- Email: [email protected]
Cardiac function in fetuses of type I diabetic mothers
Cardiac function in fetuses of type I diabetic mothers Giuseppe Rizzo, MD,. Domenico Arduini, MD,. and Carlo Romanini, MD b Rome and Ancona, Italy Cardiac function was cross-sectionally studied by means of M-mode and Doppler echocardiography in 40 fetuses of mothers with well-controlled insulin-dependent diabetes at 20 to 38 weeks of gestation. These variables were measured: interventricular septal thickness, ratio between the peak velocities during early passive ventricular filling and active atrial filling at the level of the atrioventricular valves, peak velocities, and the time to peak velocity at the level of the ascending aorta and the pulmonary artery. The values obtained were compared with our reference limits for gestation. A significant increase of interventricular septal thickness that was unrelated to maternal glycosylated hemoglobin levels was evidenced. Early passive ventricular filling/active atrial filling ratios were significantly lower in fetuses of diabetic mothers than in control fetuses. These differences were significantly related to interventricular septal thickness. No significant modifications were found in either aortic or pulmonary peak velocities or in time to peak velocity values. These findings suggest that in spite of an adequate metabolic control an interventricular septal hypertrophy that affects cardiac diastolic function develops in fetuses of diabetic mothers. (AM J OSSTET GVNECOL 1991 ;164:837-43.)
Key words: Diabetes, fetal heart, echocardiography, Doppler ultrasonography, diastolic function
Although the existence of hypertrophic cardiomyopathy in infants of diabetic mothers has been extensively described,' few data currently are available on the development of this anomaly during intrauterine life." In this study M-mode and Doppler echocardiography was used to study cardiac function in fetuses of mothers with type I diabetes.
Material and methods
Patients_ Forty-two women with type I insulindependent diabetes were evaluated in this study. Twenty-four women belonged to class B, 10 to class C, and 8 to class D, according to the modified White's classification.' All pregnancies were singleton and were accurately dated by early ultrasonographic examination. Diabetes was controlled by self-monitoring of blood glucose levels, and insulin adjustments were performed at the biweekly visit, if necessary, with the goal of maintaining the preprandial and postprandial blood glucose levels below 90 and 120 mg/ dl, respectively. The obstetric characteristics of the patients are reported in Table I. At the time of echocardiographic examination the glycosylated hemoglobin (HbA,c) was From the Laboratory of Fetal Physiology, Department of Obstetrics and Gynecologv, Universitii Cat/oliea S. Cuore, Rome," and the Department of Obstetrics and Gy!ecologv, Universit(l eli Ancona." Supported by the Italian National Council of ResN/rch (CNR grant No. 89/03875/C1'04). Received for publication June 27, 1990; revised October 12, 1990; accepted Octobl'r 18, 1990. Reprint requests: G. Rizzo, MD, 1st. ct. Ostetrica e Ginec%gica, Universitcl Cattoliea S. Cuore, Largo A. Gemelli, 8, 00168 Roma, Ita/v. 6/ j /26205
Table I. Characteristics of 40 patients with type 1 diabetes considered for study Age (yr, mean ± SD and range) Parity (mean ± SD and range) HbA" in first trimester ('le, mean ± SD and range) HbA" at Doppler recording Gestational age at Doppler recording (wk, mean ± SD and range) Gestational age at deliverv (wk, mean ± SD an'd range) Birth weight (gm, mean ± SD and range) Large-for-gestational-age newborns (>90th percentile) Fetal distress necessitating emergency cesarean section
24.9 ± 4.4
(19-35)
1.3 ± 0.5
(0-4)
6.0 ± 1.0
(4.8-7.4)
5.7 ± 0.8
(4.5-6.9)
29.7 ± 4.8
(20-38)
38.3 ± 1.9
(35-40)
3353 ± 576 (2640-4220)
4 (10%)
9 (22.5%)
measured, and the results obtained were correlated with the echocardiographic findings. All patients gave informed consent to participate in the study. Each patient was included in the study only once. Echocardiographic examination. Commercially available Ansaldo Hitachi Esacord 81 (Genoa) color Doppler equipment with 3.5 or 5 MHz convex probes was used for all measurements. The machine is equipped with a color flow mapping function, as well
837
838
Rizzo, Arduini, and Romanini
March 1991 Am J Obstet Gynecol
Fig. 1. Flow velocity waveforms through mitral valve in a control fetus (toP) and in a fetus of a diabetic mother (bottom) at 31 weeks of gestation. EI A ratio values are 0.72 in the former and 0.44 in the latter.
6
_ 51
• • • •
•
E 4 E ..... ~ 3 ~
(J)
2
4, 5 5 5,5 6 6 ,5 7
7,5 8
8 , 5 9 9, 5 10
BPD (mm)
5 4
,....
5
3
.....E
4
2
> c.:
E
,....
E E
~
~
> ~
•
6
~.
~
3 2
4, 5 5 5,5 6 6,5 7 7, 5 8 8 , 5 9 9, 5 10
BPD (mm)
o
4, 55 5, 5 6 6 ,5 7 7,5
8 8 , 5 9 9,510
BPD (mm)
Fig. 2. ISWT, LVWT, and RVWT values obtained in fetuses of diabetic mothers, as plotted on our reference range (mean and 95% confidence intervals).
Echocardiography in fetuses of diabetic mothers
Volume 164 l\umber 3
MITRAL VALVE
o
TRICUSPID VALVE
0,9
0,9
0,8
0,8 0
~
0,7
;;
0,7
L..
0,6
CI L..
...<
0,6
I.IJ
0,5
...< I.IJ
0,5 0,4
0,4
0,3
0,3
0,2
0,2
0,1
0,1
°I ..
..
18 20 22 24 26 28 30 32 3436 38 40
gestational age (weeks)
839
~
~ • • / ~ • ••
.......
°
18 20 22 2426 28 30 32 34 36 38 40
gestational age (weeks)
Fig. 3. Transtricuspid and trans mitral E/ A ratios obtained in fetuses of diabetic mothers, as plotted on our reference ranges for gestation (mean and 95% confidence intervals).
as a pulsed Doppler ultrasonography with a carner frequency that ranges from 2.5 to 5 MHz. The fetal heart was first examined with twodimensional echocardiography, and two patients were excluded from the study because of congenital heart diseases (i.e., tetralogy of Fallot and atrioventricular canal). A subcostal four-chamber view of the fetal heart was obtained in the remaining 40 patients. The Mmode beam was directed perpendicular to the interventricular septum at the level of the atrioventricular valves, and images were recorded. 5 Velocity waveforms were then recorded at the level of the atrioventricular valves, ascending aorta, and pulmonary artery, according to previously reported techniques. 6. 7 Mitral and tricuspid valves were visualized from the apical fourchamber view, as were the aortic valve from the fivechamber view and the pulmonary valve from the shortaxis view. The color flow mapping function was then superimposed, and the valve flow was evidenced. Care was taken to visualize the flow so that it was parallel to the Doppler beam. The pulsed Doppler sample volume was placed immediately distal to valve leaflets, and velocity waveforms were recorded. Recordings at a beam angle larger than 20 degrees were rejected. Doppler and M-mode images were recorded on standard V2-inch videotape for subsequent analysis. Permanent records were obtained from the videotape by means of a strip chart recorder. Ten consecutive heart cycles were selected during periods of fetal rest without breathing movements, and the values measured were averaged. These variables were measured with the aid of a computer-interfaced digitizer pad (Cardio 800, Kontron, Oxford): (1) wall thickness during end diastole at the level of the interventricular septum (ISWT), left ventricle (LVWT), and right ventricle (R VWT)5; (2) tricuspid and mitral blood flow velocities during early
ventricular filling (E wave) and during atrial systole (A wave)6; (3) the peak velocities of the aortic and pulmonary valves; (4) pulmonary and aortic time to peak velocity values, which were calculated as the time differences from the onset of the waveform to its peak velocity 7; (5) heart rate. The ratios between E-wave and A-wave peak velocities (the EI A ratio) at the level of both atrioventricular valves were then calculated as indices of ventricular diastolic function. 6 Data analysis. The variables measured were compared with our reference limits that were obtained from the cross-sectional study of 197 fetuses from normal pregnancies that were between 20 and 40 weeks' gestation. Because the size of the heart is better related to fetal dimension than to gestational age,5 reference limits for ISWT, LVWT, and RVWT were constructed considering the fetal biparietal diameter as the independent variable. In addition, gestational age was used as the independent variable to calculate normative data for Doppler measurements. 6. 7 Because all the variables considered change with gestation, the individual echocardiographic values observed in fetuses of diabetic mothers were also expressed as the number of standard deviations from the respective normal means for gestation (e.g., ~ISWT, ~mitral EI A ratio). A one-sample Student t test was used to compare echocardiographic parameters of control fetuses with those of fetuses of diabetic mothers, whereas the differences among diabetic fetuses were tested with the unpaired Student t test. Simple and multiple regression analyses were used to test the relationships between the variables considered. These analyses were performed on a Macintosh II (Apple Computers, Cupertino, Calif.) computer with the Stat View II statistical package (Abacus Concepts Inc., Berkeley, Calif.).
840
Rizzo, Arduini, and Romanini
01
-
-.5~
.... o
CI ~
=
~
1.5
-2
CI
~ 2.5 -31
.5
••
01
•
••
•
I.
LI.I
....S ...
•
-1
< ......
March 1991 Am J Obstet Gynecol
.-... -.51o
•
CI
.
0
•
LI.I
::;, C.J
·
...·c ...Qj 2.5
-2
'C
.5
.-... ~
S -2
... CI
•
• •• •
~ -2.5~
delta lVWT
., • • •
:::;o - .5 t• CI ~
< ......
-1
LI.I 'C
•
'ji -1.5 III
::;, C.J
... ...Qj -2.5
·C -2 CI
'C
I
1.5 2 2.5 3 3.5 4 4.5 5 5.5 delta ISWT
•
-31 .5 .75 1 1.21.51.8 2 2.22.52.8 3
1
0
•
CI
-31
=-1.5
-
1
•
•
LI.I
O.
1.5~
III
•
•
~
•
< ......
1.5 2 2.5 3 3.5 4 4.5 5 5.5 delta ISWT
.c:~ -.51 .... • • • < -H·~· • ...... ~. .•• • Q.
•
.~.• •• ••• .'•••• • •• • • ••• • •
01
'C
~ -1 •
.
•
-31 .5
•
•
•
• • • •• • .... •• • • • 1.5 2 2.5 3
. . ~
3.5
delta RVWT
Fig. 4. Relationships between mitral EI A ratio with ISWT (left) and LVWT (right) and between tricuspid EI A ratio with ISWT (left) and RVWT (right). All values are expressed as number of SDs by which they differed from the respective means for gestational age (~-values).
Results
Successful recordings were obtained in all cases at the level of the atrioventricular valves (Fig. 1). In 32 cases reliable data were obtained from all four cardiac sites, whereas fetal position did not allow successful recordings at the level of the pulmonary artery and ascending aorta, respectively, in three and five cases. At birth none of the infants had clinical signs of cardiomyopathy, with the exception of the two fetuses excluded for congenital heart diseases. M-mode echocardiography. The ISWT values of fetuses of diabetic mothers were significantly higher (mean difference 2.632 SDs, t = 16.437, P s 0.001) when compared with our reference limits for gestation. Similar results were obtained when LVWT (mean difference 1.626 SDs, t = 15.987, P S 0.001) and RVWT (mean difference 1.790SDs,t = 15.201,p S 0.001) values were considered (Fig. 2). However, there were no
relationships between HbA 1c and .lISWT (r = 0.194), .lLVWT (r = 0.123), and .lRVWT (r = 0.106) values. Furthermore no differences were found in M-mode parameters when fetuses of diabetic mothers assigned to White's class B were compared with those fetuses of mothers in classes C and D. Diastolic function. In the fetuses of diabetic mothers the EI A ratio results were significantly lower at the level of both mitral valves (mean difference -1.634 SDs, t = 15.414, P S 0.001) and tricuspid valves (mean difference -1.449 SDs, t = 14.038, P S 0.001) when compared with the EI A ratios of control fetuses (Fig. 3). A significant negative relationship was found between .l mitral EI A ratio and both .lISWT (constant = - 0.790, slope = - 0.321, SD = 0.567, r = 0.552, P S 0.001) and .lLVWT values (constant = -0.794, slope = -0.517, SD = 0.590,r = 0.496,ps 0.002) (Fig. 4) with a multiple r value of 0.601. Simi-
Echocardiography in fetuses of diabetic mothers
Volume 164 Number 3
841
AORTA 140,
70
..... 120 fI)
60
.....
E 100 u
"'" ='I
.oJ
u
CI ~
>
~
«I
..... E "'" > Il.
50
fI)
80 60
I-
40 30
40
20
20
10
~
•
•• • • • I .- ••••- • ••• • • • • • ••
...
. I.·
•
•
Q.
0
0 18 20 22 24 26 28 30 32 3436 38 40
18 20 22 24 26 28 30 32 3436 38 4(
gestational age (weeks)
gestational age (weeks)
PULMONARY ARTERY 60
120
'in .....
100
u
80
E
"'" ='I oJ
U
CI ~
> ~
60 40
«I
~
Q.
20
~ . I ·
~ ~
..... .§ > Il.
50
fI)
40
I- 30
20
-- ....
~ •••••• • •••• ~ • ••• • •• •
•
~
10 0
0 18 20 22 24 26 28 30 32 34 36 38 40
gestational age (weeks)
18 20 22 24 26 28 30 32 34 36 38 40
gestational age (weeks)
Fig. 5. Aortic (toP) and pulmonary (bottom) peak velocity and time to peak velocity (TPV) values in fetuses of diabetic mothers, as plotted on our reference ranges for gestation (mean and 95% confidence intervals).
larly, ~ tricuspid EI A ratio results were negatively related to ~ISWT (constant = - 0.643, slope = - 0.306, SD = 0.556, r = 0.541, P ~ 0.002) and ~RVWT values (constant = - 0.629, slope = - 0.458, SD = 0.564, r = 0.491, P ~ 0.005) (Fig. 4) with a multiple r value of 0.590. The HbA lc value results were unrelated to mitral (r = 0.141) and tricuspid (r = 0.152) ~ E/A ratios. No significant differences were found in EI A ratios when fetuses were grouped according to White's class of the mother. Systolic function. In Fig. 5 aortic and pulmonary values of peak velocity and time to peak velocity obtained in fetuses of diabetic mothers are plotted against our reference limits for gestation. No significant changes were evidenced and no significant correlations were found with MSWT, ~RVWT, and ~LVWT (r < 0.116), with HbA lc (r < 0.123) values, with White's class of the mother.
Relations to pregnancy outcome. Nine fetuses required emergency cesarean section because of fetal distress (abnormal fetal heart patterns characterized by late or repetitive variable decelerations and/ or loss of baseline variability) (Table I). When these fetuses were compared with the remaining fetuses of diabetic mothers studied, no significant differences were found in the cardiac parameters considered (Table II). Similarly, there were no differences when echocardiographic findings of large-for-gestational age fetuses (i.e., birth weight above the 90th percentile) were compared with those of appropriate-for-gestational-age fetuses (Table II). Comment A linear increase with fetal growth of both heart size and wall thickness has been evidenced in the human fetus. 5 Moreover, a concomitant improvement of the
842
Rizzo, Arduini, and Romanini
March 1991 Am J Obstet Gynecol
Table II. Echocardiographic measurements in fetuses of diabetic mothers grouped according to birth weight or occurrence of fetal distress (i.e., abnormal heart rate patterns necessitating emergency cesarean section) Fetal distress (No. = 9)
ISWT LVWT RVWT Mitral EI A ratio Tricuspid EI A ratio Peak velocity, aorta Time to peak velocity, aorta Peak velocity, pulmonary artery Time to peak velocity, pulmonary artery
2.66 1.65 1.76 -1.59 -1.54 0.33 -0.21
± ± ± ± ± ± ±
1.29 0.67 0.78 0.74 0.69 0.71 0.46
Remaining fetuses (No. = 31)
2.62 1.62 1.80 -1.66 -1.46 0.45 -0.34
± ± ± ± ± ± ±
p Value*
l.l4 0.63 0.74 0.67 0.63 0.44 0.39
NS NS NS NS NS NS NS
0.87 ± 0.67
0.79 ± 0.47
NS
-0.18 ± 0.33
0.16 ± 0.39
NS
LCAfetuses (No. = 4)
2.52 1.66 1.82 -1.62 -1.50 0.40 -0.59
± ± ± ± ± ± ±
1.44 0.90 0.75 0.72 0.67 0.88 0.62
ACAfetuses (No. = 36)
2.60 1.63 1.79 -1.65 - 1.48 0.47 -0.31
± ± ± ± ± ± ±
p
Value*
1.28 0.62 0.74 0.63 0.64 0.39 0.33
NS NS NS NS NS NS NS
0.73 ± 0.69
0.75 ± 0.44
NS
0.22 ± 0.33
0.14 ± 0.31
NS
-
LCA, Large for gestational age; ACA, appropriate for gestational age. All values are expressed as number of standard deviations by which they differed from the respective means for gestational age (Ll-values). * Unpaired t test.
cardiac performance has been described. The EI A ratio, a widely accepted index of ventricular diastolic function,S significantly increases during gestation at the level of both atrioventricular valves. These changes are believed to be secondary to the progressive maturation of ventricular compliance. 9 There is also an increase of aortic and pulmonary peak velocities and of left and right stroke volumes, to which suggests an improvement of ventricular systolic function with gestation. Previous studies have shown that in infants of diabetic mothers there is a hypertrophic cardiomyopathy that is characterized primarily by a significant thickening of the interventricular septum. 1. II Our data on a large series confirm previous sporadic reports 2 . 3 that the increase of ISWT, LVWT, and RVWT is already established during intrauterine life. The absence of correlations between the HbA lc values and the severity of cardiac wall thickness found in this investigation is in agreement with the results of neonatal studies l2 and suggests that in spite of adequate metabolic control during pregnancy, fetuses of diabetic mothers might experience an abnormal cardiac growth. However, there is no current account regarding whether the increase of cardiac wall thickness merely reflects the larger size of fetuses of diabetic mothers or whether it might also influence the fetal cardiac function. The lower EI A ratios at the level of both atrioventricular valves found in these fetuses might indicate an impaired ventricular diastolic function. The inverse relationships between the EI A values and both the interventricular septum and lateral ventricular wall thickness suggest that the abnormal cardiac thickening might impair ventricular compliance, which would affect the filling patterns of the ventricles. Previous microscopic studies on the heart in newborns of diabetic mothers corroborate this hypothesis by show-
ing a disorganization of the myofibrils, which is similar to that present in adults with hypertrophic cardiomyopathy, I. 13 a condition characterized by evident alterations of ventricular compliance and thus of Dopplermeasured diastolic function. 14 In spite of a ventricular hypercontractility reported in infants of diabetic mothers, I no significant modifications of systolic indices such as aortic and pulmonary peak velocities or time to peak velocity were evidenced in the fetuses examined. In adults diastolic dysfunction usually precedes systolic anomalies that appear later in the development of the myocardial disease. 15 Therefore it can be speculated that the fetuses studied have not reached a degree of cardiac impairment sufficient to affect systolic function. This hypothesis is supported by the data available on infants of diabetic mothers which show that echocardiographic indices of systolic function fall either in the normal range or above the normal range I and are usually abnormal only in newborns with clinical signs of cardiac impairment. 16 In conclusion our data suggest that despite careful metabolic control of pregnancy, fetuses of diabetic mothers show an increased thickness of the interventricular septum and of the ventricular walls, which may be associated with abnormalities of diastolic function. Further longitudinal studies during the perinatal period are required to clarify the clinical significance of these changes of cardiac function and their sequelae during neonatal life. REFERENCES I. Reller MD, Kaplan S. Hypertrophic cardiomyopathy in infants of diabetic mothers: an update. Am j Perinatol 1988;5:353-8. 2. Leisle j, Shen SC, Strauss L. Hypertrophic cardiomyop-
Volume 164 Number 3
3.
4. 5.
6.
7.
8. 9.
athy in a mid-trimester fetus born to a diabetic mother. J Pediatr 1982;100:631-2. Weber HS, CopelJA, GreenJ, Reece AE, Kleimann CS. Cardiac growth in the fetus of the diabetic mother [Abstract 029]. In: Proceedings of the fetal cardiac symposium, Paris, France. 1989:60. Hare JW, White P. Gestational diabetes and the White classification. Diabetes Care 1980;3:294-305. DeVore GR, Siassi B, Platt LD. Fetal echocardiography. IV. M-mode assessment of ventricular size and contractility during the second and third trimesters of pregnancy in the normal fetus. AM J OBSTET GYNECOL 1984;150: 981-8. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small for gestational age fetuses. Br J Obstet Gynaecol 1988;95:65-9. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic evaluation of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Br J Obstet Gynaecol 1990;97:603-7. Labovitz AJ, Pearson C. Evaluation of left ventricular diastolic function: clinical relevance and recent Doppler echocardiographic insights. Am HeartJ 1987; 114:836-51. Reed KL, Sahn DJ, Scagnelli S, Anderson CF, Shenker L. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coli Cardiol 1986;8:391-5.
Echocardiography in fetuses of diabetic mothers
843
10. Kenny JF, Plappert T, Saltzman DH, et al. Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1986;74:1208-16. 11. Breitweser JA, Meyer RA, Sperling MA. Cardiac septal hypertrophy in hyperinsulinemic infants. J Pediatr 1980; 96:535-9. 12. Sheehan PQ, Rowland TW, Shah BL, McGravey VJ, Reiter EO. Maternal diabetic control and hypertrophic cardiomyopathy in infants of diabetic mothers. Clin Pediatr 1986;25:266-71. 13. Gutgessel HP, Speer ME, Rosenburg HS. Characterization of the cardiomyopathy in infants of diabetic mothers. Circulation 1980;61 :441-50. 14. Takaneka K. Dabestani A, Gardin JM, et al. Left ventricular filling in hypertrophic cardiomyopathy: a pulsed Doppler echocardiographic study. J Am Coli Cardiol 1986; 7: 1263-71. 15. Aroesty JM, McKay RG, Heller GV, Royal HD, Als AV, Grossman W. Simultaneous assessment of left ventricular systolic and diastolic dysfunction during pacing induced ischemia. Circulation 1985;71 :889-900. 16. Mace S, Hirschfeld SS, Riggs T, Fanaroff AA, Meikatz IR. Echocardiographic abnormalities in infants of diabetic mothers. J Pediatr 1979;95:1013-9.
Key words: Diabetes, fetal heart, echocardiography, Doppler ultrasonography, diastolic function
Although the existence of hypertrophic cardiomyopathy in infants of diabetic mothers has been extensively described,' few data currently are available on the development of this anomaly during intrauterine life." In this study M-mode and Doppler echocardiography was used to study cardiac function in fetuses of mothers with type I diabetes.
Material and methods
Patients_ Forty-two women with type I insulindependent diabetes were evaluated in this study. Twenty-four women belonged to class B, 10 to class C, and 8 to class D, according to the modified White's classification.' All pregnancies were singleton and were accurately dated by early ultrasonographic examination. Diabetes was controlled by self-monitoring of blood glucose levels, and insulin adjustments were performed at the biweekly visit, if necessary, with the goal of maintaining the preprandial and postprandial blood glucose levels below 90 and 120 mg/ dl, respectively. The obstetric characteristics of the patients are reported in Table I. At the time of echocardiographic examination the glycosylated hemoglobin (HbA,c) was From the Laboratory of Fetal Physiology, Department of Obstetrics and Gynecologv, Universitii Cat/oliea S. Cuore, Rome," and the Department of Obstetrics and Gy!ecologv, Universit(l eli Ancona." Supported by the Italian National Council of ResN/rch (CNR grant No. 89/03875/C1'04). Received for publication June 27, 1990; revised October 12, 1990; accepted Octobl'r 18, 1990. Reprint requests: G. Rizzo, MD, 1st. ct. Ostetrica e Ginec%gica, Universitcl Cattoliea S. Cuore, Largo A. Gemelli, 8, 00168 Roma, Ita/v. 6/ j /26205
Table I. Characteristics of 40 patients with type 1 diabetes considered for study Age (yr, mean ± SD and range) Parity (mean ± SD and range) HbA" in first trimester ('le, mean ± SD and range) HbA" at Doppler recording Gestational age at Doppler recording (wk, mean ± SD and range) Gestational age at deliverv (wk, mean ± SD an'd range) Birth weight (gm, mean ± SD and range) Large-for-gestational-age newborns (>90th percentile) Fetal distress necessitating emergency cesarean section
24.9 ± 4.4
(19-35)
1.3 ± 0.5
(0-4)
6.0 ± 1.0
(4.8-7.4)
5.7 ± 0.8
(4.5-6.9)
29.7 ± 4.8
(20-38)
38.3 ± 1.9
(35-40)
3353 ± 576 (2640-4220)
4 (10%)
9 (22.5%)
measured, and the results obtained were correlated with the echocardiographic findings. All patients gave informed consent to participate in the study. Each patient was included in the study only once. Echocardiographic examination. Commercially available Ansaldo Hitachi Esacord 81 (Genoa) color Doppler equipment with 3.5 or 5 MHz convex probes was used for all measurements. The machine is equipped with a color flow mapping function, as well
837
838
Rizzo, Arduini, and Romanini
March 1991 Am J Obstet Gynecol
Fig. 1. Flow velocity waveforms through mitral valve in a control fetus (toP) and in a fetus of a diabetic mother (bottom) at 31 weeks of gestation. EI A ratio values are 0.72 in the former and 0.44 in the latter.
6
_ 51
• • • •
•
E 4 E ..... ~ 3 ~
(J)
2
4, 5 5 5,5 6 6 ,5 7
7,5 8
8 , 5 9 9, 5 10
BPD (mm)
5 4
,....
5
3
.....E
4
2
> c.:
E
,....
E E
~
~
> ~
•
6
~.
~
3 2
4, 5 5 5,5 6 6,5 7 7, 5 8 8 , 5 9 9, 5 10
BPD (mm)
o
4, 55 5, 5 6 6 ,5 7 7,5
8 8 , 5 9 9,510
BPD (mm)
Fig. 2. ISWT, LVWT, and RVWT values obtained in fetuses of diabetic mothers, as plotted on our reference range (mean and 95% confidence intervals).
Echocardiography in fetuses of diabetic mothers
Volume 164 l\umber 3
MITRAL VALVE
o
TRICUSPID VALVE
0,9
0,9
0,8
0,8 0
~
0,7
;;
0,7
L..
0,6
CI L..
...<
0,6
I.IJ
0,5
...< I.IJ
0,5 0,4
0,4
0,3
0,3
0,2
0,2
0,1
0,1
°I ..
..
18 20 22 24 26 28 30 32 3436 38 40
gestational age (weeks)
839
~
~ • • / ~ • ••
.......
°
18 20 22 2426 28 30 32 34 36 38 40
gestational age (weeks)
Fig. 3. Transtricuspid and trans mitral E/ A ratios obtained in fetuses of diabetic mothers, as plotted on our reference ranges for gestation (mean and 95% confidence intervals).
as a pulsed Doppler ultrasonography with a carner frequency that ranges from 2.5 to 5 MHz. The fetal heart was first examined with twodimensional echocardiography, and two patients were excluded from the study because of congenital heart diseases (i.e., tetralogy of Fallot and atrioventricular canal). A subcostal four-chamber view of the fetal heart was obtained in the remaining 40 patients. The Mmode beam was directed perpendicular to the interventricular septum at the level of the atrioventricular valves, and images were recorded. 5 Velocity waveforms were then recorded at the level of the atrioventricular valves, ascending aorta, and pulmonary artery, according to previously reported techniques. 6. 7 Mitral and tricuspid valves were visualized from the apical fourchamber view, as were the aortic valve from the fivechamber view and the pulmonary valve from the shortaxis view. The color flow mapping function was then superimposed, and the valve flow was evidenced. Care was taken to visualize the flow so that it was parallel to the Doppler beam. The pulsed Doppler sample volume was placed immediately distal to valve leaflets, and velocity waveforms were recorded. Recordings at a beam angle larger than 20 degrees were rejected. Doppler and M-mode images were recorded on standard V2-inch videotape for subsequent analysis. Permanent records were obtained from the videotape by means of a strip chart recorder. Ten consecutive heart cycles were selected during periods of fetal rest without breathing movements, and the values measured were averaged. These variables were measured with the aid of a computer-interfaced digitizer pad (Cardio 800, Kontron, Oxford): (1) wall thickness during end diastole at the level of the interventricular septum (ISWT), left ventricle (LVWT), and right ventricle (R VWT)5; (2) tricuspid and mitral blood flow velocities during early
ventricular filling (E wave) and during atrial systole (A wave)6; (3) the peak velocities of the aortic and pulmonary valves; (4) pulmonary and aortic time to peak velocity values, which were calculated as the time differences from the onset of the waveform to its peak velocity 7; (5) heart rate. The ratios between E-wave and A-wave peak velocities (the EI A ratio) at the level of both atrioventricular valves were then calculated as indices of ventricular diastolic function. 6 Data analysis. The variables measured were compared with our reference limits that were obtained from the cross-sectional study of 197 fetuses from normal pregnancies that were between 20 and 40 weeks' gestation. Because the size of the heart is better related to fetal dimension than to gestational age,5 reference limits for ISWT, LVWT, and RVWT were constructed considering the fetal biparietal diameter as the independent variable. In addition, gestational age was used as the independent variable to calculate normative data for Doppler measurements. 6. 7 Because all the variables considered change with gestation, the individual echocardiographic values observed in fetuses of diabetic mothers were also expressed as the number of standard deviations from the respective normal means for gestation (e.g., ~ISWT, ~mitral EI A ratio). A one-sample Student t test was used to compare echocardiographic parameters of control fetuses with those of fetuses of diabetic mothers, whereas the differences among diabetic fetuses were tested with the unpaired Student t test. Simple and multiple regression analyses were used to test the relationships between the variables considered. These analyses were performed on a Macintosh II (Apple Computers, Cupertino, Calif.) computer with the Stat View II statistical package (Abacus Concepts Inc., Berkeley, Calif.).
840
Rizzo, Arduini, and Romanini
01
-
-.5~
.... o
CI ~
=
~
1.5
-2
CI
~ 2.5 -31
.5
••
01
•
••
•
I.
LI.I
....S ...
•
-1
< ......
March 1991 Am J Obstet Gynecol
.-... -.51o
•
CI
.
0
•
LI.I
::;, C.J
·
...·c ...Qj 2.5
-2
'C
.5
.-... ~
S -2
... CI
•
• •• •
~ -2.5~
delta lVWT
., • • •
:::;o - .5 t• CI ~
< ......
-1
LI.I 'C
•
'ji -1.5 III
::;, C.J
... ...Qj -2.5
·C -2 CI
'C
I
1.5 2 2.5 3 3.5 4 4.5 5 5.5 delta ISWT
•
-31 .5 .75 1 1.21.51.8 2 2.22.52.8 3
1
0
•
CI
-31
=-1.5
-
1
•
•
LI.I
O.
1.5~
III
•
•
~
•
< ......
1.5 2 2.5 3 3.5 4 4.5 5 5.5 delta ISWT
.c:~ -.51 .... • • • < -H·~· • ...... ~. .•• • Q.
•
.~.• •• ••• .'•••• • •• • • ••• • •
01
'C
~ -1 •
.
•
-31 .5
•
•
•
• • • •• • .... •• • • • 1.5 2 2.5 3
. . ~
3.5
delta RVWT
Fig. 4. Relationships between mitral EI A ratio with ISWT (left) and LVWT (right) and between tricuspid EI A ratio with ISWT (left) and RVWT (right). All values are expressed as number of SDs by which they differed from the respective means for gestational age (~-values).
Results
Successful recordings were obtained in all cases at the level of the atrioventricular valves (Fig. 1). In 32 cases reliable data were obtained from all four cardiac sites, whereas fetal position did not allow successful recordings at the level of the pulmonary artery and ascending aorta, respectively, in three and five cases. At birth none of the infants had clinical signs of cardiomyopathy, with the exception of the two fetuses excluded for congenital heart diseases. M-mode echocardiography. The ISWT values of fetuses of diabetic mothers were significantly higher (mean difference 2.632 SDs, t = 16.437, P s 0.001) when compared with our reference limits for gestation. Similar results were obtained when LVWT (mean difference 1.626 SDs, t = 15.987, P S 0.001) and RVWT (mean difference 1.790SDs,t = 15.201,p S 0.001) values were considered (Fig. 2). However, there were no
relationships between HbA 1c and .lISWT (r = 0.194), .lLVWT (r = 0.123), and .lRVWT (r = 0.106) values. Furthermore no differences were found in M-mode parameters when fetuses of diabetic mothers assigned to White's class B were compared with those fetuses of mothers in classes C and D. Diastolic function. In the fetuses of diabetic mothers the EI A ratio results were significantly lower at the level of both mitral valves (mean difference -1.634 SDs, t = 15.414, P S 0.001) and tricuspid valves (mean difference -1.449 SDs, t = 14.038, P S 0.001) when compared with the EI A ratios of control fetuses (Fig. 3). A significant negative relationship was found between .l mitral EI A ratio and both .lISWT (constant = - 0.790, slope = - 0.321, SD = 0.567, r = 0.552, P S 0.001) and .lLVWT values (constant = -0.794, slope = -0.517, SD = 0.590,r = 0.496,ps 0.002) (Fig. 4) with a multiple r value of 0.601. Simi-
Echocardiography in fetuses of diabetic mothers
Volume 164 Number 3
841
AORTA 140,
70
..... 120 fI)
60
.....
E 100 u
"'" ='I
.oJ
u
CI ~
>
~
«I
..... E "'" > Il.
50
fI)
80 60
I-
40 30
40
20
20
10
~
•
•• • • • I .- ••••- • ••• • • • • • ••
...
. I.·
•
•
Q.
0
0 18 20 22 24 26 28 30 32 3436 38 40
18 20 22 24 26 28 30 32 3436 38 4(
gestational age (weeks)
gestational age (weeks)
PULMONARY ARTERY 60
120
'in .....
100
u
80
E
"'" ='I oJ
U
CI ~
> ~
60 40
«I
~
Q.
20
~ . I ·
~ ~
..... .§ > Il.
50
fI)
40
I- 30
20
-- ....
~ •••••• • •••• ~ • ••• • •• •
•
~
10 0
0 18 20 22 24 26 28 30 32 34 36 38 40
gestational age (weeks)
18 20 22 24 26 28 30 32 34 36 38 40
gestational age (weeks)
Fig. 5. Aortic (toP) and pulmonary (bottom) peak velocity and time to peak velocity (TPV) values in fetuses of diabetic mothers, as plotted on our reference ranges for gestation (mean and 95% confidence intervals).
larly, ~ tricuspid EI A ratio results were negatively related to ~ISWT (constant = - 0.643, slope = - 0.306, SD = 0.556, r = 0.541, P ~ 0.002) and ~RVWT values (constant = - 0.629, slope = - 0.458, SD = 0.564, r = 0.491, P ~ 0.005) (Fig. 4) with a multiple r value of 0.590. The HbA lc value results were unrelated to mitral (r = 0.141) and tricuspid (r = 0.152) ~ E/A ratios. No significant differences were found in EI A ratios when fetuses were grouped according to White's class of the mother. Systolic function. In Fig. 5 aortic and pulmonary values of peak velocity and time to peak velocity obtained in fetuses of diabetic mothers are plotted against our reference limits for gestation. No significant changes were evidenced and no significant correlations were found with MSWT, ~RVWT, and ~LVWT (r < 0.116), with HbA lc (r < 0.123) values, with White's class of the mother.
Relations to pregnancy outcome. Nine fetuses required emergency cesarean section because of fetal distress (abnormal fetal heart patterns characterized by late or repetitive variable decelerations and/ or loss of baseline variability) (Table I). When these fetuses were compared with the remaining fetuses of diabetic mothers studied, no significant differences were found in the cardiac parameters considered (Table II). Similarly, there were no differences when echocardiographic findings of large-for-gestational age fetuses (i.e., birth weight above the 90th percentile) were compared with those of appropriate-for-gestational-age fetuses (Table II). Comment A linear increase with fetal growth of both heart size and wall thickness has been evidenced in the human fetus. 5 Moreover, a concomitant improvement of the
842
Rizzo, Arduini, and Romanini
March 1991 Am J Obstet Gynecol
Table II. Echocardiographic measurements in fetuses of diabetic mothers grouped according to birth weight or occurrence of fetal distress (i.e., abnormal heart rate patterns necessitating emergency cesarean section) Fetal distress (No. = 9)
ISWT LVWT RVWT Mitral EI A ratio Tricuspid EI A ratio Peak velocity, aorta Time to peak velocity, aorta Peak velocity, pulmonary artery Time to peak velocity, pulmonary artery
2.66 1.65 1.76 -1.59 -1.54 0.33 -0.21
± ± ± ± ± ± ±
1.29 0.67 0.78 0.74 0.69 0.71 0.46
Remaining fetuses (No. = 31)
2.62 1.62 1.80 -1.66 -1.46 0.45 -0.34
± ± ± ± ± ± ±
p Value*
l.l4 0.63 0.74 0.67 0.63 0.44 0.39
NS NS NS NS NS NS NS
0.87 ± 0.67
0.79 ± 0.47
NS
-0.18 ± 0.33
0.16 ± 0.39
NS
LCAfetuses (No. = 4)
2.52 1.66 1.82 -1.62 -1.50 0.40 -0.59
± ± ± ± ± ± ±
1.44 0.90 0.75 0.72 0.67 0.88 0.62
ACAfetuses (No. = 36)
2.60 1.63 1.79 -1.65 - 1.48 0.47 -0.31
± ± ± ± ± ± ±
p
Value*
1.28 0.62 0.74 0.63 0.64 0.39 0.33
NS NS NS NS NS NS NS
0.73 ± 0.69
0.75 ± 0.44
NS
0.22 ± 0.33
0.14 ± 0.31
NS
-
LCA, Large for gestational age; ACA, appropriate for gestational age. All values are expressed as number of standard deviations by which they differed from the respective means for gestational age (Ll-values). * Unpaired t test.
cardiac performance has been described. The EI A ratio, a widely accepted index of ventricular diastolic function,S significantly increases during gestation at the level of both atrioventricular valves. These changes are believed to be secondary to the progressive maturation of ventricular compliance. 9 There is also an increase of aortic and pulmonary peak velocities and of left and right stroke volumes, to which suggests an improvement of ventricular systolic function with gestation. Previous studies have shown that in infants of diabetic mothers there is a hypertrophic cardiomyopathy that is characterized primarily by a significant thickening of the interventricular septum. 1. II Our data on a large series confirm previous sporadic reports 2 . 3 that the increase of ISWT, LVWT, and RVWT is already established during intrauterine life. The absence of correlations between the HbA lc values and the severity of cardiac wall thickness found in this investigation is in agreement with the results of neonatal studies l2 and suggests that in spite of adequate metabolic control during pregnancy, fetuses of diabetic mothers might experience an abnormal cardiac growth. However, there is no current account regarding whether the increase of cardiac wall thickness merely reflects the larger size of fetuses of diabetic mothers or whether it might also influence the fetal cardiac function. The lower EI A ratios at the level of both atrioventricular valves found in these fetuses might indicate an impaired ventricular diastolic function. The inverse relationships between the EI A values and both the interventricular septum and lateral ventricular wall thickness suggest that the abnormal cardiac thickening might impair ventricular compliance, which would affect the filling patterns of the ventricles. Previous microscopic studies on the heart in newborns of diabetic mothers corroborate this hypothesis by show-
ing a disorganization of the myofibrils, which is similar to that present in adults with hypertrophic cardiomyopathy, I. 13 a condition characterized by evident alterations of ventricular compliance and thus of Dopplermeasured diastolic function. 14 In spite of a ventricular hypercontractility reported in infants of diabetic mothers, I no significant modifications of systolic indices such as aortic and pulmonary peak velocities or time to peak velocity were evidenced in the fetuses examined. In adults diastolic dysfunction usually precedes systolic anomalies that appear later in the development of the myocardial disease. 15 Therefore it can be speculated that the fetuses studied have not reached a degree of cardiac impairment sufficient to affect systolic function. This hypothesis is supported by the data available on infants of diabetic mothers which show that echocardiographic indices of systolic function fall either in the normal range or above the normal range I and are usually abnormal only in newborns with clinical signs of cardiac impairment. 16 In conclusion our data suggest that despite careful metabolic control of pregnancy, fetuses of diabetic mothers show an increased thickness of the interventricular septum and of the ventricular walls, which may be associated with abnormalities of diastolic function. Further longitudinal studies during the perinatal period are required to clarify the clinical significance of these changes of cardiac function and their sequelae during neonatal life. REFERENCES I. Reller MD, Kaplan S. Hypertrophic cardiomyopathy in infants of diabetic mothers: an update. Am j Perinatol 1988;5:353-8. 2. Leisle j, Shen SC, Strauss L. Hypertrophic cardiomyop-
Volume 164 Number 3
3.
4. 5.
6.
7.
8. 9.
athy in a mid-trimester fetus born to a diabetic mother. J Pediatr 1982;100:631-2. Weber HS, CopelJA, GreenJ, Reece AE, Kleimann CS. Cardiac growth in the fetus of the diabetic mother [Abstract 029]. In: Proceedings of the fetal cardiac symposium, Paris, France. 1989:60. Hare JW, White P. Gestational diabetes and the White classification. Diabetes Care 1980;3:294-305. DeVore GR, Siassi B, Platt LD. Fetal echocardiography. IV. M-mode assessment of ventricular size and contractility during the second and third trimesters of pregnancy in the normal fetus. AM J OBSTET GYNECOL 1984;150: 981-8. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small for gestational age fetuses. Br J Obstet Gynaecol 1988;95:65-9. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic evaluation of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Br J Obstet Gynaecol 1990;97:603-7. Labovitz AJ, Pearson C. Evaluation of left ventricular diastolic function: clinical relevance and recent Doppler echocardiographic insights. Am HeartJ 1987; 114:836-51. Reed KL, Sahn DJ, Scagnelli S, Anderson CF, Shenker L. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coli Cardiol 1986;8:391-5.
Echocardiography in fetuses of diabetic mothers
843
10. Kenny JF, Plappert T, Saltzman DH, et al. Changes in intracardiac blood flow velocities and right and left ventricular stroke volumes with gestational age in the normal human fetus: a prospective Doppler echocardiographic study. Circulation 1986;74:1208-16. 11. Breitweser JA, Meyer RA, Sperling MA. Cardiac septal hypertrophy in hyperinsulinemic infants. J Pediatr 1980; 96:535-9. 12. Sheehan PQ, Rowland TW, Shah BL, McGravey VJ, Reiter EO. Maternal diabetic control and hypertrophic cardiomyopathy in infants of diabetic mothers. Clin Pediatr 1986;25:266-71. 13. Gutgessel HP, Speer ME, Rosenburg HS. Characterization of the cardiomyopathy in infants of diabetic mothers. Circulation 1980;61 :441-50. 14. Takaneka K. Dabestani A, Gardin JM, et al. Left ventricular filling in hypertrophic cardiomyopathy: a pulsed Doppler echocardiographic study. J Am Coli Cardiol 1986; 7: 1263-71. 15. Aroesty JM, McKay RG, Heller GV, Royal HD, Als AV, Grossman W. Simultaneous assessment of left ventricular systolic and diastolic dysfunction during pacing induced ischemia. Circulation 1985;71 :889-900. 16. Mace S, Hirschfeld SS, Riggs T, Fanaroff AA, Meikatz IR. Echocardiographic abnormalities in infants of diabetic mothers. J Pediatr 1979;95:1013-9.