AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 1 January 1990
EVALUATION OF THE HUMAN FETAL CARDIAC SIZE AND FUNCTION Jean Claude Veille, M.D., Mark Sivakoff, M.D., and Meg Nemeth, R.N.
Two-dimensional-directed M-mode echocardiography was done on 80 normal fetuses between the 17th to 42nd weeks of gestation. The M-mode beam transected the ventricles at the level of the chordae tendineae at the tip of atrioventricular valves. Right and left ventricular dimensions and free wall thicknesses correlated well with gestational age. Calculated measurements showed a good correlation of the stroke volume and cardiac output with gestational age. The right ventricular dimension, however, was significantly greater than the left ventricular one. Fractional shortening of the right and left ventricle did not change significantly with advancing gestational age. This study indicates that the human fetal right ventricle dimension, stroke volume, and cardiac output are slightly larger than that of the left ventricle. This study also suggests that the human fetus increases its cardiac output to match its growth and it does so by increasing ventricular size rather than fractional shortening or heart rate.
Until a few years ago, the only way to examine fetal circulation during normal and abnormal gestation was to evaluate the fetal cardiac function of acutely or chronically instrumented animals.1-4 Data on the human fetal circulation were limited, since most studies on human fetuses were obtained at the time of therapeutic abortion.5 Echocardiography is an established method for investigation of the neonatal, pediatric, and adult heart. High resolution real-time sonography has now made it feasible to evaluate safely human fetal cardiovascular structures and function as early as the 16th to 17th weeks of gestation.6"10 Among the published studies on the quantitative assessment of the fetal cardiac function, results on right heart dominance have been conflicting.6-811 The purpose of this study is to report the results of a prospective, cross-sectional cardiac assessment of normal fetuses using two-dimensional (2-D) directed M-mode echocardiography. The aim was to investigate the changes in ventricular size and function with fetal growth in order to understand the mechanisms by which the fetal heart increases its output to match somatic growth. Such understanding of the normal right and left ventricular function is essential if abnormal cardiovascular physiology is going to be used as an antepartum tool for the eval-
uation of the cardiovascular system of the human fetus. MATERIALS AND METHODS
A prospective study was done on 80 randomly selected pregnancies between the 17th and 41st weeks of gestation. All signed an informed consent approved by the Institutional Review Board of the hospital. One hundred and two echocardiographic M-mode recordings were obtained on the 80 patients. Ten studies were excluded because of the presence of maternal disease, such as diabetes, pregnancy-induced hypertension, or Rh isoimmunization disease. Out of the 92 remaining M-mode recordings, three were excluded because of poor visualization of the atrioventricular valves, four because of poor quality tracings. Thus, 85 tracings were available for analysis. Echocardiographic Studies
Using a Toshiba SSH-40 echo machine and either a 2.5 or a 3.5 MHz transducer, with a focus of 3 to 7 cm, echocardiograms were obtained from a constant
Department of Reproductive Biology and Division of Cardiology (Department of Pediatrics), Case Western Reserve University, Cleveland, Ohio No reprints are available.
54
Copyright © 1990 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
ABSTRACT
HUMAN FETAL CARDIAC SIZE AND FUNCTION/Veille, Sivakoff, Nemeth
Cardiac Measurements
Calculations were determined as follows: 1. Fractional shortening (% D) = [EDD - ESD/ EDD] x 100. 2. Ventricular volume (ml) = 7D3/(2.4 + D) where D is either EDD or ESD.9 3. Stroke volume (ml) = EDV - ESV where EDV is end-diastolic volume and ESV is end-systolic volume obtained using formula 2. 4. Heart rate was calculated directly from the strip chart recorder.
menstrual histories, but all had early ultrasound for dating. In these cases, the early ultrasound was used to determine gestational age. Echo data are presented versus gestational age. The observers were blinded to the gestational age during the period of actual measurements of the tracings. Reproducibility
The intraobserver reperformance reproducibility of the M-mode were done on ten studies by two observers (JCV,MS), and at different times during the study (measurement 1 and measurement 2) X 100%.8 A ± 5.1% variability was found in the measurements. The interobserver reproducibility was within 4.5%. Statistical Analysis
Multiple regression analysis was used to obtain the best correlation between cardiac measurements and gestational age. The 5th and 95th confidence limit for individual points was derived for the "Y" values. Significance was assessed using paired t test when right and left cardiac measurements were compared. RESULTS
Eighty-five successfully obtained echocardiographic tracings form the basis of this report. Right and left ventricular EDD correlated well with gestational age (r = 0.86 and 0.82, respectively). The relationship between the right and left ventricular EDD was found to correlate very well (Fig. 1; r = 0.96). The right ventricular size was statistically significantly greater than the left (p =S 0.001). Further-
Gestational Age Assessment
Dating was assessed by the last menstrual period and the biparietal diameter recorded in a conventional fashion, since this is an accurate indicator of gestational age.14 If the results from menstrual history and ultrasound measurements were different, ultrasound data were used if repeated measurements were compatible with the original ultrasound. When the fetus was found to be in a breech presentation, the head circumference and the femur length were obtained for dating. A few patients had poor
Right Ventricle End Diastolic Dimension (mm)
Figure 1. Relationship between right and left ventricular end diastolic dimension (mm) in 85 normal fetuses. The line of identity is marked (dotted line). The solid line indicates that for each right ventricle end-diastolic dimension the left ventricle is smaller.
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
2-D view of the fetal heart. This particular probe influences axial resolution, which in turn accentuates the leading edge of the signal needed to obtain M-mode echocardiograms. Using the four-chamber view, an M-mode beam was placed perpendicular to the intraventricular septum and at level of the tip of the atrioventricular valves. Previous studies have shown that maximal ventricular dimension correlates well with measurements from anatomic casts. The 2-D view was always present on the screen at the time of the M-mode recording, so exact knowledge of the beam in relation to the ventricles was seen. In the small fetuses (18 to 26 weeks) a magnification scale was used in order to obtain adequate measurements. The M-modes were recorded on a strip chart recorder at a preset speed of 50 mm/sec. Right and left ventricles were immediately labeled on the tracing in order to recognize properly the ventricle on subsequent analysis. Fetal electrocardiograms were not obtained on these fetuses, and thus the onset of the Q-wave for the end-diastolic measurements, as recommended by the American Society of Echocardiography, could not be used.12 End-diastolic dimensions (EDD) were made at the point of maximal chamber dimension as recommended in a previous article on human fetal echoes by Kleinman and Donnerstein.13 End-systolic dimensions (ESD) were measured at the nadir of the septal and posterior wall motion as previously described.11 The diameters and thickness of the right and left ventricles were measured on the same level. The thickness of the echo line was excluded from all measurements of the inner cavity dimensions as recommended by Sahn et al.11 Six consecutive cardiac cycles were measured and averaged.
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 1 January 1990
FETAL ECHOCARDIOGRAPHY Posterior Wall vs. Gestational Age
N=68 (Left) r=0.67
y=0.08x+0.16 (Right) r=0.724 y=0.07x+0.45
20
25
30
35
40
Gestational Age (weeks)
56
Figure2. Regression linesof both free walls of the right ventricle (A) and of left ventricle (o). The means were practically superimposed on each other. The 5th and 95th confidence intervals around the line were drawn as one for clarity.
Right Ventricular Stroke Volume (ml)
Figure 3. Relationship between right and left ventricular stroke volume (ml) in 85 normal fetuses. Broken line indicates the line of identity. Solid line indicates a right ventricular dominance.
COMMENTS
Fetal cardiac size and function were investigated in 80 normal fetuses at various gestational ages. Ventricular size of both the right and the left ventricles increased with gestational age. This is not surprising, since heart size is expected to increase proportionally with total body size, and is in accordance with previously published data. 71115 ' 16 Of significance was the fact that the right ventricular EDD was found to be slightly greater than that of the left ventricle. Our results are consistent not only with previous data published on chamber size of fetal lambs,4 but also with M-mode echocardiographic studies done on human fetuses that suggest a right ventricular dominance. 817 ' 18 Azancot et al19 using a Fourier transformation digitization process for analysis of ventricular shapes were able to demonstrate that most fetuses and newborns have a dominant right ventricle. The curvature of the right ventricle in utero is consistent with increased pressure and volume work compared with the ex utero heart. In our study the right ventricular to left ventricular ratio showed a slightly greater right ventricle (1.07 ± 0.12) than left. In the same article by Azancot et al the ratio was 1.07 ± 0.07. When they serially followed these fetuses in the newborn and infancy periods, the ratio dropped to 0.62 ± 0.12 and 0.45 ± 0.01, respectively. Thus, both the ratio of the major diameter of the ventricles and the Fourier analysis of shape change may reflect the pressure and volume changes associated with birth. Right ventricular predominance in the human fetus has been recently challenged. 71516 A pathoanatomic study showed equal ventricular size in human stillborns.20 In this study it is important to realize that errors can occur when measurements of heart size obtained from stillborns are made to extrapolate measurements obtained from live fetuses by echocardiograms. Rigor mortis affects the shape
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
more, the ratio of right to left ventricular size was found to be 1.07 ± 0.12. This ratio was obtained by dividing the right EDD by the left EDD for each fetus. These values were then averaged. Figure 2 correlates free walls with gestational age. Only the value of 68 studies were obtained because the epicardium could not be well visualized in all the tracings. The endocardial surface of the free walls, however, was obtained in all 85 studies. Since no difference between the right and left walls were found, all points, as well as the 5th and 95th percentiles are plotted on the same graph. Using a single M-mode dimension and the Teichholz formula, ESV, EDV, and stroke volume were estimated. The stroke volumes of both the right and the left ventricles correlated well with advancing gestational age (r = 0.83 and 0.77, respectively). A large variability was found after the 30th week of gestation. Figure 3 compares the value of the stroke volume of the right and left ventricles. Right ventricular stroke volume was significantly greater than the left ventricular stroke volume (p «£ 0.01). When compared, the cardiac output of the right ventricle was larger than the left (p =s 0.05) and also correlated well with gestational age (r = 0.83 and 0.76, respectively; Fig. 4). The results of the measurements obtained in this study for the right ventricular output are plotted against a recent estimate of the output of that chamber by Doppler (Fig. 5). No significant relationship was found between fetal heart rate or fractional shortening of either ventricle with gestational age (r = 0.30 and 0.02, respectively).
HUMAN FETAL CARDIAC SIZE AND FUNCTION/Veille, Sivakoff, Nemeth
S
1500
Z.
1000
N = 85 r = 0.83 y = 2.7E-.02x 2 - 7 4 6
1500
1000 -
500
20
5g__, : •
n
30
20
30
40
Gestational Age (weeks) Qestational Age (weeks) Relationship between right ventricular output (A), left ventricular output (B) (ml/min), and gestational age
2000 y = 2.7E-.02-X
1600
:• •
500
ir
Figure 4. (weeks).
N=85 r=76 y = 2.807 E-.02. x 2.684
y
44 = e
5164+
2746
12
.574x
20 30 Gestational Age (weeks) Figure 5. Comparison of the right ventricular (output/ ml/min) obtained by Doppler study (broken line) and the present M-mode calculated values versus the gestational ages. Note the similarity of both lines until the 32th week, at which point the lines diverge.
and the size of the heart, making a comparison of the ventricular measurements difficult. Therefore the conclusions regarding chamber size from that study have to be interpreted with caution. Furthermore, in a subsequent study on human fetuses by the same investigators using Doppler flow analysis, the right ventricle was found to be the predominating chamber.21 The discrepancy among these published works from the same authors may be due to variations in heart size between the normal and the postmortem condition. In our study, meticulous care was taken to visualize the exact point of M-Mode sampling so that at all times with the cursor at the tip of the atrioventricular valves and perpendicular to the septum in order to avoid foreshortening the ventricular cavities. Measurements obtained by the previously mentioned investigators were taken at the excursion of
the valvular leaflets, a position close to the atrioventricular valve anulus where the ratio between right and left sides could be close to one. 7 - 1516 To assess the error of the cursor placement, we compared another 15 normal fetuses, dropping the cursor at the tip of the atrioventricular valves and at the level of the atrioventricular valve. No statistical difference was found between measurements of the right and left ventricle at these two levels either during diastole or during systole. Thus, measurements of the ventricles are not statistically different from one another if measurements include the whole or the tip of the atrioventricular valve. Such views are much more difficult to obtain in that great care has to be taken so that both tips of the atrioventricular valves are showing. Such views, however, assure the exact alignment of the cursor vis-a-vis the septum and the ventricular cavities. Views of the valves are, however, of paramount importance for references purposes if ventricular size is to be assessed. Although the right ventricular EDD was statistically significantly greater than the left ventricular EDD, the ratio of 1.07 ± 0.12 for right to left ventricular EDD is so close to one, its clinical significance for the individual fetus may not be all that important. Its importance is, however, that fetuses found to have a right to left ventricular ratio of >1.31 or
EVALUATION OF THE HUMAN FETAL CARDIAC SIZE AND FUNCTION Jean Claude Veille, M.D., Mark Sivakoff, M.D., and Meg Nemeth, R.N.
Two-dimensional-directed M-mode echocardiography was done on 80 normal fetuses between the 17th to 42nd weeks of gestation. The M-mode beam transected the ventricles at the level of the chordae tendineae at the tip of atrioventricular valves. Right and left ventricular dimensions and free wall thicknesses correlated well with gestational age. Calculated measurements showed a good correlation of the stroke volume and cardiac output with gestational age. The right ventricular dimension, however, was significantly greater than the left ventricular one. Fractional shortening of the right and left ventricle did not change significantly with advancing gestational age. This study indicates that the human fetal right ventricle dimension, stroke volume, and cardiac output are slightly larger than that of the left ventricle. This study also suggests that the human fetus increases its cardiac output to match its growth and it does so by increasing ventricular size rather than fractional shortening or heart rate.
Until a few years ago, the only way to examine fetal circulation during normal and abnormal gestation was to evaluate the fetal cardiac function of acutely or chronically instrumented animals.1-4 Data on the human fetal circulation were limited, since most studies on human fetuses were obtained at the time of therapeutic abortion.5 Echocardiography is an established method for investigation of the neonatal, pediatric, and adult heart. High resolution real-time sonography has now made it feasible to evaluate safely human fetal cardiovascular structures and function as early as the 16th to 17th weeks of gestation.6"10 Among the published studies on the quantitative assessment of the fetal cardiac function, results on right heart dominance have been conflicting.6-811 The purpose of this study is to report the results of a prospective, cross-sectional cardiac assessment of normal fetuses using two-dimensional (2-D) directed M-mode echocardiography. The aim was to investigate the changes in ventricular size and function with fetal growth in order to understand the mechanisms by which the fetal heart increases its output to match somatic growth. Such understanding of the normal right and left ventricular function is essential if abnormal cardiovascular physiology is going to be used as an antepartum tool for the eval-
uation of the cardiovascular system of the human fetus. MATERIALS AND METHODS
A prospective study was done on 80 randomly selected pregnancies between the 17th and 41st weeks of gestation. All signed an informed consent approved by the Institutional Review Board of the hospital. One hundred and two echocardiographic M-mode recordings were obtained on the 80 patients. Ten studies were excluded because of the presence of maternal disease, such as diabetes, pregnancy-induced hypertension, or Rh isoimmunization disease. Out of the 92 remaining M-mode recordings, three were excluded because of poor visualization of the atrioventricular valves, four because of poor quality tracings. Thus, 85 tracings were available for analysis. Echocardiographic Studies
Using a Toshiba SSH-40 echo machine and either a 2.5 or a 3.5 MHz transducer, with a focus of 3 to 7 cm, echocardiograms were obtained from a constant
Department of Reproductive Biology and Division of Cardiology (Department of Pediatrics), Case Western Reserve University, Cleveland, Ohio No reprints are available.
54
Copyright © 1990 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
ABSTRACT
HUMAN FETAL CARDIAC SIZE AND FUNCTION/Veille, Sivakoff, Nemeth
Cardiac Measurements
Calculations were determined as follows: 1. Fractional shortening (% D) = [EDD - ESD/ EDD] x 100. 2. Ventricular volume (ml) = 7D3/(2.4 + D) where D is either EDD or ESD.9 3. Stroke volume (ml) = EDV - ESV where EDV is end-diastolic volume and ESV is end-systolic volume obtained using formula 2. 4. Heart rate was calculated directly from the strip chart recorder.
menstrual histories, but all had early ultrasound for dating. In these cases, the early ultrasound was used to determine gestational age. Echo data are presented versus gestational age. The observers were blinded to the gestational age during the period of actual measurements of the tracings. Reproducibility
The intraobserver reperformance reproducibility of the M-mode were done on ten studies by two observers (JCV,MS), and at different times during the study (measurement 1 and measurement 2) X 100%.8 A ± 5.1% variability was found in the measurements. The interobserver reproducibility was within 4.5%. Statistical Analysis
Multiple regression analysis was used to obtain the best correlation between cardiac measurements and gestational age. The 5th and 95th confidence limit for individual points was derived for the "Y" values. Significance was assessed using paired t test when right and left cardiac measurements were compared. RESULTS
Eighty-five successfully obtained echocardiographic tracings form the basis of this report. Right and left ventricular EDD correlated well with gestational age (r = 0.86 and 0.82, respectively). The relationship between the right and left ventricular EDD was found to correlate very well (Fig. 1; r = 0.96). The right ventricular size was statistically significantly greater than the left (p =S 0.001). Further-
Gestational Age Assessment
Dating was assessed by the last menstrual period and the biparietal diameter recorded in a conventional fashion, since this is an accurate indicator of gestational age.14 If the results from menstrual history and ultrasound measurements were different, ultrasound data were used if repeated measurements were compatible with the original ultrasound. When the fetus was found to be in a breech presentation, the head circumference and the femur length were obtained for dating. A few patients had poor
Right Ventricle End Diastolic Dimension (mm)
Figure 1. Relationship between right and left ventricular end diastolic dimension (mm) in 85 normal fetuses. The line of identity is marked (dotted line). The solid line indicates that for each right ventricle end-diastolic dimension the left ventricle is smaller.
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
2-D view of the fetal heart. This particular probe influences axial resolution, which in turn accentuates the leading edge of the signal needed to obtain M-mode echocardiograms. Using the four-chamber view, an M-mode beam was placed perpendicular to the intraventricular septum and at level of the tip of the atrioventricular valves. Previous studies have shown that maximal ventricular dimension correlates well with measurements from anatomic casts. The 2-D view was always present on the screen at the time of the M-mode recording, so exact knowledge of the beam in relation to the ventricles was seen. In the small fetuses (18 to 26 weeks) a magnification scale was used in order to obtain adequate measurements. The M-modes were recorded on a strip chart recorder at a preset speed of 50 mm/sec. Right and left ventricles were immediately labeled on the tracing in order to recognize properly the ventricle on subsequent analysis. Fetal electrocardiograms were not obtained on these fetuses, and thus the onset of the Q-wave for the end-diastolic measurements, as recommended by the American Society of Echocardiography, could not be used.12 End-diastolic dimensions (EDD) were made at the point of maximal chamber dimension as recommended in a previous article on human fetal echoes by Kleinman and Donnerstein.13 End-systolic dimensions (ESD) were measured at the nadir of the septal and posterior wall motion as previously described.11 The diameters and thickness of the right and left ventricles were measured on the same level. The thickness of the echo line was excluded from all measurements of the inner cavity dimensions as recommended by Sahn et al.11 Six consecutive cardiac cycles were measured and averaged.
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 7, NUMBER 1 January 1990
FETAL ECHOCARDIOGRAPHY Posterior Wall vs. Gestational Age
N=68 (Left) r=0.67
y=0.08x+0.16 (Right) r=0.724 y=0.07x+0.45
20
25
30
35
40
Gestational Age (weeks)
56
Figure2. Regression linesof both free walls of the right ventricle (A) and of left ventricle (o). The means were practically superimposed on each other. The 5th and 95th confidence intervals around the line were drawn as one for clarity.
Right Ventricular Stroke Volume (ml)
Figure 3. Relationship between right and left ventricular stroke volume (ml) in 85 normal fetuses. Broken line indicates the line of identity. Solid line indicates a right ventricular dominance.
COMMENTS
Fetal cardiac size and function were investigated in 80 normal fetuses at various gestational ages. Ventricular size of both the right and the left ventricles increased with gestational age. This is not surprising, since heart size is expected to increase proportionally with total body size, and is in accordance with previously published data. 71115 ' 16 Of significance was the fact that the right ventricular EDD was found to be slightly greater than that of the left ventricle. Our results are consistent not only with previous data published on chamber size of fetal lambs,4 but also with M-mode echocardiographic studies done on human fetuses that suggest a right ventricular dominance. 817 ' 18 Azancot et al19 using a Fourier transformation digitization process for analysis of ventricular shapes were able to demonstrate that most fetuses and newborns have a dominant right ventricle. The curvature of the right ventricle in utero is consistent with increased pressure and volume work compared with the ex utero heart. In our study the right ventricular to left ventricular ratio showed a slightly greater right ventricle (1.07 ± 0.12) than left. In the same article by Azancot et al the ratio was 1.07 ± 0.07. When they serially followed these fetuses in the newborn and infancy periods, the ratio dropped to 0.62 ± 0.12 and 0.45 ± 0.01, respectively. Thus, both the ratio of the major diameter of the ventricles and the Fourier analysis of shape change may reflect the pressure and volume changes associated with birth. Right ventricular predominance in the human fetus has been recently challenged. 71516 A pathoanatomic study showed equal ventricular size in human stillborns.20 In this study it is important to realize that errors can occur when measurements of heart size obtained from stillborns are made to extrapolate measurements obtained from live fetuses by echocardiograms. Rigor mortis affects the shape
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
more, the ratio of right to left ventricular size was found to be 1.07 ± 0.12. This ratio was obtained by dividing the right EDD by the left EDD for each fetus. These values were then averaged. Figure 2 correlates free walls with gestational age. Only the value of 68 studies were obtained because the epicardium could not be well visualized in all the tracings. The endocardial surface of the free walls, however, was obtained in all 85 studies. Since no difference between the right and left walls were found, all points, as well as the 5th and 95th percentiles are plotted on the same graph. Using a single M-mode dimension and the Teichholz formula, ESV, EDV, and stroke volume were estimated. The stroke volumes of both the right and the left ventricles correlated well with advancing gestational age (r = 0.83 and 0.77, respectively). A large variability was found after the 30th week of gestation. Figure 3 compares the value of the stroke volume of the right and left ventricles. Right ventricular stroke volume was significantly greater than the left ventricular stroke volume (p «£ 0.01). When compared, the cardiac output of the right ventricle was larger than the left (p =s 0.05) and also correlated well with gestational age (r = 0.83 and 0.76, respectively; Fig. 4). The results of the measurements obtained in this study for the right ventricular output are plotted against a recent estimate of the output of that chamber by Doppler (Fig. 5). No significant relationship was found between fetal heart rate or fractional shortening of either ventricle with gestational age (r = 0.30 and 0.02, respectively).
HUMAN FETAL CARDIAC SIZE AND FUNCTION/Veille, Sivakoff, Nemeth
S
1500
Z.
1000
N = 85 r = 0.83 y = 2.7E-.02x 2 - 7 4 6
1500
1000 -
500
20
5g__, : •
n
30
20
30
40
Gestational Age (weeks) Qestational Age (weeks) Relationship between right ventricular output (A), left ventricular output (B) (ml/min), and gestational age
2000 y = 2.7E-.02-X
1600
:• •
500
ir
Figure 4. (weeks).
N=85 r=76 y = 2.807 E-.02. x 2.684
y
44 = e
5164+
2746
12
.574x
20 30 Gestational Age (weeks) Figure 5. Comparison of the right ventricular (output/ ml/min) obtained by Doppler study (broken line) and the present M-mode calculated values versus the gestational ages. Note the similarity of both lines until the 32th week, at which point the lines diverge.
and the size of the heart, making a comparison of the ventricular measurements difficult. Therefore the conclusions regarding chamber size from that study have to be interpreted with caution. Furthermore, in a subsequent study on human fetuses by the same investigators using Doppler flow analysis, the right ventricle was found to be the predominating chamber.21 The discrepancy among these published works from the same authors may be due to variations in heart size between the normal and the postmortem condition. In our study, meticulous care was taken to visualize the exact point of M-Mode sampling so that at all times with the cursor at the tip of the atrioventricular valves and perpendicular to the septum in order to avoid foreshortening the ventricular cavities. Measurements obtained by the previously mentioned investigators were taken at the excursion of
the valvular leaflets, a position close to the atrioventricular valve anulus where the ratio between right and left sides could be close to one. 7 - 1516 To assess the error of the cursor placement, we compared another 15 normal fetuses, dropping the cursor at the tip of the atrioventricular valves and at the level of the atrioventricular valve. No statistical difference was found between measurements of the right and left ventricle at these two levels either during diastole or during systole. Thus, measurements of the ventricles are not statistically different from one another if measurements include the whole or the tip of the atrioventricular valve. Such views are much more difficult to obtain in that great care has to be taken so that both tips of the atrioventricular valves are showing. Such views, however, assure the exact alignment of the cursor vis-a-vis the septum and the ventricular cavities. Views of the valves are, however, of paramount importance for references purposes if ventricular size is to be assessed. Although the right ventricular EDD was statistically significantly greater than the left ventricular EDD, the ratio of 1.07 ± 0.12 for right to left ventricular EDD is so close to one, its clinical significance for the individual fetus may not be all that important. Its importance is, however, that fetuses found to have a right to left ventricular ratio of >1.31 or
