Hemodynamic Aspects of Diffuse Myocardial Fibrosis

Transcription

1 Hemodynamic Aspects of Diffuse Myocardial Fibrosis By EUGENE D. ROBIN, M.D., AND C. SIDNEY BURWELL, M.D. The circulatory changes associated with diffuse myocardial fibrosis have been stucliel in 11 patients. The characteristics of this form of congestive failure and its similarities to those imanifested by constrictive pericarditis and endocardial fibroelastosis are discussed. MYOCARDIAL fibrosis is defined as a diffuse replacement or invasion of the myocardium by fibrous connective tissue to such an extent that there is interference with the action of the heart. Eleven patients with this disorder have been studied. The purpose of this paper is to report the hemodynamic findings in these patients as studied by various standard methods including right heart catheterization. The changes in the circulation associated with diffuse myocardial fibrosis have not previously been studied in these ways. Aspects other than pathologic physiology will be reported in a subsequent communication.1 MATERIAL AND METHODS In 9 of the 11 patients an exploratory thoracotomy revealed a normal pericardium and a grossly diseased myocardium. In 3 of these the diagnosis of myocardial fibrosis was confirmed by microscopic study. In one patient (D.S.) neither operative nor postmortem proof of myocardial fibrosis was available. However, the clinical findings were so clear-cut that inclusion in this series seems warranted. In another patient (E.F.) myocardial fibrosis had developed in association with constrictive pericarditis. Despite surgical relief of the pericardial restriction. this patient eventually died of congestive heart failure. Postmortem examination revealed m) o- cardial fibrosis resulting from involvement of the myocardium by the inflammatory process that had produced constrictive pericarditis. This tyupe of myocardial involvement is frequent in patients with constrictive pericarditis.2 Data relating to the age of the patient, manifesta- From the Department of Medicine, Harvard Medical School, and the Medical Clinics, Peter Bent Brigham Hospital, Boston, Mass. This investigation was supported in part by a research grant (H-2243) from the National Heart Institute of the National Institutes of Health, U. S. Public Health Service, and by a grant from the Massachusetts Heart Association. TABLE 1.-Clinical Data in Patients with Myocardial Fibrosis Manifested by Intractable Right and Left Heart Failure Patient W. L. L. H. R. C. G. S. N. L. F. O'C. Wm. Wa. M. I. Wm. Wo. D. S. Age Etiologic basis of fibrosis 38 Coronary artery disease Tuberculosis? Old myocarditis Old myocarditis Coronary artery disease Coronary artery disease Diagnosis made by: and 1)ostmnortem examination O)peration and postmortem examination Postmortem exami nation and postmortem examination operation Clinical picture tions, and etiology are listed in table 1. The recorded etiologic antecedent of the fibrosis in each patient is the best judgment that could be reached with the evidence available. Cardiac catheterization was performed with the patient at rest. No sedative was administered. The catheter was introduced by, way of a vein in the antecubital space and manipulated until the tip occluded a branch of the pulmonary artery. Pulmonary "capillary" pressure was recorded.' The catheter tip was then withdrawn to a point slightly distal to the bifurcation of the pulmonary artery. At this time a short-bevel needle (20- or 21-gage) was inserted into the brachial artery. The lumen of this needle was kept patent by a slow infusion of saline solution. The patient's expired air was collected for a 3-minute period in a Douglas bag and midway during this collection samples of blood were withdrawn simultaneously from the brachial and pulmonary arteries. Pressures were then recorded 730 Circulation, Volume X II,.November 1957

2 ROBIN AND BURWELL TABLE 2.-Volumes of Blood Flow and Related Data in Patients with Myocardial Fibrosis Body surface area (M.,) 02 consumption (ml./min.) 02 saturation (No) A-V 02 difference (ml/l.) Cardiac output (L./ min.) Heart rate/ min. 1 1~~~~ 731 Cardiac Stroke inindex (L./ dex (ml./ min./m. 2) beat/m. 2) Stroke output (ml./ beat) Output per ml. O2 (L./min.) Rhythm W. L. L. H. R. C. G. S. N. L. F. 0. C. Wm. Wa. M. I. Wm. Wo. 1. S * t * Normal sinus rhythm. t Atrial fibrillation. Name W. L. L. H. R. C. G. S. N. L. F. O'C Wm.Wa. M. I. Wm.Wo. D. S. Peripheral venous mean TABLE 3. Right atrial mean Pressures and Related Data in Patients with Myocardial Fibrosis Pulmonary artery1 Pulmonary artery Pulmo- Brachial artery nary Right ventricular Syst. Diast mean = Syst. Diast. Mean in these arteries, the catheter tip was withdrawn and pressures were recorded in the right ventricle, the right atrium, and the superior vena cava successively. The volume of air expired in the 3-minute period was measured in a Tissot spirometer and the concentration of oxygen was determined by a Pauling oxygen analyzer. A respiratory quotient of 0.82 was assumed, and appropriate correction was made for the change in nitrogen concentration in the expired air.4 The oxygen content and oxygen capacity of the blood samples were determined by the method of Van Slyke and Neill.5 All pressure measurements were made by means of a Sanborn electromanometer recording on a multichannel direct-writing oscillograph. Mean pressures were obtained by electric integration. The zero point for pressure measurements was 10 cm. anterior to the back with the patient supine.6 Cardiac output was calculated by the Fick equation.' "Capillary" mean Syst. Diast. AMean i5 TV 110n 8UV RESULTS Pressure plateau Diastolic dip Volumes of blood flow and related data are shown in table 2. Cardiac outputs varied from 2.0 to 5.3 L. per minute. The customary derivatives of cardiac output (cardiac index, stroke output, and stroke index) were found to be low. To relate the output to actual oxygen consumption instead of to body surface area, the cardiac output per ml. absorbed oxygen has been calculated. In every patient this was below the expected figure of Five patients had atrial fibrillation, the rest had normal sinus rhythm. The height and contour of pressure curves in various parts of the circulatory system are shown in table 3. These data are summarized as follows:

3 732 HEMODYNAMIC ASPECTS OF DIFFUSE MYOCARDIAL FIBROSIS I E E Peripheral Right Right Pulmonary Pulmonary Venous Auricular Ventricular Artery "Capillary Mean End Diost. Diastolic Mean W.L. L.H. R.C. G.S. FO'C. D.S. WmWo. N.L. Wm.Wo. Peripherol Right Right Pulmonary Putmonary Venous Auricular Ventriculor Artery Capillary Mean End Diost. Diastolic Mean FIG. 1 Top. Intravascular pressures in 4 patients showing a plateau of the "biventricular" type. FIG. 2 Bottom. Intravascular pressures in 7 patients showing the absence of a plateau in the "left ventricular" group. 1. Every patient showed a definite elevation of pressures in the pulmonary "capillaries," the pulmonary artery, the right ventricle, the right atrium and the peripheral veins.* 2. These patients can be divided into 2 groups: (a) The "biventricular" group. Four patients (W. L., L. H., R. C., ) showed a plateau of pressures from the peripheral veins to the pulmonary "capillaries." In these patients the mean pressure in the peripheral veins, the right * Three patients showed a pulmonary "capillary" pressure higher than pulmonary artery diastolic pressure. This technical error is related to the difficulty of obtaining accurate wedge pressures in critically ill patients. atrium, the end-diastolic pressure in the right ventricle, the diastolic pressure in the pulmonary artery, and the mean pressure in the pulmonary "capillaries" did not differ by more than 5 mm. Hg (fig. 1). Since these pressures indicated abnormal function of both ventricles this group of patients is referred to as the "biventricular" group. Postmortem studies done on 2 of the patients (W. L., ) who exhibited pressure plateaus showed myocardial fibrosis involving both ventricles. (b) The "left ventricular" group. In 7 patients (G. S., N. L., 1F. O'C., Wn. Wa., 1\I. I., Wm. Wo., D. S.) the pressures relating to the function of the left side of the heart (i.e., pulmonary "capillary" and pulmonary artery pressures) exceeded the pressures in the peripheral veins and the right atrium by at least 5 mm. Hg (fig. 2). Therefore, a plateau was not. Two of these 7 patients (G. S., N. J,.) have had postmortem examinations. Both showed predominant fibrosis of the left ventricle with relative sparing of the right. The significance of these 2 groups is discussed later in this paper. 3. Pressures in the systemic arteries were within normal limits in all patients. 4. Eight out of 9 recorded right ventricular pulse contours showed a diastolic dip. The values observed during catheterization of normal subjects in this laboratory have been used as normal standards. DIscussIoN It has been our experience that limited or localized fibrosis of the myocardium produces no signs or symptoms; therefore such fibrosis presumably leads to no major hemodynamic abnormalities. Widespread fibrosis, however, does lead to hemodynamic changes and to signs and symptoms based upon them.' It appears that the hemodynamic effects of myocardial fibrosis depend on the nature, location, and extent of the fibrosis. The general pattern of hemodynamic changes in this group of patients is conisistent. This consistency stems partially from the fact that this series is a selected group in the sense that at some time in the study of each patient the possibility of constrictive pericarditis was considered.

4 ROBIN AND BURWELL 733 The data ed indicate that the cardiac failure of myocardial fibrosis is a form of lowoutput failure. As a working hypothesis it is suggested that this low-output failure results from 2 basic physiologic defects, namely, limitation of diastolic filling and impairment of systolic emptying of the heart. There is impressive evidence of limited diastolic filling. Extensive infiltration with dense fibrous tissue produces a stiffened and restricted myocardium that imposes its own altered characteristics of distensibility on both sides of the heart. This fibrous infiltration is evident by gross and microscopic examination and limited diastolic filling is manifest in the pressure plateau. The absence of a pressure plateau in some patients is not incompatible with this thesis. It appears probable that a pressure plateau is produced by myocardial fibrosis of such distribution that it limits diastolic filling to a similar degree on both sides of the heart. Isaacs et al.8 studied the pathologic physiology of experimentally produced constrictive pericarditis and showed that interference with ventricular diastolic filling was a fundamental defect in this disease. This defect was explained as due to alterations of ventricular volume-elasticity characteristics by the pericardial scar. Production of isolated left ventricular pericardial constriction was followed by pulmonary congestion, as demonstrated by both physical signs and physiologic measurements. Isolated right ventricular pericardial constriction produced a striking increase in peripheral venous pressure, right atrial pressure, and right ventricular end-diastolic pressure. This took place without a concomitant rise ili right ventricular systolic pressure, pulmonary artery pressure, or pulmonary "capillary" pressure. Constriction of both ventricles led to a pressure plateau by producing equivalent increases in all vascular pressures from the peripheral veins to the pulmonary "capillaries." When myocardial fibrosis is predominantly left-sided there are relatively high pressures in the right ventricle (systolic), pulmonary artery, and pulmonary "capillaries." In such patients the absence of a plateau may be expected. Further evidence of impairment of diastolic filling is supplied by pressure tracings. Those RV RA FIG. 3. Right ventricular and right atrial pressure tracings in a patient (R.C.) with myocardial fibrosis showing the "diastolic dip" and the "atrial M." from the right ventricle may show a "diclstolic dip" and those from the right atrium may show an "atrial M." These types of curves were originally described as occurring in constrictive pericarditis. An example of these pressure (contours is shown in figure 3. In the ventricular curve, after the maximum pressure during systole, the curve falls rapidly but does not reach zero. Having reached its minimum, it rises steeply to a level between maximum and minimum and forms a plateau that does not change further until the onset of systole. The "atrial AI" is characterized by a higher maximum than the normal, and by 2 plateau-like maxima of about equal height. _Most workers agree that such pressure contours indicate interference with diastolic filling.9 When such curves are due to constrictive pericarditis, removal of the constricting pericardium and improvement in diastolic filling is associated with disappearance of the "diastolic dip" and the "atrial WI."10 So much for the limitation of diastolic filling in patients with myocardial fibrosis. It is now pertinent to discuss the possibility of impairment of systolic emptying in these patients. It is difficult to devise a direct method of imeasuring impairment of systolic emptying and to rule out the effect of inadequate filling on systolic contraction. It seems reasonable to think that a myocardium, the muscle fibers of which are largely replaced by fibrous tissue, suffers impairment of contractility. It does not appear possible, for example, that the myocardial fibers shown in figure 4 (patients W. L. and N. L.), could contract effectively even if the stretch of the fibers were normal, but no quantitative evidence is available. Impairment of systolic emptying thus remains unconfirmed, but ap-

5 734 HEMODYNAMIC ASPECTS OF DIFFUSE MYOCARI)IAL FIBROSIS FIG. 4. Microscopic sections of myocardium in 2 patients (left, W. L.: right, N. L.) with myocardial fibrosis. pears a probable part of the hemodynamic abnormalities of myocardial fibrosis. The similarity of the hemodynamic changes in myocardial fibrosis and constrictive pericarditis is striking. A low cardiac output, elevated pressures from peripheral veins to pulmonary "capillaries," a pressure plateau, a "diastolic dip," and an "atrial M" are all findings generally in both diseases. This similarity in manifestation stems from the fact that the 2 disorders cause similar changes in cardiac dynamics. Both lead to a somewhat rigid envelope of fibrous tissue that mechanically interferes with cardiac filling. It is apparently not important that in the case of constrictive pericarditis this rigid area is located on the outside of the heart, while in the case of myocardial fibrosis it is situated within the myocardium. Indeed, endocardial fibroelastosis, a disease in which a rigid layer of fibrous tissue invades the inner layer of the heart, has been reported by Clark, Valentine, and Blount"1 to produce hemodynamic changes identical with those described above. Since it has been shown that all 3 disorders, constrictive pericarditis, myocardial fibrosis, and endocardial fibroelastosis, have similar physiologic defects, it is not surprising that they clinical pictures that are remarkably similar.' 2 It is important to ask whether the patients reported here are really reative of the wide spectrum of patients with myocardial fibrosis. Hansen, Eskildsen, and Gdtzsche'0 have reported a single case of myocardial fibrosis with findings similar to those described above. Hetzel, Wood, and Burchell"1 have reported the results of cardiac catheterization in one patient with amyloid disease and in another with idiopathic heart failure. In each case the catheterization findings were similar to those reported in the paper. Therefore, this group of patients does appear to be reative of patients with diffuse myocardial disease. SUMMARY The hemodynamic findings in 11 patients with diffuse myocardial fibrosis are described. Myocardial fibrosis can lead to low-output failure. The low output results from interference with diastolic filling and probably also

6 from interference with systolic emptying of the heart. The interference with diastolic filling is related to changes in the volume-elasticity characteristics of the fibrotic myocardium. Two other disorders-constrictive pericarditis and endocardial fibroelastosis-that have similar physiologic defects, show hemodynamic changes and a clinical picture similar to those of myocardial fibrosis. SUMMARIO IN INTERLINGUA Es describite le constatationes hemodynamic in 11 patientes con diffuse fibrosis myocardial. Iste condition pote resultar in disfallimento per basse grados de rendimento cardiac. Le basse rendimento es le consequentia de obstructiones del replenamento diastolic e probabilemente etiam del evacuation systolic del corde. Le obstruction del replenamento diastolic es relationate a alterationes del characteristicas de volumine e elasticitate del myocardio fibrotic. Duo altere disordines pericarditis constrictive e fibroelastosis endocardial-in que simile defectos physiologic es manifeste exhibi simile alterationes hemodynamic e un simile tableau clinic. REFERENCES 1 BURWELL, C. S., AND ROBIN, E. D.: To be published. 2 -, AND : Some points in the diagnosis of myocardial fibrosis. Tr. A. Am. Physicians 67: 67, HELLEMS, H. K., HAYNES, F. WV., AND DEXTER. ROBIN ANDI) BJURWELL 735) L.: Pulmonary "capillary" pressure in man. J. Appl. Physiol. 2: 24, DEXTER, L., WHITTENBERGER,.J. L., HAYNEE, F. W., GOODALE, WM. T., GORLIN, R., AND SAWYER, C. G.: Effects of exercise in circulatory dylnamics of normal individuals. J. Appl. Physiol. 3: 439, 'VAN SLYKE, D. I)., AND NEILL, J. AM.: The (letermination of gases in blood and other solutions by vacuum extraction and manometric measurement. J. Biol. Chem. 61: 523, LYONS, R. H., KENNEDY, J., AND BURWE LL, C. S.: The measurement of venous pressure by the direct method. Am. Heart J. : 675, FiCK, A.: Ueber die Messung des Blutquantums in den Herzventrikeln. Sitzungsb. d. phys.-med. Gessellsch. zu Aiirzb., p., ISAACS, J. P., CARTER, B. N., AND HALLER, J. A.: Experimental pericarditis: The pathologic physiology of constrictive pericarditis. Bull. Johns Hopkins Hosp. : 259, McKusicK, V. A., AND HARVEY, A. M.: Diseases of the pericardium. In Advances in Internal M\1edicine. Chicago, Year Book Publishers, HANSEN, T., ESKILDSEN, P., AND G6TZSCHE, H.: Pressure curves from the right auricle and right ventricle in chronic constrictive pericarditis. Circulation 3: 881, CLARK, G. M., VALENTINE, E., AND BLOUNT, S. G., JR.: Endocardial fibrosis simulating constrictive pericarditis: Report of a case with determinations of pressure in the right side of the heart and eosinophilia. New England J. Med. 254: 349, HETZEL, P. S., WOOD, E. H., AND BURCHELL, H. B.: Pressure pulses in the right side of the heart in a case of amyloid disease and in a case of idiopathic heart failure simulating constrictive pericarditis. Proc. Staff MIeet., Mayo Clin. 28: 107, 1953.