What to do when your patient wants prenatal paternity testing

By Alan E. Donnenfeld, MD, and Elizabeth S. Panke, MD, PhD

"I'm not sure who the father is. How do I find out?" Such questions about paternity are being heard more and more often in ob/gyn offices. This article provides the practical information you need to address the many issues surrounding prenatal testing.

While no reliable evidence exists about the extent of misattributed paternity in the general population, some estimates indicate that the figure may be surprisingly high: 5% to 20%.¹ According to an American Association of Blood Banks' (AABB) annual report, 314,490 family relatedness tests were performed in 2001.² While most of these were paternity tests done on children after birth, many involved prenatal testing.

Prenatal paternity testing is accurate and reliable, but whether it should be performed is another issue. Many people believe it is important to establish paternity as early as possible if there is even a small doubt about the identity of the biological father. Such early testing can provide a stable foundation for the child and our legal system places a great emphasis on determining fatherhood based on biology. When concerns about paternity arise long after birth and testing reveals that the parenting father is not the biological father, it may, unfortunately, disrupt the father-child relationship.

Some women may use test results to terminate pregnancies originating from a sexual assault or if the father is the "wrong" man. Others may use the information to notify the father and begin proceedings to establish legal custody, child support, or both. Some women say they just need to know who their child's father is. Whatever the rationale for determining paternity, you should understand that you are under no obligation to participate in paternity testing if doing so would violate your religious beliefs, conscience, or professional integrity. It's sufficient to let your patient know that prenatal paternity testing is available but that you do not become involved with the process, and to refer her to another clinician for assistance with testing.

Using DNA to determine parentage

DNA testing has privileged status because of the relative novelty of the technology, its power to uniquely identify a single individual in the universe, and its apparent technical complexity. In a forensic setting, DNA testing is typically used to determine the probability that a particular DNA pattern is present in the general population. The likelihood of a DNA pattern match with prenatal paternity testing is much higher, however, because it involves a comparison between the fetus and only two different men.

Unfortunately, in the great majority of prenatal DNA paternity tests, labs are given the opportunity to test only one alleged father. Testing of more than two alleged fathers and of men who are closely related to each other (brothers or father and son, for example) is also not uncommon. To make matters worse, when the potential fathers are close relatives, labs often are not informed of that fact. Therefore, extensive testing is crucial to achieving accurate results.

While each person (except an identical twin) has a unique DNA sequence, identical stretches of DNA surprisingly account for more than 99.9% of the information present in the human genome. DNA analysis for paternity testing is designed to identify variable regions of DNA—or polymorphisms—in the remaining 0.1% of our genetic material and compare those sequences in two or more individuals. (Length polymorphisms are variable lengths of DNA at particular loci in the genome.)

First, millions of copies of the DNA fragments of interest are produced through polymerase chain reaction (PCR) amplification. Once the fragments have been sorted by size with electrophoresis, the specific fragment lengths are identified. Tables 1 and 2 show actual paternity tests using DNA length polymorphisms. The biologic mother and the biologic father each will have one allele that is identical to the fetus, except when a new mutation occurs. When such a discrepancy is encountered on DNA paternity testing, the results usually reveal a pattern inconsistent with parentage at only a single DNA site. However, a genetic mismatch at a single DNA site should not lead to an opinion of nonpaternity because most such occurrences merely reflect genetic mutations that occur naturally in ordinary people. A minimum of two—and preferably three—genetic mismatches is necessary between an alleged father and a fetus to rule out the man's parentage.

TABLE 1
This man is not the real father

TABLE 2
This man is the real father

When testing reveals a genetic match between a man and a fetus, the likelihood of a coincidental match is calculated. This is done with a paternity index, which uses information on the frequency of any one particular allele in the general population. When the paternal allele seen in the fetus is abundant in the population at large, the paternity index is low because a coincidental match would not be uncommon. The final analysis, which combines all the genetic evidence, is a product of all the individual paternity index values. This combined paternity index (CPI) is a ratio of the chance that the tested man, given his entire analyzed genetic makeup, has produced the pregnancy compared to the likelihood of a random man producing the child. A CPI of 10,000, for example, means that the tested man is 10,000 times more likely to be the biological father than a random man in the general population. Looked at another way, the CPI is a close estimate of the number of men in the population who have the identified genetic pattern and therefore cannot be excluded from the possibility of being the father. So, the CPI of 10,000 that we've been using as an example means that approximately one individual in 10,000 has the genetic pattern in question.

CPI is one of two primary numbers used in the bottom-line results of a DNA paternity test. The other is the probability of paternity. Unfortunately, the probability of paternity can easily be misinterpreted. A 99% probability of paternity sounds convincing, but it means that approximately one out of every 100 individuals in the population could be included as the child of the tested man.

Currently, a CPI of 100 (probability of paternity 99%) is the minimum required to establish parentage for legal purposes in most states. However, some laboratories believe the minimum CPI should be raised to 1,000 (probability of paternity 99.9%) to convincingly determine the father with a reasonable degree of certainty. While that would mean performing more extensive testing at a higher cost, the results of a paternity test with a CPI of 1,000 or greater are far more reliable.

Ethical issues

To do prenatal paternity testing, DNA samples from the fetus, mother, and at least one suspected father and a court order or the informed consent of the adults are necessary. Anyone involved in such testing must be fully informed about the implications of the results, which raise ethical issues in certain situations. For example, if a married woman wants to have her "boy-friend" tested, are you obligated to inform her husband? Although there is controversy surrounding this issue, most authorities would say no, primarily because of the principle of maternal autonomy and the importance of maintaining confidentiality within the doctor/patient relationship. Pregnant women, in consultation with their obstetricians, should be permitted to make their own decisions for themselves and their fetuses.³

Clearly prenatal paternity testing may place you in an extremely uncomfortable position, and you may need to make a distinction between professional obligation and conscience. If prenatal paternity testing places an intolerable burden on your conscience because of your moral convictions or religious or ethical beliefs, it is appropriate for you to decline to participate. However, you do have an obligation to transfer your patient's care in an orderly and safe manner to a colleague whose conscience is not violated by a request for testing.³ Other ethical issues are raised when routine prenatal genetic testing unexpectedly calls into question a "father's" paternity. See "When nonpaternity is a coincidental finding" for a discussion of that aspect of this subject.

How prenatal paternity testing is done

Noninvasive prenatal paternity testing, or comparison of the DNA pattern of fetal cells from the maternal circulation with that of an alleged father, is still considered investigational. Therefore, amniocentesis or chorionic villus sampling (CVS) is required. Fetal cells from amniotic fluid (5 to 10 mL) or CVS samples are necessary for prenatal testing. (Because amniocentesis and CVS both carry a small risk of miscarriage, patients who have no medical indication for those procedures should be given the option of having paternity testing performed after birth, which is just as accurate as prenatal testing.) Samples from the mother and the alleged father typically are collected with a buccal swab. Most DNA testing laboratories provide paternity test results in 2 to 10 days after they receive all of the specimens.

An important initial step in testing is having a qualified person collect the samples. The Standards for Parentage Testing Laboratories published by the American Association of Blood Banks (AABB) state "all biologic samples for testing shall be collected by persons with no interest in the test outcome."² The individuals who are being tested must be properly identified and the identification must be documented. According to the AABB Standards, a person being tested must provide government photo identification and/or have his or her photograph taken at the time of sample collection. The collected samples also must be clearly identified, through an affixed label that also documents the date of collection and initials of the person who did the collection. Strict chain-of-custody procedures also apply: Samples must never be left out of sight, must be sealed in a tamper-proof package immediately after they are collected, and cannot be packaged or transferred by test participants. Fetal samples can either be sent directly to the paternity testing lab or to a cytogenetics lab for cell culture before they are forwarded for paternity analysis.

Choosing a DNA laboratory

More than 160 labs are advertising DNA paternity testing on the Internet, but only about 40 of them are accredited by the AABB.² In order to get accurate test results, it's imperative that you choose a qualified lab. First, make sure that the facility advertising DNA paternity testing will actually be performing the testing. Many labs contract out testing services to facilities that charge the lowest price and may be less qualified, less reputable, and lack AABB accreditation. DNA parentage testing laboratory accreditation is not mandatory, but in most states, such test results are not legally admissible unless provided by an AABB-accredited laboratory. Most DNA laboratories that choose to receive accreditation adhere to the AABB's testing procedures and standards. The State of New York and The College of American Pathologists also have established national procedures and standards for DNA parentage testing.

When choosing a lab, you should also consider the facility's track record and experience. How long has it been in business? How many DNA tests has the lab performed in the past year? What are the qualifications of the director and staff and are they available to answer questions? A lab that has been performing DNA parentage testing for many years is more likely to have an experienced staff and well-validated testing procedures. Also ask about what guarantees the lab provides for testing. Make sure that the lab requires a mismatch between the tested man and the fetus at a minimum of three genetic sites for establishment of nonpaternity. And make sure that the lab confirms paternity based on a CPI of at least 1,000 (99.9% probability of paternity).

Other important considerations when choosing a lab are the cost of testing and turn-around time for results, but test cost should not dictate quality. The laboratory fee for parentage testing typically ranges between $400 and $600 and most insurance plans do not cover it. Unless your patient has a medical indication for amniocentesis or CVS, those expenses also will not be covered by insurance.

Conclusion

Each pregnant woman's situation is extremely personal and her paternity testing experience and the results will have a profound impact on the lives of everyone involved. Make sure that your patient is thoroughly informed about DNA paternity testing before she embarks on the procedure. And if you are involved in the testing process, choose a lab that is qualified, experienced, AABB-accredited, and responsive to your inquires.

REFERENCES

1. Le Roux MG, Pascal O, David A, et al. Non-paternity rate and screening in genetic disease analysis. Lancet. 1993;341-57.

2. Standards for Parentage Testing Laboratories. American Association of Blood Banks. Fifth Edition, ISBN 1-56395-145-2, 2001 ( www.aabb.org ).

3. Chervenak FA, McCullough LB. Does obstetric ethics have any role in the obstetrician's response to the abortion controversy? Am J Obstet Gynecol. 1990;163:1425-1429.

DR. DONNENFELD is Associate Professor, Drexel University College of Medicine, and Medical Director, Genzyme Genetics, Philadelphia, Pa. DR. PANKE is Director, Genetica DNA Laboratories, Inc., Cincinnati, Ohio.

When nonpaternity is a coincidental finding

Advances in molecular genetic research are revolutionizing the prenatal evaluation of pregnant women. Testing has become routine for numerous genetic diseases, such as cystic fibrosis, sickle cell anemia, Tay-Sachs disease, Canavan's disease, alpha and beta thalassemia, and Fragile X syndrome. You should offer prenatal genetic testing to a patient if there is a family history of a particular genetic disorder that can be diagnosed with molecular genetic techniques. But what if genetic testing reveals that paternity has been misattributed? How should you respond? Consider the following situation:

Ann (who is currently 12 weeks' pregnant) and David present for a genetic evaluation because their first child was diagnosed with Werdnig-Hoffmann spinal muscular atrophy, an autosomal-recessive neuromuscular disorder that typically results in death during childhood. This disorder has been mapped and prenatal diagnosis may be possible in the current pregnancy. Ann and David consent to blood sampling for DNA extraction on themselves and their affected child. Molecular genetic analysis reveals that David is not the biological father of the child.

In 1994, the Committee on Assessing Genetics Risks of the Institute of Medicine recommended that in such cases only the woman should be informed, as "genetic testing should not be used in ways that disrupt families."¹ This approach has received widespread support. A 1992 survey of more than 500 genetic counselors in the United States and Europe revealed that 98.5% would not disclose false-paternity information to the father.² Other investigators have confirmed this approach, concluding that the protection of the mother's confidentiality overrides disclosure of true paternity.³

A 2001 editorial in The Lancet, however, concludes that a significantly stronger argument can be made in favor of disclosure of false paternity.⁴ The authors of the article in which the argument is made suggest a proactive approach. In situations where DNA testing might reveal nonpaternity, they recommend discussing that possibility before the procedure is done. Raising the issue during initial counseling would minimize the dilemmas encountered if a false paternity is revealed because the parents essentially would make the decision about disclosure before testing. Thus, a physician who knows the parents' wishes before he receives test results would not have to struggle to decide what to do with them when the report arrives.

REFERENCES

1. Institute of Medicine, Committee on Assessing Genetic Risks. Assessing genetic risks. Washington DC: National Academy Press, 1994, 276.

2. Pencarinha DF, Bell NK, Edwards JG, et al. Ethical issues in genetic counseling: a comparison of master level counselor and medical geneticist perspectives. J Genet Counsel. 1992;1:19-30.

3. Wertz DC. Ethics and genetics in international perspective: results of a survey. In: Nippert I, Neitzel H, Wolff G, eds. The New Genetics: From Research to Health Care. Heidelberg: Springer-Verlag; 1999:75-94.

4. Lucassen A, Parker M. Revealing false paternity: some ethical considerations. Lancet. 2001;357:1033-1035.

 

Alan Donnenfeld, Elizabeth Panke. What to do when your patient wants prenatal paternity testing. Contemporary Ob/Gyn Apr. 15, 2004;49:44-51.