A 52 year-old female with an elevated serum β-hCG
Morgan Darrow, MD and Danyel Tacker, PhD
A physician in the Emergency Department calls because there is a 52 year-old female patient with a screening quantitative serum β-human chorionic gonadotropin (β-hCG) level of 5 mIU/mL (normal is <5 mIU/mL; 5-25 mIU/mL is considered indeterminate). This patient came to the ED for an allergic reaction. She had experienced shortness of breath, throat swelling, and seizure-like activity upon accidental exposure to latex paint. Upon arrival to the ED, she was confused, somnolent, and unable to clearly answer questions about her history. According to the electronic medical record, her history includes seizures, depression, alcohol abuse, chronic obstructive pulmonary disease, and asthma. She also has a history of breast cancer; a full-body PET scan three months ago showed no areas of abnormal uptake. Her current medications include oxycodone, clonazepam, propranolol, sertraline, and trazodone. Due to a concern for head trauma, the ED staff wants to perform a CT scan. However, they are uncertain about the significance of the patient’s hCG result and are reluctant to proceed without consulting the lab.
The quantitative serum β-hCG is repeated and produces the same result. A follicle-stimulating hormone (FSH) level is ordered urgently and returns as 52.5 mIU/mL (reference intervals: 3.0-8.1 mIU/mL (follicular phase), 2.6-16.7 mIU/mL (periovulatory phase), 1.4-5.5 mIU/mL (luteal phase), and 26.7-133.4 mIU/mL (postmenopausal)).
Please select an answer above.
Screening pregnancy tests are almost always performed on women of reproductive age (including older patients who may be peri- or post-menopausal) prior to the initiation of medical interventions that could potentially harm a previously undetected pregnancy. While the vast majority of the results from these screening tests are straightforward to interpret, indeterminate results are occasionally encountered and pose a particular challenge in peri-menopausal patients. The prevalence of indeterminate hCG levels in women aged 41-55 has been estimated at 0.2-0.4% [1, 2]. The ambiguity associated with these results may lead to diagnostic uncertainty, confusion, and delays in care. In addition, emergency situations, such as the one encountered in this case, necessitate the rapid and accurate determination of the patient’s pregnancy status in order to proceed with the desired work-up or treatment plan.
The intended use of the standard screening test is to diagnose pregnancy by operating on the assumption that the measured hCG is being produced by placental trophoblastic tissue (whether from a normal pregnancy, an ectopic pregnancy, or gestational trophoblastic disease). However, the screening test is actually non-specific with respect to pregnancy and will also detect hCG of non-placental origin, such as the hCG secreted by some tumors (e.g. germ cell tumors) or by the pituitary [2, 3]. Clinicians who are not aware of pituitary hCG may assume that the patient is either pregnant or has an hCG-secreting tumor, leading to unnecessary and potentially harmful interventions.
The gonadotrope cells in the anterior pituitary normally secrete low levels of hCG in both men and women (typically <10mIU/mL) . This secretion has been shown to increase gradually with age in non-pregnant women . As with FSH and luteinizing hormone (LH), gonadotrope secretion of hCG is stimulated by gonadotropin releasing hormone (GnRH) and suppressed by estrogen and progesterone . Declining ovarian function in peri-menopausal women, which results in the loss of negative feedback on GnRH secretion, leads to this increased pituitary secretion of hCG, as well as FSH and, to a lesser extent, LH .
Given that FSH levels should be increased in peri-menopausal patients with pituitary hCG but not in pregnant patients, the authors of a recent study chose to evaluate FSH as a means to rule out pregnancy . They took a set of 100 patients aged 41-55 with indeterminate hCG levels (5.0-14.0 mIU/mL in this study), performed FSH testing, conducted chart reviews, and followed up to determine which of the patients were actually pregnant. They found that none of the pregnant patients had an FSH level greater than 45 mIU/mL at the time of their indeterminate hCG result. These results indicate that FSH testing can assist in determining the pregnancy status of a patient and in identifying possible cases of pituitary hCG. Alternatively, in non-urgent cases, confirmation of pituitary hCG could be achieved by observing suppression in response to treating the patient with a two-week trial of estrogen-progesterone replacement therapy .
Another consideration in this patient is her use of antidepressant medications. Many psychotropic medications, including antidepressants and antipsychotics, are capable of affecting pituitary activity through a variety of mechanisms. While there are no reports of increased hCG in response to pharmacologic action on the pituitary, it has been well-documented that some of these medications are capable of acting directly on the pituitary to increase secretion of other hormones, such as prolactin [6,7,8,9], folliculotropin , adrenocorticotropin , and thyrotropin . As such, it stands to reason that medication-induced pituitary secretion of hCG is possible and that this relatively unexplored hypothesis warrants further investigation.
- Gronowski AM, et al. 2008. “Use of Serum FSH to Identify Perimenopausal Women with Pituitary hCG.” Clin Chem 54(4): 652–656.
- Snyder JA, et al. 2005. “Diagnostic Considerations in the Measurement of Human Chorionic Gonadotropin in Aging Women.” Clin Chem 51(10): 1830–1835.
- McCudden CR, Willis MS, Grenache DG. 2008. “Persistent Low Concentration of Human Chorionic Gonadotropin in a Nonpregnant Woman.” Clin Chem 54(1): 209–214.
- Braunstein GD. 2002. “False-positive serum human chorionic gonadotropin results Causes, characteristics, and recognition.”Am J Obstet Gynecol 187: 217-24.
- Cole LA, Sasaki Y, Muller CY. 2007. “Normal Production of Human Chorionic Gonadotropin in Menopause.” N Engl J Med 356(11): 1184-1186.
- La Torre D, Falorni A. 2007. “Pharmacological Causes of Hyperprolactinemia.” Ther Clin Risk Manag 3(5): 929-951.
- Rossi L, Bonuccelli U, Maracacci G, Bindi A, De Scisciolo G, Arena R. 1983. Gynecomastia in epileptics treated with phenobarbital, phenytoin and fluoresone: two case reports. Ital J Neurol Sci 2:207-210.
- Sahota P, Prabhakar S, Kharbanda PS, Bhansali A, Jain V, Prasad Das C, Modi M. 2008. Seizure type, antiepileptic drugs, and reproductive endocrine dysfunction in Indian women with epilepsy: A cross-sectional study. Epilepsia 49(12):2069-2077.
- Jain KK. Drug-Induced Neurological Disorders. 3rd Ed. Hogrefe: Ashland OH. 2012. Electronic media accessed 29 April 2015, books.google.com.