The Hemodynamics of Common

Transcription

1 The Hemodynamics of Common (or Single) Ventricle By SHAHBUDIN H. RAHIMTOOLA, M.B., PATRICK A. ONGLEY, M.B., CH.B,. AND H. J. C. SWAN, M.B., PH.D. THE anatomy of common ventricle has been the subject of a previous communication.' The present report concerns the basic hemodynamic data in 43 cases of common ventricle and emphasizes those features of diagnostic laboratory significance that may permit accurate recognition of this condition during life. Definition and Terminology Anatomic In this paper a common (single) ventricle is defined as a ventricular chamber that receives the input of blood from both atrioventricular valves either separately or as a common atrioventricular valve" (fig. ). This common ventricular chamber may or may not communicate with a secondary, smaller outflow chamber. The two great vessels (aorta and pulmonary artery) may arise from the common ventricular chamber or, more frequently, one vessel may arise from the common ventricular chamber and the other vessel from the smaller outflow chamber.' A case has been reported in which apparently neither great vessel arose from the smaller chamber.3 In the presence of transposition of the great vessels, if the aorta lies to the left of the pulmonary artery, the term "levo" or "L-transposition" is used, and, if the aorta lies to the right of the pulmonary artery, the term "dextro" or "D-transposition" is used' (fig. ). Physiological The magnitude of a right-to-left shunt is the percentage of systemic arterial blood flow From the Mayo Clinic and Mayo Foundation, Rochester, Minnesota. 4 derived from the systemic veins, and the magnitude of a left-to-right shunt is the percentage of pulmonary arterial blood flow derived from the pulmonary veins. Method and Group Studied The 43 patients included in this report were those studied during the period May 960 to December 964 for whom satisfactory hemodynamic data are available. Of the 43 patients, 30 were males and 3 were females (ratio of male to female was.3:). Their ages ranged from 5 weeks to 7 years, with a mean of 7.6 years. The diagnosis of common ventricle in each case was established by means of selective angiocardiography. Cardiac catheterization was carried out as previously described.4' 5 Calculations of pulmonary and systemic blood flows, of resistances, and of shunts (left-to-right and right-to-left) were made in the standard manner.0 Results Incidence Common ventricle was noted in 43 of,35 patients (3.%) with congenital heart disease who were studied by cardiac catheterization at the Mayo Clinic during this period. The distribution of the 43 patients according to ages and whether or not severe pulmonary stenosis was diagnosed angiocardiographically is shown in table ; 70% of the patients at the time of study were from to 5 years old. Associated Lesions (Table ) Transposition of the great vessels occurred in 84%; three patients had incompetence of the right atrioventricular valve. The left atrioventricular valve could not be assessed by the methods of study employed. Two patients had patent ductus arteriosus, and both had severe pulmonary hypertension and some reversal of shunt through the ductus detected

2 COMMON VENTRICLE 5 Figure Left panel: Diagrammatic representation of common ventricle, with inversion of the ventricles and atrioventricular valves, a subaortic chamber, and L-transpositiom of the great arteries. Right panel: Great vessel position in L- and D-transposition. Both atrioventricular valves open into a common ventricular chamber in all cases. Table Ages of Forty-three Patients with Common Ventricle According to Time of Diagnostic Studies Severe pulmonary stenosis* Age, yr With Without Less than h 3" to to 5 0 StolS to 30 4 Total 3 *See text for discussion of "severe" pulmonary stenosis. only by indicator-dilution curves; one had an associated coarctation of the aorta and L-transposition, and the other had normally oriented great vessels. Obstruction to Aortic Flow Some measure of the degree of obstruction may be obtained by a comparison of the peak systolic pressure between the common ventricular chamber and the femoral artery (table 3). In eight patients the peak systolic pressures were equal. In nine the peak systolic pressure in the common ventricle exceeded that in the femoral artery; the pressure differences varied from 6 to 7 mm Hg. Three patients, with the highest gradients, had associated coarctation of the aorta. These small pressure differences were associated with normal systemic flows, and thus it appears that the patients in this series did not have serious obstruction to aortic flow. Right-to-Left Shunt Every patient had a right-to-left shunt that was best detected by the indicator-dilution technique and evidenced by values of systemic arterial oxygen saturation of less than 94%

3 6 6RRAHIMTOOLA ET AL. Table Associated Lesions in Forty-three Patients with Common Ventricle Associated lesiotis Position of great vessels Normal Aorta, anterior and to left (L-transposition) Aorta, anterior and to right) Aorta, directly to right ( Isolated levocardia, isolated dextrocardia, dextrocardia Isolated levocardia Isolated dextrocardia Dextrocardia with situs inversus totalis Interatrial communication With isolated levocardia or dextrocardia Heart and viscera in situs inversus Heart and viscera in situs solitus Probable common atrium ( with isolated levocardia; with isolated dextrocardia) Severe pulmonary stenosis* With isolated levocardia or dextrocardia Heart and viscera in situs inversus Heart and viscera in situs solitus Right atrioventricular valve incompetence Coarctation of aorta Patent ductus arteriosus *Angiocardiographic diagnosis; see text for discussion of "severe" pulmonary stenosis. Patients Table 3 Comparison of Peak Systolic Pressure in Common Ventricle and Femoral Artery (FA) Peak systolic pressure in common ventricle Greater than in FA Equal to FA Less than FA in 4 patients. These values were obtained in 38 patients breathing room air and in five patients under light anesthesia and breathing a mixture of gases containing 0% oxygen. The right-to-left shunt in nine patients was less than 30%; in 3 it was 30 to 50% and in the remaining it exceeded 50%. Femoral artery oxygen saturation exceeded 85% in 5. Patients % Range of difference, 6 to 7 pressure mm Hg. 4to.. i Pulmonary Stenosis Severe pulmonary stenosis was diagnosed by angiocardiography in patients. The catheter entered the pulmonary artery in only one of these patients. The severity of the stenosis was assessed by angiocardiography, with the limitations of this method being kept in mind. In five patients the pulmonary stenosis was associated with isolated levocardia or isolated dextrocardia, and the great vessels were transposed (table 4); D-transposition was present in four of them. The great vessels were oriented normally in three of t-he other seven patients in the group, and transposition was present in four, with two of the four having D-transposition. Pulmonary stenosis judged severe was associated with a large right-to-left shunt and low systemic arterial oxygen saturation. However, one patient whose angiocardiographic appearance was suggestive of severe pulmonary stenosis had a right-to-left shunt of less than 30% and a systemic arterial oxygen saturation of 9% at rest. This patient had previously had pulmonary valvotomy

4 COMMON VENTRICLE 7 Table 4 Certain Variables in Twelve Patients with Common Ventricle and Angiocardiographically Diagnosed Severe Pulmonary Stenosis Isolated levocardia or isolated dextrocardia Heart and viscera in situs inversus-dextrocardia Heart and viscera in situs solitus Site of obstruction Valvular Subvalvular Combined Right-to-left shunt Less than 30% 30 to 50% More than 50% Systemic arterial 0 saturation 5 to 85% 88% 9% Great vessel position, patients Normal L-trans- D-transposition position position Patients Table 5 Entry into Pulmonary Artery and Great Vessel Relationship in Thirty-one Patients with Common Ventricle without Severe Pulmonary Stenosis Pulmonary artery Patients Entered 9 L-transposition D-transposition 3 Normal position 4 Not entered L-transposition 0 D-transposition Total 3 Without Severe Pulmonary Stenosis The pulmonary artery was entered by cardiac catheter in 9 of the 3 patients who did not have severe pulmonary stenosis (table 5). Included in the 9 were three patients with mild subpulmonary stenosis (peak systolic gradients of 0, 4, and 0 mm Hg). Of these 3 patients, four had normally oriented great vessels, the pulmonary artery being entered in each one. Twenty-seven had transposition of the great vessels, the pulmonary artery being entered in 5. Oxygen satura- tions and pressure data regarding the pulmonary circuit of these 9 patients are available in table 6. The data are likely to be representative of the group of 3 patients without severe pulmonary stenosis. If complete mixing occurs in the common ventricular chamber, the oxygen saturation in the pulmonary artery and systemic artery (in this case the femoral artery) should be identical. This occurred in only three of the 9 patients. In more than half the patients, oxygen saturation of blood in the pulmonary artery was less than that in the femoral artery, showing that systemic venous blood is preferentially directed to the pulmonary artery and the pulmonary venous blood to the aorta, which may be considered as "favorable streaming." By contrast, five patients were considered as having "unfavorable streaming." In all but two patients in whom complete mixing did not occur, the pulmonary artery oxygen saturation was within 6% of systemic arterial oxygen saturation (fig. ). In the other two patients, both with D-transposition, the difference was 8 and 9%. The ratio of pulmonary resistance to systemic resistance (Rp/ Rs)

5 8 RAHIMTOOLA ET AL. was less than 0.7 in 68% of these patients. At the time of study there was no significant difference in the ages of the three groups whose Rp/Rs are shown (table 6). The higher the ratio of Rp/Rs, the larger was the rightto-left shunt. Comment Incidence, Age, Sex, and Associated Lesions At the Mayo Clinic, cardiac catheterization studies are performed on those patients with congenital heart disease for whom the diagnosis or severity of a lesion, or both, are in Table 6 Pertinent Data Concerning Nineteen Patients with Common Ventricle without Severe Pulmonary Stenosis in Whom the Pulmonary Artery Was Entered by the Catheter Great vessel position Patients % L-trans. D-trans. Normal 0 saturation PA>FA 5 6 PA = FA 3 6 PA<FA 58 9 Total Rp/Rs Mean age, yr. (range) < (3-7) (.5-4) > (.5-5) Total 9 00 R-L shunt Mean Rp/Rs (range) < 30% (0.-0.5) 30-50% (0.3-.5) > 50% (.0-.6) Total 9 00 Abbreviations: PA and FA --Pulmonary and femoral artery; Rp/Rs = ratio of pulmonary resistance to systemic resistance. 00 Iz <Z 80 ; < 50 / / / GREAT VESSEL POSITION *, L- TRANSPOSITION O D-TRANSPOSITION A =NORMAL z: 40 V I I I I SYSTEMIC ARTERIAL OXYGEN SATURATION - PER CENT Figure Simultaneous oxygen saturation of blood from pulmonary artery and systemic artery in 9 cases of common ventricle without severe pulmonary stenosis. Solid line is line of identity and broken lines represent + 6% saturation points. The two patients represented by open circles outside these lines had D-transposition.

6 COMMON VENTRICLE 9 doubt. Since this constitutes approximately 0% of the patients with congenital heart disease seen in the pediatric age group, our series is biased in regard to more complex lesions. Yet it is interesting that the incidence (3.%) was similar to the incidences in the postmortem series of Maude Abbott7 (.7%), Fontana and Edwards8 (3%), and Keith and associates (%). From our series, life expectancy of patients with common vertricle probably is not necessarily as short as would seem from some postmortem studies.9' 0 Mandel and Hirsch" collected 7 cases from the literature in which the patients survived beyond the age of years and added a case of their own. Our data do not allow any conclusion to be drawn regarding the association of pulmonary stenosis and increased longevity in such cases; however, Site O % Saturation Pressure (mm. of Hg.) reports in the literature8' 9 support such an association. The sex ratio and the incidence of associated lesions are similar in our series to those reported by others.", 7, 0, Hemodynamic Profile of Patients with Common Ventricle The hemodynamic feature of this condition consists of the presence of a common mixing chamber (the common ventricle) receiving the pulmonary and systemic venous blood flows via the two atrioventricular valves either separately, or as a common atrioventricular valve. The two main aspects to be considered are () the degree of mixing within this chamber and () the resistance to outflow from this chamber. Mixing Since the common ventricular chamber receives the total input of both the pulmonary Femoral Artery 9 05 /68 fpulmonary Flow/ Systemic Flow -,8 Pulmonary Artery 89 9/57 ipulmonary Resistance /Systemic Resistance c Common Ventricle 83,90 03/0-8 aleft-to-right" 0.5 Shunt = 6 % Mixed Venous 76 - "Right-to-Left" Shunt 9 % L.-: Superior Vena Cava AT= 4.5 AT= 5,5 seconds seconds Inferior Vena Cava Indicator - 5 mgm. Indocyanine Green Sampling Site - Femoral Artery Ita Common Ventricle AT= 3.0 seconds - /oxygen soturation at this site - 83% Main Pulmonary Artery Common Ventricle oxygen saturation at this site= 90% AT= 5C, seconds AT= 3.8 seconds Figure 3 Indicator-dilution curves and other pertinent data in common ventricle with L-transposition, demonstrating favorable streaming in the common ventricular chamber. Indicator injected into common ventricular chamber at site of 90% oxygen saturation is directed in larger amount to femoral artery than when it is injected at site of 83% oxygen saturation. Oxygen saturation in FA and PA is 9% and 89%, respectively. Dye curves also show significant right-to-left shunt at ventricular level and are compatible with considerable degree of intracardiac mixing.

7 0 RAHIMTOOLA ET AL. Site Femoral Artery Pulmonary Artery Common Ventricle Mixed V/erous % Saturation Pressure (mm. of Hg.) 79 /7 Puilmonary Flow/Systemic F low 87 /54 PuIlmonary Resistance/Systemic Resistance 70,9,95 /3-8 n Left-to-Right^ Shunt - 57 ufrr igh t - to- Lef t "Shunt =. = 04 74% = 46% Kt 4J Superior Vena Cova Indicator -.5 mgm. Indocyanine Sampling Site - Femoral Artery Xi Common Ventricle ;ic / oxygen saturation at this site'95% Green AT- 5.0 seconds Ca va AT= 4,0 seconds Main Pulmonary Artery Common Ventricle ygen saturation at site - 70 % this AT= 5.0 seconds AT= 3,4 seconds Figure 4 Indicator-dilution curves and other pertinent data in common ventricle with D-transposition, demonstrating unfavorable streaming in common ventricular chamber. Indicator injected into common ventricular chamber at site of 70% oxygen saturation was largely directed to femoral artery (FA), whereas when injected at site of 95% oxygen saturation it is mainly directed to pulmonary artery (PA). In this patient, oxygen saturation in FA and PA was 79% and 87%, respectively. and systemic veins, complete mixing of the pulmonary and systemic venous return may be assumed, and identical oxygen saturation of blood in the common ventricle, pulmonary artery, and aorta has been suggested as the diagnostic feature of a common ventricle.3 Complete mixing was found in only 6%6 of our patients who did not have severe pulmonary stenosis. Streaming in this chamber, therefore, occurs very frequently and, in more than half the patients (58%), it was favorable to the body in that unoxygenated blood was directed preferentially to the lungs and oxygenated blood to the systemic circuit; favorable streaming occurred most commonly with L-transposition of the great vessels. In figures 3 and 4, indicator-dilution curves and other pertinent data are shown from an 8-year-old boy with favorable streaming (fig. 3) and a -year-old girl, with unfavorable streaming (fig. 4). These indicator-dilution curves show that the stream of more oxygenated blood in the patient with L-transposition is directed preferentially to the systematic artery and in the patient with D-transposition to the pulmonary artery, and they also confirm the presence of streaming in the common ventricular chamber. In the vast majority of patients, however, a considerable amount of mixing occurred in the common ventricular chamber as the oxygen saturation of blood in the pulmonary Circulation, Volume XXXIVs July 966

8 COMMON VENTRICLE artery was usually within 6% of that in the systemic artery (fig. ). Although it is reasonable to assume that more complete mixing occurs in the presence of severe pulmonary stenosis, nevertheless this could not be accurately measured, as the catheter entered the pulmonary artery in only one of these patients. Resistance to Ventricular Outflow Apart from the resistance of the systemic vessels, obstruction to aortic flow may occur at the aortic valve or at the bulboventricular foramen,' that is, between the common ventricle and the small outflow chamber. If an obstruction to aortic flow was present in a third of the patients, it was very mild as small gradients in the patients were associated with normal systemic flows. Expectedly, in the presence of severe obstruction to aortic flow, the systemic output would be low, the lungs would be "flooded," and life expectancy would be compromised. Obstruction to pulmonary flow results from pulmonary stenosis (valvular or subvalvular), from obstructive pulmonary vascular disease, or from pulmonary venous hypertension. When obstruction is severe ( patients with pulmonary stenosis and six with pulmonary vascular disease), it results in a reduced pulmonary blood flow and a large right-to-left shunt and thus a lower systemic arterial oxygen saturation. Three patients had mild subpulmonary gradients (0, 4, and 0 mm Hg), but the gradients were associated with large pulmonary flow and left-to-right shunt of 6, 7, and 79%, respectively. In patients without severe pulmonary stenosis, the pulmonary vascular resistance was considerably elevated in only 3% (Rp/ Rs > 0.7) and was slightly elevated in 4%. In this group, the state of the pulmonary vascular bed does not appear to be related to age (table 6). The systemic arterial oxygen saturation depends mainly on three factors: () absolute volume of and the ratio of the pulmonary and systemic blood flows (this is dependent on the pulmonary and systemic vascular resistances and the severity of associated pulmonary or aortic stenosis); () systemic venous and pulmonary venous blood oxygen saturation, and (3) degree of mixing in the common ventricle. A higher oxygen saturation of systemic arterial blood results more frequently from the absence of significant obstruction to pulmonary flow and from the presence of "favorable streaming" in the common ventricular chamber (table 7). Table 7 Distribution of Forty-three Patients with Common Ventricle According to Systemic Arterial Oxygen Saturation* Systemic arterial 0 saturation More than 85% Less than 85% (5 patients) (8 patients) Severe pulmonary stenosist 0 Without severe pulmonary stenosis 3 8. Pulmonary artery entered 7 Pulmonary flow/systemic flow.6 ( )*.8 ( ))* Pulmonary resistance/systemic resistance 0.36 (0.-0.6):.0 (0.3-.6)* Favorable streaming 7 4 Unfavorable streaming 0 5 Complete mixing 0 3. Great vessel position L-transposition D-transposition 4 Normal 3 *Values obtained in 38 patients breathing room air and 5 patients under light anesthesia and breathing a mixture of gases containing 0% oxygen. tsee text for discussion of "severe" pulmonary stenosis. :Mean values with ranges given in parentheses.

9 . RAHIMTOOLA ET AL. To summarize, complete mixing is unusual in the common ventricular chamber in the absence of severe pulmonary stenosis. "Favorable streaming" occurs in more than half the patients (more commonly with L-transposition) and results in a higher systemic arterial oxygen saturation. Severe obstruction to pulmonary flow due to pulmonary stenosis or pulmonary vascular disease results in reduction of pulmonary flow, increased right-to-left shunt, and lower systemic arterial oxygen saturation. Diagnosis It has recently been stated that, after the initial clinical examination, the presence of a single or common ventricle usually is not suspected and, further, that the initial clinical impression (based on physical examination, chest roentgenograms, and electrocardiograms) is likely to be one of several diseases, none of which include common ventricle.0 Table 8 shows that in this series the primary clinical diagnosis with which the patients were referred to the laboratory for study in more than 60% of the cases was a suspected common ventricle showing the high level of clinical suspicion of this condition at this institution. The following features assist in making the diagnosis. () One must be constantly aware of the possibility of the presence of common ventricle. () All patients have a right-to-left shunt that is best demonstrated by the indicator-dilution technique. (3) In the presence of severe pulmonary stenosis, indicator-dilution curves are similar to those seen in severe pulmonary stenosis with ventricular septal defect. In the absence of severe pulmonary stenosis, indicator-dilution curves usually are compatible with considerable mixing. (4) Oxygen saturations in the common ventricular chamber (a) may vary from mixed venous to pulmonary venous levels, and (b) may be greater than those in the systemic artery. (5) If the catheter enters the pulmonary artery, the pulmonary arterial and systemic arterial oxygen saturations usually are within 6% of each other and only infrequently are they equal. (6) The systolic pressure in the common ventricle must be as high or higher than that in the ascending aorta. Selective biplane angiocardiography of a high level of technical excellence is essential for diagnosis. It allows accurate definition of the ventricular anatomy and permits localization of the atrioventricular valves. It also defines the sites of origin of the aorta and pulmonary artery and their relationship and allows recognition of the presence of significant pulmonary stenosis, either valvular or subvalvular. Summary Data from 43 patients with common ventricle for whom satisfactory hemodynamic data are available are presented; these form 3.% of the patients with congenital heart disease studied by means of catheterization at the Mayo Clinic. Selective biplane angiocardiography is essential for diagnosis of common ventricle. Associated cardiovascular anomalies are common, with transposition of the great vessels occurring in 84% of the patients. The right-to-left shunt present in all 43 patients results in desaturation of systemic arterial blood. The common ventricle receives all the systemic and pulmonary venous blood Table 8 Initial Clinical Diagnosis in Forty-three Patients with Common Ventricle Diagnosis with which referred to laboratory for study Cases Comment Common ventricle (?) 8 Twenty-one had Ltransposition Tetralogy of Fallot 6 All had severe pulmonary stenosis Transposition of the great vessels 4 All had D-transposition Tricuspid atresia One had severe pulmonary stenosis Miscellaneous 3

10 COMMON VENTRICLE flows; yet complete mixing is uncommon in this chamber, for it occurred in only 6% of the patients who did not have severe pulmonary stenosis. Fifty-eight per cent had oxygenated blood directed preferentially to the systemic circuit ("favorable streaming"). This occurred most commonly with L-transposition and resulted in a higher oxygen saturation of systemic arterial blood. Obstruction to pulmonary flow (from pulmonary stenosis or pulmonary vascular disease) results in a larger right-to-left shunt and lower oxygen saturation of systemic arterial blood. Addendum Since January 965, 7 more cases of common ventricle have been studied in our laboratory. In general, the findings in these cases have been similar to those reported in the paper, except for three; these three cases illustrate special points of interest. Patient represents an extreme example of "unfavorable streaming." In this -year-old boy with common ventricle and D-transposition, the oxygen saturation of blood in the pulmonary artery was 6% higher than that in the systemic artery. Patient, an example of severe obstruction at the bulboventricular foramen, was a -yearold girl with common ventricle, L-transposition, and a subaortic chamber, who had surgical banding of the main pulmonary artery. Simultaneous pressure measurement in the common ventricle and the subaortic chambers showed a systolic gradient of 35 mm Hg across the bulboventricular foramen. The systemic index was., the pulmonary index 3.0, Qp/Qs was 5.9, and the left-to-right shunt was 85%. The peak systolic gradient across the banded pulmonary artery was 75 mm Hg. Patient 3, an example of severe incompetence of the left atrioventricular valve, was an 8-year-old girl with common ventricle and L-transposition. Severe incompetence of the left atrioventricular valve was demonstrated by the indicator-dilution technique and by angiocardiography. A huge left atrium was present and extended to the left chest wall, forming the left border of the heart. References. VAN PRAAGH, RICHARD, ONGLEY, P. A., AND SWAN, H. J. C.: Anatomic types of single or common ventricle in man: Morphologic and geometric aspects of 60 necropsied cases. Amer J Cardiol 3: 367, KEITH, J. D., ROWE, R. D., AND VLAD, PETER: Heart Disease in Infancy and Childhood. New York, The Macmillan Co., 958, p PENMAN, H. G., AND WHITTY, R. H.: Cor triloculare biatriatum. Brit Heart J 5: 4, WOOD, E. H., AND SWAN, H. J. C.: Catheterization of the heart and large vessels. In Luisada, A. A.: Cardiology: An Encyclopedia of the Cardiovascular System, vol., New York, McGraw-Hill Book Co., Inc., 959, p SWAN, H. J. C.: Cardiac catheterization. In Lewis-Walters: Practice of Surgery, vol.. Hagerstown, W. F. Prior Co., Inc., 963, p.. 6. MARSHALL, H. W., HELMHOLZ, H. F., JR., AND WOOD, E. H.: Physiologic consequences of congenital heart disease. In Handbook of Physiology: Section, Circulation, vol.. Washington, D. C., American Physiological Society, 96, p ABBOTT, MAUDE E.: Atlas of Congenital Cardiac Disease. New York, American Heart Association, Inc., 936, p FONTANA, R. S., AND EDWARDS, J. E.: Congenital Cardiac Disease: A Review of 357 Cases Studied Pathologically. Philadelphia, W. B. Saunders Co., 96, p CAMPBELL, MAURICE, REYNOLDS, GEOFFREY, AND TROUNCE, J. R.: Six cases of single ventricle with pulmonary stenosis. Guy Hosp Rep 0: 99, VAN PRAAGH, RICHARD, VAN PRAAGH, STELLA, VLAD, PETER, AND KEITH, J. D.: Diagnosis of the anatomic types of single or common ventricle. Amer J Cardiol 5: 345, MANDEL, ARMAND, AND HIRSCH, VICTOR: Cor triloculare biatriatum: Report of a case with survival to the age of 9 years. Amer Heart J 66: 04, EDWARDS, J. E.: Congenital malformation of the heart and great vessels. In Gould, S. E.: Pathology of the Heart, ed.. Springfield, Illinois, Charles C Thomas, Publisher, 960, p HEATH, DONALD: Cor triloculare biatriatum. Circulation 5: 70,