Evaluation of menstrual cycle-related changes in 85 clinical laboratory analytes

Abstract

Objective

The menstrual cycle-related changes in clinical laboratory values were analysed by use of data obtained in the Asian multicentre study aimed at derivation of common reference intervals for 85 major clinical laboratory tests.

Methods

Among 1876 healthy female volunteers, 893 had regular menstruation. They were classified into five groups according to dates between sample collection and the start of the last menstrual cycle: early follicular phase (1–6 days), late follicular phase (7–12 days), ovulatory phase (13–16 days), early luteal phase (17–22 days), and late luteal phase (23–31 days). Multiple linear regression analysis was performed to evaluate the menstrual cycle-related changes in test results. The magnitude was expressed as a standard deviation ratio of between-phase standard deviation to between-individual standard deviation based on nested ANOVA.

Results

Aside from obvious changes for four sex hormones (oestradiol, progesterone, follicle-stimulating hormone, and luteinizing hormone), we observed statistically significant menstrual cycle-related changes in the following tests (standard deviation ratio >0.15): Na, Cl, creatine kinase, C-reactive protein, serum amyloid A, carbohydrate antigen 125, and parathyroid hormone were higher during the early follicular phase, while insulin, total cholesterol, and white blood cell were higher during the luteal phase. Significant associations of those test items with the four sex hormones were revealed.

Conclusions

The menstrual cycle-related changes in laboratory test results were revealed in some commonly tested items other than sex hormones. The findings are of interest in understanding female physiology in relation to hormonal changes, but the magnitude of changes is rather small and not very relevant in interpreting test results.

Keywords

Sex hormones three-level nested ANOVA multiple regression analysis parathyroid hormone total cholesterol C-reactive protein serum amyloid A

Introduction

In interpreting laboratory test results, it is important to consider biological sources of variations (SV) before considering pathological causes of changes in values.

Biological SVs are grouped into within-individual and between-individual SV. The former includes a host of factors such as posture, physical or emotional stress, and food intake, plus the menstrual cycle in female. Although menstrual cycle-related changes are well known for sex hormones,¹ there have been no clear-cut reports about other commonly measured laboratory tests. Munford reported in 2010 that serum total cholesterol, LDL-cholesterol, and triglycerides increase during the follicular phase while high-density lipoprotein (HDL)-cholesterol increases at the mid-cycle.² Garskins showed in 2012 that C-reactive protein (CRP) by high-sensitive assays is higher in females during the follicular phase.³ In 2014, Schisterman published a review analysing menstrual cycle-dependent changes in various cardiometabolic biomarkers. However, the literature is somewhat contradictory about the magnitude and direction of the changes.¹

In 2009, we conducted a large-scale multicentre study for the derivation of common reference intervals (RI) in East and Southeast Asia, targeting 85 major laboratory tests including health screening items (general chemistry and complete blood count (CBC)), inflammatory markers, tumour markers, major hormones, and vitamins.^4,5 Prior to donating blood, each volunteer answered a detailed questionnaire which included the date of onset of the last menstrual period.

In this study, we comprehensively analysed menstrual cycle-related changes for all the test items by use of multivariate analysis and discuss the implication of the findings in interpreting test results in females.

Materials and methods

Source data

The International Federation of Clinical Chemistry and Laboratory Medicine, the Committee on Reference Intervals and Decision Limits, and the Asian and Pacific Federation of Clinical Biochemistry, along with the Japan Society of Clinical Chemistry and the Japanese Society of Laboratory Medicine, conducted a multicentre study for derivation of common RIs.^4,5 In this study, the results of 85 test items were analysed in relation to the time in the menstrual cycle and other SVs obtained from a questionnaire, which included age, sex, occupation, body height, body weight, body mass index (BMI), blood type, alcohol consumption, smoking status, exercise, physical activity concentration, menstrual cycle data, and food habits.

The volunteers participating in this study consider themselves to be healthy. Those volunteers who had any of the following were excluded: (1) BMI ≥ 28 kg/m², (2) excessive daily consumption of alcohol (≥75 g ethanol), (3) smoking >20 cigarettes/day, (4) regular medication for chronic diseases, (5) hospitalization for any acute disease or surgery within the past two weeks, or (6) known carrier status for HBV, HCV, or HIV. Those who met the study conditions were 3314 volunteers (1438 males and 1876 females) between 20 and 65 years old. After they had fasted for 10 h or more, blood was collected after having relaxed for more than 20 min in sitting position.

Since the aim of this study is to clarify the menstrual cycle-related changes in laboratory data, we applied additional exclusion criteria to the 1876 females based on answers to the query items regarding menstruation in the questionnaire (the last period, regularity of the cycle, and average duration of each cycle) as follows: (1) postmenopausal state, (2) irregular menstrual cycles, (3) duration of menstrual cycle of ≤24 or ≥32 days, (4) no answer to the date of the last period, or (5) ≥32 days after onset of the last period. By this selection process, total of 893 female volunteers were eligible for the study.

This study was approved by the Ethical Committee of Yamaguchi University Graduate School of Medical Sciences as well as by each institutional research board of all the collaborating laboratories. The main incentive for the volunteers was free provision of results for all 85 analytes. All volunteers signed informed consent forms.

Measurements

Blood was collected into a vacuum blood collection tube containing a serum-separating agent. After the blood was centrifuged for 10 min at room temperature, serum was aliquoted and cryopreserved at ≤ −80℃. All collected samples were transferred to Beckman Coulter, Inc., Ariake Research Laboratory, Tokyo, Japan, and measured collectively.

Electrolytes were assayed by indirect potentiometry or colorimetric assay and proteins by chromogenic assay, immunoturbidimetric assay, or latex immunoturbidimetric assay, and lipids and enzymes by enzymatic assay using Unicel DxC-800 (Beckman Coulter). Tumour makers, hormones, antibodies, oestradiol, progesterone, follicle-stimulating hormone (FSH), and luteinizing hormone (LH) were measured by chemiluminescent enzyme immunoassay using the Unicel DxI-800 (Beckman Coulter). CBCs were measured using a Coulter LH750 (Beckman Coulter) haematology analyser at each local laboratory when the analyser was available. Twenty-four out of 64 laboratories had this analyser, and the actual number of subjects who had results for CBC was 564 (Table 1).

Table 1.

List of test items and analytical methods.

Test Item		Unit	Method		Transf.	Test Item		Unit	Method		Transf.
Test Item		Unit	Assay	System/Reagents	Transf.	Test Item		Unit	Assay	System/Reagents	Transf.
Na	Sodium	mEq/L	Indirect potentiometry	Unicel DxC800		TRAP-5b	Tartrate-resistant acid phosphatase 5b	U/L	EIA	Nittobo medical
K	Potassium	mEq/L	Indirect potentiometry	Unicel DxC800		BoneALP	Bone alkaline phosphatase	mg/L	CLEIA	Unicel DxI800	Log
Cl	Chloride	mEq/L	Indirect potentiometry	Unicel DxC800		Fer	Ferritin	mg/L	CLEIA	Unicel DxI800	Log
Ca	Total serum calcium	mmol/L	Indirect potentiometry	Unicel DxC800		AFP	Alfa-fetoprotein	mg/L	CLEIA	Unicel DxI800	Log
IP	Inorganic phosphate	mmol/L	Molybdate UV	Unicel DxC800		CA15-3	Carbohydrate antigen 15-3	U/mL	CLEIA	Unicel DxI800	Log
Fe	Iron	μmol/L	Direct colorimetry	Unicel DxC800		CEA	Carcinoembryonic antigen	mg/L	CLEIA	Unicel DxI800	Log
UIBC	Unsaturated iron-binding capacity	mmol/L	Direct colorimetry	Unicel DxC800		CA19-9	Carbohydrate antigen 19-9	U/mL	CLEIA	Unicel DxI800	Log
Urea	Urea	mmol/L	Enzymatic	Unicel DxC800		CA125	Carbohydrate antigen 125	U/mL	CLEIA	Unicel DxI800	Log
CRE	Creatinine	μmol/L	Enzymatic	Unicel DxC800		PSA	Prostate-specific antigen	mg/L	CLEIA	Unicel DxI800	Log
UA	Urate	μmol/L	Enzymatic	Unicel DxC800		E2	Oestradiol	pmol/L	CLEIA	Unicel DxI800	Log
CysC	Cystatin C	ng/L	Latex immunoturbidimetric	Nittobo medical		FSH	Follicle-stimulating hormone	IU/L	CLEIA	Unicel DxI800	Log
TP	Protein, total	g/L	Buret	Unicel DxC800		LH	Luteinizing hormone	IU/L	CLEIA	Unicel DxI800	Log
Alb	Albumin	g/L	Bromocresol green	Unicel DxC800		Prog	Progesterone	nmol/L	CLEIA	Unicel DxI800	Log
Glob	Globulin	g/L	TP-Alb			Testo	Testosterone	nmol/L	CLEIA	Unicel DxI800
IgG	Immunoglobulin G	g/L	Immunoturbidimetric	Nittobo medical		DHEA-S	Dehydroepiandrosterone sulphate	mmol/L	CLEIA	Unicel DxI800	Log
IgA	Immunoglobulin A	g/L	Immunoturbidimetric	Nittobo medical		Cort	Cortisol	nmol/L	CLEIA	Unicel DxI800	Log
IgM	Immunoglobulin M	g/L	Immunoturbidimetric	Nittobo medical		PRL	Prolactin	mg/L	CLEIA	Unicel DxI800	Log
IgE	Immunoglobulin E	IU/mL	CLEIA	Unicel DxC800	Log	Insulin	Insulin	mIU/L	CLEIA	Unicel DxI800	Log
CRP	C-reactive protein	mg/L	Latex immunoturbidimetric	Nittobo medical	Log	PTH	Intact parathyroid hormone	ng/L	CLEIA	Unicel DxI800	Log
SAA	Serum amyloid A	mg/L	Latex immunoturbidimetric	Eiken medical	Log	TSH	Thyroid-stimulating hormone	mU/L	CLEIA	Unicel DxI800	Log
C3	Complement component 3	mg/L	Immunoturbidimetric	Unicel DxC800		FT4	Free thyroxine	pmol/L	CLEIA	Unicel DxI800
C4	Complement component 4	mg/L	Immunoturbidimetric	Unicel DxC800		FT3	Free triiodothyronine	pmol/L	CLEIA	Unicel DxI800
RBP	Retinol-binding protein	mg/L	Latex immunoturbidimetric	Nittobo medical		Tg	Thyroglobulin	mg/L	CLEIA	Unicel DxI800
TTR	Transthyretin (prealbumin)	mg/L	Immunoturbidimetric	Unicel DxC800		EPO	Erythropoietin	IU/L	CLEIA	Unicel DxI800	Log
Glu	Glucose	mmol/L	Enzymatic	Unicel DxC800		Fol	Folate	mg/L	CLEIA	Unicel DxI800	Log
TC	Total cholesterol	mmol/L	Enzymatic	Unicel DxC800		VitB12	Vitamin B12	ng/L	CLEIA	Unicel DxI800	Log
TG	Triglycerides	mmol/L	Enzymatic	Unicel DxC800	Log	ADP	Adiponectin	mg/L	Latex immunoturbidimetric	Ostuka Phama	Log
HDL-C	HDL-cholesterol	mmol/L	Enzymatic	Unicel DxC800		PG1	Pepsinogen I	mg/L	Latex immunoturbidimetric	Eiken medical	Log
LDL-C	LDL-cholesterol	mmol/L	Enzymatic	Unicel DxC800		PG2	Pepsinogen II	mg/L	Latex immunoturbidimetric	Eiken medical	Log
AST	Asparatate transferase	U/L	IFCC Recommended	Unicel DxC800		WBC	White blood cell	10⁸/L		Coulter LH750
ALT	Alanine transferase	U/L	IFCC Recommended	Unicel DxC800		Neu	Neutrophilic	%		Coulter LH750
ALP	Alkaline phosphatase	U/L	IFCC Recommended	Unicel DxC800		Lym	Lymphocyte	%		Coulter LH750
AMY	Amylase	U/L	IFCC Recommended	Unicel DxC800		Mon	Monocyte	%		Coulter LH750
LIP	Lipase	U/L	Enzymatic	Unicel DxC800		Eos	Eosinophilic	%		Coulter LH750	Log
GGT	Gamma-glutamyl transpeptidase	U/L	IFCC Recommended	Unicel DxC800	Log	Bas	Basophilic	%		Coulter LH750
LDH	Lactate dehydrogenase	U/L	IFCC Recommended	Unicel DxC800		RBC	Red blood cell	10¹⁰/L		Coulter LH750
CK	Creatine kinase	U/L	IFCC Recommended	Unicel DxC800	Log	Hb	Haemoglobin	g/L		Coulter LH750
ASO	Antistreptolysin O	IU/mL	Immunoturbidimetric	Unicel DxC800	Log	Ht	Haematocrit	%		Coulter LH750
Tf	Transferrin	g/L	Immunoturbidimetric	Unicel DxC800		MCV	Mean corpuscular volume	fL		Coulter LH750
sTfR	Soluble transferrin receptor	nmol/L	CLEIA	Unicel DxC800	Log	MCH	Mean corpuscular haemoglobin	pg		Coulter LH750
ApoA1	Apolipoprotein AI	mg/L	Immunoturbidimetric	Unicel DxC800		MCHC	Mean corpuscular haemoglobin	%		Coulter LH750
ApoB	Apolipoprotein B	mg/L	Immunoturbidimetric	Unicel DxC800			concentration
ApoE	Apolipoprotein E	mg/L	Immunoturbidimetric	Nittobo medical		PLT	Platelet	10¹⁰/L		Coulter LH750

Transf.: Logarithmic transformation was made in analysing test results of the corresponding item.

Detailed information on the measurements including assay precision was described elsewhere^4,5 except for that of CBC, which was not included in the previous reports. Harmonization of CBC results among laboratories was achieved by use of a common calibrator (S-CAL) and the web-based monitoring system called Interlaboratory Quality Assurance Program (IQAP).⁶

Statistical analyses

On the basis of the number of days between the date of blood collection and the date of the last menstrual period as answered in the questionnaire, the menstrual cycle was classified into five phases as follows: early follicular phase (1–6 days), late follicular phase (7–12 days), ovulatory phase (13–16 days), early luteal phase (17–22 days), and late luteal phase (23–31 days).

Statistical significance of differences in test results among the five menstrual phase groups was evaluated by the Kruskal–Wallis test. Three-level nested ANOVA (3N-ANOVA) was used to examine the magnitude of possible influence of three major SVs: country (Japan, Korea, China, Taiwan, Hong Kong, Viet Nam, Malaysia, and Indonesia), age (≤30, 31–40, and ≥41 years), and menstrual cycle (five phases as shown above).

Standard deviations (SD) were computed for differences among countries (SD_cntr), among age groups (SD_age), among menstrual phases (SD_cycle), and the residual SD which corresponds to between-individual SD (SD_indiv) adjusted for the effect of country, age, and menstrual cycle.⁷⁻¹⁰ The standard deviation ratio (SDR) of each factor relative to the SD_indiv was computed as SDR_cntr, SDR_age, and SDR_cycle. SDR >0.3 was adopted as a guide to judge the need for partitioning reference values. Among the five groups, there were no significant differences in age, height, weight, BMI, and waist circumference.

Multiple linear regression analysis (MRA) was performed with each of the 85 test items set as a target variable. Five dummy variables representing each country (Taiwan and Hong Kong were merged into China) were set as explanatory variables, with Japan as the reference category, age, BMI, and four dummy variables representing each menstrual phase with early follicular phase set as reference category. In the second round of MRA analysis, dummy variables for the menstrual cycle were replaced by the four sex hormones (oestradiol, progesterone, FSH, and LH), keeping other explanatory variables in place. Because of the large data-set, the standardized partial regression coefficient (which corresponds to the partial correlation coefficient: r_p) was statistically significant at P = 0.01 when r_p was as small as 0.08. Therefore, we focused on the absolute value of r_p and judged it as significant only with |r_p| ≥ 0.15.

Test results of 33 items showed apparent skewness in the distribution, and thus they were logarithmically transformed to make the distribution closer to the Gaussian pattern (Table 1). The Kruskal–Wallis test, 3N-ANOVA, and MRA were performed by use of general purpose statistical software, StatFlex Ver. 6 (Artech Co., Ltd, Osaka, Japan).

Results

Kruskal–Wallis test for between-menstrual phase differences

We found significant between-menstrual phase differences in 17 test items among 85 by the Kruskal–Wallis test (P < 0.005), as shown in Table 2. These were Na, Cl, Glob, CRP, serum amyloid A (SAA), total cholesterol (TC), creatine kinase (CK), ferritin, carbohydrate antigen 125 (CA125), insulin, parathyroid hormone (PTH), PG1, white blood cell (WBC), oestradiol, progesterone, FSH, and LH.

Table 2.

Results of Kruskal–Wallis test (KW) and three-level nested ANOVA for menstrual cycle-related changes.

Test item	KW	Three-level nested ANOVA							Test item	KW	Three-level nested ANOVA
Test item	p	SD_cntr	SD_age	SD_cycle	SD_indiv	SDR_cntr	SDR_age	SDR_cycle	Test item	p	SD_cntr	SD_age	SD_cycle	SD_indiv	SDR_cntr	SDR_age	SDR_cycle
Na	0.0001	0.137	0.132	0.235	1.406	0.10	0.09	0.17	TRAP-5b	0.1293	0.216	0.000	0.000	0.701	0.31	0.00	0.00
K	0.0710	0.100	0.062	0.040	0.270	0.37	0.23	0.15	BoneALP	0.6918	0.431	0.174	0.512	2.700	0.16	0.06	0.19
Cl	0.0001	0.000	0.164	0.364	1.728	0.00	0.09	0.21	Fer	0.0017	7.226	0.000	1.878	18.600	0.39	0.00	0.10
Ca	0.0338	0.000	0.025	0.005	0.066	0.00	0.38	0.08	AFP	0.1213	0.134	0.287	0.235	1.089	0.12	0.26	0.22
IP	0.3988	0.000	0.041	0.000	0.134	0.00	0.31	0.00	CA15-3	0.9682	0.774	0.000	0.536	2.595	0.30	0.00	0.21
Fe	0.0102	0.223	0.000	0.865	7.584	0.03	0.00	0.11	CEA	0.1492	0.000	0.109	0.000	0.606	0.00	0.18	0.00
UIBC	0.4798	1.727	0.000	0.000	13.110	0.13	0.00	0.00	CA19-9	0.4695	0.000	0.000	0.669	7.558	0.00	0.00	0.09
Urea	0.5403	0.316	0.114	0.000	1.018	0.31	0.11	0.00	CA125	0.0026	0.592	0.000	1.229	9.059	0.07	0.00	0.14
CRE	0.2171	1.759	0.000	1.011	7.020	0.25	0.00	0.14	PSA	0.0270	0.001	0.000	0.000	0.006	0.17	0.00	0.00
UA	0.3619	11.980	2.389	6.219	46.100	0.26	0.05	0.13	E2	0.0000	0.000	0.000	72.110	148.600	0.00	0.00	0.49
CysC	0.1354	0.027	0.013	0.015	0.108	0.25	0.12	0.14	FSH	0.0000	0.000	1.495	1.837	4.066	0.00	0.37	0.45
TP	0.0170	1.682	1.170	0.418	3.790	0.44	0.31	0.11	LH	0.0000	0.000	0.000	1.772	5.062	0.00	0.00	0.35
Alb	0.0651	0.000	1.053	0.000	2.405	0.00	0.44	0.00	Prog	0.0000	0.000	0.000	5.569	6.392	0.00	0.00	0.87
Glob	0.0092	1.546	0.000	0.654	3.064	0.50	0.00	0.21	Testo	0.0463	0.000	3.248	0.000	5.322	0.00	0.61	0.00
IgG	0.0321	1.188	0.000	0.574	2.406	0.49	0.00	0.24	DHEA-S	0.8462	0.000	1.086	0.000	1.839	0.00	0.59	0.00
IgA	0.1801	0.155	0.000	0.060	0.819	0.19	0.00	0.07	Cort	0.6284	14.410	26.300	0.000	106.200	0.14	0.25	0.00
IgM	0.1002	0.083	0.142	0.083	0.569	0.15	0.25	0.15	PRL	0.0289	0.595	1.897	0.606	6.498	0.09	0.29	0.09
IgE	0.5595	14.180	12.200	0.000	108.600	0.13	0.11	0.00	Insulin	0.0019	0.565	0.184	0.336	2.084	0.27	0.09	0.16
CRP	0.0006	0.193	0.000	0.100	0.449	0.43	0.00	0.22	PTH	0.0001	8.319	3.988	2.918	16.400	0.51	0.24	0.18
SAA	0.0043	1.137	0.000	0.572	3.678	0.31	0.00	0.16	TSH	0.3338	0.108	0.078	0.000	0.904	0.12	0.09	0.00
C3	0.0088	83.670	6.957	20.780	155.900	0.54	0.04	0.13	FT4	0.3892	0.198	0.099	0.066	1.366	0.14	0.07	0.05
C4	0.0344	25.020	11.970	13.530	52.800	0.47	0.23	0.26	FT3	0.7132	0.000	0.066	0.070	0.414	0.00	0.16	0.17
RBP	0.5200	1.665	1.429	1.082	5.094	0.33	0.28	0.21	Tg	0.5713	0.641	0.000	0.000	13.350	0.05	0.00	0.00
TTR	0.1115	10.810	2.050	4.150	35.810	0.30	0.06	0.12	EPO	0.0660	1.198	0.559	0.000	5.627	0.21	0.10	0.00
Glu	0.4610	0.088	0.155	0.043	0.415	0.21	0.37	0.10	Fol	0.0064	6.825	2.677	0.691	9.987	0.68	0.27	0.07
TC	0.0042	0.000	0.245	0.162	0.762	0.00	0.32	0.21	VitB12	0.7593	41.020	26.790	0.000	177.600	0.23	0.15	0.00
TG	0.2580	0.090	0.074	0.042	0.265	0.34	0.28	0.16	ADP	0.1974	1.745	0.000	0.903	4.495	0.39	0.00	0.20
HDL-C	0.1983	0.186	0.000	0.000	0.340	0.55	0.00	0.00	PG1	0.0004	2.240	3.108	0.644	12.710	0.18	0.24	0.05
LDL-C	0.0415	0.152	0.230	0.121	0.657	0.23	0.35	0.18	PG2	0.0215	0.833	1.152	0.000	4.023	0.21	0.29	0.00
AST	0.1242	0.792	0.657	0.000	5.178	0.15	0.13	0.00	WBC	0.0000	3.969	2.745	3.446	13.220	0.30	0.21	0.26
ALT	0.2959	1.429	1.143	0.000	7.892	0.18	0.14	0.00	Neu	0.1144	1.803	0.000	1.272	8.386	0.22	0.00	0.15
ALP	0.2875	3.515	1.292	1.498	11.640	0.30	0.11	0.13	Lym	0.1927	1.293	0.141	0.694	7.277	0.18	0.02	0.10
AMY	0.7151	3.482	0.000	0.780	27.780	0.13	0.00	0.03	Mon	0.2300	1.293	0.141	0.694	7.277	0.18	0.02	0.10
LIP	0.6880	1.629	0.000	0.000	7.285	0.22	0.00	0.00	Eos	0.9323	0.343	0.174	0.233	1.814	0.19	0.10	0.13
GGT	0.6323	0.321	1.214	0.608	6.623	0.05	0.18	0.09	Bas	0.8560	0.108	0.032	0.083	0.355	0.30	0.09	0.23
LDH	0.9348	3.813	4.587	4.771	24.960	0.15	0.18	0.19	RBC	0.0707	6.950	6.799	3.893	28.810	0.24	0.24	0.14
CK	0.0007	1.130	0.000	5.270	26.220	0.04	0.00	0.20	Hb	0.4269	0.130	1.770	0.000	10.360	0.01	0.17	0.00
ASO	0.3630	14.910	6.116	0.000	51.070	0.29	0.12	0.00	Ht	0.2322	0.585	0.577	0.000	2.837	0.21	0.20	0.00
Tf	0.1319	0.089	0.000	0.000	0.417	0.21	0.00	0.00	MCV	0.6153	1.370	0.326	0.000	5.587	0.25	0.06	0.00
sTfR	0.1012	0.656	0.000	0.000	5.710	0.11	0.00	0.00	MCH	0.5752	0.441	0.000	0.237	2.165	0.20	0.00	0.11
ApoA1	0.0770	75.240	14.710	22.080	198.500	0.38	0.07	0.11	MCHC	0.3066	0.275	0.065	0.074	0.955	0.29	0.07	0.08
ApoB	0.0796	28.870	69.850	33.200	165.100	0.17	0.42	0.20	PLT	0.3867	2.048	0.000	0.460	5.625	0.36	0.00	0.08
ApoE	0.1218	1.302	0.000	0.000	11.390	0.11	0.00	0.00

Kruskal–Wallis (KW) test was performed to test for significance of menstrual cycle-related changes.

Three-level nested ANOVA was used to derive SD due to between-country, between-age, and between-menstrual cycle-related changes.

SD ratio (SDR) representing each source of variation was computed as its ratio to SD representing between-individual variations.

Analysis of differences by country, age, and menstrual cycle by 3N-ANOVA

The test results of 893 subjects were analysed using 3N-ANOVA by setting country, age, and menstrual phase as three SVs, and SD_cntr, SD_age, SD_cycle, and SD_indiv were then derived. SDR_cntr, SDR_age, and SDR_cycle were computed by taking the ratios of the respective SD to the SD_indiv. In this study, SDR_cntr > 0.3 were noted in 24 items: K, urea, TP, Glob, IgG, CRP, SAA, C3, C4, retinol-binding protein (RBP), transthyretin (prealbumin) (TTR), triglycerides (TG), HDL-C, alkaline phosphatase (ALP), ApoA1, TRAP-5b, ferritin, CA15-3, PTH, folate, adiponectin, WBC, Bas, and PLT. SDR_age > 0.3 were noted for 11 items: Ca, IP, TP, Alb, Glu, TC, LDL-C, ApoB, FSH, testosterone, and dehydroepiandrosterone sulphate (DHEA-S).

With SDR_cycle > 0.3, partitioning by menstrual phase was applied to oestradiol (0.49), progesterone (0.87), FSH (0.45), and LH (0.35). All other test items showed SDR_cycle < 0.3 (Table 2). However, for the physiological evaluation of menstrual cycle-related changes, we chose the following 30 test items with SDR_cycle > 0.15 for further analysis: Na, K, Cl, Glob, IgG, IgM, CRP, SAA, C4, RBP, TC, TG, LDL-C, CK, ApoB, Bone-ALP, alfa-fetoprotein, CA15-3, insulin, PTH, FT3, adiponectin, WBC, Neu, Bas, oestradiol, progesterone, FSH, and LH (Table 2).

MRA

We employed each individual data value as the target variable, and age, BMI, country, and the four sex hormones (oestradiol, progesterone, FSH, and LH), as explanatory variables in performing MRA. Test results for 10 test items, for which between-group differences were significant by the Kruskal–Wallis test (P < 0.005), showed significant associations (|r_p| ≥ 0.15) with either of the four sex hormones; these were Na, Cl, CRP, SAA, TC, ferritin, insulin, PTH, and WBC. In contrast, Glob, CK, CA125, and PG1 were not significantly associated with the sex hormone values (Table 3). Major findings were as follows: Oestradiol associated negatively with Na, CRP, SAA, and ferritin; progesterone associated positively with WBC and negatively with PTH; FSH associated negatively with SAA and insulin; and LH associated positively with TC but negatively with Cl.

Table 3.

Partial correlation coefficients between each test item and sources of variations by multiple regression analysis.

Test item	VNM	CHN	MYS	IDN	KOR	Age	BMI	E2	FSH	LH	Prog
Na	−0.05	0.10	0.06	−0.01	−0.01	−0.02	−0.01	−0.20	0.06	0.09	−0.14
Cl	−0.02	−0.06	0.01	0.02	0.00	0.03	0.05	−0.01	0.12	−0.20	0.12
Glob	0.30	0.00	0.13	0.22	0.05	−0.05	0.08	−0.06	0.03	0.02	0.08
CRP	0.23	0.13	0.16	0.20	0.03	0.10	0.16	−0.17	−0.13	0.10	−0.06
SAA	0.18	0.12	0.07	0.17	−0.02	0.04	0.06	−0.16	−0.16	0.13	−0.10
TC	−0.01	−0.07	0.12	0.06	−0.07	0.29	0.03	−0.08	−0.12	0.16	−0.08
CK	−0.02	−0.05	−0.01	−0.11	0.00	−0.01	0.10	−0.06	0.10	−0.05	−0.09
Fer	0.31	0.11	0.09	0.15	0.01	0.01	0.02	−0.16	−0.07	0.13	0.08
CA125	0.00	0.07	0.03	0.05	−0.05	−0.06	−0.03	0.00	0.09	−0.14	0.09
Insulin	0.04	0.08	0.13	0.07	−0.03	−0.04	0.26	−0.06	−0.21	0.11	0.08
PTH	−0.32	−0.27	−0.09	0.02	−0.05	0.17	0.11	0.05	−0.05	−0.08	−0.18
PG1	0.10	0.07	−0.09	−0.08	0.01	0.19	−0.12	0.01	0.11	0.04	−0.06
WBC		0.10	0.19		0.05	−0.17	0.15	0.11	−0.09	0.12	0.17

Multiple regression analysis was performed by setting each test item as an objective variable and the following fixed set of sources f variations as explanatory variables: dummy variables representing each country with Japan set as reference category; age, BMI, test results for oestradiol (E2), FSH, LH, and progesterone (prog).

Values in the table represent partial correlation coefficients which take values between −1.0 and 1.0.

Test items listed in this table are only those which showed possible relevance to the gonadal hormones or menstrual cycle.

CHN: China (including Taiwan and Hong Kong); IDN: Indonesia; KOR: Korea; MYS: Malaysia; VNM: Vietnam.

In the second round of MRA, we replaced the four sex hormones with four dummy variables representing the menstrual phases (early follicular phase set as the reference category) but retained all other explanatory variables. One or more menstrual phases were significantly associated with the following 12 test items: Na, Cl, CRP, SAA, CK, CA125, PTH, and WBC, plus four sex hormones. However, Glob, TC, ferritin, insulin, and PG1 were not significantly associated with the menstrual phase (Table 4). Major findings were as follows: Compared with the early follicular phase, oestradiol, progesterone, and LH were higher and Cl, CRP, SAA, and CA125 were lower during the late follicular and/or ovulatory phases. While, during the early and late luteal phases, again compared with the early follicular phase, oestradiol, progesterone and WBC were higher, and Na, CRP, SAA, CK, FSH, and PTH were lower.

Table 4.

Partial correlation coefficients between each test item and sources of variations by multiple regression analysis.

Test item	VNM	CHN	MYS	IDN	KOR	Age	BMI	LFl	Ovul	E Lt	LLt
Na	−0.04	0.11	0.04	0.00	−0.01	0.00	0.01	−0.10	−0.09	−0.12	−0.22
Cl	−0.02	−0.06	0.01	0.01	0.02	0.06	0.05	−0.15	−0.10	−0.03	0.00
Glob	0.29	0.01	0.13	0.21	0.05	−0.04	0.08	0.09	0.07	−0.02	0.04
CRP	0.24	0.14	0.15	0.22	0.04	0.06	0.19	−0.18	−0.18	−0.12	−0.15
SAA	0.19	0.14	0.07	0.18	−0.01	−0.02	0.08	−0.16	−0.16	−0.19	−0.21
TC	−0.01	−0.07	0.11	0.07	−0.08	0.25	0.03	0.10	0.08	0.04	0.00
CK	−0.01	−0.04	−0.01	−0.10	0.00	0.02	0.11	−0.08	−0.04	−0.09	−0.18
Fer	0.30	0.11	0.07	0.15	0.02	−0.02	0.03	−0.06	−0.09	−0.02	0.06
CA125	0.01	0.08	0.04	0.05	−0.03	−0.03	−0.02	−0.15	−0.14	−0.11	−0.12
E2	−0.03	−0.07	0.05	−0.01	−0.05	0.03	−0.12	0.30	0.35	0.45	0.37
FSH	−0.01	−0.03	−0.08	−0.04	−0.01	0.31	−0.04	−0.01	0.00	−0.21	−0.33
LH	−0.03	−0.02	−0.06	0.00	−0.06	−0.01	−0.06	0.24	0.29	0.10	0.00
Prog	−0.04	−0.03	0.01	−0.06	0.01	−0.04	−0.03	0.03	0.20	0.51	0.59
Insulin	0.04	0.09	0.14	0.08	−0.04	−0.11	0.26	0.01	0.09	0.11	0.09
PTH	−0.30	−0.25	−0.08	0.04	−0.06	0.16	0.12	−0.03	−0.04	−0.07	−0.16
PG1	0.09	0.07	−0.10	−0.08	0.00	0.23	−0.14	0.10	0.06	−0.04	−0.05
WBC		0.09	0.19		0.04	−0.21	0.12	−0.03	0.09	0.12	0.19

The same scheme as in Table 3 was used in listing the results of multiple regression analysis. As the fixed set of explanatory variables, four gonadal hormones were replaced with dummy variables representing menstrual phases by setting early follicular phase as reference category.

ELt: early luteal phase; LFl: late follicular phase; LLt: late luteal phase; Ovul: ovulatory phase.

Discussion

In this study, we evaluated menstrual cycle-related changes in laboratory test results in healthy females with regular cycles. The magnitude of the changes was computed as SDR after adjusting for the influence of each subject’s age and country of origin by use of 3N-ANOVA. In our previous study, we reported that regional differences were observed in inflammatory markers such as TP, IgG, CRP, SAA, C3, C4, and also nutritional markers such as TG, HDL-C, ApoA1, folate, adiponectin, tumour markers such as ferritin and CA15-3, and bone markers such as PTH and TRAP-5b.^4,5 In addition to these parameters, differences among regions were noted in measurements of nine items: K, urea, Glob, ALP, TTR, RBP, WBC, Bas, and PLT in premenopausal female. From the ANOVA results, between-age differences were detected in Ca, IP, TP, Alb, Glu, TC, LDL-C, ApoB, FSH, DHEA-S, and testosterone in premenopausal female.

Between-menstrual phase differences were significant in 15 test items by the Kruskal–Wallis test. The four sex hormones are known to fluctuate during the menstrual cycle, with SDR_cycle > 0.3, and the menstrual phase-specific RIs were derived for them. On the other hand, another 11 analytes (Na, Cl, CRP, SAA, TC, CK, ferritin, CA125, insulin, PTH, and WBC) showed significant association with the menstrual phase and/or one or more of the sex hormones. The actual profiles of menstrual cycle related change are shown in Figure 1.

Figure 1.

Menstrual cycle-related changes for selected analytes. Healthy female volunteers with regular menstrual cycles were subgrouped by the period between the first day of the last period and the blood sampling date into early follicular (EFl), late follicular (LFl), ovulatory (Ovul), early luteal (ELt), and late luteal phases (LLt). The box in the scattergram for each subgroup represents the central 50% range of values and the vertical line in the centre of the box indicates the median.

Oestrogen is well known to affect lipoprotein metabolism. For example, oestrogen promotes the clearance of chylomicron remnants, stimulates hepatic lipid production, increases very low-density lipoprotein synthesis and the production of HDL, and enhances the expression of the LDL receptor.^11–16 However, reports on relationships between lipoproteins and sex hormones across the menstrual cycle have been inconsistent.¹ We demonstrated that TC was higher in the late follicular phase than in the early follicular phase, and TC was positively correlated with LH. Similar fluctuations of TC during the menstrual cycle have been seen in several former reports.¹

The synthesis of CRP, an acute-phase protein, is induced by TNF-α, IL-1β, and IL-6, and is synthesized in the liver by macrophages.¹⁷ CRP is used as a serum marker for vascular inflammation and is a useful predictive marker for cardiovascular risk.^18–21 Serum oestrogen concentration is known to affect serum CRP concentrations, and oral administration of an exogenous oestrogen preparation to postmenopausal female increases the CRP concentration.²² SAA is also known to be a sensitive inflammatory marker,²³ and it is thought that SAA can also be used as a predictive marker for cardiovascular risk.²⁴ The oral administration of oestrogen is also known to increase the SAA concentration, as with the CRP concentration.²⁵ Serum CRP is known to fluctuate during the menstrual cycle and is accompanied by the fluctuation of endogenous oestrogen. Since SAA is likewise affected by oestrogen administration, it is also predicted to fluctuate during the menstrual cycle. In this study, we found that both SAA and CRP were inversely associated with serum oestradiol and fluctuated depending on the concentration of serum oestradiol during the menstrual cycle. Fluctuations of serum CRP concentration during the menstrual cycle have been studied previously; however, the fluctuations have not been observed in several studies.^26–28 In contrast, a recent study that chronologically evaluated the relation between concentrations of high-sensitive CRP assays and endogenous sex hormones in premenopausal female over the entire menstrual cycle showed that CRP fluctuated significantly (P < 0.001).²⁹ Our results showed that the CRP concentration in the early follicular phase was significantly higher (P < 0.01) than that in the late follicular phase and was also significantly higher (P < 0.05) than that in the ovulatory and late luteal phases. These results are consistent with the report by Gaskins et al.³ and support the hypothesis that endogenous oestrogen may exert an anti-inflammatory effect.³⁰ In addition, the concentration of SAA, similar to that of CRP, was higher in the early follicular phase than in the ovulatory and luteal phases. TNFα is a cytokine that promotes the synthesis of both CRP and SAA in the liver. TNFα is known to inhibit cholesterol efflux to HDL,³¹ suggesting the possibility that the action of oestrogen on cytokines including TNFα may regulate the fluctuations of CRP and HDL during the menstrual cycle. Previous studies reported that an increase in serum oestrogen causes a decrease in CRP, whereas an increase in progesterone causes an increase in CRP.^3,30 Because the study designs of these previous studies differ from ours, the results cannot be compared directly. Progesterone enhances neutrophil chemotaxis, increases production of some inflammatory mediators by monocytes,³² causes activation of NK cells, and decreases the production of NO synthase and the activation of T cells.³³ These results suggest that the increase in progesterone is related to anti-inflammatory effects in normal female with a normal menstrual cycle. However, the results in this study showed that CRP and SAA were not significantly associated with progesterone.

Progesterone plays an important role in regulating bone metabolism. Progestins have been demonstrated to increase gene expression of PTH and the progesterone receptor in the parathyroid gland.³⁴ In this study, however, PTH was significantly higher in the early follicular phase compared to the concentration in the late luteal phase, and PTH was negatively correlated with progesterone. Our results coincide with other previous reports.^35–38

In this study, we observed that serum concentrations of CK, insulin, WBC, Na, and Cl also fluctuated in relation to the menstrual cycle. Serum CK activity was significantly lower in the late luteal phase of the cycle, but CK activity had no correlated sex hormone values. Serum insulin was higher in the early luteal phase of the menstrual cycle; its concentrations were negatively correlated with FSH. WBC was significantly higher in the late luteal phase. IL-6 transduction by oestrogens and progesterone may affect these parameters.

There are few reports on the fluctuation of serum electrolytes during the menstrual cycle. The present study clearly shows that serum Na is lower in the luteal phase compared to the early follicular phase, and serum Na is negatively correlated with oestradiol. These findings match well with the phenomenon of surge in Na concentration after menopause when oestradiol drops sharply.³⁹ On the other hand, serum Cl was significantly lower during late follicular phase compared to the early follicular phase. Cl was negatively correlated with LH.

Conclusion

From 893 well-defined healthy females with regular menstrual cycles, we analysed menstrual phase-related changes for 85 major laboratory tests, including CBC, chemical analytes, tumour markers, and various hormones. In the four sex hormones, well-known menstrual cycle-related changes were conspicuously observed. Based on ANOVA (SDR for menstrual cycle-related changes) and multiple regression analysis (partial correlation coefficients), we revealed statistically significant menstrual cycle-related changes in the following test items: Na, Cl, CRP, SAA, CA125, and PTH are higher during the early follicular phase when oestradiol is in its nadir, while WBC is higher during the luteal phases, with no direct association with the concentration of sex hormones. However, the magnitude of these changes is rather small and not very relevant in daily interpretation of laboratory test results.

Footnotes

Acknowledgements

We are in great debt to all the colleagues in the laboratories who collaborated in the 2009 Asian study and allowed us to use the data-set for our secondary analyses to elucidate menstrual cycle-related variations of laboratory test results performed in this study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the Ethical Committee of Yamaguchi University Graduate School of Medical Sciences (193-1).

Guarantor

TK.

Contributorship

KI led the Asian multicentre study and SM collaborated in the study in recruiting healthy volunteers. KI provided the full data-set for analyses done in this study. SM, KI, and TK designed this observational study for exploring the menstrual cycle-related changes. SM analysed the data-set. SM and TK wrote the manuscript. KI edited the manuscript. All authors approved the final version.

References

Schisterman

Mumford

Sjaarda

. Failure to consider the menstrual cycle phase may cause misinterpretation of clinical and research findings of cardiometabolic biomarkers in premenopausal women. Epidemiol Rev 2014; 36: 71–82.

Mumford

Schisterman

Siega-Riz

. A longitudinal study of serum lipoproteins in relation to endogenous reproductive hormones during the menstrual cycle: findings from the BioCycle study. J Clin Endocrinol Metab 2010; 95: E80–E85.

Gaskins

Wilchesky

Mumford

. Endogenous reproductive hormones and C-reactive protein across the menstrual cycle: the BioCycle study. Am J Epidemiol 2012; 175: 423–431.

Ichihara

Ceriotti

Tam

. The Asian project for collaborative derivation of reference intervals: (1) strategy and major results of standardized analytes. Clin Chem Lab Med 2013; 51: 1429–1442.

Ichihara

Ceriotti

Kazuo

. The Asian project for collaborative derivation of reference intervals: (2) results of non-standardized analytes and transference of reference intervals to the participating laboratories on the basis of cross-comparison of test results. Clin Chem Lab Med 2013; 51: 1443–1457.

Coulter LH Series IQAP Information Guide: Hematology, https://www.beckmancoulter.com/ucm/idc/groups/public/documents/webasset/glb_bci_153779.pdf (accessed 24 July 2015).

Ichihara

Itoh

Lam

. Sources of variation of commonly measured serum analytes in 6 Asian cities and consideration of common reference intervals. Clin Chem 2008; 54: 356–365.

Ichihara

Boyd

. An appraisal of statistical procedures used in derivation of reference intervals. Clin Chem Lab Med 2010; 48: 1537−1551–1537−1551.

Ichihara

. Statistical considerations for harmonization of the global multicenter study on reference values. Clin Chim Acta 2014; 432: 108−118–108−118.

10.

Ozarda

Ichihara

Barth

. Protocol and standard operating procedures for common use in a worldwide multicenter study on reference values. Clin Chem Lab Med 2013; 51: 1027–1040.

11.

Knopp

Paramsothy

Retzlaff

. Sex differences in lipoprotein metabolism and dietary response: basis in hormonal differences and implications for cardiovascular disease. Curr Cardiol Rep 2006; 8: 452–459.

12.

Knopp

Zhu

. Multiple beneficial effects of estrogen on lipoprotein metabolism. J Clin Endocrinol Metab 1997; 82: 3952–3954.

13.

Campos

Walsh

Judge

. Effect of estrogen on very low density lipoprotein and low density lipoprotein subclass metabolism in postmenopausal women. J Clin Endocrinol Metab 1997; 82: 3955–3963.

14.

Srivastava

Baumann

Schonfeld

. In vivo regulation of low-density lipoprotein receptors by estrogen differs at the post-transcriptional level in rat and mouse. Eur J Biochem 1993; 216: 527–538.

15.

Zannis

Chroni

Krieger

. Role of apoA-I, ABCA1, LCAT, and SR-BI in the biogenesis of HDL. J Mol Med (Berl) 2006; 84: 276–294.

16.

Acton

Rigotti

Landschulz

. Identification of scavenger receptor SR-BI as a high density lipoprotein receptor. Science 1996; 271: 518–520.

17.

Gabay

Kushner

. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999; 340: 448–454.

18.

Rifai

Ridker

. High-sensitivity C-reactive protein: a novel and promising marker of coronary heart disease. Clin Chem 2001; 47: 403–411.

19.

Boekholdt

Hack

Sandhu

. C-reactive protein levels and coronary artery disease incidence and mortality in apparently healthy men and women: the EPIC-Norfolk prospective population study 1993–2003. Atherosclerosis 2006; 187: 415–422.

20.

Pradhan

Manson

Rossouw

. Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease: prospective analysis from the Women’s Health Initiative observational study. JAMA 2002; 288: 980–987.

21.

Ridker

Hennekens

Buring

. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000; 342: 836–843.

22.

Davison

Davis

. New markers for cardiovascular disease risk in women: impact of endogenous estrogen status and exogenous postmenopausal hormone therapy. J Clin Endocrinol Metab 2003; 88: 2470–2478.

23.

Uhlar

Whitehead

. Serum amyloid A, the major vertebrate acute-phase reactant. Eur J Biochem 1999; 265: 501–523.

24.

Kinlay

Schwartz

Olsson

. Inflammation, statin therapy, and risk of stroke after an acute coronary syndrome in the MIRACL study. Arterioscler Thromb Vasc Biol 2008; 28: 142–147.

25.

Abbas

Fadel

Wang

. Contrasting effects of oral versus transdermal estrogen on serum amyloid A (SAA) and high-density lipoprotein-SAA in postmenopausal women. Arterioscler Thromb Vasc Biol 2004; 24: e164–e167.

26.

Wunder

Yared

Bersinger

. Serum leptin and C-reactive protein levels in the physiological spontaneous menstrual cycle in reproductive age women. Eur J Endocrinol 2006; 155: 137–142.

27.

Capobianco

de Muro

Cherchi

. Plasma levels of C-reactive protein, leptin and glycosaminoglycans during spontaneous menstrual cycle: differences between ovulatory and anovulatory cycles. Arch Gynecol Obstet 2010; 282: 207–213.

28.

Jilma

Dirnberger

Löscher

. Menstrual cycle associated changes in blood levels of interleukin-6, α1 acid glycoprotein, and C-reactive protein. J Lab Clin Med 1997; 130: 69–75.

29.

Blum

Müller

Huber

. Low-grade inflammation and estimates of insulin resistance during the menstrual cycle in lean and overweight women. J Clin Endocrinol Metab 2005; 90: 3230–3235.

30.

Wander

Brindle

O’Connor

. C-reactive protein across the menstrual cycle. Am J Phys Anthropol 2008; 136: 138–146.

31.

Zhang

McGillicuddy

Hinkle

. Adipocyte modulation of high-density lipoprotein cholesterol. Circulation 2010; 121: 1347–1355.

32.

Bouman

Heineman

Faas

. Sex hormones and the immune response in humans. Hum Reprod Update 2005; 11: 411–423.

33.

Roberts

Walker

Alexander

. Sex-associated hormones and immunity to protozoan parasites. Clin Microbiol Rev 2001; 14: 476–488.

34.

Epstein

Silver

Almogi

. Parathyroid hormone mRNA levels are increased by progestins and vary during rat estrous cycle. Am J Physiol 1996; 270: E158–E1563.

35.

Gorai

Taguchi

Chaki

. Serum soluble interleukin-6 receptor and biochemical markers of bone metabolism show significant variations during the menstrual cycle. J Clin Endocrinol Metab 1998; 83: 326–332.

36.

Chiu

Arnaud

. Correlation of estradiol, parathyroid hormone, interleukin-6, and soluble interleukin-6 receptor during the normal menstrual cycle. Bone 2000; 26: 79–85.

37.

Thys-Jacobs

McMahon

Bilezikian

. Cyclical changes in calcium metabolism across the menstrual cycle in women with premenstrual dysphoric disorder. J Clin Endocrinol Metab 2007; 92: 2952–2959.

38.

Thys-Jacobs

Alvir

. Calcium-regulating hormones across the menstrual cycle: evidence of a secondary hyperparathyroidism in women with PMS. J Clin Endocrinol Metab 1995; 80: 2227–2232.

39.

Ichihara K, Yamamoto Y, Hotta T, et al. Collaborative derivation of reference intervals for major clinical laboratory tests in Japan. Ann Clin Biochem 2015. Epub. DOI: 10.1177/0004563215608875.