Latebreakers

by Maureen K. Baldwin, MD MPH

Any test obtained in a clinic, research, or even in a home setting has the potential to have false positive results. Setting aside user errors, such as waiting too long to read a home pregnancy test, or mis-labeling patient samples, there are uncommon but not rare instances of false positive pregnancy test results. The reasons for these false results are slightly different between urine and serum HCG tests because even though the tests are set up to detect the same molecule, the assays are different (Table 1). Also, the circumstances under which we might be obtaining urine versus serum HCG are different. Urine samples are more likely obtained for screening purposes, and serum for diagnostic testing or trend analysis. Using a diagnostic test for a screening purpose can result in higher rates of false positive results [1].

Table 1. False positive pregnancy test sources

Not possibly pregnant	Serum HCG NEG	Serum HCG low POS
Urine HCG NEG	—	Heterophile antibodies
Urine HCG POS	Contamination	Pituitary HCG with low clearance LH cross-reaction in menopause

Dear consultant, a patient has a positive urine pregnancy test and says they can’t possibly be pregnant. What do I do next?

It’s reasonable to start this evaluation with an assessment of how likely it is that the patient might be pregnant. If there is any possibility, then they are likely pregnant until proven otherwise since urine pregnancy tests are very accurate [2] . However, if the patient does not have uterus/ovaries/fallopian tubes, is menopausal, or their only sexual partner has had a vasectomy, then further steps should be taken to assess the source of the positive test. Keep in mind that you don’t want to miss rare occurrences of ectopic pregnancy, gestational trophoblastic disease, and vasectomy failure.

The first step is to repeat the urine test using a new sample on a different machine. A second step would be to confirm with a serum HCG. A negative serum HCG confirms a false urine test. A falsely positive urine test can be due to exogeneous HCG or blood contamination and repeating the test will usually resolve the question. If the serum HCG is still unexpectedly positive, then move on to the section below on unexpected positive serum HCG. Repeating the urine pregnancy test and confirming with serum HCG will resolve most unexpected positive urine pregnancy tests.

Dear consultant, a patient has a positive serum pregnancy test and says they can’t possibly be pregnant. What do I do next?

There are more potential reasons for false positive serum HCG tests than for false positive urine HCG tests, particularly when used for screening in patients at low or no risk for pregnancy. Good general advice is to start with urine testing first (and then stop checking!). However, if an unexpected positive urine test has led to testing serum HCG, or there is an unexpected serum HCG result, always start by screening for pregnancy potential and repeating the test.

The most common reason for an unexpected serum HCG is an incompletely resolved pregnancy (e.g., retained gestational tissue or the normal course of renal clearance of the large HCG molecule after the end of a pregnancy) and less commonly, gestational trophoblastic disease or a germ cell tumor. Circulating HCG is detectable for up to 6 weeks after the end of a pregnancy [3] . We commonly recommend repeating HCG in 4 weeks after medication abortion, for example, but 20% still have detectable serum HCG at that time (and that is likely normal for them if they are asymptomatic – just repeat in 1-2 weeks) [4] . Assessment of the potential for new pregnancy and interval repeat HCG are the typical next steps. If levels are plateaued, consider uterine aspiration to manage retained gestational tissue. If increasing, consider a new pregnancy. Decreasing HCG can be followed weekly until it reaches a non-pregnant level.

If a urine HCG has not yet been obtained, this can help the evaluation. A mismatch of false-positive serum HCG but negative urine HCG might occur when the individual has interfering antibodies that bind to the radio-immune assay. This so-called “phantom” or “heterophile” antibodies result in low level serum HCG detection [5] . If needed, this situation is confirmed by asking the laboratory to perform serial dilutions. Due to antibody binding, the results will not decrease proportionally compared to the dilutions. A negative urine HCG in the presence of low-level positive serum HCG confirms heterophile antibodies and further dilution testing is not necessary.

Other times, both low-level serum HCG (in the tens to low hundreds) and positive urine HCG may result when there are higher levels of circulating pituitary hormones that can bind to the assay due to similar molecular structure [6] . These include luteinizing hormone (LH), which is elevated in menopause, and elevated thyroid stimulating hormone (TSH). Additionally, since HCG is produced at low levels in multiple tissues, a false positive HCG can occur in patients with low renal clearance [7] . An unexpected positive urine and low serum HCG in the setting of menopause or chronic kidney disease should not require further evaluation. Since these HCG levels are often so low that they do not result in a positive urine pregnancy test, starting with history-based screening or urine screening is a good practice prior to common procedures that require pregnancy screening, such as dialysis.

Levels of serum HCG will be much higher in the setting of germ cell tumors and gestational trophoblastic disease. Higher unexpected serum HCG (in the thousands) requires imaging and sometimes further laboratory analyses to evaluate the source.

Dear consultant, a patient has a positive urine and serum pregnancy test that are low and not resolving, and they might be pregnant. How much testing do I need to do for confirmation?!

Some cases may require additional testing over multiple visits. However, consider the initial indication for the unexpected test and the potential risk of pregnancy versus false positive result. Take for example, the following case: A 48-year-old presents to clinic for IUD exchange for contraception. The patient has been amenorrheic for several years. Urine pregnancy test was obtained per clinic protocol and was found to be positive. It was repeated on two different machines. Last intercourse was over one month prior. Serum HCG was obtained which was trended at 121, 106, and then 91 IU/L. Ultrasound was non-diagnostic, and the differential diagnosis included presumed early pregnancy loss versus ectopic pregnancy. Uterine aspiration was performed which was atrophic. Following uterine aspiration, HCG declined only slightly to 81 IU/L. Upon further investigation of the patient’s history, she was noted to have chronic kidney disease with Cr 1.53 mg/dL. A serum FSH was obtained which was 99 IU/L. At this point, the workup could be considered complete since the patient is most certainly menopausal and this story is classic for cross-reaction of the HCG test due to elevated LH in combination with low renal clearance of endogenous pituitary HCG. Further proof of this diagnosis could be obtained by suppressing LH. In this case, administration of combined estrogen/progestin resulted in decreased HCG to 19 and then 5 IU/L. Repeat FSH at that time was 38 IU/L, confirming menopause [8] . HCG later increased to 45 IU/L upon discontinuation of the combined oral contraceptive. This case highlights the potential for false positive urine and serum HCG in patients undergoing routine screening for pregnancy who may already be menopausal and have other factors that predispose them to false positive HCG.

Dear consultant, why don’t we have to do a pregnancy test before starting contraception?

The U.S. Centers for Disease Control propose a history-based screening prior to starting contraception in their Selected Practice Recommendations that does not include HCG screening and has 99-100% negative predictive value to rule out pregnancy (Box 1) [9]. Screening for current pregnancy starts with taking a history that includes timing of intercourse related to menstrual timing, contraceptive use, and recent pregnancy history. To avoid false positive results, all situations requiring pregnancy screening should consider using this history-based format and starting with urine HCG if indicated.

Box 1. How to be reasonably certain that a patient is not pregnant

A health care provider can be reasonably certain that a person is not pregnant if there are no symptoms or signs of pregnancy and they meet any one of the following criteria:

• Is ≤7 days after the start of normal menses
• Has not had sexual intercourse since the start of last normal menses
• Has been correctly and consistently using a reliable method of contraception
• Is ≤7 days after spontaneous or induced abortion
• Is within 4 weeks postpartum
• Is fully or nearly fully breastfeeding/chestfeeding (exclusively or the vast majority [>85%] of feeds), amenorrheic, and <6 months postpartum

Abbreviations: U.S. MEC = U.S. Medical Eligibility Criteria for Contraceptive Use; U.S. SPR = U.S. Selected Practice Recommendations for Contraceptive Use.

Modified from the U.S. CDC Selected Practice Recommendations 2016 [9].

References

1. Bogduk, N., On understanding reliability for diagnostic tests. Interv Pain Med,
   2022. 1: p. 100124.
2. Cole, L.A., The hCG assay or pregnancy test. Clin Chem Lab Med, 2012. 50: p.
   617-30.
3. Cortes-Charry, R., Corredor, N., Fernandez, J., Salazar, A., Rodriguez, L., and
   Fargione, M., Determining the time required to achieve negative human chorionic
   gonadotropin value after a nonmolar pregnancy: preliminary results. J Reprod
   Med, 2014. 59: p. 209-12.
4. Raymond, E.G., Anger, H.A., Chong, E., Haskell, S., Grant, M., Boraas, C., et al.,
   &quot;False positive&quot; urine pregnancy test results after successful medication abortion.
   Contraception, 2021. 103: p. 400-403.
5. Bolstad, N., Warren, D.J., and Nustad, K., Heterophilic antibody interference in
   immunometric assays. Best Pract Res Clin Endocrinol Metab, 2013. 27: p. 647-
   61.
6. Demir, A.Y., Musson, R.E., Schols, W.A., and Duk, J.M., Pregnancy, malignancy
   or mother nature? Persistence of high hCG levels in a perimenopausal woman.
   BMJ Case Rep, 2019. 12.
7. Ostreni, I., Colatosti, A., Basile, E.J., and Rafa, O., Elevated Beta-Human
   Chorionic Gonadotropin in a Non-pregnant Female With Altered Kidney Function.
   Cureus, 2022. 14: p. e23747.
8. Baldwin, M.K. and Jensen, J.T., Contraception during the perimenopause.
   Maturitas, 2013. 76: p. 235-42.
9. Curtis, K.M., Nguyen, A.T., Tepper, N.K., Zapata, L.B., Snyder, E.M., Hatfield-
   Timajchy, K., et al., U.S. Selected Practice Recommendations for Contraceptive
   Use, 2024. MMWR Recomm Rep, 2024. 73: p. 1-77.