SUBMITTED 19 SEP 21 1 REVISIONS REQ. 25 NOV 21 & 19 JAN 22; REVISIONS RECD. 5 DEC 21 & 24 JAN 22 2 ACCEPTED 20 FEB 22 3 ONLINE-FIRST: MARCH 2022 4 DOI: https://doi.org/10.18295/squmj.3.2022.022 5 6 Establishing Trimester-Specific Hemoglobin A1c Reference Levels for 7 Pregnant Women 8 A retrospective study among healthy South Asian women with normal 9 pregnancy outcomes 10 11 *John Punnose,1 Rajeev K. Malhotra,5 Komal Sukhija,1 Rashika M. 12 Rijhwani,1 Asha Sharma,2 Naimaa Choudhary,2 Prassan Vij,3 Renuka 13 Joseph4 14 Departments of 1Endocrinology, 2Obstetrics and Gynecology, 3Reproductive Medicine and 15 4Biochemistry, St. Stephen’s Hospital, New Delhi, India; 5All India Institute of Medical 16 Science, Ansari Nager, New Delhi, India 17 *Corresponding Author’s e-mails: drpunnose@rediffmail.com; 18 drpunnose@ststephenshospital.org 19 20 Abstract 21 Objectives: This study aimed to define trimester-specific hemoglobin A1c (A1c) reference 22 intervals among healthy South Asian pregnant women. Methods: In this restrospective 23 study,1357 pregnant women without diabetes, gestational diabetes, gestational hypertension, 24 anemia, β-thalassemia, or systemic diseases were included. They had term delivery of babies 25 having weight appropriate for gestational age. A1c (using high performance liquid 26 chromatography, meeting the National Glycohemoglobin Standardization Program and 27 International Federation of Clinical Chemistry standards), hemoglobin, and RBC indices 28 were estimated at the first antenatal visit. The A1c levels were calculated in terms of non-29 parametric 2.5 and 97.5 percentiles for women in first (T1), second (T2), and third (T3) 30 trimester groups. The control group included 67 healthy non-pregnant women. Statistical 31 tests were used to obtain the normal the normal reference values for the HbA1c . and the tests 32 mailto:drpunnose@rediffmail.com mailto:drpunnose@ststephenshospital.org were considered significant when p value <0.05. Results: The median HbA1c (2.5 to 97.5 33 percentiles) was lower among the pregnant women; 4.8 (4-5.5) % or 32 (20-39) mmol/mol 34 than in the non pregnant women; 5.1 (4-5.7) % or 29 (20-37) mmol/mol ( p <0.001). These 35 were 4.9 (4.1-5.5) % or 30 (21-37) mmol/mol, 4.8 (4-5.3) % or 29 (20-34) mmol/mol, and 4.8 36 (3.9-5.6) % or 29 (19-38) mmol/mol for the T1, T2 and T3 groups, respectively; p-values:T1 37 vs T2=<0.001, T1 vs T3= 0.002, T2 vs T3= 0.111, T1 vs non pregnant group = <0.001. 38 Conclusions: Compared to normal non pregnant women, the A1c was lower in normal 39 pregnant women in South Asian population. These A1c changes were observed despite 40 having significantly higher body max index among women in the T2 and T3 groups than in 41 the T1 and non pregnant groups. To understand the factors determining the A1c decrease in 42 pregnancy and to validate the findings of this study, we recommend further prospective 43 studies among South Asian women 44 Keywords: Asian, Gestational diabetes, HbA1c , Pregnancy trimesters, Reference values. 45 46 Advances in Knowledge 47  Earlier studies stressed the need to identify ethnic- and trimester-specific HbA1c 48 reference intervals in normal pregnant women. 49  Compared to the non-pregnant state, there is significant decrease in HbA1c levels in 50 pregnancy among South Asian pregnant women; and this decrease is obvious in early 51 pregnancy. 52  Among healthy South Asian women, the suggested upper reference limits of HbA1c in 53 first, second and third trimesters are 37,34, and 38mmol/mol, respectively 54 55 Application to patient care 56  The proposed upper trimester-specific HbA1c reference values may be used as threshold 57 values to identify women prone for gestational diabetes (GDM) and other adverse 58 pregnancy outcomes. 59  Early identification of these high risk women will open up a window of opportunity to 60 introduce preventive strategies. 61  These HbA1c reference values can guide in designing further prospective studies among 62 South Asian pregnant women, and can develop alternate tests to OGTT for GDM 63 diagnosis and to establish glycemic targets in pregnancies complicated by diabetes. 64 65 Introduction 66 Glycated hemoglobin (A1c) is widely used as a standard biomarker for glycemic control 67 during management of diabetes mellitus in the general population,1 but, there is uncertainty 68 over the role of A1c for glycemic assessment during pregnancy. The accuracy of A1c 69 estimation in pregnancy is affected by several physiological changes in pregnancy like 70 increase in red cell production, younger red cell age distribution, and reduced red cell life 71 span.2 Moreover, the high prevalence of iron deficiency and the common practice of iron 72 supplementation in pregnant women (especially in developing countries) can influence the 73 A1c estimation in pregnancy.3 Despite these limitations, several prestigious organizations 74 have recommended A1c estimation in pregnancy for various reasons. The World Health 75 Organization advocates A1c estimation at the first ante natal visit to identify women with 76 ‘Diabetes in Pregnancy’(A1c > 6.5%, > 48 mmol/mol).4 77 78 Many authors recommend A1c as a screening and even as a diagnostic test for gestational 79 diabetes mellitus (GDM).2,5 In 2011, the California State Diabetes and Pregnancy Program 80 ‘Sweet Success’ adopted a new algorithm for the diagnosis and treatment of hyperglycemia 81 in pregnancy.6 Accordingly, all women with A1c values of 5.7–6.4% (39–46 mmol/mol) in 82 early pregnancy are advised to undergo GDM treatment without further confirmatory OGTT. 83 The American Diabetes Association suggests periodic A1c estimations in pregnancy as a 84 secondary measure of glycemic control after self-monitoring of glucose.7 The National 85 Institute for Health and Care Excellence in the United Kingdom proposes A1c in pregnancy 86 as a useful guide for risk stratification and prediction of pregnancy outcomes.8 This guideline 87 recommends HbA1c testing at booking and in the second and third trimesters to ensure that 88 the targets are achieved. In many population groups, the first trimester A1c is recognized as a 89 predictor of GDM later in pregnancy,2,9 as well as of adverse pregnancy outcomes.2,10 Second 90 and third trimester A1c levels are predictive of several obstetric complications: macrosomia, 91 gestational hypertension, preeclampsia, abnormal liquor volume, prematurity, and neonatal 92 deaths.2,11 93 94 However, many of these recommendations have not gained universal acceptance, due to the 95 lack of strong research evidence in obstetric population. The A1c cut off points for diagnosis 96 of ‘Diabetes in pregnancy’ (> 6.5%, > 48 mmol/mol) and GDM in the ‘Sweet 97 Success’program (5.7to 6.4%, 39-48 mmol/mol) are guided by the A1c values for diagnosis 98 of diabetes and pre-diabetes in a non-obstetric population, respectively. However, A1c levels 99 in pregnancy are lower than in non obstetric population, and it shows physiological variations 100 between trimesters.12 There are significant racial and ethnic differences in glycation of 101 hemoglobin for a level of glycemia.13 Clearly, there is a need to define ethnic- and trimester-102 specific A1c reference levels in normal pregnant women, before it is recommended for GDM 103 screening and diagnosis, risk stratification and for measuring metabolic control. 104 There is an ongoing Type 2 diabetes epidemic in the Middle East and the South Asian region 105 including the obstetric population. India has 5.7 million women with hyperglycemia during 106 pregnancy and ranks first in the world in this respect.14 However, to our knowledge, A1c 107 levels among normal pregnant South Asian women are not yet defined. Here, we identified, 108 trimester-specific A1c levels in healthy non-diabetic South Asian pregnant women who 109 delivered babies with an age-appropriate weight. 110 111 Methods 112 This retrospective study involved pregnant women who attended antenatal clinic at 113 St.Stephen’s hospital, a tertiary care hospital in Delhi, North India between January 2011 and 114 December 2016. Our center follows a universal thalassemia screening strategy for pregnant 115 women at the first antenatal visit. The protocol includes estimation of HbA, HbA2, and HbF 116 through hemoglobin (Hb) electrophoresis,with concurrent estimates of A1c. All women with 117 A1c >6.5% (48 mmol/mol) were diagnosed to have overt diabetes and those women having 118 A1c <6.5 % (48 mmol/mol) were screened for GDM through a universal one-step 75 g 119 OGTT between 24 and 28 gestational weeks or earlier if having high GDM risk factors. The 120 GDM diagnosis was made by the International Association of Diabetes and Pregnancy Study 121 Group (IADPSG) recommendations.15 All pregnant women were on iron and folic acid 122 supplementation. 123 124 Selection of the study population is shown in a flow diagram (Fig.1). As part of universal 125 thalassemia screening, 9388 pregnant women had Hb electrophoresis (A1c estimation) at the 126 first antenatal visit; all of them were evaluated for inclusion in this study. We excluded 8031 127 women due to unclear date of last menstrual period, delivery outside our hospital, diagnosis 128 of diabetes and gestational diabetes, GDM risk factors, anaemia,16 systemic diseases and 129 delivery of babies with small- and large-for gestational age babies.17 The remaining 1357 130 women in the study population were sub-categorized into three groups based on the 131 gestational age of A1c estimation: (a) first trimester <14 weeks (T1), n=513 women; (b) 132 second trimester-14-26 weeks (T2), n=550 women; and (c) third trimester 27-41 weeks (T3), 133 n=294 women. The body mass index (BMI) was calculated from the height and weight 134 recorded at first antenatal visit. The serum thyroid stimulating hormone (TSH ) was estimated 135 at first antenatal visit in all women and if elevated, was corrected with oral L-thyroxine 136 therapy (target serum TSH level below 2.5, 3 and 3 mIU/L in the first, second and third 137 trimesters respectively) 138 139 A control group of 67 non–pregnant healthy women were recruited from 750 women, who 140 attended the pre-pregnancy counseling clinic of our hospital during the study period. The age 141 and BMI of the control group was comparable to those of whole and T1 pregnancy groups 142 respectively. (Table 1&2) All had FPG < 5.5 mmol/l (< 100 mg/dl) or random plasma 143 glucose < 7 mmol/l (126 mg/dl), Hb >11 g/dl, normal HbA2 and HbF levels, no prior history 144 of gestational diabetes or abortion, no family history of diabetes in first degree relatives, and 145 had no systemic disease. The A1c levels of the study and the control groups were compared. 146 The reference intervals of A1c levels in each trimester were estimated and were compared for 147 any differences. This research protocol was approved by St.Stephen’s hospital ethics 148 committee (No.SSHEC/R0136) with a waiver for patient consent form. 149 150 Our laboratory is certified by the National Accreditation Board for Testing and Calibration 151 Laboratories and uses Bio-Rad laboratories for proficiency testing. The complete blood count 152 was done on EDTA anticoagulated blood using Beckman coulter LH 750/780 analyzer using 153 VCS technology. We used the standard protocol for the OGTT: ingestion of 75 g anhydrous 154 D-glucose dissolved in 250 ml distilled water.The sample for plasma glucose estimation was 155 collected in EDTA and sodium fluoride (grey top) vacuette. The glucose estimation was done 156 by hexokinase method on a Beckman AU680/480 analyzer. Two levels of plasma glucose 157 controls (from Bio Rad) were run daily; Level 1 - 4.53 mmol/l (81.50 mg/dl) and Level 2-158 15.57 mmol/l (280.2 mg/dl). The monthly coefficient of variation (CV) % calculated for the 159 Level -1 and Level -2 controls were 1.7 % and 1.4% respectively. The blood for A1c 160 estimation was a non fasting sample collected in EDTA vial. Estimation was done within two 161 hours of sampling by the Ion exchange High Performance Liquid Chromatography method 162 with a Bio-Rad D10TM machine (Bio-Rad laboratories, Hercules CA). The estimation was 163 traceable to the reference methods of both the National Glycohemoglobin Standardization 164 Program (NGSP) and the International Federation of Clinical Chemistry and Laboratory 165 Medicine (IFCC). The inter-assay CV was 1.3% and 1.5% for low control (mean A1c 5.45%, 166 37 mmol/mol) and high control (mean 9.95%, 86 mmol/mol) respectively. Our laboratory 167 participated in an External Quality Assurance Scheme (EQAS) for both glucose and A1c. The 168 Z Score for glucose was 0.60 and 0.65 for A1c. 169 170 All study groups had the minimum of 40 subjects mandated by IFCC for identification of 171 reference intervals.18 The data analysis was performed using SPSS version 16 (SPSS 172 Inc.,Chicago, USA) and R-software version 4.0.2. Continuous variables were presented with 173 mean and standard deviation. An unpaired student’s t- test was used to compare the mean 174 between pregnant and non-pregnant women. The homogeneity of variances was checked 175 using Leven’s test. One-way analysis of variance followed by post-hoc Tukey’s test were 176 applied to compare the mean among groups T1,T2&T3. Five normality statistical tests were 177 used to get the normal reference value of HbA1c namely: Anderson-Darling, Cramer-von 178 Mises, Kolmogorov-Smirnov, Shapiro-Francia, and Pearson chi-square. The mean + two 179 standard deviations were reported as reference values when the normality condition was 180 fulfilled or else a non-parametric method and median with percentiles (2.5th and 97.5th) were 181 reported as the normal range. The 95% confidence intervals of these percentiles were 182 determined with bootstrapping with 10000 replications using the boot package of r-software. 183 The Mann-Whitney U-test was applied to compare the distribution of A1c between pregnant 184 and non-pregnant women and between the trimesters and a p-value < 0.008 was considered 185 significant as per Bonferroni correction (0.05/number of comparisons). A p-value < 0.05 was 186 considered as significant for other statistical tests. 187 188 Results 189 Table 1 shows reference intervals for A1c for the control group, study population, and T1, 190 T2, and T3 trimester groups expressed as median and percentiles. All five statistical tests to 191 assess normality of A1c values showed violation of normality. The median A1c value of 4.8 192 % (29mmol/mol) for the whole study population and 4.9 % (30 mmol/mol)for the T1 group 193 were lower than the median value of 5.1% (32 mmol/mol) in the control group ( p <0.001 for 194 both). The A1c median values for the T1,T2 and T3 groups were 4.9 %,4.8% and 4.8 % 195 (30,29 and 29 mmol/mol) respectively, with significant differences between T1 and T2 ( p 196 =0.001), T1 and T3 (p=0.001) and no difference between T2 and T3 ( p= 0.111). 197 Fig. 2 presents the upper normal A1c level for the control,T1,T2 and T3 groups:5.7%(39 198 mmol/mol),5.5% (37mmol/mol),5.3% (34mmol/mol),and 5.6% (38 mmol/mol), 199 respectively.Table 2 shows the clinical and laboratory parameters of the whole, trimester-200 specific pregnancy groups, and the control group. The women in all trimesters were age-201 matched, and the gestational age at delivery and the birth weight were comparable (p-values 202 > 0.05 for all parameters). The Hb and RBC count were lower, and the MCV, MCH, and 203 MCHC were higher in pregnant women than in the control group; there was no difference of 204 HCT and RDW between these groups. Compared to the T1 group, there was a decrease in Hb 205 and RBC count and an increase in MCV, MCH and MCHC in T2 group; the RDW and HCT 206 remained static between groups. There was a significant rise of hemoglobin and RDW in T3 207 versus T2; HCT, MCV, and MCHC were constant. 208 209 Discussion 210 The A1c is lower during pregnancy than when not pregnant in South Asian women.The A1c 211 reference values for the first, second, and third trimesters were 4.1- 5.5% (21- 37 mmol/mol), 212 4-5.3% (20-34 mmol/mol), and 3.9-5.6% (19-38 mmol/mol), respectively. Earlier studies 213 revealed some racial differences in A1c reference intervals: (a) Caucasian women in Italy, 214 3.5-5.7% (15 - 39mmol/mol), 3.3-5.6 %(14-38 mmol/l), and 4.3-5.6 %(23-38 mmol/l) in 15-215 24, 25-27, and 28-36 gestational weeks, respectively;19 (b) Mexican women, T1 4.5-5.6% 216 (26-38mmol/mol), T24.4-5.5% ( 26-37 mmol/mol), and T34.4-5.6% (25-38 mmol/mol);20 (c) 217 Japanese women, T1 4.7-5.7% (28-39 mmol/mol), T2 4.4-5.4% (25-36mmol/mol), T3 4.6-218 5.8% (27-40mmol/mol).21 Compared to these studies, the upper A1c reference values of our 219 South Asian cohort were marginally lower. The stringent selection criteria (exclusion of 220 women having, GDM diagnosed by the most liberal IADPSG criteria and those with several 221 GDM risk factors and large or small for gestational age babies) as well as the racial 222 differences in the glycation of hemoglobin might have contributed to this modest A1c 223 difference. 224 225 Compared to first trimester, a significant decrease in A1c level was noted in the second 226 trimester, but this remained constant in the third trimester [Table 1]. The differences in A1c 227 levels between trimesters varied markedly between studies. In most populations, there was a 228 decrease in A1c level from the first to second trimester20-24 and this decrease was often 229 followed by a significant A1c rise in the third trimester (biphasic response ).19,20-24The A1c 230 rise in the third trimester was not seen in some studies,25,26 but a decrease was reported in one 231 study.12 In a Japanese study, Hashimoto et al reported that the A1c rise in late pregnancy is 232 mainly due to iron deficiencies in the third trimester.27 Significant racial differences in 233 trimester-related A1c variations were reported in a multiethnic population in the United 234 Kingdom by Hartland et al; both Caucasians and Asians had a lower A1c in the second 235 trimester than in the first trimester, but theA1c rise in the third trimester was observed only in 236 Causasian women (not in Asians, as in our study).22 237 The metabolic changes leading to the significant decline in A1c levels in mid-pregnancy was 238 apparent in a longitudinal study by Mills et al.28 This study demonstrated a significant drop in 239 plasma glucose values between 6 and 10 weeks of gestation which was followed by a 240 decrease in A1c levels in second trimester. The authors speculated that the maternal 241 metabolic and hormonal factors alter the plasma glucose concentration early in pregnancy, 242 independently of foetal glucose utilisation. Another proposed mechanism for lowering 243 plasma glucose in late first trimester is the decrease in progesterone secretion during the 244 luteoplacental shift.28 The HbA1c reduction in the second trimester is further exacerbated by 245 the physiological changes in pregnancy like high erythrocyte turnover and hemodilution. 246 Subsequent compensatory mechanisms like maternal plasma reduction and increased atrial 247 natriuretic peptide, can again raise Hb in the third trimester. 29 The high prevalence of iron 248 deficiency anemia and the common practice of universal iron supplementation in pregnancy 249 especially in developing countries, can modify A1c levels.3 We excluded women with anemia 250 and thalassemia and the changes in Hb, MCV, MCH, MCHC,and RBC over trimesters is 251 attributable to the physiological changes in pregnancy and to iron supplementation.30 (Table 252 2) 253 254 The proposed upper reference HbA1c levels in early pregnancy in this study, can be clinically 255 relevant in the early identification of women prone for GDM and adverse pregnancy 256 outcomes. This approach can open up a window of opportunity for early initiation of GDM 257 preventive strategies. Strikingly, the suggested upper reference values (5.5 % and 5.3 % in 258 first and second trimesters respectively) are lower than the generally recommended threshold 259 A1c value of 5.7 % (39 mmol/mol) for diagnosis of ‘pre-diabetes in pregnancy’.2 In an earlier 260 study, we observed that the first trimester A1c >5.5% (37 mmol/mol) was a strong predictor 261 (adjusted odds ratio 2.6, p < 0.001) of GDM later in pregnancy.12 Similarly, Rajput et al 262 studied the utility of A1c estimation between 24 and 28 weeks of gestation for GDM 263 diagnosis in 607 Asian Indian pregnant women.7 In that study, the A1c of 5.25% 264 (34mmol/mol) was a reliable cut off value for identification of GDM women when IADPSG 265 criteria was applied for GDM diagnosis. The A1c threshold values identified for GDM 266 diagnosis in first and second trimesters in these studies agree well with the corresponding 267 upper reference values of our study. Further, Maine et al also assessed the relationship of A1c 268 level in the first trimester with adverse pregnancy outcomes among a cohort of multiethnic 269 pregnant women residing in Spain.17 270 271 The risk for eclampsia, LGA and macrosomia increased at A1c threshold values of 5.3 %, 5.4 272 % and 5.7 % (34,36 and 39 mmol/mol) respectively for the South Asian pregnant women in 273 this cohort. These cut off values are near (though not exact) to the first trimester A1c upper 274 reference value of 5.5% (37mmol/mol) in our study. The studies above suggest that the risks 275 for GDM and other adverse pregnancy events start in A1c levels lower than the ‘prediabetic’ 276 level of 5.7 % (39 mmols/mol).We recommend further prospective studies to validate the 277 proposed trimester specific A1c reference levels for prediction and identification of various 278 adverse events among South Asian pregnant women. 279 280 Our study has several limitations. The A1c reference values of this study are derived from a 281 cross-sectional analysis of different women who attended our antenatal clinic over three 282 trimesters. A longitudinal study on the sequential changes of HbA1c levels of a cohort of 283 same women over different trimesters would have been ideal. The impact of this limitation is 284 alleviated significantly in this study: Age, gravidity, family history of DM, history of GDM 285 and abortion, gestational age at delivery, birthweight, Hb, HbA2 and HbF of women in 286 different trimesters and the BMI between control and T1 groups were comparable (Table 3). 287 The BMI rise in T2 and T3 groups are due to physiological gestational weight gain.The lack 288 of data on iron, folate and B12 status of women in different trimesters is a limitation, but the 289 RBC indices of these women do not suggest any major deficiencies of these factors. The 290 strengths of this study include the large study population, with identification and exclusion of 291 GDM by universal OGTT based screening as per IADPSG guidelines. All women with GDM 292 risk factors,anaemia and thalassemia (the common hemoglobinopathy of the region) were 293 excluded in this study. Being a single center hospital based study, the blood samples were 294 sampled and processed under optimal conditions in one laboratory. 295 296 Conclusion 297 The trimester specific hemoglobin A1c levels are not yet defined for healthy South Asian 298 pregnant women. This study evaluated the upper reference limits for first, second and third 299 trimesters as 37, 34, and 38 mmol/mol, respectively. These trimester-specific A1c values can 300 be of clinical relevance for prediction and diagnosis of GDM and for risk stratification of 301 other adverse events among South Asian pregnant women. Further prospective studies to 302 validate the proposed A1c reference intervals are recommended. 303 304 Conflicts of interest 305 The authors declare that they have no conflict of interest 306 307 Funding 308 The present study is an unfunded project. 309 310 Acknowledgements 311 We gratefully acknowledge the immense help received from Sheeba Samuel and Sapna 312 Robinson (diabetes nurses) as well as Aashish Kumar, Nikita Tyali, and Deepali Garg 313 (diabetes educators) at the Dept of Endocrinology, St. Stephen’s hospital for the collection of 314 data for this study 315 316 Author contributions 317 JP conceptualised the idea and prepared the manuscript. RM carried out statistical analysis 318 and contributed substantially in discussion and preparation of the manuscript. KS and RMR 319 contributed in discussion and provided constructive criticism regarding manuscript. AS, PV, 320 NC assisted in clinical data collection, its analysis and contributed to manuscript preparation 321 and discussion RJ assisted in analysis of laboratory data and contributed to the manuscript. 322 All authors approved the final version of the manuscript. 323 324 References 325 1. American Diabetes Association. Glycemic Targets: Standards of Medical Care in 326 Diabetes—2021 Diabetes Care 2021; 44 (Supplement 1): S73-S84. 327 https://doi.org/10.2337/dc21-S006 328 2. Hughes RCE, Rowan J, Florkowski CM. Is there a role for HbA1c in pregnancy? 329 Curr Diab Rep 2016; 6: 5-14. DOI 10.1007/s11892-015-0698-y. 330 3. FOGSI General Clinical Practice Recommendations. Management of Iron deficiency 331 anemia in pregnancy available at The Federation of Obstetric & Gynecological 332 Societies of India. From: www.fogsi.org/wp-content/uploads/2017/07/gcpr-333 recommendation-ida.pdf. Accessed; Nov 2021. 334 4. World Health Organization. Diagnostic Criteria and Classification of Hyperglycaemia 335 First Detected in Pregnancy 2013. 336 https://care.diabetesjournals.org/content/44/Supplement_1/S73 https://care.diabetesjournals.org/content/44/Supplement_1/S73 https://doi.org/10.2337/dc21-S006 http://www.fogsi.org/wp-content/uploads/2017/07/gcpr-recommendation-ida.pdf http://www.fogsi.org/wp-content/uploads/2017/07/gcpr-recommendation-ida.pdf From:http://apps.who.int/iris/bitstream/10665/85975/1/WHO_NMH_MND_13.2_eng337 .pdf. Accessed; Nov 2021. 338 5. Rajput R, Yogesh Y, Rajput M, Nanda S. Utility of HbA1c for diagnosis of 339 gestational diabetes mellitus. Diabetes Res Clin Pract. 2012; 98:104-7. 340 6. California State Diabetes and Pregnancy Program sweet Success. From: 341 http://perinatology.com/Reference/CDAPP%20SS%20Guidelines%202012.pdf. 342 Accessed; Nov 2021. 343 7. American Diabetes Association. Management of Diabetes in Pregnancy: Standards of 344 Medical Care in Diabetes—2021 Diabetes Care 2021; 44 (Supplement 1): S200-S210. 345 https://doi.org/10.2337/dc21-S014 346 8. NICE guideline. National Institute for Health and Care Excellence (2015), Diabetes in 347 pregnancy: management Diabetes in pregnancy: management from preconception to 348 the postnatal from preconception to the postnatal period. From: 349 https://www.nice.org.uk/guidance/ng3. Accessed; Nov 2021. 350 9. Punnose J, Malhotra RK, Sukhija K, Mathew A, Sharma A, Choudhary N. Glycated 351 haemoglobin in the first trimester: A predictor of gestational diabetes mellitus in 352 pregnant Asian Indian women. Diabetes Res Clin Pract. 2020 ;159: Article107953 . 353 doi:10.1016/j.diabres.2019.107953. 354 10. Hughes RC, Moore MP, Gullam JE, Joanna EG, Muhammed K, Rowan J. An early 355 pregnancy HbA1c >5.9% (41 mmol/mol) is optimal for detecting diabetes and 356 identifies women at increased risk of adverse pregnancy outcomes. Diabetes Care. 357 2014; 37(11):2953–9. 358 11. Mañé L, Flores-Le Roux JA, Gómez N, Chillarón JJ, Llauradó G, Gortazar L, et al. 359 Association of first-trimester HbA1c levels with adverse pregnancy outcomes in 360 different ethnic groups Diabetes Res Clin Pract. 2019;150:202-210. doi: 361 10.1016/j.diabres.2019.03.017.Epub 2019 Mar 15. 362 12. Nielsen LR, Ekbom P, Damm P, Glumer C, Frandsen MM, Jensen DM, et al. 363 HbA1clevels are significantly lower in early and late pregnancy. Diabetes Care 2004; 364 5:1200–1. doi: 10.2337/diacare.27.5.1200 365 13. Venkataraman K, Kao SL, Thai AC, Salim A, Lee JJM, Heng D et al. Ethnicity 366 modifies the relation between fasting plasma glucose and HbA1c in Indians, Malays 367 and Chinese. Diabet Med. 2012; 29 (7):911-917. 368 http://apps.who.int/iris/bitstream/10665/85975/1/WHO_NMH_MND_13.2_eng.pdf http://apps.who.int/iris/bitstream/10665/85975/1/WHO_NMH_MND_13.2_eng.pdf http://perinatology.com/Reference/CDAPP%20SS%20Guidelines%202012.pdf https://care.diabetesjournals.org/content/44/Supplement_1/S200 https://care.diabetesjournals.org/content/44/Supplement_1/S200 https://doi.org/10.2337/dc21-S014 https://pubmed.ncbi.nlm.nih.gov/31794807/ https://pubmed.ncbi.nlm.nih.gov/31794807/ https://pubmed.ncbi.nlm.nih.gov/31794807/ https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Ma%C3%B1%C3%A9+L&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Flores-Le+Roux+JA&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=G%C3%B3mez+N&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Chillar%C3%B3n+JJ&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Llaurad%C3%B3+G&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Gortazar+L&cauthor_id=30880095 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Venkataraman+K&cauthor_id=22283416 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Kao+SL&cauthor_id=22283416 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Thai+AC&cauthor_id=22283416 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Salim+A&cauthor_id=22283416 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Lee+JJ&cauthor_id=22283416 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Heng+D&cauthor_id=22283416 14. Guariguata L, Linnenkamp U, Beagley J, Whiting DR, Cho NH. Global estimates of 369 the prevalence of hyperglycemia in pregnancy. Diabetes Res Clin Pract 370 2014;103:176–185. 371 15. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P ,et al. 372 International Association of Diabetes and Pregnancy Study Group. 373 Consensus.International Association of Diabetes and Pregnancy Study Groups 374 recommendations on the diagnosis and classification of hyperglycemia in pregnancy. 375 Diabetes Care 2010; 33:676–682. 376 16. World Health Organization, Vitamin and Mineral Nutrition Information System, 377 WHO, Geneva, 2011 (WHO/NMH/NHD/MNM/11.1). From: 378 https://www.who.int/vmnis/indicators/haemoglobin/en/ Accessed; Sept 2020. 379 17. National Neonatology Forum. National Neonatal-Perinatal Database (2002-03). 380 From: www.newbornwhocc.org/pdf/nnpd_report_2002-03.PDF. Accessed; Nov 2021 381 18. Solberg HE. The IFCC recommendation on estimation of reference intervals. The Ref 382 Val program. Clin Chem Lab Med.2004; 42 (7):710–4. 383 19. Mosca A, Paleari R, Dalfrà MG, Di Cianni G, Cuccuru I, Pellegrini G, et al. 384 Reference intervals for hemoglobin Alc in pregnant women: data from an Italian 385 multicenter study. Clin Chem. 2006; 52(6):1138–43. 386 20. Sánchez- González CM, Castillo-Mora A, Alvarado-Maldonado IN, Ortega-González 387 C, Martínez Cruz N, Arce –Sánchez L et al . Reference intervals for hemoglobin A1c 388 (HbA1c) in healthy Mexican pregnant women: a cross-sectional study. BMC 389 Pregnancy and Childbirth. 2018; 18:424-429. doi: 10.1186/s12884-018-2057-x 390 21. Hiramatsu Y, Shimizu I, Omori Y, Nakabayashi M. Determination of reference 391 intervals of glycated albumin and hemoglobin A1c in healthy pregnant Japanese 392 women and analysis of their time courses and influencing factors during pregnancy. 393 Endocr J. 2012;59 (2):145–51. 394 22. Hartland AJ, Smith JM, Clark PMS, Webber J, Chowdhury T, Dunne F. Establishing 395 trimester- and ethnic group-related reference ranges for fructosamine and HbA1c in 396 non-diabetic pregnant women. Ann Clin Biochem 1999; 36:235–7. 397 23. Phelps RL, Honig GR, Green D, Metzger BE, Frederiksen MC, Freinkel N. Biphasic 398 changes in haemoglobin A1c concentrations during normal human pregnancy. Am J 399 Obstet Gynecol 1983;147: 651-653. 400 https://www.who.int/vmnis/indicators/haemoglobin/en/ 24. Worth R, Potter JM, Drury J, Fraser RB, Cullen DR. Glycosylated haemoglobin in 401 normal pregnancy: a longitudinal study with two independent methods. Diabetologia 402 1985; 28:76. 403 25. O’Kane MJ, Lynch PLM, Moles KW, Magee SE. Determination of a diabetes control 404 and complications trial-aligned Hb A1c reference range in pregnancy. Clin Chim Acta 405 2001; 311:157–9. 406 26. Parentoni LS, de Faria EC, Bartelega MJLF, Moda VMS, Facin ACC, Castilho LN. 407 Glycated haemoglobin reference limits obtained by high performance liquid 408 chromatography in adults and pregnant women. Clin Chim Acta 1998; 274:105–9. 409 27. Hashimoto K , Noguchi S, Morimoto Y, Hamada S, Wasada K, Imai S, et al. A1C but 410 not serum glycated albumin is elevated in late pregnancy owing to iron deficiency 411 Diabetes Care 2008; 31(10):1945-8. doi: 10.2337/dc08-0352. Epub 2008 Jul 3. 412 28. Mills JL, Jovanovic L, Knopp R, Aarons J, Conley M, Park E , Lee YJ, et al : 413 Physiological reduction in fasting plasma glucose concentration in the first trimester 414 of normal pregnancy: The Diabetes in Early Pregnancy Study. Metabolism 1998; 415 47(9):1140–1144. doi: 10.1016/s0026-0495(98)90290-6 416 29. Lurie S, Mamet Y. Red blood cell survival and kinetics during pregnancy. Eur J 417 Obstet Gynecol Reprod Biol.2000; 93:185–192. 418 30. Milman N, Byg KE, Agger AO. Hemoglobin and erythrocyte indices during normal 419 pregnancy and postpartum in 206 women with and without iron supplementation. 420 Acta Obstet Gynecol Scand. 2000; 79(2):89-98. doi: 10.1034/j.1600-421 0412.2000.079002089.x.PMID:10696955. 422 423 Table 1: The median and percentiles of HbA1c in non pregnant women, whole study 424 population, and in first (T1), second (T2) and third (T3) trimester groups. 425 Study Group N HbA1c % Median (0.95 CI) Percentile (0.95 CI)a Range (min- max) P value Type of distributi on 2.5th (0.95 CI) 97.5th (0.95CI) Non pregnant 67 5.1(4.9-5.2) 4.0 (3.9- 4.6) 5.7 (5.5- 6.0) 3.9-6.0 <0.001b Non- Gaussian( 5)f Pregnant (whole group) 135 7 4.8 (4.8-4.8) 4.0 (3.9- 4.1) 5.5(5.4-5.5) 3.2-5.9 <0.001b Non- Gaussian( 5) https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Hashimoto+K&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Noguchi+S&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Morimoto+Y&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Hamada+S&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Wasada+K&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/?size=200&term=Imai+S&cauthor_id=18599529 https://pubmed.ncbi.nlm.nih.gov/10696955/ https://pubmed.ncbi.nlm.nih.gov/10696955/ T1 0-13weeks 513 4.9 (4.8-4.9) 4.1(4.0- 4.2) 5.5(5.4-5.5) 3.7-5.8 <0.001c Non- Gaussian( 5) T2 14-26 weeks 550 4.8 (4.7-4.8) 4.0 (3.9- 4.1) 5.3 (5.2- 5.4) 3.2-5.5 <0.001d Non- Gaussian( 5) T3 27-41 weeks 294 4.8 (4.7-4.8) 3.9(3.8- 4.1) 5.6(5.4-5.7) 3.6-5.9 0.002d 0.111e Non- Gaussian( 5) n= number of women, a CI: Confidence Intervals , b P value for comparison of non-pregnant 426 with pregnant groups, c P value for comparison with non pregnant group , d P value for 427 comparison with T1 group , e P value for comparison of T2 and T3 groups, f Values in 428 parentheses indicate number of tests for goodness of fit with p<0.05 429 430 Table 2: Comparison of clinical and Laboratory parameters in mean + standard 431 deviation. (A) Whole study population versus control group (B) Between First (T1), 432 Second (T2) and Third (T3) trimesters. 433 Parameter A Whole study population B Women in different Trimesters women all trimest ers n=1357 Non pregnant control group n= 67 P value 0-13 weeks n = 513 (T1) 14-26 weeks n = 550 (T2) 27-41 weeks n= 294 (T3) P valu e T1 vs T2 P value T1 vs T3 P value T2 vs T3 Age (years) 26.67 3.51 26.662. 89 0.956 4 26.83  3.50 26.69  3.50 26.343. 57 0.790 0.135 0.351 GA at HbA1c estimation (weeks) -- -- -- 9.53 2.47 18.813. 62 31.23  3.02 - -- -- Hemoglobin- g/L 120.3 7.4 122.8 8.7 0.027 121.7  7.5 119 6.8 120.5 8.0 <0.0 01 0.081 0.016 MCV- fl 88.36 5.61 84.097. 86 <0.00 1 87.735.8 4 88.705. 52 88.85 5.22 0.016 0.025 0.937 MCH - pg 29.38 2.41 28.182. 87 <0.00 1 29.06  2.38 29.61  2.23 29.532. 76 0.001 0.030 0.895 HCT- % 0.36  0.03 0.37 0.03 0.063 0.36  0.03 0.36  0.03 0.36 0.03 0.366 0.974 0.375 MCHC- g/L 332.5 10.2 328.5 8.9 0.003 330.9 9.7 333.61 0.1 333.1 11 <0.0 01 0.015 0.762 RDW- % 14.75 2.16 14.761. 48 0.977 14.551.6 8 14.742. 19 15.182. 75 0.337 <0.00 1 0.021 RBC – 1012 L 4.100. 40 4.430.5 6 <000 1 4.140.41 4.070.3 9 4.100. 39 0.009 0.298 0.605 GA at Delivery (weeks) -- -- -- 38.53  1.02 38.56  1.02 38.531. 00 0.921 0.999 0.925 Birth Weight- kg -- -- -- 2.890.29 2.87 0.36 2.850.2 8 0.562 0.199 0.660 For (A) Unpaired students t-test was applied to compare the mean value between the groups 434 and for (B) One-way analysis of variance followed by post-hoc Turkey’s test. GA = 435 Gestational age, MCH = Mean corpuscular hemoglobin, MCV = Mean corpuscular volume, 436 MCHC = Mean corpuscular hemoglobin concentration, RDW = RBC diameter width. n= 437 number of women. 438 439 Figure 1: Flow diagram on selection of study population. OGTT= Oral Glucose Tolerance 440 Test, GDM= Gestational Diabetes mellitus; having fasting plasma glucose (PG) between 441 5.1-6.9 mmols/L, 1-hour PG > 10 mmol/L , 2-hour PG between 8.5 – 11.1 mmol/L in 442 OGTT , IADPSG = International Association of Diabetes and Pregnancy Study Group, 443 Pre-gestational diabetes = Diabetes diagnosed before pregnancy , Diabetes in pregnancy = 444 Overt diabetes first diagnosed in pregnancy; HbA1c > 48 mmol/mol or FPG > 7 mmol/L or 445 2-h PG > 11.1 mmol/L 446 447 448 Figure 2: Median and percentile (2.5 to 97.5) for hemoglobin A1c (%) for women in non-449 pregnant and different trimesters. 450