YH Lam, MHY Tang

Introduction

The prenatal diagnostic and counselling service for Hong Kong was established at Tsan Yuk Hospital in 1981. In the past 16 years, we accepted referrals from other government-based units and private doctors. The number of referrals has grown from 500 to over 3,000 per year. The most common indications for referral are advanced maternal age for prenatal diagnosis of Down syndrome, fetal abnormalities detected on routine ultrasound examination and couples who are thalassaemia carriers (Table I). The protocol for prenatal screening for Down's syndrome and the prenatal diagnosis of homozygous α-thalassaemia-1 have changed dramatically in the past few years. In this review, we like to describe the basis for these changes.

Prenatal Screening for Fetal Down Syndrome

Down syndrome is the commonest inborn cause of mental retardation. Its birth incidence is around 1:1000.¹ Prenatal screening and diagnosis of the condition in early pregnancy allow the parents the option of pregnancy termination. This provides a means of prevention. The incidence of Down syndrome birth increases with advancing maternal age. Pregnant women are often screened by their age and those at or above 35 years old at the expected date of confinement are offered amniocentesis for diagnosis. With this approach, around 5-10% of the pregnant population will be subjected to amniocentesis and only 30-40% of the Down syndrome pregnancies will be detected. Amniocentesis is an invasive procedure that is associated with a 0.5% risk of miscarriage. Every effort should be taken to minimize the number of amniocentesis done in order to detect an affected pregnancy.

In the 1980's, maternal serum human chorionic gonadotrophin (hCG) level was found to be elevated, and serum alpha-fetoprotein (AFP) and estriol levels were depressed in women carrying a Down's syndrome pregnancy at 15-20 weeks of gestation. Although these maternal marker levels overlap extensively between normal and affected pregnancies, the markers in combination with maternal age serve to identify a group of high risk women for amniocentesis. Large prospective studies including more than 80,000 women in the Caucasian population showed that serum markers and maternal age identify over 60% of the Down syndrome pregnancies at a false positive rate of 5%.^2-6 This is a great improvement compared to screening with maternal age alone. In many Western countries, serum screening has become the established method to screen for fetal Down syndrome.

In 1994, we established a similar screening programme at Tsan Yuk Hospital, later extending it to Kwong Wah Hospital.⁷ Because the value of estriol amongst the other serum markers is controversial and because of the additional cost involved, we had only incorporated hCG and AFP into our screening programme. Women less than 35 years old were offered serum screening for fetal Down syndrome. The uptake of screening was 75%. We prospectively measured their serum AFP and hCG levels between 15 to 20 weeks of gestation. They were assigned a risk of having a Down syndrome pregnancy taking into account their age and the AFP and hCG levels. All those at high risk (>= 1:250) were offered amniocentesis. Between June 1994 and December 1996, 9177 women were screened. 183 women were screened positive and were offered amniocentesis and eight of these pregnancies were affected by Down syndrome. Six Down's syndrome pregnancies were missed by the screening programme. The detection rate and false positive rate were 57% and 2%, respectively.⁷

In order to assess the value of serum screening in women aged >= 35 we obtained blood samples from women between 18 to 20 weeks of gestation just prior to amniocentesis. Between January 94 to May 95, 1784 samples were obtained and frozen. These samples were retrieved retrospectively and assayed for AFP and hCG. The individual risk for Down syndrome was calculated as mentioned previously. Using a risk cut-off of 1:250, serum AFP and hCG and maternal age detected 23 of the 31 Down's syndrome cases in this group of 'old' women. The detection rate and false positive rate were 71% and 16%, respectively.

The results of both the prospective serum screening trial on young women and the retrospective screening trial on old women agreed well with data on the Caucasian population.^8-10 These data suggested that less amniocentesis need to be performed to detect a case of Down syndrome pregnancy. Fewer women carrying a normal pregnancy need to be subjected to the risk of amniocentesis. Thus, we have modified our policy of prenatal screening of Down's syndrome at Tsan Yuk Hospital since 1997. Women of all ages are offered second-trimester serum AFP and hCG screening. Because of the traditional belief of advanced age being a high risk factor, we offer women at or above 35 years old the option of undergoing amniocentesis directly without serum screening.

Prenatal diagnosis of homozygous α-thalassaemia-1

In Hong Kong, 5% of the population are ct-thalassaemia carriers.¹¹ Couples with heterozygous α-thalassaemia-1 face a 25% risk of having a fetus with homozygous α-thalassaemia-1. Affected fetuses are severely anaemic and hydropic and they usually die soon after birth. Mothers carrying an affected pregnancy have 60% chance of developing life threatening obstetric complications such as pre-eclampsia and postpartum haemorrhage.¹² Prenatal diagnosis of homozygous α-thalassaemia-1 is indicated. This is conventionally performed by DNA study of cells obtained from chorionic villus sampling¹³ or amniocentesis.¹⁴ These procedures are associated with a small but not negligible risk to the fetus.¹⁵ In the past 10 years, we have studied alternate means of prenatal diagnosis of the condition.

Because of severe anaemia and hypoxia, fetuses affected by homozygous α-thalassaemia-1 develop hydropic changes such as subcutaneous edema, pleural and pericardial effusion and ascites. These signs were previously reported to be only apparent after 20 weeks of pregnancy.^16-18 In a recent study of around 100 at-risk pregnancies in the early second-trimester, we found that ultrasonographic features of hydrops fetalis were detected in 7% at 12-14 weeks and 33% at 17-18 weeks.¹⁹ The at-risk pregnancies can be examined serially by ultrasound scan to detect these changes. However, this prenatal diagnostic approach does not allow an early diagnosis in majority of the cases. Other subtle signs of fetal anaemia such as cardiac and placental enlargement may be more useful for prenatal diagnosis in the early stage of pregnancy.

We have shown previously that the placental volume is increased in pregnancies affected by homozygous α-thalassaemia-1 after 20 weeks of gestation.¹⁶ In a recent prospective study of 231 at-risk pregnancies, we found a similar increase in placental thickness in affected pregnancies between 10-21 weeks.²⁰ Using a cut-off of mean plus 25D placental thickness, ultrasound assessment was 88% sensitive and 96% specific of disease. Its predictive value was further improved by ultrasound measurement of the fetal heart (Fig. 1 and 2). We compared the use of ultrasound measurement of the fetal cardiothoracic ratio and placental thickness at 13 to 18 weeks in another study of 98 at-risk pregnancies.²¹ The cardiothoracic ratio predicted disease better than the placental thickness. Using a cardiothoracic ratio cut-off of >= 0.5, 75% of affected pregnancies were detected at 13-14 weeks and all cases were detected at 17-18 weeks, with false positive rates of 7 and 8%, respectively. In our experience, a normal placental thickness and cardiothoracic ratio at or after 17 weeks is reassuring, and invasive test to confirm normality is not necessary Pregnancies with ultrasound findings of placental thickening (> mean + 2SD) and cardiothoracic ratio > 0.5 at or beyond 12 weeks are highly suggestive of disease.²² We prefer to perform cordocentesis and haemoglobin study²² to confirm the diagnosis in these cases because the test is cheaper and takes only 1 day, in contrast to DNA Southern blot studies which take 8-10 days. In this manner, the unaffected pregnancies need not be subjected to the risk of invasive testing, without delaying the diagnosis of the affected pregnancies. Fewer fetuses will be lost unnecessarily. Hence from 1996, women referred for the prenatal diagnosis of homozygous α-thalassaemia- 1 are offered the options of undergoing choronic villus sampling at 10-12 weeks, or amniocentesis at 18 weeks, or have serial ultrasound examinations at 12-13 weeks, 16-18 weeks and 30 weeks to detect the signs of fetal anaemia (cardiomegaly, placentomegaly or hydropic changes), followed by cordocentesis only in those cases with abnormal findings.

Fig. 1 Ultrasound measurement of the fetal heart at the plane of cardiac 4-chamber in a normal fetus at 12 weeks, CT ratio = cardiothoracic ratio.

Fig. 2 Ultrasound measurement of the fetal heart in a detus affected by homozygous α-thalassaemia-1 at 12 weeks.

Conclusion

Prenatal screening and diagnosis of Down syndrome and thalassaemias account for most of the workload of prenatal diagnosis in Hong Kong. We have refined the existing protocols for screening of fetal Down's syndrome and the prenatal diagnosis of homozygous α-thalassaemia-1 according to the available evidence. This will reduce the number of unaffected pregnancies subjected to the Prenatal Diagnosis and Screening risk of invasive testing to detect a case of affected pregnancy. We hope that the programme will become more cost-effective in the long run.

References

1. Ghosh A, Woo JS, Chan CL, et al. Down syndrome and maternal age in Hong Kong Chinese. Asia Oceania J Obstet Gynaecol 1985;11:93-8.

2. Haddow JE, Palomaki GE. Knight GJ, et al. Prenatal screening for Down's syndrome with use of maternal serum markers. N Engl J Med 1992;327:588-93.

3. Phillips OP, Elias S, Shulman LP, Anderson RN, Morgan CD, Simpson JL. Maternal serum screening for fetal Down's syndrome in women less than 35 years of age using alpha-fetoprotein, hCG and unconjugated estriol: a prospective 2-year study. Obstet Gynecol 1992;80:353-8.

4. Wald NJ, Kennard A, Densem JW, Cockle HS, Chard T, Butler L. Antenatal maternal serum screening for Down's syndrome: result of a demonstration project. Br Med J 1992;305:391-4.

5. Wenstrom KD, Williamson RA, Grant SS, Hudson JD, Getchell JP. Evaluation of multiple-marker screening for Down's syndrome in a statewide population. Am J Obstet Gynecol 1993;169:793-7.

6. Goodbum SF, Yates JRW, Raggatt PR, et al. Second trimester maternal serum screening using alpha-fetoprotein, human chorionic gonadotrophin and unconjugated oestriol: experience of a regional programme. Prenat Diagn 1994;14:391-402.

7. Lam YH, Ghosh A, Tang MHY, et al. Second-trimester maternal serum alpha-fetoprotein and human chorionic gonadotrophin screening for Down's syndrome in Hong Kong. Prenat Diagn 1998. In press.

8. Mooney RA, Peterson J, French CA, Saller DN, Aryan DA. Effectiveness of combining maternal serum alpha-fetoprotein and hCG in a second-trimester screening program for Down syndrome. Obstet Gynecol 1994;84:298-303.

9. Crossly JA, Aitken DA, Berry E, Connor JM. Impact of a regional screening programme using maternal serum a-fetoprotein (AFP) and human chorionic gonadotrophin (hCG) on the birth incidence of Down's syndrome in the west of Scotland. J Med Screen 1994;1:180-3.

10. Wenstrom KD, Desai R, Owen J, Dubard MB, Boots L. Comparison of multiple marker screening with amniocentesis for the detection of fetal aneuploidy in women >= 35 years old. Am J Obstet Gynecol 1995;173:1287-92.

11. Lau YL, Chan LC, Chan YYA, et al. Prevalence and genotypes of α- and β- thalassaemia carriers in Hong Kong - implications for population screening. N Engl J Med 1997;336:1298-301.

12. Liang ST, Wong VCW, So WWK, Ma HK, Chan V, Todd D. Homozygous a-thalassaemia: clinical presentation, diagnosis and management. A review of 46 cases. Br J Obstet Gynaecol 1985;92:680-4.

13. Rubin EM, Kan YW. A simple sensitive prenatal test for hydrops fetalis caused by homozygous α-thalassaemia. Lancet 1985;i:75-7.

14. Chan V, Ghosh A, Chan TK, Wong V, Todd D. Prenatal diagnosis of homozygous a-thalassaemia by direct DNA analysis of uncultured amniotic-fluid cells. Br Med J 1984;288:1327-9.

15. Canadian Collaborative CVS-Amniocentesis Clinical Trial Group. Multicentre randomised clinical trial of chorionic villus sampling and amniocentesis. Lancet 1989;i:1-6.

16. Ghosh A, Tang MHY, Liang ST, Ma HK. Ultrasound evaluation of pregnancies at risk for homozygous a-thalassaemia-1. Prenat Diagn 1987;7:307-13.

17. Tan SL, Tseng P, Thong PW. Bart's hydrops fetalis - clinical presentation and management - an analysis of 25 cases. Aust NZ J Qbstet Gynaecol 1989;29:233-7.

18. Tongsong T, Wanapirak C, Srisomboon J, Piyamongkol W, Sirichotiyakul S. Antenatal sonographic features of 100 alphathalassaemia hydrops fetalis fetuses. J Clin Ultrasound 1996;24:73-7.

19. Lam YH, Ghosh A, Tang MHY, Lee CP, Sin SY. Second-trimester hydrops fetalis in pregnancies affected by homozygous α-thalassaemia-1. Prenat Diagn 1997;17:267-9.

20. Ghosh A, Tang MHY, Lam YH, Fung E, Chan V. Ultrasound measurement of placental thickness to detect pregnancies affected by homozygous a-thalassaemia-1. Lancet 1994;344:988-9.

21. Lam YH, Ghosh A, Tang MHY, Lee CP, Sin SY. Early ultrasound prediction of pregnancies affected by homozygous α-thalassaemia-l. Prenat Diagn 1997;17:327-32.

22. Lam YH, Tang MHY. Prenatal diagnosis of haemoglobin Bart's disease by cordocentesis at 12-14 weeks' gestation. Prenat Diagn 1997; 17:501-4.