Printer Friendly

Cessation of menstruation improves the correlation of FPG to hemoglobin Alc in Caucasian women.

BACKGROUND: Anemia is known to cause spurious hemoglobin Alc (HbAlc) results. The effect of menstruation on HbAlc was tested by correlating it to FPG in non-anemic premenopausal and in menopausal women.

METHOD: Non-diabetic, non-obese middle-aged Caucasian women were classified as premenopausal or menopausal. Hemogram, FPG, and Alc results were obtained.

RESULTS: Hemoglobin concentrations were lower in the premenopausal group. FPG showed a poor correlation to Alc value overall (r = 0.251, p = 0.001) which was improved by multiplying the Alc % by the total hemoglobin concentration to create an absolute Alc value (r = 0.362, p = 0.000). When the data was sorted by menopause status, the correlation of FPG to Absolute Alc improved (r = 0.463,P = 0.000) in the menopausal women, but remained low (r = 0.283, p = 0.005) in the premenopausal women.

CONCLUSIONS: Menstruation may be a significant factor affecting the accuracy of Alc concentrations.

ABBREVIATIONS: Alc = hemoglobin Alc; BMI = body mass index; CI = confidence interval, FPG = fasting plasma glucose; HbAlc = hemoglobin Alc; MBG = mean blood glucose; MCV. mean corpuscular volume; NGSP = National Glycohemoglobin Standardization Program; RDW = red blood cell distribution width.

INDEX TERMS: Absolute Alc; Alc; glucose; menopause; menstruation.

In the past 24 years, the prevalence of diagnosed diabetes has increased 59% for American women. In 2004, 7.8% of white females between the ages of 45 and 64 had diagnosed diabetes, with an average age at diagnosis of 46.4 years. (1) Chronic complications of diabetes include retinopathy, kidney disease, neuropathy, and cardiovascular disease, and can be delayed and diminished by tight glycemic control for both type 1 and type 2 diabetics. (2-5) Fasting plasma glucose (FPG) is the preferred screening tool for diabetes mellitus, and a FPG [greater than or equal to] 126 mg/dL (7.0 mmol/L) that is reproduced on a separate occasion is diagnostic for diabetes. (6) Hemoglobin Alc has long been recognized as the best test of long term glycemic status for both type 1 and type 2 diabetics, and elevations in Alc are linked to higher incidence of chronic complications. (2-4,7) Glycation is a continuous process, and the long life of red blood cells gives the Alc test a look back capability over the lifetime of those cells. (8)

Technical problems with Alc measurement that cause a spurious result have been discussed in depth, and encompass the inclusion of labile Alc in the total Alc, the inability to distinguish glycated hemoglobin from carbamylated hemoglobin, and the positive and negative effects of variant hemoglobins on Alc concentration. (8,9) The National Glycohemoglobin Standardization Program was established to certify methods as traceable to the Diabetes Control and Complications Trial Reference Method. (10,11)

Nathan and colleagues' determined a linear relationship between mean blood glucose (MBG) and Alc in 21 diabetics, showing MBG (mg/dL) = 33.3*HbAlc--86, r = 0.958.

Recently Rohlfing and colleagues (12) examined 1439 patients and refined the relationship to mean plasma glucose MPG (mg/dL) = 35.6*Alc--77.3, r = 0.82. When they examined fasting glucose alone, the correlation to Alc was lower, r = 0.69. Pathologic conditions can cause discrepant Alc values, invalidating the correlation to plasma glucose. Alc values are lower in hemolytic disease and hemoglobinopathies with decreased red blood cell survival. 13 Iron deficiency anemia causes increased Alc values in diabetics 11,15 and non-diabetics. (16) Spuriously low values of Alc were found in diabetic subjects with anemia, and correction of the anemia raised the Alc. (17) Subtler differences remain to be discovered. Kilpatrick and coworkers reported an intra-individual variance of six percent for glycated hemoglobin, but an 85% inter-individual variance. (18) Rates of glycation differ among individuals, and Alc concentration is also affected by the duration of the diabetic status. (19)

Menstruation has not been investigated as a significant cause of Alc inaccuracy, even though it is a leading cause of iron deficiency anemia. The average age of menopause in American women is 51, close to the average age of diagnosis for diabetes. Many women experience significant blood loss during perimenopause, a several year span of time prior to menopause. This work looks at middle-aged, non-obese, non-diabetic Caucasian women, distinguishing them as pre menopausal or menopausal, and compares the hemoglobin values between the groups. It correlates Alc to FPG, and Hemoglobin to FPG in both groups. It introduces a parameter called "Absolute Alc", the product of total hemoglobin * Alc and tests its correlation to FPG in both groups.



Two hundred women between the ages of 40 and 54 were recruited by newspaper, television, and radio announcement in the Pensacola, Florida area between November 2004 and May 2005. Participants were interviewed to determine their health status, race, age, and menopause status. Some women had a previous hysterectomy and were not aware of their menopause status; these women were grouped with the menopausal women. Participants were weighed and height measured, and BMI was calculated as weight in kg/height in [m.sup.2]. Participants were excluded if they were outside the specified age range, obese (BMI [greater than or equal to] 30), were known diabetics or had a fasting plasma glucose (FPG) [greater than or equal to] 126 mg/dL (7.0 mmol/ L), or if they took medication to lower their cholesterol. Since race other than Caucasian is a risk for diabetes, and only 7.8% of the enrollees were not Caucasian, these women were excluded from this analysis. 169 subjects remained. For the correlation with FPG and Alc, women with hemoglobin < 11.0 g/dL or MCV < 80 fL were disqualified (n = 5). All of the subjects were in good health. This study was approved by the Institutional Review Board at the University of West Florida; all subjects provided informed consent.

Laboratory analysis

Subjects were instructed to fast between ten and 12 hours, and blood was collected in the morning. Blood for glucose analysis was collected in serum separator tubes, and centrifuged within 60 minutes of collection. Glucose was performed on a Dade RXL using the Hexokinase G-6-PD method. Hemoglobin Alc was collected in EDTA and was performed on a Tosoh Alc 2.2+. This instrument is an ion exchange HPLC method approved by the NGSP, it is not affected by carbamylation or HbAS, it separates labile Alc from stable Alc, and printouts alert the operator when the instrument is unable to identify unusual elution peaks. Hemoglobin concentrations were per formed by the Cyanmethemoglobin method and MCV and RDW by electrical impedance on a Beckman Coulter HMX. All assays were performed at West Florida Hospital, Pensacola, Florida within one day of collection by a qualified technical staff. The between run imprecision (CV) for glucose, Alc, hemoglobin, MCV and RDW were 1.7%,2.9%,0.9%,0.7%, and 1.2%, respectively. Absolute Alc value was calculated as the Alc %* total hemoglobin in g/L.

Data analysis

Descriptive statistical analysis was performed using Microsoft Excel. Student's t-test of unequal variances was performed for age, FPG, hemoglobin. Student's t-test of equal variance was performed for Alc, Absolute Alc, and MCV Confidence intervals are 95%. Correlations and box plots were generated with SPSS v.12. Pearson correlation coefficients (r) represent two tailed significance.


All of the women were between 40 and 54 years old, Caucasian, non-diabetic, and normal to overweight. The women were grouped as premenopausal (n = 100) or postmenopausal (n = 69). None of the women reported anemia in their health history. One woman had a hemoglobin concentration of 5.9 g/dL; as an extreme outlier she was excluded. The premenopausal women had a mean hemoglobin of 13.6 g/dL (95% Confidence interval 13.5-13.8). The menopausal women had a mean hemoglobin of 13.9 g/L (95% Confidence interval 13.7-14.1),p = 0.003. Median hemoglobins were 13.6 g/dL for the premenopausal women and 13.8 g/dL for the menopausal women. Boxplot analysis showed that the pre menopausal women had lower hemoglobin concentrations in general, ranging from 9.8 g/dL to 15.4 g/dL compared to the menopausal women with 12.6 g/dL to 15.9 g/dL (Figure 1). Since iron deficiency anemia is known to prolong red cell survival and increase Alc results, four additional premenopausal women were excluded from the following analysis based on low hemoglobin or MCV


The two sets of women were similar with respect to BMI, FPG, Alc, MCV, and RDW; they differed in age (p < 0.001) and Hemoglobin concentration (p = 0.009). Comparison of the two groups is shown in Table 1. FPG showed a weak correlation to Alc value overall (r = 0.25 = 0.001). FPG showed a better correlation to Alc % in the menopausal group (r = 0.218, p = 0.034 for the premenopausal group, r = 0.296, p = 0.013 for the menopausal group), shown in Figure 2, A and B.


Total hemoglobin correlated with FPG in the menopausal women (r = 0.385, p = 0.001), but not with the premenopausal women (r = 0.130, p = 0.209). Absolute Alc values were generated in g/L (Alc %* total hemoglobin) shown in Table 1; the two groups of women were significantly different (p = 0.008). Absolute Alc correlated significantly to FPG (r = 0.362, p = 0.000). The correlation between Absolute Alc was stronger in the postmenopausal women (r = 0.463, p = 0.000) than the premenopausal women (r = 0.283, p = 0.005), and is shown in Figure 2, C and D.


Previous studies have found a less than perfect correlation between Alc and FPG even in non-diabetics. As manufacturers correct the technical factors affecting accuracy of those assays, pre-analytical factors come more into focus. Hemoglobin is glycated in a glucose-dose dependent fashion to form Alc. Total hemoglobin concentration has not been considered in this equation except in the instances where turnover of red cell mass is noticeably altered. Menstrual blood loss is an innocuous factor here, in that most of the premenopausal women would not be classified as anemic. Comparing the hemoglobin ranges of premenopausal to menopausal women shows the effect of that menstrual loss (Figure 1).

This study showed a positive correlation between total hemoglobin and FPG in the menopausal women (r = 0.385, p = 0.001). This is not unexpected. Serum hematocrit and serum ferritin levels have been shown to be positively associated with plasma glucose levels in both diabetics and in subjects with insulin resistance syndrome. (20-25) In a study of nondiabetic non-obese Caucasian women in a Pennsylvania population, (26) hemoglobin correlated significantly with FPG in women aged 54 to 70 years (r = 0.344,p = 0.000, n = 119, unpublished results). Incorporating the total hemoglobin and Alc into Absolute Alc in the present study improved its correlation to FPG from 0.296 to 0.463. This suggests that both the glucose and the hemoglobin concentrations are relevant in the glycation process.

Total hemoglobin did not correlate to glucose in the premenopausal women (r = 0.130,p = 0.209). This data is also consistent with findings from the Pennsylvania study: hemoglobin did not correlate to FPG in women aged 40 to 50 years (r = 0.208)p = 0.103, n = 63, unpublished data). Incorporating the total hemoglobin and Alc into Absolute Alc improved its correlation to FPG, but to a lesser extent than in the menopausal women (from 0.218 to 0.283). This finding suggests that menstrual blood loss has a pre-analytical effect on Alc. Since menstruation is not a pathologic process it has been overlooked as a source of variation in Alc accuracy. Women with heavy periods can become iron deficient; in fact 20% of women will become iron deficient sometime during their reproductive years. (27) The paradox is this: While the iron deficiency should make the Alc concentration higher, the blood loss and replacement with younger cells should make the Alc concentration lower. In this study, five percent of the premenopausal women were suspected to have iron deficiency and therefore inaccurate Alc results. Even after they were excluded from the analysis the premenopausal group still had a poor correlation to FPG when compared to the non-menstruating group, consistent with a variable pattern of red blood cell turnover.

This study suggests that menstruation during middle age can have a significant effect on the accuracy on Alc, and that effect goes away at menopause. This is a significant finding, because half of the women who develop type 2 diabetes do so in perimenopause. (28) The average age for diagnosis of diabetes in white American women is about 46 years,' the average age for her menopause is 51, with perimenopause occurring as early as 35 or as late as 55. (28) Women in perimenopause suffer from irregular menstruation. In a study of 500 perimenopausal women 18% reported more frequent and/or heavier periods with irregular bleeding between periods. (29) Clearly the accuracy of the Alc result is dependent upon consistent hemoglobin concentration. This study showed a significant difference in hemoglobin concentration between premenopausal and menopausal women with similar MCV and RDW.

This work introduces the parameter "Absolute Alc", derived as the product of total hemoglobin and %Alc. It is intriguing that this parameter showed a superior correlation to FPG. It is logical that Alc formation is dependent upon the concentration of both glucose and hemoglobin. This parameter is very appealing, given the association of insulin resistance with increased iron stores, (23-25) and after thorough study may prove to be a valuable diagnostic tool. It is important to interpret these results within the specifications of this study. The subjects were not diabetic, so the range of glucose and Alc was limited. Furthermore, only 169 subjects were tested. Mean blood glucose would be a stronger tester than FPG, but that data was not collected. Subjects with hemoglobinopathies may have different results. This study uses Absolute Alc as a tool to examine the differences between premenopausal and menopausal Caucasian women, and it is clear that these groups are different from each other. It will be important to test the validity of Absolute Alc and the effect of menstruation on Alc in non-Caucasian women, as these women represent a significant at-risk population. The prevalence of diabetes in African Americans is 1.8 times that of whites, the prevalence in Hispanics is 1.7 times that of whites, and American Indians are 2.2 times as likely to have diabetes as whites of a similar age. (30)

It will be interesting to test Absolute Alc in other groups, including those with iron deficiency, men, and older women. Coban and colleagues (16) have already speculated that iron deficient individuals may have increased Alc in part due to a higher glucose to hemoglobin ratio. One study restricted to men showed a low variation in Glycohemoglobin, (31) which is consistent with the expectation of stable hemoglobin levels. Older women of all races have an increased prevalence of diabetes (30) and will benefit from improvements in the indicators of glycemic control.

Strengths of this study lie in its limitation of variables: the subjects are non-diabetics so overall glucose variation was limited, none of the subjects were obese so that blood volume was comparable in the subjects, and they all were the same sex and similar in age. It is unlikely that any participants had a hemoglobinopathy based on race and the ability of the HPLC analyzer to flag unexpected eluate peaks. One weakness of this study is that each set of data--FPG and Alc-was obtained only once from each participant. Women were dichotomized as premenopausal or menopausal based on patient reporting. Since this study revolved around blood loss from menstruation, women who had hysterectomies were included with the menopausal women. There was no method to determine which non-menopausal women were in peri-menopause, therefore all of the menstruating women were analyzed as one group. Hormonal changes may have some effect on glucose and Alc; this was not explored.

Alc results are valuable in long term maintenance and assessment of diabetes, and both the American Diabetes Association (6) and the American Heart Association recommend a target goal Alc < 7% to reduce the risk for micro and macrovascular disease in women. (32) The Standards of Medical Care in Diabetes do not include an assessment of total hemoglobin in diabetes management, and men and women have the same target Alc goal. Clinicians and researchers should be aware that perimenopausal patterns of bleeding may cause inconsistent Alc values, and incorporate patient history of menstrual and other bleeding, self monitoring blood glucose results, and hemogram results (including hemoglobin, MCV and RDW) with Alc for a thorough assessment of glycemic control.


The author would like to thank Jorge Salgado, Mary Gimenez, and Frances Connolly Forrest for help with data collection; Sherman Bonomelli and Swarna Krothapalli of the Medical Technology Program at the University of West Florida; and Joan Simmons, Stephanie Risinger, Marsha Dumas, Rita Casey and the laboratory staff at West Florida Hospital. This project was funded by a grant by the American Society for Clinical Laboratory Sciences.


(1.) National Center for Chronic Disease Prevention and Health Promotion, Diabetes Public Health Resource. Available from Accessed 2005 Dec.

(2.) Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin-dependent diabetes mellitus. N Eng J Med 1993;329:977-86.

(3.) UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patient with type 2 diabetes (UKPDS 33). Lancet 1998;352:837-53.

(4.) Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group. Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. N Engl J Med 2003;348:2294-303.

(5.) Gaede P, Vedel P, Larsen N, Jensen GVH, Parving H-H, Pederson O. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med 2003;348:383-93.

(6.) American Diabetes Association. Standards of medical care in diabetes mellitus. Diabetes Care 2005;28(Suppl 1):S4-S36.

(7.) Nathan DM, Singer DE, Hurxthal K, Goodson JD. The clinical information value of the glycosylated hemoglobin assay. New Engl J Med. 1984;310:341-6.

(8.) Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002;48:436-72.

(9.) Bry L, Chen PC, Sacks DB. Effects of hemoglobin variants and chemically modified derivatives on assays for glycohemoglobin. Clin Chem 2001;47:153-63.

(10.) Little RR, Rohlfing CL, Wiedmeyer HM, Myers GL, Sacks DB, Goldstein DE. The national glycohemoglobin standardization pro gram: a five-year progress report. Clin Chem 2001;47:1985-1992.

(11.) National Glycohemoglobin Standardization Program. Available from Accessed 2005 Dec.

(12.) Rohlfing CL, Wiedmeyer H-M, Little RR, England JD, Tennill A, D Goldstein. Defining the relationship between plasma glucose and HbA,. Diabetes Care 2002; 25:275-8.

(13.) Winter WE, Signorino MR. Diabetes mellitus: pathophysiology, etiologies, complications, management, and laboratory evaluation. Washington, DC: AACC Press. 2002;137pp.

(14.) Tarim O, Kucukerdogan A, Gunay U, Eralp O, Ercan I. Effects of iron deficiency anemia on hemoglobin Alc in type 1 diabetes mellitus. Pediatr Int 1999;41:357-62.

(15.) El-Agouza I, Abu Shahla A, Sirdah M. The effect of iron deficiency anaemia on the levels of haemoglobin subtypes: possible consequences for clinical diagnosis. Clin Lab Haematol 2002;24:285-289.

(16.) Coban E, Ozdogan M, Timuragaoglu A. Effect of iron deficiency anemia on the levels of hemoglobin Alc in non-diabetic patients. Acta Haematologica 2004;112:126-8.

(17.) Camargo JL, Gross JL. Conditions associated with very low values of glycohaemoglobin measured by an HPLC method. J Clin Pathol 2004;57:346-9.

(18.) Kilpatrick ES, Maylor P W, Keevil BG. Biological variation of glycated hemoglobin. Implications for diabetes screening and monitoring. Diabetes Care 1998;21:261-4.

(19.) Hudson PR, Child DF, Jones H, Williams CP Differences in rates of glycation (glycation index) may significantly affect individual HbAlc results in type 1 diabetes. Ann Clin Biochem 1999;36:451-9.

(20.) McCarty ME Hyperinsulinemia may boost both hematocrit and iron absorption by up-regulating activity of hypoxia-inducible factor-1alpha. Med Hypothesis 2003;61:567-73.

(21.) Ford ES, Cogswell ME. Diabetes and serum ferritin concentration among US adults. Diabetes Care 1999;22:1978-83.

(22.) Fernandez-Real JM, Lopez-Mermejo AL, Ricart W Perspectives in diabetes: cross-talk between iron metabolism and diabetes. Diabetes 2002;51:2348-54.

(23.) Haap M, Fritsche A, Mensing HJ, Haring, H-U, Stumvoll, M. Association of high serum ferritin concentration with glucose intolerance and insulin resistance in health people. Ann Int Med 2003;139:869-71.

(24.) Jehn M, Clark JM, Guallar E. Serum ferritin and risk of the metabolic syndrome in US adults. Diabetes Care 2004;27:2422-28.

(25.) Bozzini C, Girelli D, Olivieri O, Martinelli N, Bassi A, DeMatteis G, and others. Diabetes Care 2005;28:2061-3.

(26.) Behan KJ, Amin RW Normoglycemia may encompass two subpopulations with respect to vascular risk in non-obese Caucasian women. Lab Med 2005; 36:2-7.

(27.) Schreiber W Iron, porphyrin, and bilirubin metabolism. In: Kaplan LA, Pesce AJ , Kazmierczak SC eds. Clinical Chemistry Theory, Analysis, Correlation 4" edition. St. Louis: Mosby 2003;657-674.

(28.) Corio LE, Kahn LG. 2000. The change before the change. New York: Bantam Books. 2000;434pp.

(29.) Seltzer VL, Benjamin F, Deutsch S. Perimenopausal bleeding patterns and pathologic findings. JAMWA 1990;45:132-134.

(30.) National Diabetes Statistics. Available from Accessed 2006 Apr 29.

(31.) Rohlfing C, Wiedmeyer H-M, Little RR, Grotz VL, Tennill A, England J, Madsen R, Goldstein D. Biological variation of glycohemoglobin. Clin Chem 2002;48:116-8.

(32.) Mosca L. Heart disease prevention in women. Circulation 2004;109: e158-60.

Clin Lab Sci 2006;19(4):225

Kristina Jackson Behan is Associate Professor, University of West Florida, Department of Biology, Division of Life and Health Sciences.

Address for correspondence: Kristina Jackson Behan PhD MT(ASCP), Program in Clinical Laboratory Sciences, Department of Biology, Division of Life and Health Sciences, University of West Florida, 11000 University of West Florida, Pensacola FL 32514. (850) 474-3060, (850) 474-2749 (fax).

The peer-reviewed Research and Reports Section seeks to publish reports of original research related to the clinical laboratory or one or more subspecialties as well as information on important clinical laboratory-related topics such as technological clinical and experimental advances and innovations. Literature reviews are also included. Direct all inquiries to David G Fowler PhD CLS(NCA), Clin Lab Sci Research and Reports Editor, Dept of Clinical Laboratory Sciences, University of Mississippi Medical Center, 2500 North State St, Jackson MS 39216. (601) 984-6309, (601) 815-1717 (fax).
Table 1. Descriptive statistics of non-anemic subjects

 Premenopausal Menopausal
 Mean (95% CI) Mean (95% CI) p value

Age 46.1 (45.3-47.0) 50.0 (49.3-50.7) <0.001 **
BMI 23.2 (22.6-23.8) 23.6 (22.8-24.3) 0.190
FPG mg/dL 88.0 (88-90) 88.0 (86-90) 0.500
Alc % 5.4 (5.3-5.4) 5.4 (5.3-5.5) 0.081
Hemoglobin g/dL 13.6 (13.5-13.8) 13.9 (13.7-14.1) 0.009 **
MCV fL 90.5 (89.8-91.2) 90.1 (89.2-91.0) 0.311
RDW 12.3 (12.2-12.5) 12.2 (12.1-12.4) 0.189
Absolute Alc g/L 7.29 (7.18-7.41) 7.55 (7.39-7.71) 0.008 **
n 95 69

* indicates significance at <0.05; ** indicates significance at <0.01
COPYRIGHT 2006 American Society for Clinical Laboratory Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:RESEARCH AND REPORTS; fasting plasma glucose
Author:Behan, Kristina Jackson
Publication:Clinical Laboratory Science
Geographic Code:1USA
Date:Sep 22, 2006
Previous Article:Two non-invasive diagnostic tools for invasive aspergilosis: (1-3)-[beta]-D-Glucan and the galactomannan assay.
Next Article:An overview of the human immunodeficiency virus featuring laboratory testing for drug resistance.

Related Articles
Aging and glycation: the importance of hemoglobin A1c testing.
Biological variation of glycohemoglobin.
Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus.
Serum selenium concentrations and diabetes in U.S. adults: National Health and Nutrition Examination Survey (NHANES) 2003-2004.
[HbA.sub.1c] for diabetes diagnosis.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters