Suggestions for the reporting of sensitive genetic test results

A Transfusion Medicine Fellow in Canada reports a case of a newborn infant who was referred to an academic hospital for management of low birth weight, neonatal hypoglycemia, and severe thrombocytopenia. The thrombocytopenia was initially managed with transfusion of pooled random donor platelets (pre-storage leukoreduced), but the transfusions were only transiently effective in increasing the patient's platelet count. Investigations for neonatal alloimmune thrombocytopenia were undertaken, and PCR revealed the baby to have a platelet genotype of HPA-5a/5b, while the mother had a platelet genotype of HPA 5a/5a. Monoclonal Antibody-specific Immobilization of Platelet Antigens (MAIPA) revealed anti-HPA-5b in the maternal serum. By the time the aforementioned results were available, the baby's platelet count sufficiently normalized to allow it to be returned to the referring hospital, and no further transfusions were needed. The inquiring physician wonders how others would deal with the "stated father's" platelet genotype results which showed him to be HPA-5a/5a. Assuming that the stated father's platelet genotype results are correct (and since the family has left the city and platelet genotype testing cannot be easily repeated), the implication is that the stated father might not be biologically related to the baby! This has medical implications for future pregnancies. However, it is also clearly sensitive information, and raises questions of confidentiality and the fiduciary relationship (or lack thereof) between the blood bank medical director and the baby's mother. To whom would you convey these lab results? Would you convey them to the stated father, the baby's mother, the family physician, or other individual? By what means would you convey the information? Would you convey the information by phone, letter, email, face-to-face, or some other method? What should the 'message' be?

The following comments have been received.

ADDENDA Jan. 19, 2006

1. A transfusion medicine physician in Boston reports his opinion that if in the case under discussion there was no requiest for paternity testing, then the test results should NOT be reported as such. Indeed, he reminds us that proper paternity testing is only done with careful identification of the people being tested and a strict control of the chain of custody of the specimens. In this case, he would explain that the mother made an antibody to the baby's platelets and this also puts future pregnancies at risk and that those pregnancies should be followed by a high-risk obstetrician. He would suggest that if the woman becomes pregnant again, that the Blood Bank physician can provide and explain the test results to the obstetrician.
   
   
2. Dr. Paul Holland (attribution used with permission) reports that when he was at the NIH, he was involved with several family studies that examined the linkage of blood groups to various inherited disorders. He recalls that an NIH colleague (who was also involved with the family studies) once remarked that about 5-10% of the family studies revealed non-paternity in one blood group system or another. As Dr. Holland recalls, they provided the information to the physician who had ordered the tests, so that the family could be informed, or so he assumed. Dr. Holland acknowledges that they did not want there to be false assignment of linkage in any of the family studies, and they usually suggested that the tests be repeated with new samples to ensure that a mixup of samples did not lead to a false assumption of non-paternity. He cautions that if a family member has to be contacted, it should be the mother. Dr. Holland adds that in the current case under discussion, he would suggest that new samples be drawn to verify the genotype findings, as those results have import for future pregnancies. Also, if the mother were to give birth to another baby who needs platelet transfusions because of NAIT, Dr. Holland suggests using the mother as a platelet donor for the baby, and washing those platelets to remove the anti-HPA-5b. However, if the "stated father" is the biological father of the next baby, and if his platelet genotype is really HPA-5a/5a, there should be no risk of NAIT from anti-HPA-5b.
   
   
3. A transfusion medicine physician in the North Eastern US reports that questionable paternity is considered from time to time in his HLA lab, when reviewing HLA results for possible living related kidney or stem cell transplant donors. From time to time they "discover" that siblings and fathers do not appear to be related! In his experience there are different ways to handle these situations, but he cautions against over interpreting the results of a platelet genotype study, since the sample in question was not collected for paternity testing, so that it is unlikely that the sample was collected under the same rigor as would be required for paternity testing. Without assurances to rule out a specimen identification error, it would be inappropriate to infer anything about paternity from these test results. Furthermore, he recommends that the Canadian physician contact his institution's HLA lab to see if that lab has dealt with this situation in the past, and if so, to be consistent with his institution's existing practices.

ADDENDA Jan. 20, 2006

4. A transfusion medicine physician in New York State reports that he would be very cautious in the interpretation of non-paternity based on a platelet genotype test result for several reasons:
   1. possibility of specimen misidentification,
   2. possibility of an undiscovered anomaly in platelet antigen inheritance that has not yet been described,
   3. possibility that the baby was conceived by artificial insemination using a sperm donor who was not the stated father.

Given the aforementioned uncertainty, the New York physician would issue a written report of the factual test results, without interpretation of paternity, and he would call and explain the subtle nuances of the report to the woman's obstetrician and/or pediatrician. That having been said, he would advise the obstetrician and/or pediatrician that communicating the findings to the woman in a most sensitive fashion would probably be the right thing to do. Based on the woman's response, repeating the laboratory tests, and perhaps other tests of paternity might be indicated.

5. Editor's note: The following stories may be of interest to the e-Network Forum and germane to the present discussion about discovering non-paternity when doing genotype testing:
   • Paternity Misrepresentation: A Florida Court Rules That a Husband Waited Too Long to Disprove Fatherhood, and Reaffirms His Status as the Child's Father (FindLaw Legal News and Commentary, Dec. 27, 2005)
   • Far more men than women favor routine paternity testing at birth (University of Washington News, Sept. 28 2004)
   • Non-paternity rates from paternity testing laboratories (Child Support Analysis, Jan. 4, 2006)
   • Child Support Analysis Weblog Archive for October 2004 (Nov. 1, 2004)

ADDENDA Jan. 21, 2006

6. Dr. Ronald E. Domen, Professor of Pathology, Medicine, and Humanities/Associate Dean for Graduate Medical Education/Medical Director, Blood Bank and Transfusion Medicine/Milton S. Hershey Medical Center at Penn State University College of Medicine (attribution used with permission) reports that in the series of ethics cases that he presents to his residents, one case deals with HLA testing in a living, related transplant scenario where one of the siblings is found to not be the father's child and this is unknown to the siblings and the putative father. This case always generates a great deal of discussion on what to do with this information. Should the results be ignored? What if the father has always had suspicions and pulls you aside and asks if this is his child? What if the child in question obtains a copy of the test results and figures it out? What if it is the patient (rather than a potential donor) who has the questionable paternity and this information may be important for genetic/therapeutic reasons? Dr. Domen comments that the fact is, information has been obtained (albeit not through a "normal" paternity work up) that brings into question the patenity of one or more members of the family. This information potentially has serious and important implications for those involved. Does not a physician (HLA lab director, transplant physician, treating physician, etc.) have an ethical obligation to convey these results in some fashion? In the case under discussion by the e-Network Forum, therapeutic decisions are being made based on the platelet antibody/antigen testing, so why in the case of paternity, are the results/chain of custody in question when it comes to making a genetic assessment? Are we being too paternalistic when we decide to ignore the results as none of our business? At the very least, it seems to Dr. Domen, the results should be privately discussed with the mother and/or the patient/child as appropriate. The mother of a minor child may or may not want to then discuss the issue with the husband. An adult sibling may or may not decide that the information is important. In any event, additional testing could be performed if all involved want to pursue that route. Dr. Domen adds that it is estimated that 10% or more of children have parents different from who they think they have. He is not convinced that the best ethical and medical approach is to ignore this possibility when laboratory tests raise the issue of paternity. Patients have a right to know the results of tests performed on them.

ADDENDA Feb. 8, 2006

7. A colleague in Wisconsin reports that his laboratory performs testing on about 500 cases of suspected neonatal alloimmune thrombocytopenia (NAIT) each year. However, they do not usually receive blood from an affected child for platelet genotyping. They have observed that the mother can have a platelet specific antibody without the father having a platelet antigen with which the mother's antibody is incompatible. Such a finding could be due to a previous pregnancy with a different partner (before she met the current husband) or exposure from a prior transfusion, but would not explain the clinical presentation of the baby. However, they have seen at least one case similar to the one under current discussion by the e-Network Forum. That case also involved HPA-5. Their policy is to report what they find without making any insinuations about paternity. They leave it to the patient's physician to decide how to use this information. Often times the physician will call to consult on how such a typing discrepancy might occur, and what the results might mean for future pregnancies.

Submit comments to the e-Network Forum at enetworkforum@cbbsweb.org

Ira A. Shulman, MD
CBBS e-Network Forum Senior Editor & Moderator

W. Tait Stevens, MD
CBBS e-Network Forum Editor & Moderator

Elizabeth M. St. Lezin, MD
CBBS e-Network Forum Associate Editor & Moderator