Abstract
Background
‘Hyperglycaemia in pregnancy’ (HIP) is one of the most common antenatal complications, affecting about one in six pregnancies globally. HIP is sub-classified into two categories, namely ‘gestational diabetes mellitus’ (GDM) and ‘overt diabetes mellitus’ (ODM). Pregnancy is characterised by the accumulation of adipose tissue and a growing placenta, acting as endocrine organs, thus intensifying the hyperglycaemic environment and building up oxidative stress by dysregulation of metabolic pathways, instigating peripheral neuropathy.
Purpose
Due to paucity in existing literature on neurological influences of GDM and ODM, this follow-up study was planned to detect subclinical peripheral neuropathy by nerve conduction studies (NCS) and its correlation with biochemical parameters among them.
Methods
Thirty-nine pregnant women were divided into three groups: control, GDM and ODM. The NCS (sural and ulnar nerves) and biochemical parameters, that is, fasting plasma glucose (FPG), glycated haemoglobin (HbA1c), serum fasting insulin, homeostatic model assessment of insulin resistance (HOMA-IR), serum chromium, serum N-carboxy-methyl lysine, total cholesterol, low-density lipoprotein, triglycerides and high-density lipoprotein, were recorded during different stages of gestation, that is, 24–28 weeks (first visit), 32–38 weeks (second visit) and 6–12 weeks after parturition (third visit).
Results
Nerve conduction studies reveal significant alterations in diabetic pregnant groups compared to control pregnant women, particularly in sensory latencies and amplitudes of the sural and ulnar nerves. Significant hyperglycaemia (FPG, HbA1c), hyperinsulinaemia, and elevated HOMA-IR in GDM and ODM groups confirm insulin resistance and poor glycaemic control during pregnancy and postpartum. Chromium levels were markedly lower in diabetics (p = .001).
Conclusion
This study necessitates ongoing metabolic and neurological monitoring in GDM and ODM after childbirth. Early screening and focused interventions, including micronutrient supplementation and lifestyle modifications, may help avert progression to overt neuropathy and mitigate long-term complications.
Keywords
Introduction
WHO first defined gestational diabetes mellitus (GDM) in 1965 as ‘hyperglycaemia of diabetic levels occurring during pregnancy’. 1 Historically, the term GDM encompasses the entire spectrum of hyperglycaemic disorders, that is, during pregnancy or from pre-gestational diabetes to hyperglycaemia first detected in pregnancy. ‘hyperglycaemia in pregnancy’ (HIP)/diabetes in pregnancy (DIP) is now sub-classified by the ‘International Association of Diabetes and Pregnancy Study Groups’ (IADPSG) into two separate categories, namely ‘overt diabetes mellitus during pregnancy’ (ODM) and GDM. 2
HIP is one of the most common antenatal complications, affecting about one in six pregnancies globally, about one in four pregnancies in Southeast Asia, and GDM contributes to 80% of cases of DIP.3, 4
GDM and ODM are hyperglycaemic disorders, first revealed during pregnancy. ODM meets the diagnostic threshold of diabetes in non-pregnant adults and is a more severe form of hyperglycaemia associated with worse maternal and foetal outcomes and a higher risk of postpartum diabetes. 3
The probable causes of the pathogenesis of GDM and ODM may include:
Pregnancy is characterised by growing insulin resistance (IR), starting from the second trimester to parturition. The body adapts to this change by undergoing β-cell proliferation and increasing synthesis of insulin in order to prevent hyperglycaemia. Inability to counter the growing IR during pregnancy can lead to pregnancy-induced diabetes. Pregnancy is characterised by the accumulation of adipose tissue and a growing placenta, acting as endocrine organs in pregnancy. Secretion of inflammatory cytokines such as tumour necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) may be implicated in altering insulin signalling and promoting IR in both normal pregnancy and GDM. The hyperglycaemic environment intensifies oxidative stress, as there is excessive production of free radicals and downregulation of the scavenger system. Setting up a vicious cycle of release of inflammatory cytokines in GDM.5, 6
The role of chromium in the pathogenesis of HIP/carbohydrate metabolism was first demonstrated in trials involving patients who were put on complete parenteral nutrition and developed glucose intolerance over time. The exact mechanism of action of chromium on the insulin signalling cascade is still not fully understood. However, a number of experimental studies point out that low-molecular-weight chromium-binding substance (chromodulin) (LMW-Cr) binds to activate the insulin receptor and enhances its tyrosine kinase activity, thus amplifying the insulin signalling cascade, which helps in potentiating insulin action on sensitive tissues.7, 8
Diabetic neuropathy patients can either have axonal loss or demyelinating disease, with or without symptoms of polyneuropathy, which may be timely detected for improvement.
The NCS enables sensory and motor abnormalities associated with neuropathy to be diagnosed, even if the dysfunction is subclinical. Incidence of pregnancy-related carpal tunnel syndrome (PRCTS), diagnosed by the electrodiagnostic criteria, was found to be 48%.9, 10
There is very limited literature regarding the influence of GDM and ODM on maternal peripheral nerves, specifically ‘follow-up study’; thus, this study, first of its kind, was aimed at the evaluation of ‘nerve conduction studies’ (NCS) and various biochemical parameters, that is, serum advanced glycated end products levels, glycated haemoglobin, insulin levels, lipid profile and chromium level, in healthy pregnant women and pregnant women with GDM and ODM during different stages of gestation, that is, (a) 24–28 weeks (first visit), (b) 32–38 weeks (second visit) and (c) 6–12 weeks after parturition (third visit), to assess if peripheral neuropathy was present in an otherwise transient state of GDM. Hence, we aimed to conduct this longitudinal study to assess a correlation between NCS and several biochemical parameters and to determine the intensity of neurological involvement.
Methods
Study Setting and Recruitment of Participants
This longitudinal case-control study was approved by the Institutional Ethics Committee. All pregnant women attending the outpatient department (OPD) and ward of the Department of Obstetrics & Gynaecology were screened for their gestational age and glycaemic status for enrolment in the study. Women diagnosed as ODM, GDM and ‘euglycaemic pregnant women’ (control) in 24–28 weeks of gestation were enrolled in the study. After explaining the study in the Hindi/local language and resolving their queries, written informed consent was obtained from the participants. All the participants were instructed to immediately discontinue the procedure anytime during the recording if they wished to.
Inclusion Criteria
Pregnant women with normal blood glucose levels aged 19–38 years in 24–28 weeks of gestation after getting written informed consent to participate in this study.
Diagnostic criteria for recruitment:
The diagnostic criteria for ODM and GDM used in this study were the ‘International Association of Diabetes in Pregnancy Study Group (IADPSG) diagnostic criteria 2008’ for detection of DIP in 24–28 weeks of gestation, based on fasting 75-g oral glucose tolerance test (OGTT) and glycated haemoglobin (HbA1c). 2
Exclusion Criteria
Pregnant women with chronic kidney diseases, hypertension of pregnancy, psychiatric illness, prior head trauma, cardiovascular disease, central nervous system (CNS) disorder, neuromuscular disorder, limb injury, mental disability, hypothyroidism, long-term use of certain medications that may influence NCS parameters and alcoholism were excluded from the study.
Study Flow
Refer to Figure 1.
Consort Diagram Showing Flow of Study-Outlining Recruitment, Allocation, Drop-offs, Follow-up, and Analysis Phases for ‘Gestational Diabetes (GDM), Overt Diabetes (ODM) and Euglycaemic (Control)’ Pregnant Women Groups.
Study Parameters
The nerve conduction study parameters were recorded at the Department of Physiology, and biochemical parameters, except chromium, were measured at the Department of Biochemistry at a tertiary care institute. Chromium level was measured in an institute of national repute externally. Demographic and anthropometric parameters were recorded, followed by NCS and biochemical parameters.
Demographic and Anthropometric Parameters
Age, gestation weeks, height, pre- and post-gestational weight, body mass index (BMI), family history and others.
Electrophysiological-NCS Parameters11, 12
It was recorded using Neuropack S1 MEB-9400 electrodiagnostic equipment (Nihon Kohden, Tokyo, Japan) in a dark room where the temperature was maintained at 24°C–26°C throughout the procedure. The NCS was done on the non-dominant extremities (this was the left side in all) of the subject, while she was lying supine at rest. A standardised technique and a blinded examiner were used to attain recordings of action potentials in two nerves. The normative data values are as follows: for sural nerve (sensory)—latency (LAT): 2.7 ± 0.3 ms, amplitude (AMP): 20.9 ± 8.0 µV and sensory conduction velocity (SCV): 52.5 ± 5.6 m/s; for ulnar nerve (sensory)—LAT: 2.83 ± 0.4 ms, AMP: 5.54 ± 2.3 µV, SCV: 54.17 ± 5.7 m/s; and for ulnar nerve (motor)—LAT: 2.59 ± 0.4 ms, AMP: 8.51 ± 2.0 mV and motor conduction velocity (MCV): 61.45 ± 5.7 m/s.
Biochemical Parameters
After an overnight fasting of 8 h, venous blood (10 mL) was collected by puncturing the antebrachial vein using vacutainers with sodium fluoride (glucose analysis), ethylenediaminetetraacetic acid (EDTA) (HbA1c analysis) and serum separator tubes (lipid, insulin, N-carboxymethyl-lysine [N-CML] and chromium analysis).
Sample storage: Serum samples for chromium and N-CML estimation were stored at –80° in a freezer for later analysis.
FPG estimation:
Plasma glucose (normal level is 70–99 mg/dL) was estimated using the enzymatic colourimetric method on a fully automated analyser (Beckman AU 680, Beckman Coulter Diagnostics, CA, USA). The test method used was the hexokinase enzymatic method.
Estimation of HbA1c:
HbA1c (normal HbA1c level is <5.7%) was estimated using high-performance liquid chromatography (HPLC) (equipment name: HLC, automated glycohaemoglobin analyser, Tosoh India Pvt Ltd, Mumbai).
Estimation of serum fasting insulin (FI):
FI (normal level is <25 mIU/L) was estimated using chemiluminescence immunoassay (CLIA) with the help of ADVIA Centaur XP Immunoassay System (Siemens, USA).
13
Estimation of Homeostatic Model Assessment of Insulin Resistance (HOMA-IR):
Estimating HOMA-IR (<1.0 indicates optimal insulin sensitivity) is to assess the degree of IR in an individual. IR is decreased sensitivity of tissues to the metabolic effects of insulin, such as decreased insulin-mediated glucose uptake or reduced gluconeogenesis. HOMA-IR is a representation of hepatic IR. It is derived by using the following formula: IRHOMA = FI (IU/L) × fasting glucose (mg/dL)/405.
14
Serum chromium:
It was measured by using inductively coupled plasma mass spectrometry (ICP-MS) (equipment name: NexIon 2000 Make, Perkin Elmer) in the Nutrition Laboratory, DIPAS, DRDO, New Delhi. The reference values for serum chromium are 0.1–0.2 µg/dL in adults.
8
Serum N-CML:
N-CML is a type of advanced glycated end product (AGE product) found in serum. It was analysed using a competitive inhibition enzyme-linked immunosorbent assay (ELISA) technique using a commercially available kit (manufacturer: ELK Biotechnology Co. Ltd, China; catalogue: ELK7896). Sensitivity of the kit was 25.9 ng/mL, and the detection range was 78.13–5,000 ng/mL. The reference values for serum N-CML (AGE product) are 433.56–679.50 ng/mL in adults.
15
Lipid profile:
Total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C), triglycerides (TGs) and high-density lipoprotein-cholesterol (HDL-C) were measured using enzymatic method with commercially available kits (DiaSys Diagnostic Systems, Germany) as per the manufacturer’s instructions, using a fully automated analyser (Beckman AU 680, Beckman Coulter Diagnostics, CA, USA). Normal fasting levels considered are TGs <150 mg/dL, LDL-C <100 mg/dL, HDL-C 40–60 mg/dL and TC <200 mg/dL.
16
Statistical Analysis
The statistical analysis was performed using STATA software version 12.0 (StataCorp LLC, TX, USA). The normality of the continuous variables was assessed using the skewness–kurtosis test.
Comparison of continuous variables between the three groups was done using one-way analysis of variance (ANOVA) or the Kruskal–Wallis test, as appropriate.
A comparison of continuous variables for all three visits within the same group was done using repeated-measures ANOVA or the Friedman test, as appropriate.
Correlation was performed using Pearson’s, Spearman’s or point biserial correlation test, as appropriate.
A p value ≤.01 was considered statistically significant.
Results
This longitudinal case-control study included ODM, GDM and ‘euglycaemic pregnant women’ (control) in 24–28 weeks of gestation. The research team recorded the parameters from the participants during three follow-ups as Visit 1: 24–28 weeks of gestation, Visit 2: 32–38 weeks of gestation, and Visit 3: 6–12 weeks after parturition.
The baseline characteristics, such as age, height and pre-gestational weight, did not differ significantly among controls, GDM and ODM groups. However, a significant difference was observed in gestational age at Visit 2 (p = .03), possibly reflecting earlier interventions or complications in diabetic pregnancies. Family history of diabetes was significantly higher in GDM and ODM groups (p = .01), suggesting a strong genetic predisposition. Though postpartum BMI tended to be higher in diabetic groups, it was not statistically significant. The incidence of numbness symptoms was minimal (Table 1).
Comparison of Demographic, Anthropometric and Clinical Parameters Between Controls, Gestational Diabetes Mellitus (GDM) and Overt Diabetes Mellitus (ODM) Groups During Visits.
Nerve conduction studies reveal significant alterations in diabetic groups compared to controls, particularly in sensory latencies and amplitudes of the sural and ulnar nerves. Prolonged latencies and reduced amplitudes in GDM and ODM indicate early axonal and demyelinating changes, suggestive of subclinical neuropathy. Notably, sural nerve latency increased and amplitude decreased across visits in ODM, indicating progressive sensory neuropathy. Motor responses showed higher ulnar compound muscle action potential (CMAP) amplitudes and conduction velocities in diabetics, possibly reflecting compensatory changes or early hyperexcitability (Table 2). The post hoc analysis revealed significant sensory nerve conduction abnormalities in ODM compared to both controls and GDM, particularly in sural and ulnar nerves. Latency was prolonged and amplitude reduced, reflecting axonal degeneration. GDM showed milder but significant delays in ulnar SNAP latency and reduced amplitudes, indicating early subclinical neuropathy. Motor findings showed increased ulnar CMAP amplitudes in ODM, possibly due to collateral reinnervation. Conduction velocity differences were less consistent (Table 3). Within-group comparisons over three visits revealed significant longitudinal changes in the GDM group, especially in ulnar sensory latency and amplitude, indicating evolving subclinical neuropathy. Notably, ulnar sensory latency differed significantly between Visit 2 and Visit 3 (p = .033), and amplitude declined between Visit 1 and Visit 2 (p = .013). These findings suggest early functional deterioration in GDM despite glycaemic normalisation. In contrast, controls and ODM groups showed minimal progression across visits (Table 4).
Comparison of Nerve Conduction Study (NCS) Parameters Between Controls, Gestational Diabetes Mellitus (GDM) and Overt Diabetes Mellitus (ODM) Groups During Each Visit.
Intergroup Comparison of Nerve Conduction Study (NCS) Parameters During Each Visit.
Intragroup Comparison of Nerve Conduction Study (NCS) Parameters Between Visits.
Significant hyperglycaemia (FPG, HbA1c), hyperinsulinaemia and elevated HOMA-IR in GDM and ODM groups confirm IR and poor glycaemic control during pregnancy and postpartum. Chromium levels were markedly lower in diabetics (p = .001). Elevated N-CML levels and AGE accumulation in GDM and ODM were present significantly (p = .01). Dyslipidaemia, particularly elevated LDL and TGs, was more significantly pronounced in these groups, indicating heightened cardiometabolic risk (p = .001) (Table 5). The intergroup comparisons showed significantly elevated FPG, HbA1c, insulin and HOMA-IR in both GDM and ODM versus controls, reflecting progressive glucose intolerance and IR during pregnancy. Postpartum, glycaemic differences between GDM and ODM diminished, indicating partial metabolic normalisation in GDM but persistent impairment in ODM. Serum chromium was significantly lower in ODM, emphasising its role in insulin action (p = .001) (Table 6). Within-group comparisons show significant postpartum improvements in glycaemic markers (FPG, HbA1c, insulin, HOMA-IR) in GDM and ODM groups, indicating enhanced insulin sensitivity after delivery. Chromium levels increased significantly only in the control group, while remaining suboptimal in diabetics, suggesting persistent micronutrient imbalance in diabetes. N-CML levels significantly declined only in GDM, indicating a transient reduction in oxidative stress. Lipid profiles showed significant reductions in TC and TGs in all groups, particularly in diabetics, though LDL-C and HDL-C changes were not significant (Table 7).
Comparison of Biochemical Parameters Between Controls, Gestational Diabetes Mellitus (GDM) and Overt Diabetes Mellitus (ODM) Groups During Visit 1 and 3 or 1, 2 and 3.
Intergroup Comparison of Biochemical Parameters During Visit 1 and 3 or 1, 2 and 3.
Intragroup Comparison of Biochemical Parameters Between Visits.
Correlation analysis revealed that poor glycaemic control (FPG, HbA1c, HOMA-IR) negatively correlated with sensory and motor amplitudes, especially SNSAMP and UNMAMP at both visits. 17 Notably, HbA1c and FI negatively correlated with UNMAMP and SNSAMP, consistent with metabolic injury to peripheral nerves. A stronger negative association of NCS parameters with N-CML, and a weak but consistent negative correlation of serum chromium with motor latency, is observed (Table 8).
Correlation Between Nerve Conduction Study (NCS) and Biochemical Parameters of All Three Groups (N = 39).
Discussion
This longitudinal investigation assessed glycaemic, biochemi-cal and neurophysiological metrics in pregnant women exhibiting normal glucose tolerance, GDM and ODM during antenatal and postpartum visits. The results highlight the intricate metabolic and neurophysiological changes linked to HIP and its possible long-term consequences.
Baseline characteristics like age, height and pre-gestational weight were similar across all groups, consistent with previous studies that reported no reliable anthropometric differences at baseline in similar populations. 18 However, the notably earlier gestational age at Visit 2 in the diabetic cohorts likely reflects clinical decisions for earlier monitoring or interventions, supporting previous findings related to obstetric complications linking GDM with metabolic risks and obstetric interventions. 19 The increased prevalence of family history within these groups points to a genetic susceptibility to dysglycaemia, aligning with the results from Alfadhli. 18
The biochemical profiles confirmed significant hypergly-caemia (FPG, HbA1c), hyperinsulinaemia and raised HOMA-IR levels in both GDM and ODM groups during pregnancy and after delivery, indicating a persistent state of IR, which is consistent with findings reported by Metzger et al. 20 Glycaemic normalisation postpartum in GDM, but not in ODM, indicates that the former may be reversible, while the latter suggests chronicity. Consistent with our study findings, Vincent 21 also reported reduced serum chromium levels among diabetic participants, thus reflecting its role in enhancing insulin action and insulin signalling. This action might be due to chromium enhancing the action of insulin while mitigating oxidative damage. Persistently low chromium levels after birth in both GDM and ODM indicate an ongoing micronutrient imbalance that might contribute to continuing metabolic dysfunction. These findings support early intervention to mitigate long-term maternal and foetal complications associated with dysglycaemia in pregnancy. Another study by Uribarri et al. 22 reported that elevated levels of N-CML, an AGE product, observed in GDM and ODM signal increased oxidative stress and a heavy glycation burden, both of which are vital in the development of diabetic complications. While N-CML levels dropped significantly in the GDM cohort postpartum, they remained elevated in ODM, pointing to progressive oxidative damage. The dyslipidaemia noted, especially high levels of LDL and TGs, further reinforces the cardiometabolic strain observed in diabetic pregnancies. 23 Lipid profiles improved postpartum in all cohorts, but the persistent abnormal patterns in ODM indicate ongoing cardiovascular risk. Also, these trends highlight postpartum metabolic recovery and residual risk.
Electrophysiological evaluations showed early peripheral nerve dysfunction in both GDM and ODM, with the sensory component of sural and ulnar nerves being predominantly affected. Prolonged latencies and diminished amplitudes in these nerves suggest early axonal and demyelinating changes.17, 24 GDM exhibited milder yet notable abnormalities, suggesting early subclinical neuropathy. These results are consistent with findings by Yagihashi et al., 25 who documented structural alterations in peripheral nerves even at the onset of diabetes. The increased ulnar CMAP amplitude in ODM may reflect compensatory collateral reinnervation. 25 Longitudinal within-group analysis emphasised progressive alterations in nerve conduction within the GDM cohort across visits, particularly the worsening of ulnar sensory latency and amplitude, indicating evolving subclinical neuropathy.26, 27 Despite improvements in metabolism after delivery, nerve dysfunction advanced in GDM, implying that normalisation of glycaemia may not completely reverse neuronal damage, as previously observed by Vinik et al. 24 These results suggest that GDM may represent an intermediate stage of diabetic neuropathy, meriting early screening and intervention.
Correlation analysis showed negative associations between glycaemic metrics (FPG, HbA1c, insulin) and sensory and motor amplitudes, pointing to metabolic damage and suggesting early axonal degeneration with hyperglycaemia. 17 A strong negative correlation with N-CML, an AGE product, emphasises oxidative stress-induced neuronal damage. 28 Additionally, a weak yet consistent association between low chromium and prolonged latencies suggests a protective effect through improved insulin sensitivity and reduced oxidative damage. 21
Limitations
Limited sample size and single-centre study might limit the generalisability of the results of the study.
Conclusion
This longitudinal follow-up case-control study on healthy pregnant women and pregnant women with GDM and ODM during different stages of gestation, that is, (a) 24–28 weeks (first visit), (b) 32–38 weeks (second visit) and (c) 6–12 weeks after parturition (third visit), was conducted to assess the neurological status and biochemical changes among North Indian pregnant women. The findings of this study accentuate the necessity for ongoing metabolic and neurological monitoring in GDM and ODM after childbirth. Early screening and focused interventions, including micronutrient supplementation and lifestyle modifications, may help avert progression to overt neuropathy and mitigate long-term complications. Postpartum across all groups, but persistently abnormal trends in ODM signal residual cardiovascular risk.
Footnotes
Acknowledgement
Nutrition Laboratory, DIPAS, DRDO, New Delhi.
Authors’ Contributions
Akriti Kapila, Sunita Mittal and Latika Mohan formulated and conceptualised the idea, aims and goals. Sunita Mittal was the principal investigator. Ashwini Mahadule and Gauri Mittal contributed to the formal analysis and visualisation. Anissa Atif Mirza and Anupama Bahadur contributed to the resource and supervision.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The source of funding for this study was institutional, i.e. All India Institute of Medical Sciences, Rishikesh, since this was an intramural project.
Statement of Ethics and Informed Consent
This study was approved by the Institutional Ethics Committee, AIIMS Rishikesh (Certificate no. AIIMS/IEC/19/925, dated 19 July 2019, and, for extension of duration due to COVID-19, AIIMS/IEC/22/43, dated 18 February 2022).
Participants were explained about the study in their local language, and, if they agreed to take part and signed the informed consent form, they were recruited into the study.
