Clin Chem Lab Med 2011;49(6):1029–1032  2011 by Walter de Gruyter • Berlin • Boston. DOI 10.1515/CCLM.2011.158 2010/535 Article in press - uncorrected proof Short Communication A case of consistent discrepancies between urine and blood human chorionic gonadotropin measurements Arjan Albersen 1, *, Evelien Kemper-Proper 1 , Marc H.M. Thelen 2 , Noush A. Kianmanesh Rad 3 , Robert F. Hoedemaeker 4 and Lianne S.M. Boesten 1 1 Department of Clinical Chemistry, IJsselland Ziekenhuis, Capelle aan den IJssel, The Netherlands 2 Department of Clinical Chemistry, Amphia Ziekenhuis, Breda, The Netherlands 3 Department of Gynecology, IJsselland Ziekenhuis, Capelle aan den IJssel, The Netherlands 4 PATHAN Foundation, Rotterdam, The Netherlands Abstract Background: Our laboratory was confronted with two suc- cessive urine samples from a single patient which tested pos- itive for human chorionic gonadotropin (hCG) when tested with both qualitative and quantitative assays, combined with no detectable hCG in corresponding plasma samples. Methods: Serial dilution and recovery experiments were per- formed in order to investigate the presence of interfering substances or a high-dose hook effect. The ovarian cysts that were removed from this patient were immunohistochemically stained using polyclonal anti-human hCG antibodies. Fur- thermore, a urine sample was sent to the USA hCG Refer- ence Service for hCG variant analysis. Results: Dilution and recovery experiments in urine and plasma samples were unremarkable. The biopsy stained neg- ative for human hCG and free b-subunit. hCG isoform anal- ysis in the urine sample revealed that approximately 87.5% of the immunoreactive hCG lacked the b-subunit C-terminal peptide (CTP). Conclusions: We report a rare case in which two successive urine samples test positive for hCG whereas in corresponding plasma samples hCG is undetectable. The majority of the total hCG contained a degraded form of b-subunit that lacks the CTP. This hCG variant, possibly of pituitary origin, is thought to have an extreme fast clearance rate possibly explaining the discordance between the hCG results in urine and plasma samples. Keywords: chorionic gonadotropin; false-positive reactions; point-of-care systems; pregnancy tests. *Corresponding author: Arjan Albersen, Department of Clinical Chemistry (AKL), IJsselland Ziekenhuis, Post box 690, 2900 AR Capelle a/d IJssel, The Netherlands Phone: q3110 2583139, Fax: q3110 2585393, E-mail: aalbersen@ysl.nl Received September 15, 2010; accepted January 9, 2011; previously published online March 17, 2011 Qualitative detection of human chorionic gonadotropin (hCG) in urine is a common laboratory procedure for ruling out (ectopic) pregnancy in women of childbearing age with vaginal bleeding and/or abdominal pain. We report a case where two consecutive urine hCG point-of-care tests (hCG- POCT) were positive without detectable hCG in correspond- ing plasma samples. The patient, a 41-year-old woman (para 3), presented with vaginal blood loss and was hospitalized. Her history revealed irregular vaginal blood loss for the past 1.5 months after removal of an intrauterine contraceptive device (Mirena  , Bayer Schering Pharma Oy, Finland), and for which her for- mer physician prescribed progestagen (medroxyprogeste- rone, dd 10 mg P.O.). Serum, plasma and urine samples (sample 1) were obtained for routine laboratory diagnostics, and a urine hCG-POCT (QuickVueqOne-step hCG combo  , Quidel Corp., San Diego, CA, USA) was performed. A quan- titative plasma hCG (hCGqb assay) was measured (Mod- ular, Roche Diagnostics, Rotkreuz, Switzerland) to confirm the positive urine hCG-POCT. hCG concentrations in plasma were not detectable (-0.1 IU/L). Three hours later, another plasma and urine sample (sample 2) from the same patient arrived at the laboratory. Once more, the urine hCG-POCT was positive and no hCG could be detected in plasma, con- firming the discrepant results between urine and plasma hCG analyses. Sonography revealed a thin endometrium and no gestational sac in the uterus. Estrogens were prescribed (2.0 mg dd. P.O.), the bleeding ceased and the patient was discharged from the hospital. She returned 3 days later for another sonography which revealed a cyst in right (2.7 cm=3.6 cm) and left (1.9 cm=3.1 cm) ovary. Urine and blood samples were collected, and both the urine hCG- POCT and the quantitative plasma hCG assays were nega- tive. Nine days later, both ovarian cysts were removed by laparoscopy. During this procedure no abnormalities were observed other than the two excised cysts. During the first hospitalisation, the laboratory was con- fronted with two independent consecutive hCG discrepancies between plasma and urine. To rule out an accidental urine sample mix-up, we compared the urine test-strip results between samples 1 and 2 (Table 1). All analytes measured were comparable, making an accidental switch in urine sam- ples unlikely. In both cases, blood samples were collected by the same laboratory technician and showed unremarkable results, apart from the absence of hCG. Quantitative analyses of the urine samples on the Roche platform showed hCG concentrations of 519 IU/L and 46 IU/L for urine samples 1 and 2, respectively. In contrast, the serum samples showed negative results using the hCG-POCT (suit- Brought to you by | University of Georgia Libraries Authenticated Download Date | 5/27/15 11:03 PM