New Triple Screen Test for Down Syndrome: Combined U rine Analytes and Serum AFP Ray O. Bahado-Singh, MD, 1 * Utku Oz, MD, 1 Ertug Kovanci, MD, 1 Dilek Cermik, MD, 1 Divinia Flores, MD, 1 Joshua Copel, MD, 1 Maurice Mahoney, MD 1,2 and Laurence Cole, MD 1 1 Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, Connecticut 2 Department of Genetics, Yale University School of Medicine, New Haven, Connecticut Abstract In this study we report a new triple test that combines serum AFP, urine -core fragment of hCG, total urine estriol, and maternal age for calculating individual Down syndrome odds in the second trimester. The urine -core fragment/estriol ratio was used as a single screening variable. Analyte levels were measured prospectively in 10 Down syndrome cases and 346 normals. Individual Down syndrome odds were calculated by multiplying the product of the Down syndrome likelihood ratios of serum AFP and urine -core/estriol levels by the age-related midtrimester risk. The screening efficiency of an algorithm that combines urine -core/estriol with maternal age was compared to one that included serum AFP data. A 90% detection rate for Down syndrome was obtained at a 4.65% false positive rate. This was superior to the 75% sensitivity at 5% false positive rate observed when -core/estriol and age alone were used. This newtriple test has a higher screening efficiency than that generally reported for the traditional serum triple screen and other urine tests, and it also provides information on the risk of neural tube defects. If confirmed in larger trials, the new algorithm could be used as an alternative to the traditional serum triple screen. J. Matern.–FetalMed. 7:111–114, 1998.  1998 Wiley-Liss, Inc. Key words: urine -core fragment; estriol; AFP; Down syndrome IN TRO DUCTIO N Recent studies report that a single maternal urinary analyte, -core fragment of hCG, has a Down syndrome screening efficiency equal to or superior to the traditional maternal serum triple screen test [1–4]. Sensitivity rates were noted to be 62%, at a 5% false positive rate. In Down syndrome pregnancies, maternal urine -core levels are elevated, whereas total estriol is reduced. The -core/estriol ratio has a high detection rate [3,5]. The latter study noted a 75% Down syndrome detection at a 0.5% false positive rate using this ratio analyte. These results appear to represent a substantial improvement over the traditional serum triple analyte screen. A major limitation of urine screening algorithms is the lack of information on the fetal risk of neural tube defects. Although AFP has the lowest Down syndrome detection rate of the three common serum analytes [6], it cannot be excluded from current screening protocols because of its unique ability to detect open neural tube defects. The purpose of our study was to determine the Down syndrome screening efficiency of a new protocol that combines maternal serum AFP and urine -core fragment of hCG and total urine estriol. SUBJECTS AN D M ETHO DS Serum AFP and a urine sample (for -core fragment and total estriol levels) were collected prospectively prior to amniocentesis. The method of specimen handling and measurement of the urine analytes has been previously described [4,5]. This was done as part of an ongoing prospective study to determine the Down syndrome screen- ing efficiency of maternal urine analytes in midtrimester pregnancies between 14–24 weeks. The dating was based on the last menstrual period (LMP), or an ultrasound. If there was a greater than 5 days difference between the LMP dates and the first trimester scan, or greater than 10 days *Correspondence to: Ray O. Bahado-Singh, M.D., Division of Mater- nal Fetal Medicine, Department of Obstetrics and Gynecology, Yale University School of Medicine, 333 Cedar Street, P .O. Box 208063, New Haven CT06520-8063. Received 20 October 1997; Revised 11 December 1997; accepted 12 December 1997 The Journal of Maternal-Fetal Medicine 7:111–114 (1998)  1998 Wiley-Liss, Inc.