| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Obstetrics and Gynecology, The University of Hong Kong, Hong Kong, China, and Tsan Yuk Hospital, Hong Kong, China.
Address reprint requests to: Terence T. Lao, MBBS, FRCOG, Queen Mary Hospital, Department of Obstetrics and Gynecology, 102 Pokfulam Road, Hong Kong SAR, China; E-mail: laotth{at}hkucc.hku.hk.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: In a prospective observational study, 762 nondiabetic Chinese women with singleton pregnancies, whose initial visit hemoglobin concentration and mean cell volume were 10 g/dL or more and 80 fL or more, respectively, recruited at 28–30 weeks, had blood drawn for repeat measurement of hemoglobin concentration and iron parameters. These women were categorized by their initial visit hemoglobin concentration into quartiles, and the incidence of GDM was analyzed together with the maternal characteristics and iron status.
RESULTS: The final study sample comprised 730 women. Compared with the rest, the group in the highest hemoglobin quartile (more than 13 g/dL) had a significantly higher incidence of GDM (18.7% versus 10.9%, P = .007), as well as greater age, weight, and serum ferritin and iron concentrations. Logistic regression was used to examine the effects of high body mass index (more than 25 kg/m2), advanced age (older than 34 years), parity of 1 or more, and hemoglobin in the highest quartile, on the incidence of GDM. Only advanced age (odds ratio 3.79, 95% confidence interval 2.33, 6.17) and hemoglobin in the highest quartile (odds ratio 1.73, 95% confidence interval 1.08, 2.78) emerged to be significant factors.
CONCLUSION: A high maternal hemoglobin (more than 13 g/dL) at the initial prenatal visit in Chinese women is an independent risk factor for GDM. This may reflect a better nutritional status in these women, as suggested by the increased iron status.
There is now accumulated evidence in the literature that high maternal hemoglobin is associated with increased incidence of adverse pregnancy outcomes, which include low birth weight and small-for-gestational-age newborns, preterm birth, increased perinatal mortality, and preeclampsia.1–6 Although there is no universal criterion of what constitutes a high hemoglobin concentration, the data in aforementioned studies indicate that a hemoglobin concentration of 14 g/dL in the first trimester, and between 12.5 g/dL and 13.5 g/dL for the rest of pregnancy, are appropriate cutoff values. However, a high maternal hemoglobin concentration has never been categorically labeled as a risk factor for pregnant women, and there are no data on the association of a high maternal hemoglobin concentration with complications other than those aforementioned.
Recently, a case-control study in Chinese women with a body mass index (BMI) of more than 26 kg/m2 has shown that those who developed the World Health Organization category of impaired glucose tolerance, with the 2-hour glucose value of the 75-g oral glucose tolerance test between 8.0 mmol/L (144 mg/dL) and 10.9 mmol/L (196 mg/dL),7 during pregnancy had significantly increased hemoglobin concentration compared with BMI-matched controls.8 Indeed, in the same population, the incidence of gestational diabetes mellitus (GDM) in women with iron-deficiency anemia was not only lower than that in the nonanemic women, but it was also significantly lower than that in women with thalassemia traits.9 In the nonpregnant population, an association between hemoglobin concentration and red cell count with diabetes mellitus has been reported before. Diabetic subjects were found to have increased total red cell count compared with age- and sex-matched controls.10 In one prospective study on 10,059 men, hemoglobin was significantly related to the incidence of diabetes mellitus.11 This association may be more significant in women, as other studies indicated that hemoglobin was positively and significantly related to fasting glucose in women but not in men,12,13 and that red cell count was positively related to glucose intolerance in women.13,14 It is, therefore, possible that the association between higher maternal hemoglobin with gestational glucose intolerance reported previously8 was part of the same phenomenon found in the nonpregnant women.
To confirm whether a higher maternal hemoglobin concentration is associated with increased risk of GDM in previously nondiabetic women, this study was conducted to address two issues. The first is to determine whether a high maternal hemoglobin concentration, defined as a hemoglobin value in the highest quartile of the study cohort at initial visit before 14 weeks’ gestation, is associated with increased incidence of GDM. The second is whether women with a high initial visit hemoglobin concentration have increased iron stores at 28–30 weeks’ gestation, even though high-dose iron supplementation has not been prescribed. This is because it has been demonstrated in the same population that women who developed GDM in the third trimester had evidence of increased iron stores.15,16
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
- or ß-thalassemia, for which termination of pregnancy would be arranged. Patients having hemoglobin level less than 10 g/dL at any time during pregnancy are diagnosed to have anemia. They will be further investigated to determine the cause of anemia, and they will be treated accordingly. At 28–30 weeks, the hemoglobin level is repeated to identify patients who have subsequently developed anemia. Women with risk factors for the development of GDM, such as advanced maternal age (older than 34 years), relevant past obstetric and family history, prepregnant weight of 75 kg or more, and recurrent/ significant glycosuria, will have the 75-g oral glucose tolerance test arranged after the initial visit. The result of the oral glucose tolerance test is interpreted by the World Health Organization criteria, and GDM is diagnosed if the oral glucose tolerance test 2-hour glucose value is 8.0 mmol/L (144 mg/dL) or more.7 In addition, all low-risk women have blood drawn for random blood glucose testing at the beginning of the third trimester (28–30 weeks). Those with abnormal spot glucose (more than 5.8 mmol/L if less than 2 hours postprandial and more than 5.0 mmol/L if more than 2 hours postprandial) will also undergo the 75-g oral glucose tolerance test as described before.17
In a prospective observational study approved by the hospital ethics committee, we recruited, over a 4-month period, consecutive women with singleton pregnancies, who had normal hemoglobin and mean corpuscular volume at the initial antenatal visit at or before 14 weeks’ gestation, and who attended one of the antenatal clinics at 28–30 weeks. Women with low hemoglobin and mean corpuscular volume were excluded because the high dose of iron treatment given beforehand would have influenced their iron stores and possibly their pregnancy outcome. After informed consent, an additional 4-mL sample of blood was taken for the study of iron parameters at the time of the repeat measurement of hemoglobin and mean corpuscular volume and random glucose screening. The result of the iron status was blinded to the women and the obstetric staff. Their records were marked for the later retrieval of data and the serial number registered, but the management of the remaining part of their pregnancies was not influenced by the study, and those diagnosed with GDM were managed as per protocol. The exclusion criteria were antenatal visit after 14 weeks’ gestation, preexisting anemia or hemoglobinopathies, GDM diagnosed before 28 weeks’ gestation (as the maternal hemoglobin and iron concentrations could have been influenced by diet therapy), and multifetal pregnancy.
The additional blood sample was collected into a plain plastic tube and then sent to the laboratory. The serum was aliquoted for the later batched assay of serum ferritin concentration (Microparticle Enzyme Immunoassay, IMx System of Abbott Laboratories, Abbott Park, IL), and serum iron and transferrin (measured as the total iron-binding capacity, Calorimetric Method, Sigma, St. Louis, MO). The transferrin saturation was calculated as the result of serum iron divided by the total iron-binding capacity. After delivery, the records were retrieved for analysis and matched with the results of the iron parameters. The hemoglobin concentration at the first visit for the entire cohort was used to generate the 75th, 50th, and 25th percentile values, and the patients were categorized into four groups according to the quartile ranking of their hemoglobin result. The four groups were compared for the maternal and infant characteristics, the incidence of GDM, and hemoglobin and iron parameters. Statistical analysis was performed using a commercial computer package (SPSS for Windows, SPSS Inc., Chicago, IL). Categoric variables were compared with the 
2 test, and correlation between the incidence of GDM with the hemoglobin quartiles was tested by Pearson correlation. Odds ratios (ORs) with 95% confidence intervals (CIs) were generated as indicated. For continuous variables, the values that are normally distributed were expressed as mean ± standard deviation and tested by one-way analysis of variance with the Duncan test set at a significance level of 5% for post hoc analysis. For values that are non-normally distributed, data were expressed as mean, median (95% CI of mean). Stepwise multiple logistic regression analysis was used to determine the role of hemoglobin in the incidence of GDM, adjusting for the possible confounding factors found in the initial analysis.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The maternal characteristics of the four groups are shown in Table 1
. Significant difference was found among the four groups in the maternal age and weight, but not in the height or calculated BMI. Using an age of 34 years or older to define advanced maternal age, there was a significant correlation between the incidence of advanced age and hemoglobin quartiles. Using a BMI of 25 kg/m2 to define overweight, the incidence of overweight was also significantly correlated with the hemoglobin quartiles. There was a significant difference in the incidence of multiparous women among the four groups, but there was no correlation with the quartiles. The incidence of GDM was significantly different among the four groups and was correlated with the quartiles.
|
).
|
. The mean hemoglobin concentration was significantly different among the four groups as expected, but the hemoglobin concentration at 28–30 weeks and that on day 3 postpartum were also significantly different among the four groups, with the highest concentration in the highest quartile. This was not associated with any difference in the estimated blood loss at delivery. On the other hand, there was no significant difference in the values of the mean corpuscular volume among the four groups either at initial visit or at 28–30 weeks.
|
). The serum ferritin concentration was highest in the highest quartile and lowest in the lowest quartile, whereas serum iron concentration was similarly higher in the first three quartiles compared with the lowest quartile.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In nonpregnant subjects, the association between high hemoglobin concentration and red cell count with diabetes mellitus is attributed to the increased proportion of glycosylated hemoglobin in diabetic subjects. As glycosylated hemoglobin has increased oxygen affinity, a higher level of glycosylated hemoglobin can lead to tissue hypoxia, which in turn results in increased red cell count.10,18 However, the effect of glycosylated hemoglobin on the oxygen affinity pattern of hemoglobin at the concentrations normally found in vivo in diabetic subjects has been challenged.19 Nevertheless, the majority opinion is that the increased oxygen affinity of glycosylated hemoglobin does result in sufficient tissue hypoxia, which leads to an increase in hemoglobin and red cell count.10,18,20,21 Indeed, the total red cell count was positively correlated with the proportion of glycosylated hemoglobin in diabetic subjects.10 Furthermore, a correlation between reticulocyte count with glycemic control, paralleling the hemoglobin-A1c level, has been demonstrated in diabetic subjects after the exclusion of those with overt anemia and low serum iron concentration.20 This process is probably mediated by erythropoietin, as anemia occurs secondary to erythropoietin depletion in type 1 diabetic subjects with autonomic neuropathy and nephropathy compared with matched type 1 diabetic controls without autonomic neuropathy.22,23 These findings, therefore, suggest that the increased hemoglobin and red cell count in diabetic subjects represented the effect of chronic hyperglycemia.
In this study, GDM was diagnosed in the last trimester. Although it was possible that some of these women could have had previously undiagnosed diabetes mellitus, and that their diabetic state would have remained asymptomatic and undiscovered till the last trimester, the number of these women would have been extremely small. In the great majority, the effect of increased glycosylation of hemoglobin could not have influenced the hemoglobin concentration in the first antenatal visit. On the other hand, hemoglobin concentration is related to body iron stores. The patients in the highest hemoglobin quartile not only had increased serum concentrations of ferritin and iron, but also the highest hemoglobin concentration for the third trimester and day 3 postpartum, which was unlikely to be explained by the effect of inadequate hemodilution. Thus, the relationship between hemoglobin concentration and incidence of GDM found in this study could be a reflection of the relationship between increased maternal iron status with GDM as reported before.15,16
During pregnancy, maternal iron stores are depleted as reflected by a declining maternal serum ferritin concentration with advancing gestation, and despite iron supplementation commenced before midpregnancy.24–26 We speculate that the higher iron stores in patients in the highest quartile are consequent to their better nutritional status because there was a positive correlation between the hemoglobin quartiles with maternal weight and BMI. Also, the higher maternal age and incidence of advanced age, in the presence of a similar incidence of multiparous women, were compatible with the known behavior of the more affluent women in our society, who tend to have their pregnancy later in the reproductive age. One possible contributory factor is that these patients had increased their intake of iron out of their own initiative, as iron supplementation in high doses was thought to have played a role in the association between high hemoglobin concentration and fetal growth retardation.1
Our finding of an association between high hemoglobin concentration at initial visit and the subsequent development of GDM had not been reported before. Our results indicate that in Chinese women without underlying hemoglobinopathies, an initial visit hemoglobin concentration of more than 13 g/dL is an independent risk factor of subsequent GDM. Further studies in other ethnic groups are warranted to determine whether a high hemoglobin concentration can be used as a predictor for GDM and other pregnancy complications.
|
Footnotes |
|---|
Received August 3, 2001. Received in revised form November 20, 2001. Accepted December 18, 2001.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Huisman A, Aarnoudse JG. Increased 2nd trimester hemoglobin concentration in pregnancies later complicated by hypertension and growth retardation. Early evidence of a reduced plasma volume. Acta Obstet Gynecol Scand 1986; 65:605–8.[Medline]
3. Murphy JF, O’Riordan J, Newcombe RG, Coles EC, Pearson JF. Relation of haemoglobin levels in first and second trimesters to outcome of pregnancy. Lancet 1986; i:992–5.
4. Knottnerus JA, Delgado L, Knipschild PG, Essed GGM, Smits F. Maternal hemoglobin and pregnancy outcome. Lancet 1986;ii:282.
5. Rasmussen S, Øian P. First- and second-trimester hemoglobin levels. Relation to birth weight and gestational age. Acta Obstet Gynecol Scand 1993;72:246–51.[Medline]
6. Steer P, Alam MA, Wadsworth J, Welch A. Relation between maternal haemoglobin concentration and birth weight in different ethnic groups. Br Med J 1995;310: 489–91.
7. World Health Organization Expert Committee on Diabetes Mellitus. Technical report series 646. Geneva. World Health Organization, 1980:8–12.
8. Lao TT, Ho LF. Impaired glucose tolerance and pregnancy outcome in Chinese women with high body mass index. Hum Reprod 2000;15:1826–9.
9. Lao TT, Pun TC. Anaemia in pregnancy—Is the current definition meaningful? Eur J Obstet Gynecol Reprod Biol 1996;68:53–8.[Medline]
10. Graham JJ, Ryall RG, Wise PH. Glycosylated haemoglobin and relative polycythaemia in diabetes mellitus. Diabetologia 1980;18:205–7.[Medline]
11. Medalie JH, Papier C, Herman JB. Diabetes mellitus among 10,000 adult men: I. Five-year incidence and associated variables. Israel J Med Sci 1974;10:681–97.[Medline]
12. Rao GMM, Morghom LO. Erythrocyte and hemoglobin levels in diabetic women. Eur J Appl Physiol 1984;52: 272–5.
13. Harlan LC, Harlan WR, Landis JR, Goldstein NG. Factors associated with glucose tolerance in adults in the United States. Am J Epidemiol 1987;126:674–84.
14. Rao GMM. Erythrocyte count changes in obesity and diabetes. Indian J Zool 1987;10:1–6.
15. Lao TT, Tam KF. Maternal serum ferritin and gestational impaired glucose tolerance. Diabetes Care 1997;20: 1368–9.[Abstract]
16. Lao TT, Chan PL, Tam KF. Gestational diabetes mellitus in the last trimester—A feature of maternal iron excess? Diabet Med 2001;18:218–23.
17. Lao TT, Lee CP. Gestational ‘impaired glucose tolerance’: Should the cut-off be raised to 9 mmol-1? Diabet Med 1998;15:25–9.[Medline]
18. Roberts AO, Story CJ, Ryall RG. Erythrocyte 2,3-bisphosphoglycerate concentrations and haemoglobin glycosylation in normoxic Type 1 (insulin-dependent) diabetes mellitus. Diabetologia 1984;26:389–91.[Medline]
19. Samaja M, Melotti D, Carenini A, Pozza G. Glycosylated haemoglobins and the oxygen affinity of whole blood. Diabetologia 1982;23:399–402.[Medline]
20. Fekete T, Sopon E. Glycaemic control and reticulocyte count in diabetic patients. Horm Metabol Res 1986;18: 141.[Medline]
21. Giugliano D. Glycosylated haemoglobin and reticulocyte count in diabetes. Diabetologia 1982;22:223.[Medline]
22. Rocerca BM, Todero L, Caputo S, Cotroneo P, Damiani P, Manto A. Blunted erythropoietin response to anaemia in type 1 diabetes patients. Diabetes Care 1999;22:647–51.
23. Winkler AS, Marsden J, Chaudhurit KR, Hambley H, Watkins PJ. Erythropoietin depletion and anaemia in diabetes mellitus. Diabet Med 1999;16:813–9.[Medline]
24. Kaneshige E. Serum ferritin as an assessment of iron stores and other haematologic parameters during pregnancy. Obstet Gynecol 1981;57:238–42.
25. Thomsen JK, Prien-Larsen JC, Devantier A, Fogh-Andersen N. Low dose iron supplementation does not cover the need for iron during pregnancy. Acta Obstet Gynecol Scand 1993;72:93–8.[Medline]
26. Milman N, Agger AO, Nielsen OJ. Iron status markers and serum erythropoietin in 120 mothers and newborn infants. Acta Obstet Gynecol Scand 1994;73:200–4.[Medline]
This article has been cited by other articles:
|
|
 |
|
  X. Chen, T. O. Scholl, and T. P. Stein Association of Elevated Serum Ferritin Levels and the Risk of Gestational Diabetes Mellitus in Pregnant Women: The Camden Study Diabetes Care, May 1, 2006; 29(5): 1077 - 1082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. T. Lao and L.-F. Ho Impact of Iron Deficiency Anemia on Prevalence of Gestational Diabetes Mellitus Diabetes Care, March 1, 2004; 27(3): 650 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. T. Lao, K.-Y. Tse, L. Y. Chan, K.-F. Tam, and L.-F. Ho HBsAg Carrier Status and the Association Between Gestational Diabetes With Increased Serum Ferritin Concentration in Chinese Women Diabetes Care, November 1, 2003; 26(11): 3011 - 3016. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
