PEDIATRICS Vol. 106 No. 4 October 2000, pp. 650-653

Neonatal Outcome of Preimplantation Genetic Diagnosis by Polar Body Removal: The First 109 Infants

Charles M. Strom, MD, PhD, Rebecca Levin, BA, Sam Strom, Christina Masciangelo, MS, Anver Kuliev, PhD, and Yury Verlinsky, PhD

From the Reproductive Genetics Institute, Department of Obstetrics and Gynecology, Illinois Masonic Medical Center, Chicago, Illinois.

	ABSTRACT

Top Abstract Methods Conclusion References

Context.  Our center developed the technique of preimplantation genetic diagnosis (PGD) by sequential polar body removal (PBR) for the diagnosis of Mendelian disorders and aneuploidies. This study examines the obstetric and neonatal outcome of the first 109 live births after PGD by PBR.

Objective.  To determine if there were any observable effects of PGD by PBR on perinatal morbidity and mortality, birth defects, and growth parameters.

Design.  Data on perinatal outcome were gathered for the first 109 infants by parental reporting and confirmed by telephone interview and chart review when indicated. In infants >6 months old, a follow-up telephone interview was performed establishing the developmental milestones attained by the child.

Setting.  A research center conducting an institutional review board-approved research protocol in PGD.

Patients.  All patients who had PGD by PBR who had clinical pregnancies.

Main Outcome Measures.  Gestational age, mode of delivery, perinatal mortality, birth weight, birth length, the presence of birth defects, and developmental milestones.

Results.  There was no significant decrease in birth length or weight, or the frequency of small for gestational age infants. No specific pattern of birth defects was observed.

Conclusion.  Thus far, there are no observable detrimental effects of PGD by PBR on children born after the procedure.  Key words:  preimplantation genetic diagnosis, polar body removal, outcome.

Preimplantation genetic diagnosis (PGD) was developed for couples at high genetic risk to avoid establishing pregnancies with genetic diseases. PGD is performed by blastomere biopsy or polar body removal (PBR) for Mendelian or chromosomal disorders.¹

Because the aneuploidy rates in oocytes and embryos of women of advanced maternal age (>35 years old) undergoing in vitro fertilization-embryo transfer (IVF) are extremely high (30%-40%), PGD for aneuploidy detection has been offered to IVF patients of advanced maternal age. All PGD is performed under an Illinois Masonic Medical Center Institutional Review Board-approved research protocol after obtaining informed consent.^2-4

Our center has pioneered the use of PBR for the purposes of PGD for both Mendelian disorders and aneuploidies. For Mendelian disorders, the mother must be heterozygous for a mutation, either autosomal dominant, autosomal recessive, or X-linked. Primordial germ cells will contain 1 chromosome carrying the affected allele and another carrying the normal allele. At the onset of meiosis, the immature oocyte doubles its genetic material, yielding 2 chromosomes containing the normal allele and 2 chromosomes containing the mutated allele. At the conclusion of meiosis I, the oocyte extrudes half of its chromosomes in the form the first polar body. We remove the first polar body before fertilization and analyze it for the presence of the normal and the mutated gene.

Subsequently, fertilization is performed by intracytoplasmic sperm injection. After fertilization, the oocyte completes the second meiotic division and the second polar is extruded containing 1 set of chromosomes, leaving the egg with 1 copy of each chromosome. We then remove the second polar body for analysis. If the first polar body contains only the mutated gene, it is demonstrated that both copies of the mutated gene have been extruded. In this case the second polar body will have the normal allele. Thus the oocyte is predicted to contain the normal allele. Conversely, if the first polar body contains only the normal gene, the second polar body will contain 1 affected gene and the oocyte is shown to be affected. If a cross-over has occurred, the first polar body will contain both the mutated and the normal allele. That means that the immature oocyte contains 1 copy each of the normal and the mutated gene. In this case second polar body analysis is necessary to complete the diagnosis. If the second polar body contains the normal allele, the oocyte will contain the affected allele, and if the second polar body contains the mutated allele, the oocyte is predicted to be normal. Only embryos developing from oocytes shown to contain the normal gene are transferred to the mother to establish a pregnancy.

For aneuploidy detection, the first and second polar bodies are removed simultaneously and subjected to fluorescent in situ hybridization with probes to chromosomes 13, 16, 18, 21 and 22. If the first polar body contains 2 copies of each of these chromosomes and the second polar body contains a single copy of each chromosome, the oocyte must retain 1 copy of each of these chromosomes. Embryos developing from such oocytes can be transferred without the risk of a trisomy or monosomy attributable to maternal nondisjunction for these chromosomes. If the first polar body contains only 1 copy of a chromosome and the second polar body has 1 copy of the same chromosome, that oocyte will have retained 2 copies of that particular chromosome and its fertilization with a normal sperm would lead to a trisomy for that chromosome.

This study is a follow-up of the first 97 pregnancies and resulting 109 live births after PGD by PBR for both Mendelian disorders and aneuploidy detections. This analysis showed no diagnostic errors and that there were no adverse affects on the infants.

	METHODS

Top Abstract Methods Conclusion References

Records were reviewed for all pregnancies achieved after PGD by PBR. Some of the clinical outcomes included in these data have been reported elsewhere.^2-7 Because of the geographical dispersion of our patients, it was not feasible to examine each infant. All patients were sent a birth questionnaire at the time of delivery. After obtaining consent from parents, pediatricians were contacted for records. In infants >6 months old, a follow-up telephone call was made to all patients to assess the most recent developmental information.

The birth questionnaire gathered data on the type of delivery, estimated gestational age at birth, the birth length, weight, and the presence of any birth defects or other complications. The telephone interview reconfirmed the birth data and included questions regarding attainment of developmental milestones, the most recent weight, length, and the presence of any physical or developmental anomalies. Growth percentiles were calculated using standard Denver growth curves. In special cases, medical records were reviewed with the parents' consent to eliminate spurious diagnoses such as bilirubin lights, apnea, and bradycardia in a premature infant. Commonly occurring anatomic variations, such as a single umbilical artery with normal renal ultrasound and function in 1 case were not included.

	RESULTS AND DISCUSSION

There were 91 infants born after PGD for chromosomal disorders and 18 infants born after PGD for Mendelian disorders, including cystic fibrosis (11 infants), sickle cell disease (1 infant), long chain acyl CoA dehydrogenase deficiency (1 infant) and thalassemia (5 infants). In all of these cases, mutation-free embryos were selected based on sequential polar body detection of the unaffected maternal allele in the oocyte.^3-7 The results of PGD for Mendelian disorders were confirmed postnatally in all resulting pregnancies. In all cases the maternal allele present in the newborn was correctly predicted by the preimplantation genetic analysis. There were no exceptions.

In the follow-up of aneuploidy detections, no cases of trisomy 21, trisomy 18, or trisomy 13 were found at prenatal diagnosis, and no diagnosis of aneuploidy was made in the resulting cohort of the newborns.

For 2 of the 97 pregnancies (2%), no birth or follow-up data were available. Birth data were available for 98% of the cohort. Developmental information was available for all 44 children >6 months old. As shown in Table 1, there were 80 singleton pregnancies (defined as a single gestational sac present on ultrasound), 9 twin, and 7 triplet gestations, of which 3 were selectively reduced to twins. Four couples delivered triplets at gestational ages varying between 32 and 36 weeks' gestation. All triplets were discharged from the hospital within 1 week of delivery and none required mechanical ventilation or oxygen for >2 hours. One gestation with 5 fetuses miscarried in the first trimester.

The delivery information for the 80 singleton pregnancies is summarized in Table 2. Sixty (75%) of the singleton pregnancies progressed to term, defined as >37 completed gestational weeks. Ten infants were delivered in the 37th week. There was 1 neonatal death caused by a placental abruption at 39 weeks' gestation. This newborn had no liver or kidney function and died at 6 days of age of multiple organ failure.

The mean birth weight percentile for all the singleton term births was 47% and the mean birth length was 57%. The percentage of low birth weight infants (defined as less than the 10th percentile for gestational age) for the entire singleton cohort was 9%, indicating that PBR did not seem to cause a decrease in the size of the newborns.

The rate of cesarean section (CS) in term infants was 40% (23 of 57). These data are comparable to 2 published studies of CS rates in IVF patients, one reporting 47.3% in 169 deliveries,⁸ and the other 41% in 260 IVF deliveries.⁹ The CS rate for all deliveries was also 40% (30 of 76).

Table 3 is a summary of 6 valid reports of birth defects. Two of these could be considered major defects. One newborn (previously reported) had a transverse limb reduction defect. This gestation had an amniotic band detected prenatally by ultrasound examination. Although this pregnancy had been exposed to chorionic villus sampling (CVS) for confirmation of diagnosis, the prenatally observed amniotic bands are known to be a cause of limb reduction.¹⁰ Examination of photographs of this child by one of the authors (C.M.S.) and an independent dysmorphologist's examination have confirmed the diagnosis of amniotic band syndrome in this infant. Although CVS has been reported to be associated with limb reduction abnormalities, none of these reports have shown amniotic bands associated with CVS.¹⁰

A second newborn had seizures in the newborn period and imaging studies revealed 3 discrete cerebral infarcts. The child was born at term with Apgars scores of 7 at 1 minute and 8 at 5 minutes. Although there is no developmental information available for this child, this cause of this problem is much more likely to be prenatal or perinatal infarct rather than a developmental or teratogenic abnormality.

The other 4 malformations reported were minor: 2 neonates had small strawberry hemangiomas, 1 on both arms, and for the other the location was not specified. The parents could not be contacted for further details. The incidence of congenital hemangiomas in our PGD cohort was 2 per 109 births (2%). This rate is within the reported incidence of hemangiomas in newborns, 1.1% to 2.6%^12-14 and lower than the incidence reported after amniocentesis of 7.4%.¹⁴

One child had a thickened tricuspid valve not requiring surgery or medication, and, finally, another child had bilateral syndactyly of 2 toes on each foot, which likely represents a genetic abnormality and not a developmental or teratogenic abnormality.

Other than the presence of 2 cases of hemangiomas, there is no specific pattern of birth defects in this cohort. Further data will need to be accumulated before reaching a conclusion regarding an association of PGD with hemangiomas. Because the numbers are still low, no statistical conclusions can be made at this time, but the initial data are reassuring that there is no specific pattern of birth defects associated with infants born after PGD.

The current health status of 44 children >6 months old was assessed. The mean age at follow-up was 11 months with a range of 6 to 30 months. The average percentile weight at the most recent pediatric visit was 40% and the average percentile length was 47% indicating no adverse effect on postnatal growth. One mother reported developmental delays. This was a twin gestation delivered at 36 weeks with no perinatal complications. At the time of the interview, 1 of the twins was 30 months old, suffered from a 6-month speech delay, and was receiving speech therapy. There were no motor delays in either twin.

According to the Birth Defects Encyclopedia "As many as 16% of all deliveries involve birth defects. Birth defects that can have an impact on function occur in a little over 7% of deliveries."¹⁵ In our cohort 2 children (2%) had birth defects impacting function and another 4 (4%) had minor abnormalities. Therefore, PBR does not appear to increase the rate of subsequent birth defects.

At the moment, we are not aware of any similar data available from other centers on the outcome of PGD except for the report of 1 case of acardius acranius in a triplet pregnancy with a twin-reversed arterial perfusion after PGD for Duchenne muscular dystrophy performed by blastomere biopsy.¹⁶ Based on this preliminary dataset, it may be concluded that PGD by PBR has no demonstrable adverse effects on the pregnancies, deliveries, or health of the infants.

We plan to continue longitudinal follow-up of this cohort to insure that subtle abnormalities do not become apparent as the children mature.

The financial cost for PGD is a combination of the costs of in vitro fertilization plus the cost of the genetic analysis. In our center, the cost of a typical PGD cycle is $8500. However, many of our aneuploidy patients required IVF regardless of the PGD. In these cases the cost of the PGD is approximately an additional $2000 to the IVF costs.

For Mendelian disorders, assuming the couple is fertile and would not otherwise require IVF, the alternatives would be a routine pregnancy followed by prenatal diagnosis by CVS or amniocentesis. Both those procedures cost approximately $1000 in our center exclusive of the cost of the genetic diagnosis. The emotional and financial costs of terminating affected pregnancies must be added to the cost of this alternative.

Aside from the financial cost, IVF carries some intrinsic risks. As noted above, IVF carries with it an increased risk of CS. There is also an increased risk of multiple gestations in IVF patients. In our cohort there was a 9% incidence of twins, a 7% incidence of triplets, and 1 case of 5 fetuses that spontaneously miscarried. The issue of potential multiple gestations is discussed with each patient before embarking on PGD. We include a discussion of multifetal pregnancy reduction and encourage patients to consider transferring no more than 3 embryos to avoid problems of higher multiple gestations. As IVF techniques improve, hopefully fewer embryos will need to be transferred in each cycle and the multiple gestation rates can be lowered without adversely affecting pregnancy rates.

PGD provides an important new option for couples at high genetic risk. The process of prenatal diagnosis with the potential for pregnancy termination is extremely stressful and often traumatic for couples attempting to have healthy children. Although more expensive than conventional prenatal diagnosis, many couples choose PGD in hopes of having pregnancies free from the concerns of having to terminate an affected pregnancy. In women of advanced maternal age, PGD by PBR is an effective method to eliminate the age-related increased risk of aneuploidies. Although PGD by PBR cannot eliminate their risk of having children with Down syndrome or other aneuploidies, it should theoretically reduce the risk to that of a 20-year-old woman.

	CONCLUSION

Top Abstract Methods Conclusion References

In summary, the data presented here demonstrate that PGD by PBR is a safe and accurate technique for couples at high genetic risk to avoid having children with genetic abnormalities without the anxiety of awaiting prenatal diagnosis and the potential of having to terminate affected fetuses.

	FOOTNOTES

Received for publication Jun 18, 1999; accepted Jan 26, 2000.

Reprint requests to (C.M.S.) Reproductive Genetics Institute, Department of Obstetrics and Gynecology, Illinois Masonic Medical Center, 836 Wellington Ave, Chicago, IL 60657. E-mail: dnaguy{at}aol.com

	ABBREVIATIONS

PGD, preimplantation genetic diagnosis; PBR, polar body removal; IVF, in vitro fertilization-embryo transfer; CS, cesarean section; CVS, chorionic villus sampling.

	REFERENCES

Top Abstract Methods Conclusion References

1. Handyside AH, Kontogianni EH, Hardy K, Winston RML Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature. 1990; 344:768-770 [CrossRef][Medline]
2. Verlinsky Y, Rechitsky S, Freidine M, Birth of a healthy girl after preimplantation gender determination using a combination of polymerase chain reaction and fluorescent in situ hybridization analysis. Fertil Steril. 1996; 65:358-360 [Medline]
3. Strom CM, Verlinsky Y, Milayeva S, Preconception genetic diagnosis of cystic fibrosis. Lancet. 1990; 336:306-307 [CrossRef][Medline]
4. Verlinsky Y, Rechitsky S, Evsikov S, Preconception and preimplantation diagnosis for cystic fibrosis. Prenat Diagn. 1992; 12:103-110 [Medline]
5. Strom CM, Ginsberg N, Rechitsky S, Three births after preimplantation genetic diagnosis for cystic fibrosis with sequential first and second polar body analysis. Am J Obstet Gynecol. 1998; 178:1298-1306 [CrossRef][Medline]
6. Verlinksy Y, Cieslak J, Ivakhenko V, Preimplantation diagnosis of common aneuploidies by first- and second-polar body FISH analysis. J Assist Reprod Genet. 1998; 15:285-289 [CrossRef][Medline]
7. Strom CM, Strom S, Levine E, Obstetric outcome in 102 pregnancies after preimplantation genetic diagnosis. Am J Obstet Gynecol. 2000; 182:1629-1632 [CrossRef][Medline]
8. Hook EB, Cross PK, Shreinemachers DM Chromosome abnormality rates at amniocentesis and in liveborn infants. JAMA. 1983; 249:2034-2048 [Abstract/Free Full Text]
9. Levy T, Goldman JA, Dicker D, Ashkenazi J, Feldberg D Very early pregnancy wastage in in vitro fertilization and embryo transfer (IVF-ET). J In Vitro Fertil Embryo Transf. 1991; 8:250-3 [CrossRef][Medline]
10. Maman E, Lunenfeld E, Levy A, Vardi H, Potashnik G Obstetrical outcome of singleton pregnancies conceived by in vitro fertilization and ovulation induction compared with those conceived spontaneously. Fertil Steril. 1998; 70:240-245 [CrossRef][Medline]
11. Reubinoff BE, Samueloff A, Ben-Haim M, Friedler S, Schenker J, Lewin A Is the obstetric outcome of in vitro fertilized singleton gestations different from natural ones? A controlled study. Fertil Steril. 1997; 67:1077-1083 [CrossRef][Medline]
12. Pratt AG Birthmarks in infants. Arch Dermatol. 1953; 67:302-305 [Abstract/Free Full Text]
13. Jacobs AH The incidence of birthmarks in neonates. Pediatrics. 1976; 58:218-222 [Abstract/Free Full Text]
14. Silverman RA Hemangiomas and vascular malformations. Pediatr Clin North Am. 1991; 38:811-834 [Medline]
15. Center for Birth Defects Information Services, Inc. Buyse ML, ed. Birth Defects Encyclopedia. New York, NY: Alan R. Liss, Inc; 1990:IX
16. Burton BK, Shulz CJ, Angle B, Burd L An increased risk of hemangiomas in infants born following chorionic villus sampling (CVS). Prenat Diagn. 1995; 15:209-214 [Medline]