Hemoglobin A1c determinations in diabetic pregnancy

Translating HbA1c measurements into estimated average glucose values in pregnant women with diabetes

AIMS/HYPOTHESIS

This study aimed to examine the relationship between average glucose levels, assessed by continuous glucose monitoring (CGM), and HbA_1c levels in pregnant women with diabetes to determine whether calculations of standard estimated average glucose (eAG) levels from HbA_1c measurements are applicable to pregnant women with diabetes.

METHODS

CGM data from 117 pregnant women (89 women with type 1 diabetes; 28 women with type 2 diabetes) were analysed. Average glucose levels were calculated from 5-7 day CGM profiles (mean 1275 glucose values per profile) and paired with a corresponding (±1 week) HbA_1c measure. In total, 688 average glucose-HbA_1c pairs were obtained across pregnancy (mean six pairs per participant). Average glucose level was used as the dependent variable in a regression model. Covariates were gestational week, study centre and HbA_1c.

RESULTS

There was a strong association between HbA_1c and average glucose values in pregnancy (coefficient 0.67 [95% CI 0.57, 0.78]), i.e. a 1% (11 mmol/mol) difference in HbA_1c corresponded to a 0.67 mmol/l difference in average glucose. The random effects model that included gestational week as a curvilinear (quadratic) covariate fitted best, allowing calculation of a pregnancy-specific eAG (PeAG). This showed that an HbA_1c of 8.0% (64 mmol/mol) gave a PeAG of 7.4-7.7 mmol/l (depending on gestational week), compared with a standard eAG of 10.2 mmol/l. The PeAG associated with maintaining an HbA_1c level of 6.0% (42 mmol/mol) during pregnancy was between 6.4 and 6.7 mmol/l, depending on gestational week.

CONCLUSIONS/INTERPRETATION

The HbA_1c-average glucose relationship is altered by pregnancy. Routinely generated standard eAG values do not account for this difference between pregnant and non-pregnant individuals and, thus, should not be used during pregnancy. Instead, the PeAG values deduced in the current study are recommended for antenatal clinical care.

Method for chronological recording of antigen appearance in human head-hair shafts and its use for monitoring glycation products in diabetes

We describe immunochemical assays of non-enzymatic glycation products in human head-hair protein extracts and hair cross sections using Western blots and a novel "dot-block" methodology. In the latter, groups of approximately 15 hair fibers, clipped at about 1 mm proximal to the scalp-skin were aligned, wound around, and attached to 3 mm diameter araldite screw rods. Up to 40 such rods were next embedded lengthwise in additional araldite polymer creating a solid block and the top surface of the block was sectioned off to the half-diameters of the screw rods thus exposing accurately transected hair cross sections at regular ( approximately 0.5 cm) intervals. Early- and advanced-glycation products (EGAs and AGEs, respectively) were determined in the exposed cross sections in-situ using specific antibodies and ECL densitometry as in conventional Western blots. Both Western blots and this technique demonstrated 3.1 fold EGAs increases in the proximal 2 cm of hair of diabetics as compared to non-diabetics. Dot-blocks, in addition, were less variable and demonstrated exponential EGAs decreases along fibers distally, with calculated intercepts (at the hair roots) of 4.9 fold increases in diabetics as opposed to non-diabetics and half-lives of 6.0, 5.9 and 9.0 months in hair of non-diabetics, gestational diabetics and diabetic patients, respectively. Correlations in amounts of BG vs. HbA1(c), BG vs. EGAs, and HbA1(c) vs. EGAs, using dot-block and clinical lab data were all significant (p<0.05). Acute onset T1D patients, defined as previously unsuspected patients diagnosed upon hospitalization due to diabetic complications, exhibited nearly identical EGAs levels in their proximal 0-9 cm hair as did T1D patients with long-established diabetes, thus supporting the notion of long and insidious T1D etiology. Removal of 1-2 microm layers from dot-block surfaces enabled their re-use for multiple assays. Applied anti-AGEs antibodies demonstrated slight decreases or no significant changes in CML and MGI along hair shafts of normal and diabetic subjects. Fluctuations in EGAs and AGEs along hair shafts, indicating alterations in glycemic control were also observed. We conclude that the dot-block method has a potential for early diagnosis and monitoring of diabetes, and more generally, as a long term "biological record" of various chronic medical conditions.

Glucose monitoring and insulin therapy during pregnancy

Excellent blood glucose control is necessary to reduce the excess fetal morbidity and mortality associated with the diabetic pregnancy. This article outlines the roles of glucose monitoring and insulin therapy in intensive treatment regimens during gestation. The discussion includes recommended monitoring frequency, glycemic standards, types of insulin and mechanism of action, goals and timing of insulin therapy, as well as the complications of insulin therapy.

Use of fructosamine and glycated haemoglobin to verify self blood glucose monitoring data in diabetic pregnancy

Relationships between fructosamine and HbA1, and mean blood glucose over the previous 1-8 weeks, determined from self blood glucose monitoring with memory meters, were studied prospectively throughout 16 pregnancies in Type 1 diabetic women. Fructosamine correlated best (Spearman rank) with mean blood glucose over the previous 2 weeks in the first and second trimesters (0.5) and over the previous 1 week in the third trimester (0.39). HbA1 correlated best with mean blood glucose over the previous 8 weeks in the first and second trimesters (0.56), but over the previous 2 weeks in the third trimester (0.524) probably because of increased erythropoiesis in late pregnancy. From Deming regression models, 95% prediction intervals for mean blood glucose for fructosamine and HbA1 values were calculated, showing that fructosamine predicted levels of mean blood glucose more precisely than HbA1. These intervals can be used to estimate an individual pregnant diabetic woman's mean blood glucose from her fructosamine or HbA1 results and to verify self blood glucose monitoring data. In well-controlled diabetic pregnancy, both fructosamine and HbA1 reliably indicated trends in blood glucose but fructosamine estimated blood glucose levels more precisely.

The influence of maternal weight, smoking, vascular complications and glucose regulation on the birth weight of infants of type 1 diabetic women

The influence of maternal blood glucose regulation, weight gain, pre-pregnancy weight, vascular complications and smoking on the birth weight of infants was investigated in 72 type 1 (insulin-dependent) pregnant diabetic women. In patients with vascular complications (n = 23) the birth weight was significantly lower than in patients without vascular involvement (n = 49) (mean 380 g, P less than 0.05). The mean RBWR (relative birth weight ratio) of infants of patients who smoked more than 10 cigarettes per day was statistically significant lower compared to the mean RBWR of infants of non-smokers (P less than 0.025). A significant correlation was present between haemoglobin A1c (HbA1c) and RBWR (r = 0.28, P less than 0.02) and between maternal net weight gain and RBWR (r = 0.36, P less than 0.005) and this correlation became even stronger when only patients without vascular lesions were considered (n = 49) (r = 0.51, P less than 0.005) and (r = 0.50, P less than 0.005), respectively. In contrast no correlation was found between pre-pregnancy weight and RBWR. The study suggests that factors other than maternal hyperglycaemia stimulate fetal growth and may explain why fetal macrosomia may occur despite of strict blood glucose regulation.

Unexplained variability of glycated haemoglobin in non-diabetic subjects not related to glycaemia

We have studied levels of glycated haemoglobin in a sample of 223 people aged over 40 years without known diabetes mellitus screened in a community study. Each had a glucose tolerance test and glycated haemoglobin measured by four methods - agar gel electrophoresis with and without removal of Schiff base, affinity chromatography and isoelectric focusing. The correlation coefficients between 2 h blood glucose and levels of glycated haemoglobin were between 0.43 and 0.64. This poor correlation was not explained on the basis of assay or biological variability of either 2 h blood glucose or glycated haemoglobin. Multiple regression analysis showed that other assays of glycated haemoglobin contributed to the variance of any single glycated haemoglobin value by 0.1%-52.9% (median 12.8%) compared to the variance of 18.6%-41.4% (median 30.8%) explained by 2 h blood glucose alone, suggesting that in a non-diabetic population, the degree of glucose intolerance may explain only one third of the variance of glycated haemoglobin levels, but other factors operate to produce consistent changes in levels of glycated haemoglobin. Investigation of 42 subjects with consistently high (20 subjects) or low (22 subjects) levels of glycated haemoglobin relative to their 2 h blood glucose level showed no difference in age, gender, body mass index, haemoglobin levels or smoking, although 50% of low glycators had impaired glucose tolerance. Neither ambient blood-glucose levels, as estimated on two five-point blood-glucose profiles, nor dietary intake of carbohydrate, starch, sugars, fibre or alcohol, explained the difference between high and low glycators. The determinants of the consistent interindividual differences in levels of glycated haemoglobin in non-diabetic subjects remain to be determined.

Verified self-monitored blood glucose data versus glycosylated hemoglobin and glycosylated serum protein as a means of predicting short- and long-term metabolic control in gestational diabetes

Glycosylated hemoglobin and glycosylated serum protein have been suggested as tools for evaluation of long- and short-term glycemic control, respectively. Twenty-six patients with gestational diabetes were prospectively studied to determine the relationship of glycosylated hemoglobin and glycosylated serum protein to metabolic control. To verify the accuracy of blood glucose data, a memory-based reflectance meter was used for subjects with gestational diabetes who tested 6.5 +/- 1 times per day. Our analysis revealed that despite a statistically positive correlation between glycosylated hemoglobin, glycosylated serum protein, and verified data, their use as a clinical tool is limited because of their poor predictability.

Haemoglobin A1c levels in normal and diabetic pregnancies

Serial measurements of the HbA1c levels were performed during pregnancy in 4 groups of patients attending Antenatal Clinics: 36 normal pregnancies; 16 pregnancies in established insulin-dependent diabetic patients; 9 patients with gestational diabetes diagnosed during that pregnancy; and 21 patients who had been diagnosed as having gestational diabetes in at least one previous pregnancy. In the normal pregnancy HbA1c levels showed a small but significant increase from the end of the first trimester to delivery despite blood glucose levels remaining constant throughout. In the insulin-dependent and gestational diabetic patients, blood glucose levels remained significantly higher than in the normal throughout pregnancy but only in insulin-dependent diabetic patients and the newly diagnosed untreated gestational diabetic patients were the HbA1c levels significantly higher than in the normal. In those patients who had previous pregnancies complicated by gestational diabetes, blood glucose levels were significantly higher than in the normal but HbA1c levels were not. This dissociation between blood glucose and HbA1c levels in gestational diabetic pregnancies in particular limits the value of HbA1c levels in monitoring antidiabetic treatment in such pregnancies.

Screening for carbohydrate intolerance in pregnancy: a comparison of two tests and reassessment of a common approach

The usefulness of glycosylated hemoglobin as a prenatal screening test for carbohydrate intolerance was studied in 806 consecutive subjects by correlating glycosylated hemoglobin to 1-hour post-50 gm Glucola plasma glucose levels, 3-hour oral glucose tolerance tests, and perinatal and maternal outcomes. Sixty-seven subjects whose 1-hour post-50 gm Glucola plasma glucose levels were greater than or equal to 150 mg/100 ml underwent 3-hour oral glucose tolerance tests; 12 were diagnostic of carbohydrate intolerance. Compared to carbohydrate-tolerant control subjects, gravid patients with carbohydrate intolerance were older, more obese, had higher 1-hour post-50 gm Glucola plasma glucose and glycosylated hemoglobin levels, and infants with increased birth weight percentiles, depressed 5-minute Apgar scores, and an increased incidence of shoulder dystocia and perinatal mortality. Three of 10 carbohydrate-intolerant patients who were evaluated post partum were found to have previously undiagnosed diabetes. Division of measurements of 1-hour post-50 gm Glucola plasma glucose and glycosylated hemoglobin into normal, borderline, and suspicious groups demonstrated a reduction in discriminatory capability of glycosylated hemoglobin as compared to the 1-hour post-50 gm Glucola plasma glucose. We conclude that laboratory screening for carbohydrate intolerance should be a standard element of the prenatal evaluation; gravid patients found to have carbohydrate intolerance should be reevaluated post partum to rule out overt diabetes, and the 1-hour post-50 gm Glucola plasma glucose test is the preferred means of routine screening for carbohydrate intolerance in pregnancy.

Obstetric care in diabetic pregnancy

In diabetic pregnancy near-normalization of maternal blood glucose levels improves the perinatal outcome. Strict metabolic control can be achieved by self-monitoring of blood glucose in ambulant praxis. The obstetric supervision may now therefore be organized on an out-patient basis aiming at early recognition of pregnancy complications such as preeclampsia and deviation in fetal growth. For uncomplicated and well-controlled diabetes without vascular complications the obstetric care should be individualized and routine programmes for obstetric surveillance, such as fetal heart rate monitoring and determination of fetal maturity, are usually not necessary. Special attention should, however, be paid to patients with poor metabolic control or vascular complications, particularly in the presence of disturbances of intra-uterine growth.

Glycosylated serum protein levels in diabetic and nondiabetic pregnant patients: an indicator of short-term glycemic control in the diabetic patient

Measurement of the level of nonenzymatic glycosylation of blood proteins with more rapid turnover times than hemoglobin has been suggested as an indicator of time-averaged glucose control in nonpregnant diabetic patients. Using affinity chromatography, we have measured the levels of glycosylated serum proteins during pregnancy in 14 normal volunteers and 15 insulin-dependent diabetic patients. No relationship was noted between the percentage of glycosylated serum proteins in serum from normal patients and the gestational age at the time of sampling in the second and third trimesters of pregnancy. When the relative frequency distribution of glycosylated serum protein levels in normal patients was compared with that in diabetic patients, a significant difference was noted between the two groups, with a higher percentage of glycosylated serum protein levels in diabetic patients being at elevated values compared to those in normal patients. Normal patients had measured glycosylated serum protein levels of 12.5% +/- 2.2% whereas diabetic patients had glycosylated serum protein levels of 14.0% +/- 3.6%. When peak fasting serum glucose and high Chemstrip glucose levels were compared with glycosylated serum proteins in the diabetic population, a significant correlation for each was noted. The best correlation resulted from a comparison of an average Chemstrip glucose level (mean of 49 glucose values during the previous week) and the glycosylated serum protein value obtained at the end of that week. This inexpensive assay can be adapted to any clinical laboratory and should provide an objective means to evaluate short-term glycemic control, complementing the evaluation provided by self-glucose monitoring (immediate control) and intermittent assay of glycosylated hemoglobin (long-term control).

Glycosylated hemoglobin as a screening test for carbohydrate intolerance in pregnancy

The usefulness of glycosylated hemoglobin as a prenatal screening test for carbohydrate intolerance was studied in 806 consecutive subjects by correlating glycosylated hemoglobin to 1-hour post-50 gm Glucola plasma glucose (1 degree G) levels, and 3-hour oral glucose tolerance tests (3 degrees GTT). Sixty-seven subjects whose 1 degree G greater than or equal to 150 mg/100 ml received a 3 degrees GTT; 12 were diagnostic of carbohydrate intolerance. Compared to carbohydrate-tolerant controls, carbohydrate-intolerant gravid patients had higher 1 degree G (p less than 0.001) and glycosylated hemoglobin (p less than 0.05) levels. Linear regression analysis of 1 degree G and glycosylated hemoglobin demonstrated r = 0.35 (p less than 0.0001). Compared to the glycosylated hemoglobin test, the 1 degree G screening test has greater specificity, sensitivity, and predictive value for a positive diagnosis. Consequently, the 1 degree G is a better routine screening test for carbohydrate intolerance than is glycosylated hemoglobin.

Seven years experience of home management in pregnancy in women with insulin-dependent diabetes

Fifty-eight of a consecutive series of 75 pregnancies in women with insulin-dependent diabetes went into the third trimester. Diabetes was managed by home blood glucose monitoring and women were not routinely admitted at any stage before delivery. The mean number of in-patient days before delivery was 15 for the whole series but has been reduced to 9 during the past four years. Each woman performed an average of 171 blood glucose measurements during her pregnancy. Mean blood glucose (including post-prandial levels) fell significantly from 7.9 mmol/l in the first trimester to 7.3 in the second and 6.4 in the third. Mean percentage of haemoglobin A1 was within the normal range in the second and third trimesters. The cesarean section rate was high at 66% but there were no perinatal deaths. Three infants had congenital abnormalities. We conclude that home blood glucose monitoring is a safe and effective way of managing pregnant diabetic women as out-patients. The cost of meters and sticks is repaid many times over in the saving of hospital costs. In addition, home blood glucose monitoring is popular with the patients and many choose to continue it after delivery.

Haemoglobin A1c in pregnant diabetic patients at delivery and haemoglobin F1 in cord blood from their newborns

Haemoglobin A1c has been measured in diabetic and non-diabetic pregnant women at delivery as well as haemoglobin F adducts in paired cord blood of their infants. Haemoglobin A1c and haemoglobin F adduct estimates covariate in a group of diabetic and non-diabetic pregnant women/infants studied by ion exchange chromatography, but not when studied by isoelectric focusing. Haemoglobin A1c and birth weight ratio covariate when haemoglobin A1c is determined by isoelectric focusing but not when estimated by ion exchange chromatography. Similar results appear from the studies of others indicating that mothers' haemoglobin A1c covariates with cord blood haemoglobin F adducts and with the birth weight ratio, though correlation is scant because of extensive and efficient control of diabetic pregnant women.

Biphasic changes in hemoglobin A1c concentrations during normal human pregnancy

Analyses of hemoglobin A1c concentrations were performed throughout gestation in 377 nondiabetic women. We observed significant biphasic changes in hemoglobin A1c concentrations, with an initial gradual decline to a nadir level at 24 weeks, followed by a subsequent slow reascension to peak near term. All these changes fell within the usual range of normal values for hemoglobin A1c. Values for plasma glucose estimated 1 hour after a 50 gm oral glucose load in 1,756 normal gravid women showed similar biphasic excursions, with the nadir occurring 4 weeks earlier, i.e., at 20 weeks' gestation. We conclude that the small but significant changes in hemoglobin A1c during the course of normal gestation reflect, with an appropriate displacement in time, the biphasic alterations in mean blood sugar that characterize the sequential changes in glucoregulation during normal pregnancy.

Relationship between glycosylation of haemoglobin and the duration of diabetes: a study during the third trimester of pregnancy

The relationship between blood glucose and glycosylated haemoglobin (HbAlc) had been investigated during an 8 week period in 53 Type 1 (insulin-dependent) diabetic women studied during the third trimester of pregnancy. Blood glucose estimations (fasting and 2h post-prandially) were made an average of 41 times in each patient during this period and HbAlc was determined once at the end of the study. There was a significant correlation between both the mean blood glucose over the preceding 8 weeks and the standard deviation of the fasting blood glucose with HbAlc (r = 0.69, p less than 0.001; r = 0.46, p less than 0.001, respectively). A "glycosylation index" was calculated for each patient (HbAlc divided by the mean blood glucose value). There was a significant correlation between the "glycosylation index" and duration of diabetes (r = 0.68, p less than 0.001). In contrast, there was no correlation between red cell 2,3-diphosphoglycerate and HbAlc or "glycosylation index". These findings suggest that increasing duration of diabetes influences the post-translational formation of HbAlc and that isolated HbAlc values need to be interpreted with caution in the pregnant diabetic.