20060044.883_887.tp

Clin Chem Lab Med 2006;44(7):883–887 ᮊ 2006 by Walter de Gruyter • Berlin • New York. DOI 10.1515/CCLM.2006.160 Establishment of detailed reference values for luteinizing
hormone, follicle stimulating hormone, estradiol, and
progesterone during different phases of the menstrual cycle
on the Abbott ARCHITECTᮋ analyzer
Reto Stricker1,*, Raphael Eberhart1, Marie-
Normal reproductive endocrine function involves a Christine Chevailler1, Frank A. Quinn2, Paul
wide variety of hormones controlled by a number of Bischof3 and Rene´ Stricker1
intricate feedback mechanisms (1). This is particularlytrue during the menstrual cycle, in which serum levels of luteinizing hormone (LH), follicle-stimulating hor- 2 Abbott Diagnostics, Abbott Park, IL, USA mone (FSH), estradiol, and progesterone follow a 3 Laboratoire d’Hormonologie Maternite´, Department cyclical pattern closely coordinated by the hypothal- of Obstetrics and Gynaecology, University of mic-pituitary-gonadal axis (1, 2). Because LH, FSH, estradiol, and progesterone play such a central rolein female reproductive function, measurement of theserum concentration of these hormones is important Abstract
in the diagnosis, treatment, and monitoring of manydifferent conditions (1). According to the Clinical and During a normal menstrual cycle, serum levels of Laboratory Standards Institute (CLSI; formerly the luteinizing hormone (LH), follicle-stimulating hor- National Committee for Clinical Laboratory Standard- mone (FSH), estradiol, and progesterone can vary ization, NCCLS), ‘‘for a decision making process to widely between cycles for the same woman, as well occur, reference values are needed for all tests in the as between different woman. Reliable reference val- clinical laboratory’’ (3). Assay manufacturers typically ues based on the local population are important for provide reference range data for their assays. Ideally, correct interpretation of laboratory results. The pur- laboratories should either verify these ranges or pose of our study was to determine detailed reference determine their own reference ranges, based on the values for these hormones throughout the menstrual populations they serve. However, performing detailed cycle using the Abbott ARCHITECT system. From 20 reference range studies can be logistically complex, volunteers (age 20–36 years) with normal cycles andno use of oral contraceptives, samples were taken and are beyond the capabilities of many clinical every day during their cycle. Volunteers received laboratories. This is especially true for reproductive three vaginal ultrasound examinations (days 10 and hormones, for which serum concentrations vary sig- 13, and 1 or 2 days after ovulation) to measure follic- nificantly throughout the menstrual cycle, as well as ular and corpus luteum development. Hormone levels from woman to woman, and collection of multiple were measured using the corresponding ARCHITECT samples across many days is required to establish assay and were synchronized to the LH peak. Median, reference values with any degree of resolution (4).
and 5th and 95th percentile values were determined The purpose of our study was to obtain detailed ref- for each day of the cycle, as well as for early follicular erence values for LH, FSH, estradiol, and progester- (days y15 to y6), late follicular (days y5 to y1), LH one for each day and phase of the menstrual cycle peak (day 0), early luteal (q1 to q4), mid-luteal (days q5 to q9), and late luteal (days q10 to q14) phases From 20 apparently healthy female volunteers (age of the cycle. Based on our data, we were able to range 20–36 years) with normal menstrual cycles and establish detailed reference values for LH, FSH, estra- no use of oral contraceptives or other medications, diol, and progesterone, which should aid in the inter- serum samples were collected on every day of the pretation of results for these reproductive hormones cycle. Each sample was then assayed for LH, FSH, estradiol, and progesterone using the Abbott ARCHI- Park, IL, USA). Reagent list numbers for these assays Keywords: ARCHITECT analyzer; estradiol; follicle-
were: LH, 6C25; FSH, 6C24; estradiol, 2K25; and pro- stimulating hormone; luteinizing hormone; progester- gesterone, 6C26. The ARCHITECT i2000 paramagnetic microparticles and chemiluminescentdetection technology (5). Assays were performed *Corresponding author: Reto Stricker, DIANALAB SA, 6 ruede la Colline, Geneva, Switzerland according to the manufacturer’s instructions. Analyt- ical performance characteristics (e.g., imprecision, Stricker et al.: Hormone reference values Reference range data for LH and FSH by day in normal cycling women.
sensitivity, method comparison) for these assays n-19, 5th and 95th percentile values are reported for have previously been reported (6–9). Per the manu- illustrative purposes only, and these data should be facturer, the assays are standardized accordingly: LH, interpreted with caution. Data were analyzed using WHO 2nd IS 80/552; FSH, WHO 2nd IRP 78/549; estra- Microsoft Excel 2000 (Microsoft, Redmond, WA, diol, gravimetric, ID-GCMS verified; progesterone, USA). The study protocol was reviewed and approved gravimetric, USP grade progesterone. For analysis of by the internal institutional Review Board.
serum hormone concentrations, variations in the Data for normal cycling women by day of cycle, cycle length were normalized by defining ‘‘day 0’’ with median, and 5th and 95th percentiles, are shown based on the peak LH value for each woman. To in Table 1 and represented graphically in Figure 1.
ensure a normal cycle was taking place, volunteers Reference values for different phases of the menstrual received three vaginal ultrasound examinations (days cycle are shown in Table 2. During the early follicular 10 and 13, and 1 or 2 days post-ovulation) to measure phase (days y15 to y6) median, and 5th and 95th follicular and corpus luteum development. Ovulation percentile values were: LH, 3.96 (2.01 and 7.80) IU/L; occurred in all cycles used for data analysis. Owing FSH, 6.42 (3.91 and 8.76) IU/L; estradiol, 149.74 (77.99 to the highly specialized nature of the sample set and 266.08) pmol/L; and progesterone, 0.64 (0.32 and required to perform this study, obtaining a large num- 1.91) nmol/L. For the late follicular phase (days y 5 ber of specimens from many different women repre- to y1), median, and 5th and 95th percentile values senting each day of their menstrual cycle was were: LH, 5.68 (3.24 and 14.68) IU/L; FSH, 4.66 (2.97 logistically challenging. In recommendations for the and 6.99) IU/L; estradiol, 450.49 (195.43 and 1146.91) determination of reference values, the CLSI recogniz- pmol/L; and progesterone, 0.64 (0.32 and 1.59) es that collecting large numbers of certain sample nmol/L. During the LH peak (defined as day 0), medi- types may be ‘‘difficult, if not impossible’’; in such an, and 5th and 95th percentile values were: LH, 41.19 cases, their guidance is to report ‘‘percentiles appro- (21.16 and 73.63) IU/L; FSH, 12.82 (5.30 and 15.07) priate to the number of values obtained’’ (3). In the IU/L; estradiol, 671.06 (482.00 and 1425.39) pmol/L; present study, reference values were defined by cal- and progesterone, 2.54 (1.24 and 4.13) nmol/L. For the culating the median, and 5th and 95th percentiles in early luteal phase (days q1 to q4), median, and 5th sample sets of nG19, the minimum number needed and 95th percentile values were: LH, 7.86 (3.65 and to define these percentiles (3). For sample sets with 18.03) IU/L; FSH, 5.62 (3.02 and 11.09) IU/L; estradiol, Stricker et al.: Hormone reference values Reference range data for estradiol and progesterone by day in normal cycling women.
Hormone values in daily serum samples across the menstrual cycle for normal women: (A) luteinizing hormone; (B) follicle-stimulating hormone; (C) estradiol; and (D) progesterone. Solid lines represent median values; dotted lines rep-resent 5th and 95th percentiles.
Stricker et al.: Hormone reference values Reference values during different phases of the menstrual cycle.
313.42 (178.14 and 566.43) pmol/L; and progesterone, important for physicians and laboratories to be aware 13.67 (3.15 and 39.65) nmol/L. During the mid-luteal of the potential sources of differences that exist phase (days q5 to q9), median, and 5th and 95th per- between immunoassay methodologies, and confirm centile values were: LH, 3.88 (1.05 and 9.73) IU/L; FSH, that the reference values they use for reproductive 3.14 (1.48 and 5.31) IU/L; estradiol, 495.82 (275.95 and hormones are valid for the populations they serve.
761.67) pmol/L; and progesterone, 36.25 (21.21 and Based on our study, we were able to develop detailed 54.28) nmol/L. For the late-luteal phase (days q10 to reference ranges for LH, FSH, estradiol, and proges- q14), median, and 5th and 95th percentile values terone on the Abbott ARCHITECT analyzer during the were: LH, 2.75 (0.51 and 6.98) IU/L; FSH, 2.85 (1.14 and normal menstrual cycle. These data should aid labo- 5.64) IU/L; estradiol, 327.36 (100.52 and 787.14) ratories and physicians in the interpretation of results pmol/L; and progesterone, 13.99 (1.96 and 49.18) for these reproductive hormones in a wide variety of In general, our data are in agreement with those provided by the assay manufacturer. However, directcomparison is difficult, as our data were analyzed to References
a higher degree of resolution, allowing establishmentof reference values in greater detail throughout the 1. Gronowski AN, Landau-Levine ME. Reproductive endo- menstrual cycle. It should be noted that gonadotropin crine function. In: Burtis CA, Ashwood ER, editors. Tietz hormones are known to be highly heterogeneous (10, fundamentals of clinical chemistry. Philadelphia: WB 11). This heterogeneity is presumably one of the fac- tors that leads to assay differences reported in the 2. Marshall JC. Hormonal regulation of the menstrual cycle literature (12). These differences can, in turn, affect and mechanisms of ovulation. In: DeGroot LJ, Jameson interpretation of results from different assay manu- JL, editors. Endocrinology. Philadelphia, PA: WB Saun- facturers or laboratories. For example, FSH concen- 3. Clinical and Laboratory Standards Institute. How to define tration on day 3 of the menstrual cycle (along with and determine reference intervals in the clinical labora- estradiol and inhibin B) is commonly used by physi- tory; approved guideline, 2nd ed. Document c28-A2. Vil- cians to evaluate ovarian reserve. Taieb and col- leagues, in their study of six different immunoassays 4. Dighe AS, Moy JM, Hayes FJ, Sluss PM. High-resolution for FSH, noted statistically significant differences for reference ranges for estradiol, luteinizing hormone, and day-3 FSH values, depending on which immunoassay follicle stimulating hormone in men and women using theAxSYM assay system. Clin Biochem 2005;38:175–9.
was used. Their data prompted them to recommend ‘‘that it is advisable to refer patients to selected lab- Wild D, editor. The immunoassay handbook, 3rd ed.
oratories using analytical methods for which they have defined reference values «’’ (12). Similarly, 6. Yang DT, Owen WE, Ramsay CS, Xie H, Roberts WL. Per- measurement of serum estradiol is important in eval- formance characteristics of eight estradiol immunoas- uating a variety of conditions, including abnormal says. Am J Clin Pathol 2004;122:332–7.
7. Hendriks HA, Kortlandt W, Verweij WM. Analytical per- menstrual cycles. However, measurement of steroid formance comparison of five new generation immuno- hormones (such as estradiol) by immunoassay is assay analyzers. Ned Tijdschr Klin Chem 2000;25:170–7.
notoriously difficult, and differences in assay perform- 8. Verheecke P, Quinn FA. Laboratory evaluation of repre- ance significant enough to affect clinical utility have sentative disease state assays on the Abbott ARCHITECTᮋ been reported (6). For these reasons, it is especially i2000ᮋ analyzer. J Assoc Lab Automat 2000;5:30–3.
Stricker et al.: Hormone reference values 9. Rufo G, Brookhart P, Barnes W, Frels M, Sheu M, White gonadotropins through the normal menstrual cycle and M, et al. Performance characteristics of the Abbott in the post-menopause state. Mol Hum Reprod 1998;47: ARCHITECT progesterone assay wabstractx. Clin Chem 12. Taieb J, Olivennes F, Birr AS, Benattarc C, Righini C, 10. Stanton PG, Burgon PG, Hearn MT, Robertson DM.
Frydman R, et al. Comparison of day 3 FSH values as Structural and functional characterization of hFSH and determined by six different immunoassays. Hum Reprod hLH isoforms. Mol Cell Endocrinol 1996;125:133–41.
11. Anobile CJ, Talbot JA, McCann SJ, Padmanabhan V, Received February 1, 2006, accepted April 18, 2006

Source: https://www.dianalabs.ch/documents/ajouts/Hormones.pdf