Comparison of fetal growth in singleton, twin, and triplet pregnancies

Atsushi Kuno1, Masashi Akiyama, Toshihiro Yanagihara and Toshiyuki Hata

Department of Perinatology, Kagawa Medical University, 1750-1 Ikenobe, Miki, Kagawa 761-0793, Japan


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The objective of this longitudinal retrospective study was to evaluate differences of the fetal growth and fetal organ growth among singleton small for gestational age (S-SGA), singleton appropriate for gestational age (S-AGA), twin (Tw-AGA), and triplet (Tri-AGA) infants. Ultrasonographic examinations were performed on 35 S-AGA, 18 S-SGA, 52 Tw-AGA and 12 Tri-AGA fetuses. Circumferences of head (HC), abdomen (AC), spleen (SC) and adrenal gland (AGC) and lengths of femur diaphysis (FDL), liver (LL), estimated weight (EWT) were measured every 2 weeks after 15 weeks of menstrual age until delivery. There was no significant difference in predicted HC values in S-AGA, Tw-AGA and Tri-AGA fetuses; these values were lowest in S-SGA fetuses. As the number of fetuses in the uterus increased with advancing menstrual age, the slope of the growth curve for predicted AC value became lower, but there was no significant difference between Tri-AGA and S-SGA fetuses. There was no significant difference in predicted FDL values among Tw-AGA, Tri-AGA and S-SGA fetuses; those values were significantly lower than that in S-AGA fetuses. There was no significant difference in predicted EWT value between Tw-AGA and Tri-AGA fetuses, which were intermediate between those for S-AGA and S-SGA fetuses. There were no significant differences in predicted SC and AGC values between S-AGA and Tw-AGA fetuses, respectively. However, in S-SGA fetuses, the slopes of the growth curve for SC and AGC were lower than those in the other two groups with advancing menstrual age. There were slight differences in predicted LL values between S-AGA, S-SGA and Tw-AGA fetuses. These results suggest that in AGA fetuses, there was a slight difference in growth pattern among singleton, twin, and triplet pregnancies.

Key words: appropriate for gestational age infant/fetal growth/multiple fetuses/organ growth/singleton fetus


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
With the introduction of ovulation-inducing agents, the incidence of multiple pregnancies has increased (Holcberg et al., 1982Go; Loucopoulos et al., 1982). Moreover, the implementation of multiple embryo transfer in in-vitro fertilization (IVF) programmes may further increase multiple pregnancies (Weissman et al., 1990Go). Multiple pregnancies are known to be at increased risk of a variety of complications during the antepartum and intrapartum periods (Landy and Keith, 1998Go). These complications include premature delivery, fetal growth restriction, congenital anomalies, placenta previa, abruptio placentae, cord accidents, and malpresentations. Due to the complexity of multiple pregnancies, obstetric management is more difficult, and thus a means for evaluating its effectiveness is essential (Xu et al., 1995Go).

The use of fetal growth curves generated from uncomplicated singleton pregnancies to estimate intrauterine growth of multiple pregnancies is controversial (Weissman et al., 1990Go). Some studies recommended the use of specific charts generated for twin pregnancies (Grennert et al., 1978Go; Sokol et al., 1984Go), whereas others conclude that growth curves for singletons may be applied safely for twins, but not for triplets (Sheh et al., 1987Go). To the best of our knowledge, there has been no report on the evaluation of singleton, twin, and triplet appropriate for gestational age fetal growth using a longitudinal ultrasound study.

With recent advances in ultrasound imaging, especially improvements in resolution and focusing, fetal intracranial, intrathoracic, and intra-abdominal organs can now be clearly identified (Hata and Deter, 1992Go). Numerous reports concerned with fetal organ measurements on liver, spleen, and adrenal gland in singleton pregnancies have been presented (Aoki et al., 1992Go; Hata et al., 1992Go, 1993Go; Senoh et al., 1994Go). However, there has been no report on these organ growth measurements made with ultrasound in multiple pregnancies.

The objective of our present study was to construct reference limits for various fetal growth parameters from head circumference, abdominal circumference, femur diaphysis length, estimated weight, splenic circumference, adrenal gland circumference, and liver length, based on a longitudinal study of 35 singleton appropriate for gestational age (S-AGA), 18 singleton small for gestational age (S-SGA), 52 twin appropriate for gestational age (Tw-AGA), and 12 triplet appropriate for gestational age (Tri-AGA) infants, and to evaluate the alterations in those growth parameters among these four fetal groups during gestation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
We studied 35 S-AGA (18 females and 17 males), 18 S-SGA (14 females and four males), 52 Tw-AGA (27 females and 25 males) in 28 pregnancies (nine monochorionic diamniotic and 19 dichorionic pairs), and 12 Tri-AGA (eight females and four males) in five pregnancies (five trichorionic triplets). Four twins (two small for gestational age and two large for gestational age infants) and three triplets (two small for gestational age and one large for gestational age infants) were excluded from the study. One S-AGA pregnancy was excluded from the study when maternal systemic lupus erythematosus appeared during pregnancy, and one S-SGA pregnancy due to fetal abnormalities. The all pregnancies were in middle-class Japanese women from the Kagawa area. All women were non-smokers, with neither indication of maternal complication nor evidence of drug ingestion. There were no significant differences for the height, weight and parity of the maternal population among the four groups (Gardosi et al., 1992Go). All the obstetric deliveries were made in our university hospital, and comprehensive paediatric assessment (within 24 h of delivery) revealed no evidence of genetic diseases or congenital anomalies. There were nine monochorionic diamniotic and 16 dichorionic pairs that were conceived naturally. Two pregnancies were induced artificially by induction of ovulation; one pregnancy in the Tw-AGA group and five in the Tri-AGA group were IVF pregnancies.

Birth characteristics of subjects are shown in Table IGo. Birth weight of 35 S-AGA were in the normal range (between the 10th and 90th percentiles) of the standard growth curve for the Japanese (Sato et al., 1982Go), and 18 S-SGA were below normal ranges. Birth weights in Tw-AGA and Tri-AGA infants were compared with the twin weight growth curve for the Japanese (Fukuda, 1989Go), and all fetal weights were appropriate for gestational age. The study was approved by the local ethical committee of Kagawa Medical University, and standardized informed consent was obtained from each patient.


View this table:
[in this window]
[in a new window]
 
Table I. Birth characteristics of subjects
 
Fetal age determination was estimated from the first day of the last menstrual period and confirmed by the first-trimester and early second-trimester ultrasound examinations (crown–rump length, biparietal diameter, and femur diaphysis length measurements) (Tsuzaki et al., 1982Go; Iwamoto, 1983Go) or determined from the date of conception (IVF pregnancies) plus 2 weeks (the ultrasound age estimates confirmed these age determinations).

Ultrasound examinations were carried out at 2 week intervals beginning at ~15 weeks of menstrual age continuing until delivery. The number (mean ± SD) of examinations of individual patients ranged from 5 to 13 (9.3 ± 2.1) in the S-AGA group, from 8 to 13 (10.1 ± 1.5) in the S-SGA group, from 5 to 12 (8.0 ± 2.0) in the Tw-AGA group, and from 8 to 10 (9.5 ± 0.9) in the Tri-AGA group. At each examination, each fetus was identified by its position in the uterus, its size, or its sex in multiple pregnancies. Measurements of the biparietal diameter, head circumference, abdominal circumference and femur diaphysis length were obtained at each examination for each fetus using procedures described previously (Deter et al., 1981Go). The estimated weight was determined from values for the biparietal diameter, abdominal circumference, and femur diaphysis length as described previously (Shinozuka et al., 1987Go). The methods used to obtain splenic circumference, liver length, and adrenal gland circumference, measurements have been described in detail elsewhere (Aoki et al., 1992Go; Hata et al., 1993Go; Senoh et al., 1994Go). Unfortunately, fetal organ measurements in triplet pregnancy could not be done, because of fetal crowding and its complexity in the uterus.

Results are expressed as mean ± SD. Statistical analysis for comparison of maternal age, parity, and Apgar score among the groups was done using a Kruskal–Wallis one-way analysis of variance. Maternal height, maternal weight, birth age, birth weight, neonatal crown–heel length, neonatal head circumference, neonatal abdominal circumference, and neonatal thigh circumference were compared using an analysis of variance and Newman–Keuls multiple comparison test. P < 0.05 was considered to be significant.

For each parameter in each fetal group data set regression analysis was carried out, testing the regression of the measurement value on menstrual age, using polynomials of the first through the third degree (Dunn and Clark, 1974Go; Rohatgi, 1976Go; Bertagnoli et al., 1983Go). Different models were tested and independent variable deletion carried out by analysis of variance applied to the regression was followed by calculation of the step-down method coefficients (Snedecor and Cochran, 1967Go). The choice of the optimal model was based on the following criteria: largest R2, all coefficients different from 0, and low standard deviation of regression (SDR ) (Bertagnoli et al., 1983Go).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The results of the mathematical modelling of the data in S-AGA, S-SGA, Tw-AGA, Tri-AGA fetuses are shown in Table IGo. The predicted values of each parameter derived from these functions and their variability at different menstrual ages in AGA fetuses are presented in Tables III–IXGoGoGoGoGoGoGo. Comparisons of predicted values for each parameter among the groups are presented graphically in Figures 1–7GoGoGoGoGoGoGo.


View this table:
[in this window]
[in a new window]
 
Table III. Head circumference value (cm) in each group
 

View this table:
[in this window]
[in a new window]
 
Table IV. Abdominal circumference value (cm) in each group
 

View this table:
[in this window]
[in a new window]
 
Table V. Femur diaphysis length value (cm) in each group
 

View this table:
[in this window]
[in a new window]
 
Table VI. Estimated weight value (g) in each group
 

View this table:
[in this window]
[in a new window]
 
Table VII. Adrenal gland circumference value (cm) in each group
 

View this table:
[in this window]
[in a new window]
 
Table VIII. Splenic circumference value (cm) in each group
 

View this table:
[in this window]
[in a new window]
 
Table XI. Liver length value (cm) in each group
 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Change in head circumference with menstrual age. S-AGA = singleton appropriate for gestational age fetus; S-SGA = singleton small for gestational age fetus; Tw-AGA = twin appropriate for gestational age fetus; Tri-AGA = triplet appropriate for gestational age fetus.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 2. Change in abdominal circumference with menstrual age. For abbreviations see Figure 1Go.

 


View larger version (12K):
[in this window]
[in a new window]
 
Figure 3. Change in femur diaphysis length with menstrual age. For abbreviations see Figure 1Go.

 


View larger version (14K):
[in this window]
[in a new window]
 
Figure 4. Change in estimated weight with menstrual age. For abbreviations see Figure 1Go.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 5. Change in splenic circumference with menstrual age. For abbreviations see Figure 1Go.

 


View larger version (14K):
[in this window]
[in a new window]
 
Figure 6. Change in adrenal gland circumference with menstrual age. For abbreviations see Figure 1Go.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 7. Change in liver length with menstrual age. For abbreviations see Figure 1Go.

 
There was no significant difference in predicted head circumference values in S-AGA, Tw-AGA and Tri-AGA fetuses (Figure 1Go). However, in S-SGA fetuses, it was lower than those in the other three groups. For abdominal circumference values, as the number of fetuses in the uterus increased with advancing menstrual age, the slope of the growth curve for predicted abdominal circumference became lower (Figure 2Go). There was no significant difference in predicted abdominal circumference value between Tri-AGA and S-SGA fetuses. For femur diaphysis length values, there was no significant difference in predicted femur diaphysis length values among Tw-AGA, Tri-AGA and S-SGA fetuses (Figure 3Go). However, those values were significant lower than that in S-AGA fetuses. For estimated weight values, there was no significant difference in predicted estimated weight value between Tw-AGA and Tri-AGA fetuses (Figure 4Go). These values had an almost intermediate value between those of S-AGA fetuses and S-SGA fetuses.

There were no significant differences in predicted splenic circumference (Figure 5Go) and adrenal gland circumference (Figure 6Go) values between S-AGA and Tw-AGA fetuses. However, in S-SGA fetuses, the slopes of the growth curve for splenic circumference and adrenal gland circumference were lower than those in the other two groups with advancing menstrual age. There were slight differences in predicted liver length values between S-AGA, S-SGA and Tw-AGA fetuses, respectively (Figure 7Go).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
According to the birth data from live-born triplet infants (Elster et al., 1991Go), from 26 to 35 weeks, the average triplet newborn has a weight corresponding to approximately the 30th percentile level compared with singletons. After 35 weeks, triplet birth weights fall progressively behind those of singletons, reaching the 10th percentile at 38 weeks. Other data (Jones et al., 1991Go) on triplet growth when compared with previously reported singleton growth (Brenner et al., 1976Go) support the concept that multiple and singleton gestations exhibit similar growth through 30 weeks. Beyond 30 weeks gestation, the singleton fetus appears to accelerate in growth in a sigmoid manner. The triplet gestation lacks this acceleration in growth shown by singleton gestations. However, triplet growth remains linear throughout the third trimester rather than plateauing as previously suggested (Jones et al., 1991Go). Similarly, the mean birth weight of triplets was slightly below the 10th percentile for singletons at 38 weeks gestation or later (Yuval et al., 1995Go). They concluded that the growth of triplet fetuses as estimated from live-born birth weights is slower than that of singletons. However, the subjects of these three investigations included many small for gestational age triplet infants studied as live-born babies. Therefore, their standard triplet growth curve may be lower than that of singletons, especially in the third trimester of pregnancy. In this study, we selected appropriate for gestational age fetuses as a subject in multiple pregnancies, to avoid this bias, and we tried to evaluate whether growth differences exist among S-AGA, Tw-AGA and Tri-AGA infants in utero. Consequently, there was no significant difference in predicted estimated weight value between Tw-AGA and Tri-AGA fetuses. However, these values were slightly lower than that of S-AGA fetuses and slightly higher than that of S-SGA fetuses. The reason for these differences is currently unknown. It has been suggested (Hata et al., 1991Go; Xu et al., 1995Go) that the deposition of soft tissue seen in normal singletons during the third trimester occurs to a much lesser extent in normal twins and triplets. The current study may give strong support to the this work (Hata et al., 1991Go; Xu et al., 1995Go), and further study is needed to clarify whether this decrease in soft tissue deposition in multiple pregnancies represents a true growth abnormality or merely a physiological adaptation to the energy demands associated with the support of growth in multiple fetuses.

There have been only two reports on the growth of triplet fetuses measured by ultrasonography between 15 weeks gestation to delivery (Weissman et al., 1990Go; Fountain et al., 1995Go). However, these two prenatal ultrasound studies were carried out in the USA and UK, respectively. Available data for Asian populations are limited. To the best of our knowledge, there have been no reports on examination of multiple dimensions of fetal growth in multiple pregnancies in the Asian population. It has been stressed (Lei and Wen 1998Go) that a different standard of ultrasonography based fetal growth is needed for different populations. Therefore, the ultrasonography-based growth curve constructed in this longitudinal Japanese population provides an additional tool for the evaluation of fetal growth and development in multiple pregnancies.

With respect to fetal organ measurements by ultrasonography in normal twin pregnancy compared to normal singleton pregnancy, there were no significant differences in predicted splenic circumference and adrenal gland circumference values between S-AGA and Tw-AGA fetuses in this study. However, there was a slight difference in predicted liver length value between S-AGA and Tw-AGA fetuses. Similarly, the predicted abdominal circumference value in Tw-AGA fetuses was slightly lower than that in S-AGA fetuses. The liver occupies most of the upper abdominal cavity, so corresponding differences for liver length and abdominal circumference between S-AGA and Tw-AGA fetuses may occur. Our results suggest that growth curves of organ measurements for singletons may be applied safely for twins.


View this table:
[in this window]
[in a new window]
 
Table II. Models fitted to the data
 

    Notes
 
1 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Aoki, S., Hata, T. and Kitao, M. (1992) Ultrasonographic assessment of fetal and neonatal spleen. Am. J. Perinatol., 9, 361–367.[ISI][Medline]

Bertagnoli, L., Lalatta, F., Gallicchio, R. et al. (1983) Quantitative characterization of the growth of the fetal kidney. J. Clin. Ultrasound, 11, 349–356.[ISI][Medline]

Brenner, W.E., Edelman, D.A. and Hendricks, C.H. (1976) A standard of fetal growth for the United States of America. Am. J. Obstet. Gynecol., 126, 555–564.[ISI][Medline]

Deter, R.L., Harrist, R.B., Hadlock, F.P. and Carpenter, R.J. (1981) The use of ultrasound in the assessment of normal fetal growth: a review. J. Clin. Ultrasound, 9, 481–493.[ISI][Medline]

Dunn, O.J. and Clark, V.A. (1974) Applied Statistics: Analysis of Variance and Regression. Wiley, New York, pp. 252–305.

Elster, A.D., Byeyl, J.L. and Craven, T.E. (1991) Birth weight standards for triplets under modern obstetric care in the United States, 1984–1989. Obstet. Gynecol., 77, 387–393.[Abstract]

Fountain, S.A., Morrison, J.J., Smith, S.K. and Winston, R.M.L. (1995) Ultrasonographic growth measurements in triplet pregnancies. J. Perinat. Med., 23, 257–263.[ISI][Medline]

Fukuda, M. (1989) Clinical study of twins, Part 1. Intrauterine growth curve of twins. Acta Neonat. Jpn., 26, 366–371.

Gardosi, J., Chang, A., Kalyan, B. et al. (1992) Customised antenatal growth charts. Lancet, 339, 283–287.[ISI][Medline]

Grennert, L., Person, P. and Gennser, G. (1978) Intrauterine growth of twins judged by BPD measurements. Acta Obstet. Gynecol. Scand., 78, 28–32.

Hata, T. and Deter, R.L. (1992) A review of fetal organ measurements obtained with ultrasound: normal growth. J. Clin. Ultrasound, 20, 155–174.[ISI][Medline]

Hata, T., Deter, R.L. and Hill, R.M. (1991) Reduction of soft tissue deposition in normal triplets. J. Clin. Ultrasound, 19, 541–545.[ISI][Medline]

Hata, T., Deter, R.L., Aoki, S. et al. (1992) Mathematical modeling of fetal splenic growth: Use of the Rossavik Growth Model. J. Clin. Ultrasound, 20, 321–327.[ISI][Medline]

Hata, T., Deter, R.L., Nagata, H. et al. (1993) Mathematical modeling of fetal organ growth using the Rossavik Growth Model: II. Adrenal grand. Am. J. Perinatol., 10, 97–100.[ISI][Medline]

Holcberg, G., Biale, Y., Lewenthal, H. and Insler, V. (1982) Outcome of pregnancy in 31 triplet gestations. Obstet. Gynecol., 59, 472–476.[Abstract]

Iwamoto, K. (1983) Estimation of gestational age with ultrasonic measurement of the fetus in each trimester. Acta Obstet. Gynaecol. Jpn., 35, 2330–2338.

Jones, J.S., Newman, R.B. and Miller, M.C. (1991) Cross-sectional analysis of triplet birth weight. Am. J. Obstet. Gynecol., 164, 135–140.[ISI][Medline]

Landy, H.J. and Keith, L.G. (1998) The vanishing twin: a review. Hum. Reprod. Update, 4, 177–183.[Abstract/Free Full Text]

Lei, H. and Wen, S.W. (1998) Ultrasonographic examination of intrauterine growth for multiple fetal dimensions in a Chinese population. Am. J. Obstet. Gynecol., 178, 916–921.[ISI][Medline]

Loucopoulos, A. and Jewelewicz, R. (1982) Management of multifetal pregnancies: Sixteen years' experience at the Sloane Hospital for Women. Am. J. Obstet. Gynecol., 143, 902–905.[ISI][Medline]

Rohatgi, V.K. (1976) An Introduction to Probability Theory and Mathematical Statistics. Wiley, New York, pp. 506–512.

Sato, A., Akama, M., Yamanobe, H. et al. (1982) Intrauterine growth of live-born Japanese infants between 28 and 42 weeks of gestation. Acta Obstet. Gynaecol. Jpn., 34, 1535–1538.

Senoh, D., Hata, T. and Kitao, M. (1994) Fetal liver length measurement does not provide a superior means for prediction of a small for gestational age fetus. Am. J. Perinatol., 11, 344–347.[ISI][Medline]

Shinozuka, N., Okai, T., Kohzuma, S. et al. (1987) Formulas for fetal weight estimation by ultrasound measurements based on neonatal specific gravities and volumes. Am. J. Obstet. Gynecol., 157, 1140–1145.[ISI][Medline]

Sheh, Y.G., Graham, D., Stinton, S.K. and Render, T.D. (1987) Biparietal diameter growth in uncomplicated twin gestation. Am. J. Perinatol., 4, 229–232.[ISI][Medline]

Snedecor, W.G. and Cochran, W.G. (1967) Statistical Methods, edn 6. Ames, Iowa State University Press, 135–197.

Sokol, M.L., Tamura, R.K., Sabbagha, R.E. et al. (1984) Diminished biparietal diameter and abdominal circumference growth in twins. Obstet. Gynecol., 94, 235–238.

Tsuzaki, T., Iwamoto, K. and Maeda, K. (1982) Significance in ultrasonographic measurement of fetal limb bones. Acta Obstet. Gynaecol. Jpn., 34, 315–320.

Weissman, A., Jakobi, P., Yoffe, N. et al. (1990) Sonographic growth measurements in triplet pregnancies. Obstet. Gynecol., 75, 324–328.[Abstract]

Xu, B., Deter, R.L., Milner, L.L. and Hill, R.M. (1995) Evaluation of twin growth status at birth using individual growth assessment: Comparison with conventional methods. J. Clin. Ultrasound, 23, 277–286.[ISI][Medline]

Yuval, Y., Seidman, D.S., Achiron, R. et al. (1995) Intrauterine growth of triplets as estimated from liveborn birth weight data. Ultrasound Obstet. Gynecol., 6, 345–348.[ISI][Medline]

Submitted on October 26, 1998; accepted on January 29, 1999.