| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From theTaiwan Down Syndrome Screening Group: the Department of Obstetrics and Gynecology, Taipei Chang Gung Memorial Hospital, Taipei; Lin-Kou Chang Gung Memorial Hospital, Tau-Young; Kashung Chang Gung Memorial Hospital, Kashung; and National Taiwan University Hospital, Taipei, Taiwan; and the Endocrine Unit, Department of Clinical Biochemistry, Harold Wood Hospital, Romford, Essex, United Kingdom.
Address reprint requests to: Jenn J. Hsu, MD, Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital, 199, Tung-Hwa North Road, Taipei, Taiwan, E-mail: jjhsu{at}ms6.hinet.net
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: Free ß-hCG and total E3 were measured in the urine samples of 28 Taiwanese Down syndrome pregnancies and 268 unaffected singleton pregnancies at 14–25 weeks. Results were normalized to urine creatinine concentrations and converted to multiples of the median (MoM) levels. Gestational ages were estimated by ultrasound measurements.
Results: Median values of free ß-hCG, total E3, free ß-hCG to total E3 ratio, and the free ß-hCG to total E3 MoM ratio in Down syndrome pregnancies were 4.75 MoM, 0.66 MoM, 8.99 MoM, and 9.51, respectively. At a 5% false-positive rate, the observed detection rates were 36% (ten of 28) with total E3, 71% (20 of 28) with free ß-hCG, 68% (19 of 28) with free ß-hCG/total E3, and 71% (20 of 28) with free ß-hCG/total E3 MoM. When combined with maternal age, the expected detection rates were 65% with total E3, 71% with free ß-hCG, 76% with free ß-hCG/total E3, 80% with free ß-hCG/total E3 MoM, and 89% when combining free ß-hCG, total E3, and maternal age.
Conclusion: Urine free ß-hCG and total E3 are useful markers for Down syndrome screening during the second trimester in Taiwanese women.
Second-trimester Down syndrome screening using multiple serum markers has become a routine antenatal test.1–3 Although screening tests using alpha-feto-protein (AFP), hCG, and unconjugated estriol (E3) are widespread in the United States,3 a screening strategy combining free ß-hCG and AFP with maternal age is well established in either Asians1,4,5 or whites.2,6 This simple two-analyte approach has achieved a 68–75% detection rate with a 5– 6% false-positive rate.1,2,6
Most current protocols for Down syndrome screening are based on measuring the concentrations of biochemical markers in maternal serum. However, mass screening based on urine testing has mostly been neglected. ß-core, a major metabolite of hCG in the urine of pregnant women, has been reported as a potential marker for Down syndrome screening.7–11 However, free ß-hCG can also be detected in maternal urine because of its small molecular weight and high renal clearance rate.12 A few studies have indicated that free ß-hCG could be another effective urine marker for the detection of fetal Down syndrome.13–15
Low levels of total E3 were found in the urine of pregnant women whose infants had Down syndrome.16 A screening strategy using total E3 and ß-core has been proposed recently, using either the two analytes in a bivariate algorithm or their ratios in a univariate algorithm, and is reported to be of great value in improving detection.9,10 The Taiwanese population is 98% ethnically Chinese and 2% Polynesian in origin. Because of racial differences between Asians and whites, it is unknown whether these urine screening strategies are equally applicable in Taiwanese. Therefore, the aim of this study was to investigate whether second-trimester free ß-hCG and total E3 in the maternal urine are useful markers for Down syndrome screening in a Taiwanese population.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Urine samples were centrifuged and stored at -40 C before assay. The mean storage times in affected and unaffected pregnancies were 54.8 and 50.7 weeks, respectively. Free ß-hCG was measured in duplicate with an immunoradiometric assay (CIS Ltd., High Wy-combe, Bucks, UK). All samples were diluted 1:5 with zero diluent before analysis. Total E3 was measured in duplicate using the Ortho-Clinical Diagnostics Estriol (total) II radioimmunoassay kit (Ortho-Clinical Diagnostics, Amersham, UK). All samples were diluted 1:50 with normal male serum before analysis. Immunoassay results were normalized to urine creatinine concentrations, which were measured according to the Jaffe principle after a 1:20 urine dilution with 0.9% normal saline. Optimally regressed weekly medians for free ß-hCG, total E3, and their ratios were calculated by weighted nonlinear regression from the observed weekly medians. Assay results were converted to multiples of the median (MoM) levels for the appropriate gestation.
Standard statistical modeling methods were used to assess the performance of each analyte, their ratios, and the analytes in combination.18 Population statistics of both the affected and unaffected populations were used to calculate the likelihood ratios using single analytes or their ratios in a univariate algorithm, and two analytes in a bivariate algorithm.19 The expected detection rate was calculated at a given false-positive rate from the log-Gaussian model using the maternal age distribution of an affected population by Cuckle et al20 and a normal population of ours.21 Statistical analyses were done with the statistical software package SPSS Graduate Pack 8.0 for Windows 95 (SPSS Inc., Chicago, IL). We used Kolmogorov-Smirnov tests to assess the normal distributions of the analytes and Pearson correlation coefficients to assess the correlations among these various indices. Receiver operating characteristic (ROC) curves for each analyte, their ratios, and combinations were established by plotting the detection rate versus the respective false-positive rate. P < .05 was considered statistically significant.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
There were also no significant correlations between free ß-hCG/total E3 or free ß-hCG/total E3 MoM and maternal age in Down syndrome or unaffected pregnancies (correlation coefficients were 0.08, 0.11, 0.09, and -0.04, respectively). The distributions of free ß-hCG/total E3 and free ß-hCG/total E3 MoM after logarithmic transformation also showed Gaussian distributions in both affected and unaffected pregnancies (P = .535, P = .574; and P = .916, P = .939; respectively). The median values of free ß-hCG/total E3 and free ß-hCG/total E3 MoM in affected pregnancies were significantly higher than those in unaffected pregnancies (9.51 versus 1.01 MoM and 8.99 versus 1.15; P < .001 in both cases).
Figure 1
illustrates the free ß-hCG and total E3 MoM values in the urine of all 28 women with Down syndrome pregnancies and the 268 with unaffected pregnancies. The distribution in the upper left quadrant indicates that high levels of free ß-hCG and low levels of total E3 are strongly associated with Down syndrome pregnancies in Taiwanese women. When the observed detection rates of each analyte and their ratios were examined at a fixed 5% false-positive rate, total E3 would identify 36% (ten of 28; 95% confidence interval [CI] 18, 54) of affected pregnancies (0.43 MoM cutoff), free ß-hCG 71% (20 of 28; 95% CI 55, 88) (3.38 MoM cutoff), free ß-hCG/total E3 68% (19 of 28; 95% CI 51, 85) (6.26 MoM cutoff), and free ß-hCG/total E3 MoM 71% (20 of 28; 95% CI 55, 88) (4.93 MoM cutoff).
|
depicts the ROC curves for each analyte and their ratios in a univariate algorithm and two analytes in a bivariate algorithm with maternal age. When the observed statistics for each analyte and their ratios were used, the expected detection rate in the mathematical model at a 5% false-positive rate rose from 65% with total E3, to 71% with free ß-hCG, 76% with free ß-hCG/total E3, 80% with free ß-hCG/total E3 MoM, and 89% when combining free ß-hCG, total E3, and maternal age.
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
As in previous reports,9,10,17 we found a reduced total E3 level in the urine of Asian women with Down syndrome pregnancies. Our total E3 MoM value of affected pregnancies (0.66) agrees with those of Kellner et al10 (0.64) and Hsu et al17 (0.65) but is higher than that of Cole et al9 (0.33). In our study, 36% of affected cases had total E3 at or lower than the fifth percentile, slightly better than the results of Kellner et al10 (22%) and Hsu et al17 (35%) but lower than that of Cole et al9 (42%).
Previous reports have shown that urine ß-core and total E3 combined with maternal age achieved a detection rate of 79–80%.9,10 In our study, when combining urine free ß-hCG, total E3, and maternal age, the detection rate (89%) increased compared with free ß-hCG and age. This result is better than the finding of Kellner et al10 (6% increase), who compared multivariate analysis of ß-core, total E3, and maternal age with ß-core and age.
Recent studies have proposed using the raw ß-core and total E3 concentrations without normalizing to creatinine as a simple ratio, then using this ratio as if it were an individual analyte.9,10 In our study, the free ß-hCG to total E3 ratio increased detection efficacy more than free ß-hCG or total E3 alone. The detection rate of the ratio method combined with maternal age was also higher than that of free ß-hCG and age, but lower than that of two analytes and age at a fixed false-positive rate. This observation is therefore contrary to the results of Cole et al9 and Kellner et al,10 who found that detection rates were comparable between the ratio method and the bivariate method.
There is no clear advantage in using serum hCG/AFP MoM.1,4 In our study, the detection rate using free ß-hCG/total E3 MoM was better than that using total E3 and free ß-hCG/total E3, but the same as that of free ß-hCG. When combined with maternal age, the detection rate of this MoM ratio was higher than that of the ratio method but lower than that of the bivariate method.
Several studies have indicated that urine screening could replace serum screening because of logistical advantages and increased sensitivity.7–9 Ours and previous studies have confirmed that urine free ß-hCG could be a useful marker in Asians14 and whites.11,13,15 The addition of total E3 with free ß-hCG and maternal age increased the performance of urine screening with either the ratio method or the bivariate method. Our study offers initial proof of the feasibility of Down syndrome screening using maternal urine in an Asian population. However, a large-scale collaborative study is needed for further evaluation.
|
Footnotes |
|---|
Received September 30, 1998. Received in revised form December 3, 1998. Accepted December 30, 1998.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Macri JN, Anderson RW, Krantz DA, Larsen JW, Buchanan PD. Prenatal maternal dried blood screening with 
-fetoprotein and free ß-human chorionic gonadotropin for open neural tube defect and Down syndrome. Am J Obstet Gynecol 1996;174:566–72.[Medline]
3. Palomaki GE, Knight GJ, McCarthy JE, Haddow JE, Donhowe JM. Maternal serum screening for fetal Down syndrome in the United States: A 1995 survey. Am J Obstet Gynecol 1997;176:1046–51.[Medline]
4. Hsu JJ, Ou YC, Chen KC, Hsieh TT, Soong YK. High maternal serum free beta-hCG levels in Down syndrome pregnancies: A preliminary report. Chang Keng I Hsueh 1996;19:36–41.
5. Hsu JJ, Hsieh TT, Hsieh FJ. Down syndrome screening in an Asian population using alpha-fetoprotein and free ß-hCG: A report of the Taiwan Down syndrome Screening Group. Obstet Gynecol 1996; 87:943–7.[Abstract]
6. Spencer K, Coombes EJ, Mallard AS, Milford Ward A. Free beta human choriogonadotropin in Down’s syndrome screening: A multicentre study of its role compared with other biochemical markers. Ann Clin Biochem 1992;29:506–18.
7. Cuckle HS, Iles RK, Chard T. Urinary ß-core human chorionic gonadotrophin: A new approach to Down’s syndrome screening. Prenat Diagn 1994;14:953–8.[Medline]
8. Canick JA, Kellner LIE, Sailer DN, Palomaki GE, Walker RP, Osanthanondh R. Second trimester levels of maternal urinary gonadotropin peptide in Down syndrome pregnancy. Prenat Diagn 1995;15:739–44.[Medline]
9. Cole LA, Acuna E, Isozaki T, Palomaki GE, Bahado-Singh RO, Mahoney MO. Combining ß-core fragment and total oestriol measurements to test for Down syndrome pregnancies. Prenat Diagn 1997;17:1125–33.[Medline]
10. Kellner LH, Canick JA, Palomaki GE, Neveux LM, Saller DN, Walker RP, et al. Levels of urinary beta-core fragment, total oestriol, and the ratio of the two in second-trimester screening for Down syndrome. Prenat Diagn 1997;17:1135–41.[Medline]
11. Hallahan TW, Krantz DA, Tului L, Alberti E, Buchanan PD, Orlandi F, et al. Comparison of urinary free beta (hCG) and beta-core (hCG) in prenatal screening for chromosomal abnormalities. Prenat Diagn 1998;18:893–900.[Medline]
12. Wehmann RE, Nisula BC. Metabolic clearance rates of the subunits of human chorionic gonadotropin in man. J Clin Endocrinol Metab 1979;48:753–9.[Abstract]
13. Spencer K, Aitken DA, Macri JN, Buchanan PD. Urine free beta hCG and beta core in pregnancies affected by Down’s syndrome. Prenat Diagn 1996;16:605–13.[Medline]
14. Hayashi M, Kozu H, Takei H. Maternal urinary free ß-subunit of human chorionic gonadotrophin: Creatinine ratios and fetal chromosomal abnormalities in the second trimester of pregnancy. Br J Obstet Gynaecol 1996;103:577–80.[Medline]
15. Cole LA, Jacobs M, Isozaki T, Palomaki GE, Bahado-Singh RO, Mahoney MJ. Screening for Down syndrome using urine hCG free beta-subunit in the second trimester of pregnancy. Prenat Diagn 1997;17:1107–11.[Medline]
16. Cuckle HS, Iles RK, Sehmi IK, Oakey RE, Davies S, Ind T. Urinary multiple marker screening for Down’s syndrome. Prenat Diagn 1995;15:745–51.[Medline]
17. Hsu JJ, Spencer K, Aitken DA, Crossley J, Choi T, Ozaki M, et al. Urinary free beta hCG, beta-core fragment and total oestriol as markers of Down syndrome in the second trimester of pregnancy. Prenat Diagn (in press).
18. Royston P, Thompson SG. Model based screening by risk with application to Down’s syndrome. Stat Med 1992;11:257–68.[Medline]
19. Reynolds T, Penney M. The mathematical basis of multivariate risk screening: With special reference to screening for Down’s syndrome associated pregnancy. Ann Clin Biochem 1990;27:452–8.
20. Cuckle HS, Wald NJ, Thompson SG. Estimating a woman’s risk of having a pregnancy associated with Down’s syndrome using her age and serum alpha-fetoprotein level. Br J Obstet Gynaecol 1987;94:387–402.[Medline]
21. Sheu BC, Shyu MK, Lee CN, Kuo BJ, Tseng YY, Hsieh FJ. Maternal age-specific risk of Down syndrome in an Asian population: A report of the Taiwan Down Syndrome Screening Group. Prenat Diagn 1998;18:675–82.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
