JB .H Juomch Gimesneent e(t1 9a9l.8: )S 4te3t:t8e5r-ia9 0hydrogenophila © Jpn Soc Hum Genet and Springer-Verlag 198958
original
Nobuo Yaegashi · Masato Senoo · Shigeki Uehara
Hisako Suzuki · Tohru Maeda · Keiya Fujimori
Fumiki Hirahara · Akira Yajima
Received: July 23, 1997/Accepted: December 1, 1997
summary
The purpose of this study was to calculate the expected incidence of chromosomal abnormalities detected by amniocentesis in Japanese women aged 35 years or older . Genetic amniocentesis data from 5484 pregnant women aged 35 years or older, whose only risk was advanced age, were collected from four clinics in Japan. The data were analyzed using a logistic regression model. Of the 5484 fetuses, 117 (2.1%) were diagnosed with chromosomal abnormalities.
The chromosomal abnormalities were as follows: 42 trisomy 21, 13 trisomy 18, 7 trisomy 13, 10 47,XXY, 4 47,XXX, 1 47,XYY, 27 various structural abnormalities, and 13 various types of mosaicism. The incidence of trisomy 21, lethal autosomal aneuploidies (trisomy 18, trisomy 13), and sex chromosome abnormalities (XXY, XXX, XYY) increased with maternal age.
The regression parameters and their standard errors were calculated to determine the expected incidence of chromosomal abnormalities for each maternal age.
The predicted incidence of chromosomal abnormalities obtained in this study is the first such data in Japan and is likely to be useful for counseling pregnant women. The incidence of trisomy 21 is similar to the rates previously published in Western countries. The incidence of chromosomal abnormalities is not affected by racial or geographic factors.
keyword
Advanced maternal age, amniocentesis, prenatal diagnosis , chromosomes, Down syndrome
Introduction
It is well known that the frequency of Down syndrome (trisomy 21) and other chromosomal abnormalities increases with advancing maternal age (Polani et al. 1976; Schreinemachers et al. 1982; Ferguson-Smith and Yates 1984; Hook et al. 1984). Obstetricians, family physicians, and clinical geneticists are questioned by pregnant women about the possibility that their offspring may have cytogenetic abnormalities. In fact, even if not questioned, failure to provide such information to pregnant women in putative high-risk situations has been cited as a basis for legal liability. In genetic counseling before prenatal testing , it is important to provide pregnant women with an accurate estimate of their risk of abnormalities.
Especially for older women, physicians predict the incidence of chromosomal abnormalities for the patient’s age and obtain the patient’s informed consent before testing. To detect chromosomal abnormalities , genetic amniocentesis is commonly performed in the second trimester as one of several testing methods for prenatal diagnosis , as it is highly accurate, safe, and does not increase the risk of fetal loss or damage (NICHD National Registry for Amniocentes Study Group, 1976).
In Japan, the number and proportion of pregnancies among women aged 35 years or older has increased over the last 20 years, reaching 10% of all births in 1995 (Maternal and Child Health Organization 1996), resulting in a significant increase in the demand for genetic counseling and prenatal testing . However, data employed in counseling on the estimated risk of chromosomal abnormalities are mainly derived from Western countries such as Europe, North America, and Australia. Because biological and environmental differences may affect the incidence of chromosomal abnormalities , it seemed worthwhile to pool available data obtained from Japanese mothers to estimate the incidence of all significant cytogenetic abnormalities in 1-year maternal age intervals.
In this report, we obtained pooled genetic amniocentesis data from four genetic clinics in Japan, where tests were performed in the second trimester only because of advanced maternal age. We used logistic regression to analyze the relationship between maternal age and the incidence of cytogenetic abnormalities and estimated the incidence of abnormalities by maternal age at amniocentesis in the Japanese population.
material and method
We collected data on Japanese cases of cytogenetic amniocentesis from four genetic clinics where cytogenetic amniocentesis was routinely performed by obstetricians during the late stages of pregnancy (14-18 weeks): Kitasato University Hospital, Fukushima Prefectural University Hospital, Yokohama City University Hospital, and Tohoku University Hospital, and pooled the results. Gestational age was estimated from the last menstrual period and corrected by fetal ultrasound measurements. Contributors were asked to provide information on pregnancies that were tested solely because of maternal age and to exclude pregnancies with additional chromosomal indications, such as previous chromosomal abnormalities or parental chromosomal translocations.
Amniotic cells were cultured in Ham’s F10 or Chang’s medium containing 10% or 15% fetal bovine serum at 37°C under 5% CO2 for more than 2 weeks. After adding colcemid, the cells were cultured for another 5 hours, harvested, treated with hypotonic KCl solution (75 μmM), fixed with methanol/acetic acid fixative (3:1), and attached to a slide glass.
After air- or mechanically drying the glass slides, G-banding was performed using standard methods. The chromosomal abnormalities studied herein include structural abnormalities such as aneuploidies, translocations and inversions, as well as mosaicism on the sex chromosomes as well as on the autosomes. However, normal variants of chromosomal structure, such as pericentric inversion of chromosome 9, expanded heterochromatin on various chromosomes, expanded satellites, etc., were removed. If it was unclear whether mosaicism was an artifact, a second amniocentesis or fetal blood sampling was performed to be sure. To analyze the relationship between maternal age at amniocentesis and the logarithm of the incidence of chromosomal abnormalities , logistic regression analysis was performed using the formula y = exp( a + bx ), where x is maternal age at amniocentesis and y is the incidence of cytogenetic abnormalities at age x . The values of parameters a and b were obtained by a maximum likelihood regression approach . The correlation was assessed as statistically significant if the maternal age coefficient b was significantly different from zero ( P < 0.05) when considering the standard error. The expected incidence of chromosomal abnormalities at amniocentesis was calculated for each maternal age interval of 1 year using y = exp(a + bx ).
Result
Table 1 shows chromosomal abnormalities in a survey of 5484 second-trimester amniocentesis in pregnancies monitored by maternal age >35 years. Of these, 117 were diagnosed with chromosomal abnormalities , for a total incidence of 2.1% (117/5484). There were 62 non-mosaic autosomal and sex chromosomal aneuploidies (42 trisomy 21, 13 trisomy 18, and 7 trisomy 13) and 15 (10 XXY, 4 XXX, and 1 XYY), respectively. No cases of X-monosomy (Turner syndrome) were diagnosed.
Table 1 Number and incidence of chromosomal abnormalities by maternal age
| Maternal age at amniocentesis | Fetal number | Aneuploidy (excluding mosaicism) | Other | Total | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 47, +21 | 47, +18 | 47, +13 | XXY | XXX | XYY | 全異数性 | ||||
| 35 | 736 | 1(1.36)a | 1(1.36) | 1(1.36) | 2(2.72) | |||||
| 36 | 850 | 4(4.71) | 2(2.35) | 2(2.35) | 8(9,41) | 4(4.71) | 12(14.12) | |||
| 37 | 824 | 8(9.71) | 8(9.71) | 4(4.85) | 12(14.56) | |||||
| 38 | 800 | 2(2.50) | 2(2.50) | 2(2.50) | 6(7.50) | 7(8.75) | 13(16.25) | |||
| 39 | 713 | 8(11.22) | 2(2.81) | 1(1.40) | 1(1.40) | 12(19.17) | 8(11.22) | 20(28.05) | ||
| 40 | 626 | 6(9.58) | 5(7.99) | 1(1.60) | 1(1.60) | 13(20.77) | 10(15.97) | 23(36.74) | ||
| 41 | 417 | 3(7.19) | 3(7.19) | 2(4.80) | 2(4.80) | 10(23.98) | 1(2.40) | 11(26.38) | ||
| 42 | 255 | 6(23.53) | 1(3.92) | 3(11.76) | 1(3.92) | 11(43.14) | 2(7.84) | 13(50.98) | ||
| 43 | 146 | 2(13.70) | 1(6.85) | 1(6.85) | 1(6.85) | 5(34.23) | 2(13.70) | 7(47.95) | ||
| 44 | 87 | 2(22.99) | 2(22.99) | 1(11.49) | 3(34.48) | |||||
| 45 | 22 | 1(45.45) | 1(45.45) | 1(45.45) | ||||||
| 46 | 7 | |||||||||
| 47 | 1 | |||||||||
| Total | 5484 | 42(7.66) | 13(2.37) | 7(1.28) | < td>10(2.82)4(0.73) | 1(0.18) | 77(14.04) | 40(7.29) | 117 (21.33) | |
aNumber (and incidence per 1000).
aNumbers (and incidence per 1000).
For trisomy 21, the maternal age coefficient b was positive ( b = 0.245) and significantly different from zero (95% confidence interval (CI): 0.100-0.389, P = 0.0045), considering the standard error ( SE = 0.063), indicating that the incidence of trisomy 21 increases exponentially with increasing maternal age ( Fig. 1 , Table 2 ).
Table 2 Coefficients and standard errors of logistic regression calculation*
| Determinant item | Native language | |||||
|---|---|---|---|---|---|---|
| a | SE | b | SE | 95% 电影区間 | P値 | |
| Trisomy 21 | -7.605 | 2.485 | 0.245 | 0.063 | 0.100-0.389 | 0.0045 |
| +18, +13 | -9.661 | 3.464 | 0.284 | 0.085 | 0.075-0.493 | 0.0159 |
| Sex chromosomes | -10.087 | 4.196 | 0.290 | 0.105 | 0.020-0.560 | td>0.0400 |
| All aneuploidy** | -7.719 | 2.085 | 0.261 | 0.052 | 0.144-0.379 | 0.0007 |
95%CI: 95% confidence interval
*Model y = exp(a + bx) where x = maternal age at the time of amniocentesis, y = incidence of chromosomal abnormalities.
** Excludes mosaics and structural anomalies.
95%CI: 95% confidence interval
* Model y = exp(a + bx) where x = maternal age at time of amniocentesis and y = incidence of chromosomal abnormality.
** Excludes mosaicism and structural abnormalities.
For trisomy 18 and trisomy 13, an increased incidence in older mothers was observed for each trisomy, similar to that for trisomy 21, but the relationship was not statistically significant. When these two aneuploidies were calculated together as lethal autosomal aneuploidies, the incidence increased with advancing maternal age ( b = 0.284, SE = 0.085, 95% CI: 0.075-0.493, P = 0.0159) ( Figure 2 , Table 2 ).
Figure 1 Rate of trisomy 21 (per 1,000 fetuses) (vertical axis) and maternal age (horizontal axis)
Figure 2. Rate of lethal autosomal aneuploidies (trisomy 18 and trisomy 13) (per 1,000 fetuses) (vertical axis) and maternal age (horizontal axis)
For each sex chromosome aneuploidy, as for trisomy 21, a trend toward increased incidence in older mothers was observed, but the correlation was not statistically significant. When the three types of sex chromosome aneuploidy were calculated together as sex chromosome aneuploidy, the incidence significantly increased with increasing maternal age ( b = 0.290, SE = 0.105, 95% CI: 0.020-0.560, P = 0.040) ( Figure 3 , Table 2 ). For all aneuploidies, the maternal age coefficient b was also positive ( b = 0.261) and was significantly different from zero (95% CI: 0.144-0.379, P = 0.0007) considering the standard error (SE = 0.052) ( Figure 4 , Table 2 ). No non-mosaic trisomies other than the six listed were detected.
Figure 3. Rate of sex chromosome abnormalities (XXY, XXX, XYY) (per 1,000 fetuses) (vertical axis) and maternal age (horizontal axis)
Figure 4. Percentage of all aneuploidies excluding mosaicism and structural abnormalities (vertical axis) and maternal age (horizontal axis)
Table 3 Regression smoothed abnormality incidence/1000 at time of amniocentesis (by maternal age)
| Maternal age at amniocentesis | Regression smoothed expected incidence rate (per 1000 people) | |||
|---|---|---|---|---|
| 21 Toriソミー | +18, +13 | Sex chromosomes | Fully heteronumeric | |
| 35 | 2.62 | 1.32 | 1.07 | 4.12 |
| 36 | 3.37 | 1.76 | 1.42 | 5.35 |
| 37 | 4.31 | 2.33 | 1.90 | 6.94 |
| 38 | 5.50 | 3.10 | 2.54 | 9.02 |
| 39 | 7.03 | 4.12 | 3.40 | 11.70 |
| 40 | 8.98 | 5.47 | 4.54 | 15.20 |
| 41 | 11.47 | 7.26 | 6.07 | 19.73 |
| 42 | 14.66 | 9.65 | 8.11 | 25.61 |
| 43 | 18.73 | 12.82 | 10.84 | 33.25 |
| 44 | 23.93 | 17.03 | 14.48 | 43.16 |
| 45 | 30.57 | 22.62 | 19.36 | 56.04 |
Based on these regression analyses, we calculated the expected incidence of all aneuploidies except trisomy 21, lethal autosomal aneuploidies (trisomy 18 and trisomy 13), sex chromosome aneuploidies (47,XXY; 47,XXX; 47,XXYY), and mosaicism and structural abnormalities in mid-trimester amniocentesis in Japanese people ( Table 3 ). There were no significant differences between the coefficients obtained in this study and those obtained from previous studies in other countries (Polani et al. 1976; Schreinemachers et al. 1982; Ferguson-Smith and Yates 1984; Hook et al. 1984); for example, the equation for trisomy 21 y = exp(-14.7 + 0.262 x ) obtained from a North American study (Hook et al. 1984).
All other clinically significant cytogenetic diagnoses are shown in Table 4. The total number of structural abnormalities was 27 (1 unbalanced Robertsonian translocation, 10 balanced reciprocal translocations, 4 balanced Robertsonian translocations, and 12 inversions). There were 11 mosaic autosomal aneuploidies and 2 sex chromosome aneuploidies. In cases with mosaic autosomal trisomy, aneuploidies of chromosomes other than 21, 18, and 13 were diagnosed.
Table 4 Other chromosomal abnormalities by maternal age
| Maternal age at amniocentesis | Translocation | Reversal | Mosaicism | Total | |||
|---|---|---|---|---|---|---|---|
| Unbalanced | Balanced | Autosomal | 45, X/46, XY | ||||
| Robertson type | Mutual | Robertson type | |||||
| 35 | 1(1.36)a | 1(1.36) | |||||
| 36 | 1(1.18) | 1(1.18) | 1(1.18) | 1(1.18) | 4(4.71) | ||
| 37 | 3(3.64) | 1(1.21) | 4(4.85) | ||||
| 38 | 1(1.25) | 2(2.50) | 2(2.50) | 1(1.25) | 1(1.25) | 7(8.75) | |
| 39 | 1(1.40) | 5(7.01) | 2(2.81) | 8(11.22) | |||
| 40 | 2(3.19) | 2(3.19) | 1(1.60) | 4(6.39) | 1(1.60) | 10(15.97) | |
| 41 | 1(2.40) | 1(2.40) | |||||
| 42 | 1(3.92) | 1(3.92) | 2(7.84) | ||||
| 43 | 1(6.85) | 1(6.85) | 2(13.70) | ||||
| 44 | 1(11.49) | 1(11.49) | |||||
| Total | 1(0.18) | 10(1.82) | 4(0.73) | 12(2.19) | 11(2.01) | 2(0.36) | 40(7.29) |
aNumber (and incidence per 1000).
aNumbers (and incidence per 1000).
For other chromosomal abnormalities, including translocations, inversions, and mosaicism , no significant relationship was found between maternal age and incidence.
Investigation
This study clarified the incidence of chromosomal abnormalities at amniocentesis in Japan by maternal age.
These are the first data published in countries and regions outside of Europe and the United States. Regression analysis clearly showed that the incidence of trisomy 21, lethal autosomal abnormalities , and sex chromosome aneuploidies increased with increasing maternal age, as in previous reports (Polani et al. 1976, Schreinemachers et al. 1982, Ferguson-Smith and Yates 1984, Hook et al. 1984, Hassold et al. 1996). In addition, equations for the expected incidence of each chromosomal abnormality were obtained. These data will be useful for genetic counseling of Japanese women.
We calculated the parameters of the formula [ y = exp(a + bx )] ( Table 2 ) and the smoothed expected rates derived from the equation for trisomy 21, lethal autosomal trisomies (trisomy 18 and trisomy 13), sex chromosome aneuploidies, and all aneuploidies at each maternal age of 35 years or older ( Table 3 ). We compare the formula for trisomy 21 [ y = exp(-7.6 + 0.245 x )] obtained from the present study with the formula [ y = exp(-14.7 + 0.262 x )] obtained from European and North American studies (Hook et al. 1984). No significant differences in the parameters were found, suggesting that the prevalence of trisomy 21 is not influenced by either racial or geographic factors. Recent studies have shown that aneuploidies resulting from meiosis-II as well as meiosis-I are correlated with maternal age (Plachot et al. 2011 ). 1988; Macas et al. 1990, Dailey et al. 1996; Hassold et al. 1996). Based on these facts, it was suggested that nondisjunction may be caused by factors that are only induced by aging and may be a universal phenomenon in human reproduction.
Because cytogenetic abnormalities diagnosed prenatally by amniocentesis show a higher rate of spontaneous abortion and/or stillbirth than cytogenetically normal fetuses (Hook 1983; Hook et al. 1988, 1989; Halliday et al. 1995; Macintosh et al. 1995), the incidence of prenatal testing differs significantly from the incidence at birth. Compared with previous reports on non-mosaic aneuploidies diagnosed at birth (Hook et al. 1984; Mutton et al. 1991), the abnormal karyotypes and their sequences revealed in the study were almost the same, with some exceptions, but the incidence of each non-mosaic aneuploidy was significantly higher, for example, the incidence of trisomy 21 in Japan was 1.42/1000 births (Maeda et al. 1991). These high incidences are caused by selective abortions occurring in the middle or late stages of pregnancy, and fetuses with aneuploidies may be stillborn. During counseling, risk estimates derived from prenatal testing data should be discussed, and counselors should provide estimates of the contribution of chromosomal abnormalities to natural mortality.
In all non-mosaic aneuploids diagnosed by mid-trimester amniocentesis, only six trisomies were detected. Although monosomy X and trisomy 16 were detected as aneuploidies in chorionic villus sampling (CVS) (Hook et al. 1988) or earlier in amniocentesis (Lockwood and Neu 1993), no such aneuploidies were detected in this study. Because cytogenetic CVS and early amniocentesis are usually performed in the first trimester (CVS at 9–11 weeks of gestation; early amniocentesis at 11–14 weeks of gestation), most fetuses with aneuploidies may have spontaneously aborted before amniocentesis in the second trimester. Considering additional cases, such aneuploidies may be detected by mid-trimester amniocentesis as exceptional cases.
The proportion of pregnant women aged 35 years or older has been gradually increasing and currently accounts for more than 10% of pregnant women in Japan (Maternal and Child Health Organization 1996). These women receive information about the high risk of chromosomal abnormalities due to advanced age from the mass media and medical sources, and visit obstetric clinics to receive genetic counseling or to request prenatal testing . If an older pregnant woman does not have information about the risk, it should be explained to her obstetrician. During genetic counseling for women at risk because of advanced age, the obstetrician or other medical professional should present the estimated possibility of chromosomal abnormalities and allow the patient to choose whether or not she would like to undergo prenatal testing based on this possibility . It is considered that this possibility should be derived from data at the time of the study, such as the predicted rate revealed in this study.
This study provides useful data for counseling Japanese women. However, more cases should be investigated to more accurately predict the rate of each chromosomal abnormality in mid-trimester amniocentesis.
References
- Dailey T, Dale B, Cohen J, Munne S (1996) Association of maternal age with nondisjunction in meiosis-II human oocytes. Am J Hum Genet 59: 176-184
- Ferguson-Smith MA, Yates JRW (1984) Maternal age-specific rates of chromosomal abnormalities and factors influencing them: report of a European collaborative study of 52,965 amniocentesis. Trials 4: 5-44
- Halliday JL, Watson LF, Lumley J, Danks DM, Sheffield LJ (1995) New estimates of risk of Down syndrome at birth following chorionic villus sampling, amniocentesis, and birth in women of uniquely defined maternal advanced age. Pregnancy 15: 455-465
- Hassold T, Abruzzo M, Adkins K, Griffin D, Merrill M, Millie E, Sakar D, Shen J, Zaragoza M (1996) Incidence, origin, and pathogenesis of human aneuploidies. Environ Mol Mutagen 28: 167-175.
- Fook EB (1983) Chromosomal abnormalities and spontaneous fetal death after amniocentesis: further data and correlation with maternal age. Am J Hum Genet 35: 110-116.
- Hook EB, Cross PK, Regal RR (1984) Frequency of 47,121, 47,118, and 47,113 at the upper end of maternal age: results on 56,094 fetuses compared with prenatal and birth data. Hum Genet 68: 211-220
- Hooke EB, Cross PK, Jackson L, Pergament E, Brambati B (1988) Maternal age-specific rates of 47,121 and other cytogenetic abnormalities diagnosed early in pregnancy in chorionic villus biopsy specimens: comparison with rates expected from observations at amniocentesis. Am J Hum Genet 42: 797-807
- Hook EB, Topol BB, Cross PK (1989) The natural history of cytogenetically abnormal fetuses detected by mid-trimester amniocentesis: new data and estimates of excess and relative risks of late fetal death associated with 47,121 and other abnormal karyotypes. Am J Hum Genet 45: 855-861
- Lockwood DH, Neu RL (1993) Cytogenetic analysis of 1375 amniotic fluid samples from pregnancies less than 14 weeks gestation. Pregnancy Diagnostics 13: 801-805
- Macas E, Floersheim Y, Hotz E, Imthurn B, Keller PJ, Walt H (1990) Aberrant chromosomal arrangements in human oocytes. Hum Reploid 5: 703-707
- Macintosh MCM, Wald NJ, Chard T, Hansen J, Mikkelsen M, Therkelsen AJ, Petersen GB, Lundsteen C (1995) Elective abortion of Down syndrome fetuses in women aged 35 years or older. Br J Obstet Gynaecol 102: 798-801
- Maeda, Ohno, Matsuno, A, Yoshihara, and Yabe, N (1991) Cytogenetic study of 14,835 consecutively born individuals. Jpn J Human Genes 36: 117-129
- Maternal and Child Health Organization (1996) Number of live births by maternal age. In: Maternal and Child Health Statistics of Japan 1996 (ed.), Tokyo, pp. 47-53.
Mutton DE, Alberman E, Ide R, Bobrow M (1991) Results of the first year (1989) of the national registry of Down’s syndrome in England and Wales. Br Med J 303: 1295-1297. - NICHD National Registry for Amniocentissis Study Group (1976) Second trimester amniocentesis for prenatal diagnosis: safety and accuracy. Japan Automobile Manufacturers Association 236: 1471-1476
- Plachot M, de Grouchy J, Junca AM, Mandelbaum J, Salat-Baroux J, Cohen J (1988) Chromosome analysis of human oocytes and embryos: does delayed fertilization increase chromosomal imbalances? Human Reproduction 3: 125-127
- Polani PE, Alberman E, Berry AC, Blunt S, Singer JD (1976) Chromosomal abnormalities and maternal age. Lancet II: 516-517
- Schreinemachers DM, Cross PK, Hook EB (1982) Rates of trisomies 21, 18, 13, and other chromosomal abnormalities in approximately 20,000 prenatal studies compared with estimated birth rates. Hum Genet 61: 318-324
関連記事
-
NEW
-
NEW
-
NEW
-
NEW
-
NEW
-
NEW
