20,27), so that it cannot be explained by neoplasia of histyocytes(20). Although the HISTIOCYTE-like appearance was pronounced in histologic preparations, the ultrastructural feaures of histiocites are not present in the histiocite-like cardiac muscle cells(3).

Mitochondrial abnormalities have been described in association with trygliceride accumulation in the heart and liver of patients with REYE's syndrome(14). Changes indicative of Reye's syndrome were specifically sought and not found in these patients(3).

The quality of DIETARY LIPID is a factor of potential importance in the pathogenesis of myocardial fiber abnormalities, including focal fatty degeneration of myocardial fibers accompanied by miofibrillar loss, mitochondrial vacuolation, and increased miticondrial concentration. These focal degenerative changes precede the development of infarcts(14).

In CARNITINE DEFICIENCY there is only a moderate increase in mitochondria. The skeletal muscle always show an intensive fat storage(22).


V) Cardiac rhabdomiomas or glycogen deposits:

Pathologic changes have been interpreted as RHABDOMIOMAS(1,23) or a nodular form of glycogen storage disease(23). These histiocyte-like cells do bear some resemblance to the large, clear cells in cardiac rhabdomiomas. The latter cells do have myofibrills arranged either radially or in narrow, peripheral zones parallel to the plasma membranes and their cytoplasm is clear rather than extensively vacuolated. Rhabdomyomas do not originate from, or invade cardiac valves(3). Cardiac rhabdomyoma cells have intercellular junctions, with desmosomes and nexuses, all along their periphery. These cells usually have few mitochondria and large amounts of glycogen; they contain clearly recognizable myofibrils as well as peculiar banded structures known as leptofibrils(3). Leptofibrils have been suggested to form in myocytes whose myofibrillar development has been arrested or diverted at a relatively early stage. They have been thought to be prodromal phase of normomeric myofibrils, but most authors believe that they have a mechanical function in muscle-collagen junctions(12).

The cells in oncocytic cardiomyopathy had not differentiated beyond a primitive stage when they changed into oncocytes(the same reasoning given in the pathogenesis of cardiac rhabdomyoma)(12).


VI) Myocardial degeneration

The lesion is myofilament loss and mitochondrial hyperplasia in myocardial cells and can be considered as a peculiar focal myocardial degeneration(2,14). The degenerative changes sometimes found in myocardial tissue (necrosis, inflammation) and the accumulation of lipid-filled cells may represent cellular injury of an unrecognized type rather than neoplasia(20). The change in shapes of these cells appeared to be made possible by the loss of myofibrils and by the immaturity of the connective tissue framework that surrounds the cardiac muscle cells in the very young children(3). The marked increase in cell size is due to cytoplasmic distention by both swelling and proliferation of mitochondria(3). The accumulation of glycogen and fat, both of which normally serve as sources of energy for contraction, may reflect the loss of contractile function in these cells, and lipid deposition is an inconstant feature of the histiocyte-like cells(3). The "positive" fat stains may at least be due to staining of membrane lipid of the markedly increased number of mitochondria(4). These cells are cardiac muscle cells that have assumed a very primitive type of morphology. Increased automacity, with a pronounced tendency to produce ectopic cardiac rhythms, may be a consequence of this peculiar morphology(3). In a series of 26 infants and children who had been diagnosed at necropsy as having myocardial necrosis of various cases(post-exchange transfusion, congenital heart defects, cardiac surgery, myocarditis, vasculitis, inhalation of toxic), in no instance large, foamy, histiocyte-like cardiac muscle cells were found(3).

CHRONIC HEART FAILURE as a cause of this lesion may be minimized on clinical and structural grounds(14).

Irreversible ischemic changes occur during fatal episodes of paroxysmal atrial tachycardia(14). It is possible that these cardiac cellular alterations represent a peculiar form of myocytolisis occurring in small children as a result of prolonged, severe ARRYTHMIAS(3). It seems unlikely that ischemia alone would result in selective destruction of myofibrills in fibers that remain intact following injury and show only a configurational change accompanied by increased lipid droplets and concentration of mitochondria(14). Reversible myocardial trygliceride accumulation occurs, perhaps on the basis of sublethal hypoxic injury(14). It is possible that sublethal myocardial hypoxia, caused by repeated brief episodes of paroxysmal atrial tachycardia, might produce subendocardial patches of progressive myocardial fiber degeneration in areas where residual lipid from a previous episode had not been completely cleared. Loss of myofibrills might also result from a noxious effect of the lipid, particularly of free fatty acids, on mitochondrial function(14). However it should be emphasized that it is not possible, at present, to distinguish between primary and secondary disorders of mitochondrial on the basis of morphological changes(14). The presence of cardiac hypertrophy and the gross and microscopic features of the lesions( which have a different distribution from that of the ischemic lesions) favor the concept that these arrhythmias are a consequence rather than a cause of the heart disease(3), and the ischemic lesions result from prolonged arrhythmias and associated poor myocardial perfusion(3). Histiocyte-like cardiac muscle cells have not been described in other morphologic studies of myocardial ischemia and necrosis or of cardiomyopathies in children(3). The ultrastructural features of myocardial ischemia differ from those of the histiocyte-like muscle cells(3).

The SYMPATHOMIMETIC agents used to support some of these patients’s blood pressure may also have contributed to the lipid accumulation and to myocardial necrosis(14).

These patients had increased cardiac weight, even though most of them died within a few days after developing their terminal illnesses(3). Thus, cardiac HYPERTROPHY must have developed in these patients well before the onset of arrhythmias(3).


VII) Infección:

Nothing in the clinical or pathological studies suggested an INFECTION(2).

Several of the patients had prodromes characterized by illnesses that definitely were or could have been caused by viruses(roseola, chikenpox, upper respiratory tract infections,enlarged mesenteric lymph nodes, pulmonary infections), while others had recently received vaccinations against viral diseases(3).

The proposal that this lesion may represent virus-induced transformation of myocardial fibers is based solely upon the finding of rare intranuclear inclusions by light microscopy and the finding of an unusual filamentous branching pattern of nuclear chromatin in electron micrographs of occasional myocardial fibers. There was no inflammation, however, and no unequivocal viral particles could be found. Moreover, lipid accumulation is not a feature in any of the viral myocarditides with which we are familiar(14). The lesions observed in this syndrome have not been described in viral myocarditis(3). Viral cultures(even of the heart(4)) were negative(2,20,4,12).

The results of viral serology were not considered diagnostic for a rubella infection(36,12).

Morphologic evidence that might suggest active or healed myocarditis in the hearts was present in 28-56% of the infant(12). Central nervous system and ocular associated abnormalities are in keeping with an intrauterine infection(12). But it would not explain the marked predominance of females(12). No patient had any cardiac malformation(in particular, patent ductus arteriosus)(12). However, the rubella virus could damage the cells and cause oncocytic change without cell death and consequent malformation during the critical period of organogenesis(12).

Oncocytic transformation of myocytes should have occurred prior to four week's gestation, leaving transformed early myocytes(with a paucity and disarray of contractile elements) in sites from which they normally disappear(12).

The morphologic changes in rubella consist in a peculiar myocarditis characterized by granular and vacuolated swelling and eosinophilic changes of myocytes with loss of cross striations and nuclear pyknosis but with an absence of inflammation(12).

Some authors have suggested that an oncocyte should be defined solely by the ultraestructural criterion of mitochondrial hyperplasia(12). Cardiac muscle has been described as an epithelial tissue developmentally by reason of the ultrastructural presence of apical junctional complexes(12). The cardiac lesion in an ONCOCYTIC CARDIOMYOPATHY, and the agent was a virus, such as rubella. A different and rarer viral effect, possibly related to the maternal level of immunity and the timing of the insult(12).

The accumulation of small lymphoid cells in close spatial relationship to subendocardial foamy cells may be suspected to have resulted from a local derangement of lymph circulation, as seen for example, in lymphangiomas(16).


VIII) Degeneración postnatal:

La muerte celular en algunas áreas del sistema de conducción aurículoventricular, sobre todo en la mitad izquierda del haz de His, es considerado un proceso degenerativo normal que moldea y da forma al haz de His(3). Las células afectadas son grandes y pálidas y recuerdan algo a las células anómalas descritas, aunque estas no son componentes del sistema de conducción y los cambios morfológicos son distintos(3). El estadio final del proceso de morfogénesis normal lleva a la desaparición de las células mediante fagocitosis por macrófagos(3).


IX) Factores genéticos:

The peculiar sex distribution is unexplained(3). The rarity of male cases does not seem to permit GENETIC attribution(5). The female predominance has led some authors to suggest a possible X-linked disorder( 36). No evidence of a familial incidence(2,3), except in one series with two families affected with four cases(25). From these family histories it would seem likely that foamy myocardial transformation of infancy may be an inherited condition, possibly due to an autosomal recessive gene rather than an X-linked condition(25). A simple autosomal recessive condition, however, cannot explain the female predominance of the reported cases(25).

Sengers et al (1975) described 7 children from 3 families with hypertrophic cardiomyopathy and mitochondrial myopathy(37). This report obviously deals with another type of cardiomyopathy. All this patients suffered from cataracts, furthermore heart and skeletal muscle showed extensive deposition of fat droplets(22,37).



Arrhythmias develop only in the terminal phase of the syndrome(3).

When a sufficient amount of abnormal substance has accumulated in the affected region(11). The small tumors do grow(23).

Some of the clinical difficulty and progression of cardiac electrical instability could be attributed to extracardiac problems such as dehydration (vomiting, diarrhea) or fever(23).

The lesions are in a dynamic state of development/ progression(4).

The presence of supraventricular tachycardia in one patient on the day after birth supports a conclusion that the cardiomyopathy was present at birth (12).


Other findings