Correspondence

A False Negative Pregnancy Test in a Patient with a Hydatidiform Molar Pregnancy

N Engl J Med 2003; 349:2172-2173November 27, 2003

Article

To the Editor:

Determination of pregnancy status is vital in the workup of a patient presenting with vaginal bleeding or lower abdominal pain. We report a case of false negative results on urine and serum pregnancy tests as a result of the “high-dose hook effect.”

A 26-year-old woman (gravida 1, para 1) presented to the emergency department, reporting lower abdominal pain, vomiting, and several weeks of irregular vaginal bleeding. On examination, she was anxious and in moderate discomfort, and her heart rate and blood pressure were elevated. Her abdomen was soft, with a tender lower abdominal mass. Pelvic examination revealed a cervical os that was open 2 cm, without active bleeding, and a pelvic mass the size of a 23-week-old fetus. Results of urine and serum tests for the beta subunit of human chorionic gonadotropin (β-hCG) were negative. Ultrasonography revealed a heterogeneous, intrauterine mass with cystic components, a finding consistent with molar pregnancy (Figure 1Figure 1Ultrasonographic Image Revealing a Heterogeneous Intrauterine Mass with a Characteristic Vesicular Pattern, a Finding Consistent with the Presence of a Hydatidiform Mole.). The laboratory was informed of the preliminary diagnosis, and after multiple dilutions, repeated testing of the serum showed a β-hCG level of 1.5 million mIU per milliliter. The patient immediately underwent dilation and curettage. Pathological examination revealed a complete hydatidiform mole.

Gestational trophoblastic disease occurs in 0.75 to 1.1 of every 1000 pregnancies in the United States.1 Early identification may prevent potentially serious complications, including genital tract hemorrhage, cardiac and ventilatory failure, and central nervous system dysfunction.2 Previous reports have described falsely low β-hCG levels in patients with a hydatidiform mole2-4; here, we report a serum test with a negative result.

Current urine and serum pregnancy tests use antibodies directed against β-hCG for immunologic identification. Most use an immunometric, or sandwich, assay in which two antibodies, directed against different parts of the beta chain, sandwich the antigen, resulting in a positive test for pregnancy. Falsely low or false negative results can occur when an extremely high level of substrate overwhelms the assay system. This high-dose hook effect has been documented with immunoassays for β-hCG, ferritin, prostate-specific antigen, prolactin, thyrotropin, and CA-125.5 These inaccurate results can be overcome by diluting the serum or urine sample.4

A falsely low β-hCG result in gestational trophoblastic disease may affect assessments of the response to therapy, recurrence, or the persistence of disease. A false negative β-hCG result may also delay the diagnosis. Fortunately, in the current case, other findings led to prompt evaluation and diagnosis.

Dilution of the serum or urine sample in the appropriate setting can prevent delays and complications from a falsely low or negative pregnancy test. Clinicians should understand the possibility of inaccurate results and should know how to respond.

Jeffrey A. Tabas, M.D.
University of California, San Francisco, School of Medicine, San Francisco, CA 94143
jtabas@itsa.ucsf.edu

Matthew Strehlow, M.D.
Stanford University School of Medicine, Stanford, CA 94304

Eric Isaacs, M.D.
University of California, San Francisco, School of Medicine, San Francisco, CA 94143

5 References

1. 1

   Hayashi K, Bracken MB, Freeman DH Jr, Hellenbrand K. Hydatidiform mole in the United States (1970-1977): a statistical and theoretical analysis. Am J Epidemiol 1982;115:67-77
   Web of Science | Medline

2. 2

   Flam F, Hambraeus-Jonzon K, Hansson LO, Kjaeldgaard A. Hydatidiform mole with non-metastatic pulmonary complications and a false low level of hCG. Eur J Obstet Gynecol Reprod Biol 1998;77:235-237
   CrossRef | Web of Science | Medline

3. 3

   Wisdom SJ, Irvine GA, Gray C, Dryburgh F, Greer IA. Moles and sandwiches. Br J Obstet Gynaecol 1994;101:455-456
   CrossRef | Medline

4. 4

   Wheeler CA, Davis S, Degefu S, Thorneycroft IH, O'Quinn AG. Ovarian choriocarcinoma: a difficult diagnosis of an unusual tumor and a review of the hook effect. Obstet Gynecol 1990;75:547-549
   Web of Science | Medline

5. 5

   Wolf BA, Garrett NC, Nahm MH. The “hook effect“: high concentrations of prostate-specific antigen giving artifactually low values on one-step immunoassay. N Engl J Med 1989;320:1755-1756
   Web of Science | Medline

Citing Articles (2)

Citing Articles

1. 1

   Christine E. Richter, Natalia Buza, Pei Hui, Dan-Arin Silasi, Masoud Azodi, Alessandro D. Santin, Thomas J. Rutherford, Peter E. Schwartz. (2011) Urinary hCG Screening in the Gynecologic Oncology Population. Journal of Gynecologic Surgery 27:3, 143-146
   CrossRef

2. 2

   Susan B. Promes, Flavia Nobay. (2010) Pitfalls in First-Trimester Bleeding. Emergency Medicine Clinics of North America 28:1, 219-234
   CrossRef