International Blood Research & Reviews

Aims: To determine trimester-specific reference intervals and compare glycated haemoglobin (HbA1c), fasting plasma glucose (FPG) and full blood count (FBC) parameters between second- and third-trimester non-diabetic pregnant women.

Study Design: Cross-sectional comparative study.

Place and Duration of Study: Antenatal clinic of Adeoyo Maternity Teaching Hospital, Yemetu, Ibadan, Oyo State, Nigeria, between August and September 2025.

Methodology: Two hundred and fifty (250) consenting pregnant women aged 18–45 years (138 in the second trimester, 112 in the third trimester) were consecutively recruited. Women with a pre-existing diagnosis of type 1 or type 2 diabetes were excluded, and non-diabetic pregnancy status was further operationally defined using each participant’s routine antenatal gestational diabetes mellitus (GDM) screening record: only women who had undergone the standard 75-g oral glucose tolerance test (OGTT) at 24–28 weeks (or, for third-trimester enrolees screened earlier, at booking) and screened negative by International Association of Diabetes and Pregnancy Study Groups (IADPSG)/WHO 2013 criteria (fasting plasma glucose <5.1 mmol/L, 1-hour <10.0 mmol/L, 2-hour <8.5 mmol/L) were retained. Socio-demographic, obstetric and lifestyle data were obtained with a structured questionnaire. Fasting plasma glucose was determined by the glucose oxidase method, HbA1c by turbidimetric inhibition immunoassay, and FBC parameters by an automated haematology analyser; daily internal quality control was run on all three platforms, with assay coefficients of variation of <5% for FPG, <3% for HbA1c and <5% for FBC parameters, within manufacturer-stated limits. Data were analysed in SPSS version 25 using an independent-samples t-test, Pearson correlation, and percentile estimation (2.5th–97.5th) for trimester-specific reference intervals; mean differences are reported with 95% confidence intervals (CI) and Cohen’s d as a measure of effect size; P < 0.05 was considered significant.

Results: HbA1c was higher in the third trimester (5.2 ± 0.4%) compared with the second trimester (4.9 ± 0.4%; mean difference 0.3%, 95% CI 0.20–0.40, t(248) = 5.90, P < 0.001, Cohen’s d = 0.75), and FPG was higher in the third trimester (4.7 ± 0.6 mmol/L) compared with the second trimester (4.5 ± 0.5 mmol/L; mean difference 0.2 mmol/L, 95% CI 0.06–0.34, t(248) = 2.87, P = 0.004, d = 0.37). Haemoglobin, haematocrit and platelet count were significantly lower in the third trimester compared with the second trimester (11.1 ± 0.9 vs 10.6 ± 1.1 g/dL; 33.3 ± 2.9 vs 31.8 ± 3.4%; and 231 ± 50 vs 202 ± 51 × 10⁹/L, respectively; t(248) = −3.95, P < 0.001, d = −0.50 for haemoglobin; t(248) = −3.76, P < 0.001, d = −0.48 for haematocrit; t(248) = −4.55, P < 0.001, d = −0.58 for platelet count), while WBC count was significantly higher in the third trimester (10.1 ± 1.9 × 10⁹/L) compared with the second trimester (9.3 ± 1.9 × 10⁹/L; t(248) = 3.31, P = 0.001, d = 0.42). Contrary to our original analysis, recomputation of the test statistics from the reported group means and standard deviations showed that red blood cell count was also significantly, though modestly, lower in the third trimester (4.0 ± 0.3 vs 4.1 ± 0.3 × 10¹²/L; mean difference −0.10, 95% CI −0.18 to −0.02, t(248) = −2.62, P = 0.009, d = −0.33), rather than showing no significant difference as previously reported. HbA1c and FPG correlated positively with WBC (r = 0.30 and r = 0.27, respectively; P = 0.001) and weakly, negatively with haemoglobin, haematocrit and platelet count. Trimester-specific 2.5th–97.5th percentile reference intervals were derived for all parameters, though the third-trimester subgroup (n = 112) fell short of the 120 reference individuals per partition recommended by CLSI EP28-A3c, so these intervals should be regarded as provisional pending confirmation in a larger sample.

Conclusion: Normal, non-diabetic pregnancy may be accompanied by trimester-dependent, physiologically adaptive changes in glycaemic and haematological indices, consistent with plasma volume expansion and a mild inflammatory shift rather than pathology. Population- and trimester-specific reference intervals, rather than generic or non-pregnant cut-offs, are recommended for interpreting antenatal HbA1c, FPG and FBC results in this setting, although the provisional intervals reported here require validation in a larger, adequately powered multicentre sample before routine clinical adoption.