Search Results (114)

Background and Objectives: Abnormal plantar pressure distribution is an important biomechanical risk factor for diabetic foot ulceration in individuals with type 2 diabetes (T2D) and peripheral neuropathy. Although several factors have been linked to increased forefoot plantar pressure, factors associated with the forefoot-to-rearfoot pressure ratio remain unclear. This study aimed to investigate the clinical factors associated with the forefoot-to-rearfoot pressure ratio and peak forefoot pressure in this population. Materials and Methods: This study was a cross-sectional analysis of 84 individuals with T2D and peripheral neuropathy enrolled in a randomized controlled trial. Vibration perception threshold (VPT), passive ankle dorsiflexion range of motion, ankle stiffness, plantar pressure distribution, body mass index (BMI), HbA1c, duration of diabetes, and foot deformities were assessed. Multiple linear regression analyses were performed to identify factors associated with plantar pressure outcomes. Results: The model for forefoot-to-rearfoot pressure ratio explained 51.2% of the variance (adjusted R² = 0.467); ankle stiffness (β = 0.505, p < 0.001), foot deformity (β = 0.194, p = 0.030), and ankle dorsiflexion range of motion (β = −0.363, p = 0.001) were independent predictors. The model for peak forefoot pressure explained 31.4% of the variance (adjusted R² = 0.251); ankle stiffness (β = 0.337, p = 0.002), foot deformity (β = 0.278, p = 0.009), and BMI (β = 0.226, p = 0.029) were independent predictors. HbA1c, diabetes duration, and VPT were not significant in either model. Conclusions: In individuals with T2D and peripheral neuropathy, plantar pressure distribution, particularly forefoot loading, appears to be more closely related to biomechanical and structural factors than to metabolic indicators. These findings suggest that assessment of ankle stiffness, joint mobility, and foot posture may provide clinically relevant information for identifying biomechanical factors associated with unfavorable plantar loading in individuals with T2D and peripheral neuropathy. Full article

Background/Objectives: Abnormal nocturnal dipping (ND) in diabetes mellitus is linked to adverse outcomes. Its association with vitamin D (VD), cardiovascular autonomic neuropathy (CAN), and diabetic kidney disease (DKD) remains unclear. We aim to evaluate the association of VD status with ND loss as well as CAN in type 2 diabetes mellitus (T2DM) across different DKD stages. Methods: 83 T2DM patients were included in this cross-sectional study: 36 with early-stage DKD (urine albumin-to-creatinine ratio [UACR] 30–299 mg/g; Group 1) as well as 47 with advanced-stage DKD (UACR ≥ 300 mg/g; Group 2). Results: Patients with advanced DKD exhibited higher CAN prevalence (72.70% vs. 52.80%; p < 0.05), lower 25(OH)D (26.90 ± 9.90 vs. 29.60 ± 8.00 ng/mL; p = 0.05), and lower systolic ND (0.82 ± 8.40 vs. 4.50 ± 7%; p < 0.05). Among patients with advanced DKD, lower 25(OH)D levels, classified according to the Institute of Medicine (IOM) criteria (normal ≥ 20 ng/mL), were associated with reduced systolic ND (1.90 ± 7.80 vs. −5.60 ± 9.30%; p < 0.05), while the Endocrine Society criteria showed a progressive reduction in systolic ND across VD classification (insufficiency 20–29.9, deficiency <20, and sufficiency ≥30 ng/mL), (−5.60 ± 9.30 vs. −0.20 ± 7.30 vs. 3.30 ± 8%, respectively; p < 0.05). Patients with definite/severe CAN exhibited lower systolic ND (−4.38 ± 7.41 vs. 3.34 ± 8.00%; p < 0.01) than those with absent/early CAN. In addition, lower VD levels were associated with greater CAN severity according to both the Endocrine Society criteria (p < 0.01, r = 0.40), and the IOM criteria (p < 0.05, r = 0.30). In the regression model, VD remained an independent predictor of absence of diastolic ND (Rsqr = 0.393; p < 0.05). Conclusions: These findings suggest a potential link between vitamin D status, autonomic dysfunction, and ND in patients with T2DM and advanced DKD. Full article

Background: Memantine, an uncompetitive and voltage-dependent N-methyl-D-aspartate (NMDA) receptor antagonist, is clinically established for moderate-to-severe Alzheimer’s disease. Its pharmacodynamic profile, low-to-moderate affinity, rapid open-channel block, and strong voltage dependency allows selective inhibition of pathological NMDA overactivation while preserving physiological neurotransmission. Increasing evidence shows that these same mechanistic principles operate in the peripheral nervous system, where NMDA receptors contribute to excitotoxicity, oxidative stress, neuroinflammation, and maladaptive nociceptive signaling. Purpose: To synthesize emerging preclinical and clinical evidence demonstrating memantine’s modulatory and neuroprotective actions in peripheral neurons and glia and to outline implications for drug repurposing across neurology, pain medicine, oncology, supportive care, and ophthalmology. Methodology: A narrative integration of mechanistic studies, in vivo preclinical models, and heterogeneous clinical trials evaluating memantine’s effects on peripheral sensory neurons, autonomic neurons, Schwann cells, retinal ganglion cells, and neuromuscular junction physiology. Evidence was examined across conditions involving excitotoxicity, oxidative injury, mitochondrial dysfunction, apoptotic signaling, neuroinflammation, and neuropathic pain amplification. Results: Memantine consistently attenuates peripheral excitotoxic calcium influx, suppresses NOX-2–mediated ROS generation, stabilizes mitochondrial membrane potential, modulates Bax/Bcl-2 signaling, and reduces neuroinflammatory cytokine activity. It also inhibits dorsal horn wind-up selectively under neuropathic conditions. These convergent mechanisms yield protective effects across chemotherapy-induced peripheral neuropathy (CIPN), diabetic neuropathy, traumatic nerve injury, phantom limb pain, retinal ganglion cell excitotoxicity, and organophosphate-induced neuromuscular toxicity. Clinical evidence includes improved multimodal neuropathy outcomes in diabetic neuropathy when combined with gabapentin, reduced phantom limb pain prevalence and intensity at six months, and a five-fold reduction in post-mastectomy neuropathic pain with pre-emptive administration. Conclusions: Memantine should be conceptually reframed as a system-wide neuroprotective agent with substantial translational potential beyond the CNS. Priorities for future development include NR2B-selective peripheral NMDA antagonists, peripherally restricted formulations, single-cell transcriptomic mapping of peripheral NMDA receptor subtypes, and adequately powered PNS-specific randomized trials. Full article

Background: Peripheral neuropathy (PN) is one of the most common disorders of the peripheral nervous system, affecting around 10% of the general adult population. PN results in pain, paraesthesia, and sensory loss, negatively impacting patients’ quality of life. The condition is characterised by damage to peripheral nerves and is often linked to risk factors, among which diabetes is the most common cause. The incidence rate of patients suffering from PN has significantly increased in the last three decades, yet there are currently no pharmacological treatments to reverse or prevent the development of PN. Therefore, neurotropic B vitamins, specifically vitamins B1, B6, and B12, which are essential for protecting nerves from damage, nerve regeneration, and maintaining nerve health, are promising treatment options. Other biofactors with therapeutic potential are vitamin D3, which has been described as a regulator of neuroplasticity and neuroinflammation with indications to promote neuronal regeneration following injury, as well as alpha-lipoic acid (ALA), a well-described treatment option in PN. Methods: The regenerative capacity of vitamins B1, B6, B12, and D3 and ALA was determined following neurite degeneration in vitro induced by either hydrogen peroxide (H₂O₂) or homocysteine. Results: In line with previous studies, the results show that a combination of vitamins B1, B6, and B12 has a significantly better effect on neurite extension following homocysteine-induced degeneration. In H₂O₂-insulted neural cells, the combination of all three neurotropic B vitamins was also shown to be superior to that of individual B vitamin treatment, demonstrated by longer neurite length after insult. Vitamin D3 also demonstrated a neuro-regenerative effect on neurites in healthy and insulted neural cells, whereas ALA only appears to provide a protective effect, preventing ongoing damage and supporting neurite outgrowth in healthy cells. No synergistic effects were seen with any treatment combinations. Conclusions: We have established an in vitro model of homocysteine-induced degeneration, and using this model alongside H₂O₂-induced degeneration, we provide evidence of the neurite regenerative capacity of vitamins B1, B6, B12, and D3. Full article

(1) Background: Neuropathies are common complications of type 1 diabetes (T1D) and previous research demonstrated concomitant demyelinization involving myelin basic protein (MBP). T1D patients have a significantly increased risk of multiple sclerosis (MS) and vice versa. The causes are unknown. (2) Methods: Proteonomic searches using BLAST were used to identify similarities between T-cell receptor (TCR) sequences from T1D and MS patients. LALIGN was used to explore whether T1D-related proteins mimic MBP. Enzyme-linked immunoadsorption assay (ELISA) was used to test for cross-reactivity between T1D-related antibodies or sera with MBP. Ultraviolet spectroscopy was used to test whether T1D-specific T-cell receptor (TCR) sequences recognized MBP. (3) Results: BLAST revealed that a subset of T1D and MS TCR are nearly identical. Sequence similarities were found by LALIGN between MBP and T1D-associated antigens. Antibodies against insulin, PTPRN(IA-2), and GAD1 recognized MBP. Most TCR sequences derived from T1D patients recognized MBP as an antigen. Most T1D-derived sera recognize MBP, but healthy sera do not. BLAST revealed that coxsackievirus proteins do not mimic MBP, but many Clostridia proteins do. (4) Conclusions: Cross-reactivity between TID-related antibodies and TCR with MBP may help to explain the pathogenesis of T1D neuropathies and the increased co-occurrence of these two diseases. Full article

Background: Gabapentinoids (gabapentin and pregabalin) are medications used to treat epilepsy and diabetic neuropathy. Reports of gabapentinoid-associated adverse drug reactions (ADRs) are increasing. This study aimed to explore these potential signals, with a focus on ocular adverse effects. Methods: A mixed-methods design was adopted, utilising quantitative and qualitative approaches encompassing: (I) a social listening approach using Reddit; (II) a pharmacovigilance retrospective disproportionality analysis study using serious reports to the FDA Adverse Event Reporting System (FAERS) (2015–2025); and (III) a literature review (2014–2025). Results: Reddit data revealed several user-reported adverse ocular effects, including blurred vision, diplopia and photosensitivity, across 78 and 40 identified reactions for gabapentin and pregabalin, respectively. Pharmacovigilance findings identified positive signals for eye disorders with both gabapentinoids compared to active control diazepam, amitriptyline and carbamazepine). Incidental findings highlighted potential discrepancies between FAERS data and undesired effects listed within the manufacturer’s literature, with pregabalin exhibiting higher reporting odds than gabapentin for cataract (ROR: 4.55; 95% CI: 2.51–8.26) and blindness (ROR: 2.63; 95% CI: 1.85–3.73). The literature review reinforced eye disorder concerns, specifically retinal effects with gabapentinoids, noting the absence of robust, high-quality research. Nevertheless, it identified a plausible biological mechanism involving the CACNA2D1 receptor in retinal cells of animal models, with evidence of an effect following oral administration of pregabalin. Conclusions: This study suggests an increase in eye disorder reports associated with gabapentinoid use. Further clinical and epidemiological studies are warranted to validate these real-world signals. Full article

Diabetic peripheral neuropathy (DPN) causes progressive sensorimotor dysfunction of the lower extremities, impairing mobility and increasing fall risk. Phase-specific neuromuscular activation during repeated sit-to-stand (STS) transfers in DPN remains poorly characterized. This study compared lower-limb muscle activation during repeated STS transfers in fifteen ambulatory DPN patients and fifteen age- and gender-matched healthy controls. Bilateral surface electromyography (sEMG) was recorded from the vastus lateralis, vastus medialis, biceps femoris, gluteus maximus, and gluteus medius; normalized to maximum voluntary contractions (MVCs), and analyzed separately for STS and stand-to-sit phases. No significant between-group differences were observed for any muscle during either phase (all p > 0.05). Effect sizes were small-to-medium, with confidence intervals indicating uncertainty around the estimated between-group differences. Notably, gluteus maximus activation during STS demonstrated the largest between-group effect size, favoring the DPN group (d = 0.54), although this difference was not statistically significant. This finding may indicate a possible proximal hip-extensor strategy that requires confirmation in larger studies. Both groups demonstrated significantly greater vastus lateralis and gluteus maximus activation during STS than stand-to-sit, reflecting greater neuromuscular demand during rising. Overall, the non-significant between-group findings, together with the effect-size and confidence-interval analysis, suggest that MVC-normalized EMG amplitude alone may have limited sensitivity for detecting neuropathic neuromuscular alterations during transitional tasks in ambulatory patients with DPN. Full article

Type 2 diabetes (T2D) is a prevalent disease worldwide that increases the risk of cardiovascular (CV) complications, disability and mortality. While excessive alcohol consumption is harmful, the effects of moderate wine consumption remain debated. This review evaluates whether moderate wine intake affects the risk of developing T2D and its impact on subjects with T2D. Twenty-eight studies were analysed. Evidence suggests an association between moderate wine consumption and the risk of developing T2D, with a J-shaped relationship, and reduced risk observed at low levels. This effect appears more pronounced with red wine, likely related to its higher polyphenol content, and when consumed with meals. On the other side, in patients with T2D, moderate wine consumption has been associated with a reduced risk of CV complications, nephropathy and mortality. It has also been linked to improved lipid profiles and reduced inflammatory markers, without adversely affecting body weight or glycaemic control in well-managed patients. These effects may be enhanced within a Mediterranean dietary pattern, suggesting synergistic actions. However, alcohol intake may increase the risk of hypoglycemia, particularly in patients receiving glucose-lowering therapies. It should be avoided by vulnerable individuals, and those with comorbidities such as MASLD and other significant liver diseases, peripheral neuropathy or other severe conditions. In conclusion, moderate wine consumption may be associated with a reduction in the risk of developing T2D and with several CV benefits in patients with T2D. Vulnerable patients should abstain and individuals who currently do not drink alcohol should not start drinking. If wine is consumed, intake should always remain moderate (as low as possible), within healthy meals and only after individual clinical assessment. Full article

Diabetic neuropathy (DN) is a multifactorial complication of diabetes mellitus driven by chronic hyperglycemia, insulin resistance, and disturbed metabolic homeostasis, leading to progressive injury of both the peripheral and central nervous systems. This study investigated whether L-serine supplementation could attenuate DN through dose-dependent metabolic and neuroprotective mechanisms in a high-fat diet (HFD) plus streptozotocin (STZ)-induced diabetic rat model. Male Wistar rats (n = 8 per group) were allocated to five groups: normal control (NC), diabetic control (DC), pioglitazone (PIO; 1.5 mg/kg/day), low-dose L-serine (S1; 200 mg/kg/day), and high-dose L-serine (S2; 400 mg/kg/day). After 60 days of oral gavage, behavioural testing, glucose and insulin profiling, HOMA-IR calculation, brain histopathology, nerve growth factor (NGF) immunohistochemistry, and LC–MS/MS-based proteomic analysis of cerebral tissue were performed. Diabetic rats exhibited marked hyperglycaemia (355.33 ± 4.72 mg/dL), hyperinsulinaemia, severe insulin resistance (HOMA-IR 16.8 ± 3.2; a 14-fold increase), impaired thermal nociception, motor dysfunction, and pronounced neuronal degeneration. L-serine supplementation significantly improved metabolic status: S1 reduced HOMA-IR by 77.4% and S2 by 87.5% relative to diabetic controls (p < 0.001). High-dose L-serine produced greater improvements in thermal sensitivity, motor coordination (rotarod latency 26.67 ± 1.52 s vs. 16.1 ± 0.85 s in DC; p < 0.05), and NGF expression (8.6-fold increase vs. DC). Histopathology confirmed attenuation of neuronal injury and gliosis in both treatment groups. Exploratory, group-level proteomic profiling identified dose-specific molecular signatures: S1 was predominantly associated with carbohydrate, lipid, and biosynthetic pathways, whereas S2 was associated with synaptic, neurotransmission-related, and proteostasis pathways. Within the constraints of an exploratory design—group-level pooled proteomics, analysis of cerebral rather than peripheral-nerve tissue, and only two doses—these findings indicate that L-serine attenuates the metabolic and behavioural features of experimental diabetic neuropathy and generates the testable hypothesis of dose-dependent neuro-metabolic remodelling. The proteomic signatures are hypothesis-generating and require orthogonal validation before any mechanistic or translational inference can be drawn. Full article

Background/Objectives: Orthostatic hypotension (OH) is a clinically relevant manifestation that may reflect cardiovascular autonomic dysfunction in type 2 diabetes (T2D), yet its correlates remain incompletely characterized. This cross-sectional study evaluated clinical, neuropathic, and sudomotor factors associated with OH and explored balance-related outcomes as secondary analyses. Methods: In this cross-sectional study, 124 adults with T2D aged ≥60 years underwent standardized orthostatic blood pressure testing. Peripheral neuropathy was assessed using the Michigan Neuropathy Screening Instrument (MNSI), and sudomotor function was assessed by electrochemical skin conductance measured with Sudoscan. Balance, mobility, and fear of falling were evaluated as exploratory secondary outcomes. Active antihypertensive treatment was recorded at the time of assessment and considered a potential confounder. Multivariable logistic regression was used to identify factors associated with OH. Results: OH was associated with longer diabetes duration (OR = 1.11/year, p = 0.002), higher objective neuropathy severity (MNSI-B; OR = 1.27, p = 0.049), and increased urinary albumin-to-creatinine ratio (OR = 1.01, p = 0.035). Sudomotor parameters did not differ significantly between OH groups in univariate analyses and were not retained in the final parsimonious model. Exploratory analyses showed no significant univariate differences in balance or fear-of-falling outcomes by OH status. Model discrimination was acceptable (AUC = 0.787), whereas calibration was imperfect according to the Hosmer–Lemeshow test; therefore, model performance should be interpreted as apparent and explanatory rather than predictive. Conclusions: In older adults with T2D, OH was associated with longer disease duration, greater neuropathy burden, and microvascular involvement. Sudoscan-derived measures were not independently associated with OH in this cohort. Because of the cross-sectional design and residual medication confounding, all findings should be interpreted as associations only. These results support routine orthostatic evaluation alongside neuropathy and albuminuria assessment, while predictive modeling requires external validation in larger cohorts. Full article

Background and Objectives: Diabetic peripheral neuropathy (DPN) is a common and debilitating complication of obesity and type 2 diabetes (T2D) affecting up to 50% of patients with long-standing disease. While chronic hyperglycemia plays a central role in its pathogenesis, intensive glycemic control provides only partial protection, suggesting the involvement of additional metabolic pathways. The primary objective of this systematic review was to evaluate the role of lipid metabolism disturbances and advanced lipidomic alterations in the development and progression of DPN in patients with obesity and T2D. Secondary objectives included identifying specific lipid species associated with DPN and exploring their potential pathophysiological and clinical implications. Methods: This systematic review included 8 studies that met the inclusion criteria and was conducted according to PRISMA guidelines and registered in PROSPERO/2026/CRD420261288920. Study quality was assessed using the Newcastle–Ottawa Scale. Results: Large population-based cohorts reported a consistent association between hypertriglyceridemia and DPN prevalence, with triglyceride levels >204 mg/dL associated with an approximately 40% increased risk. Lipidomic analysis revealed alterations in acylcarnitine, sphingolipids, and phospholipids. However, the evidence remains limited and heterogeneous, and neuropathy-specific outcomes were insufficiently evaluated in interventional studies. Conclusions: Lipid metabolism disturbances, particularly hypertriglyceridemia and specific lipidomic alterations, may contribute to DPN beyond the effects of hyperglycemia. Although not yet clinically actionable, lipidomic alterations may represent promising future biomarkers and therapeutic targets in DPN. However, the current evidence is limited by heterogeneity and predominantly observational designs. Further well-designed longitudinal and interventional studies are needed to clarify causal relationships and clinical relevance. Full article

Background: Growth differentiation factor-15 (GDF-15) is a stress-responsive cytokine associated with inflammation, metabolic dysfunction, and cardiovascular disease. Its role as a biomarker of microvascular complications in type 2 diabetes (T2D) remains incompletely defined. Objective: To evaluate circulating GDF-15 levels and their association with microvascular complications in patients with T2D. Methods: This cross-sectional study included 160 participants divided into three groups: T2D (n = 93), obesity without carbohydrate disorders (n = 36), and healthy controls (n = 31). Microvascular complications (neuropathy, nephropathy, retinopathy) were assessed. Multivariable logistic regression and receiver operating characteristic (ROC) analysis were performed. Results: GDF-15 levels were significantly higher in T2D compared with non-diabetic individuals (267.5 ± 168.9 vs. 118.3 ± 55.5 pg/mL, p < 0.001). Higher GDF-15 was associated with neuropathy (odds ratio (OR) 1.985; 95% confidence interval (CI) 1.431–2.753) and nephropathy (OR 1.673; 95% CI 1.243–2.254) in unadjusted models. After adjustment, only nephropathy remained independently associated (OR 1.405; 95% CI 1.026–1.923). ROC analysis showed moderate discriminative ability for nephropathy (area under the curve (AUC) = 0.763). Conclusions: Circulating GDF-15 levels are significantly elevated in patients with T2D and are associated with microvascular complications, with the strongest independent association observed for diabetic nephropathy. These findings suggest that GDF-15 may represent a promising biomarker reflecting metabolic stress and risk of diabetic complications. Full article

Background: Diabetic foot ulcer (DFU) is a major complication of type 2 diabetes (T2D), frequently resulting in disability, lower-limb amputation, and substantial healthcare burden. Early identification of patients at high risk of progressing to severe DFU is essential for timely intervention, yet evidence on associated risk factors remains limited in Bangladesh. This study aims to identify demographic, clinical, and behavioral predictors of severe DFU to support early management strategies. Methods: A cross-sectional study was conducted among 159 DFU patients attending the Rajshahi Diabetic Association General Hospital, Bangladesh. Data on demographic characteristics, clinical variables, and behavioral factors were obtained through structured questionnaires and standardized examinations. Severe DFU was defined as Wagner grades 3–5, while grades 0–2 were considered non-severe. Firth’s penalized logistic regression was used to identify determinants of severe DFU. Model performance was assessed using ROC analysis, calibration belt analysis, and decision curve analysis (DCA). Results: Among the 159 participants, 101 (63.5%) presented with severe DFU. Patients with severe DFU had significantly higher BMI (26.1 vs. 23.7 kg/m²), treatment costs (50,000 vs. 20,000 BDT), and were older (57 vs. 54 years). Severe DFU was also associated with higher prevalence of peripheral arterial disease (PAD) (29.7% vs. 3.4%), prior amputation (31.7% vs. 3.4%), peripheral neuropathy (PN) (86.1% vs. 58.6%), and poor glycemic control (71.3% vs. 30.7%) (all p < 0.05). Firth’s regression identified older age (aOR 1.08), poor glycemic control (aOR 3.90), PN (aOR 3.41), PAD (aOR 7.54), and previous amputation (aOR 13.67) as independent predictors of severe DFU. Conclusions: Older age, uncontrolled glycemia, PN, PAD, and prior amputation were significantly associated with severe stages of DFU. Early detection and targeted management of these factors are critical to reducing complications and lowering the healthcare burden. Full article

Background/Objectives: Enteric glial cells (EGCs) are key players in regulating enteric neurons and gastrointestinal functions including disturbed gut motility in diabetic patients. Enteric neuronal damage has been shown in type 1 diabetes, but EGCs’ vulnerability to hyperglycaemic insults requires more investigation. Therefore, we aimed to study the quantitative changes in the EGC network enmeshing enteric plexuses, intestinal smooth muscle and mucosa in streptozotocin-induced acute (1-week) and chronic (10-weeks) diabetic rat models. Methods: Fluorescent immunohistochemistry using Sox10 glial and HuC/HuD pan-neuronal markers, immunogold electron microscopy and ELISA were performed on different gut segments. Results: In the submucosal ganglia of the ileum and colon, the density of Sox10-immunoreactive EGCs was significantly reduced in acute and increased in chronic hyperglycaemic rats without any changes in the duodenum. In the myenteric ganglia, regionally distinct alterations of glial density were noted in acute hyperglycaemia; however, a remarkable decrease was observed in chronic animals. Alterations of neuronal density did not follow the pattern of glial changes, resulting in shifts in the glia/neuron ratio. The presence of Sox10-HuC/HuD-immunoreactive cells and their diabetes-related quantitative changes were also revealed in enteric plexuses. The density of Sox10-labelling gold particles was significantly increased in the duodenal myenteric glia of diabetic rats. Muscular EGC density increased only in the colon after acute hyperglycaemia and changed in all segments after chronic hyperglycaemia. Glial fibrillary acidic protein levels decreased in the small intestine of chronic hyperglycaemic rats. Conclusions: Our present findings reveal time-dependent and regionally distinct changes in the EGC network in response to hyperglycaemia, contributing to diabetic enteric neuropathy and gut motility disturbances. Full article

Background/Objectives: Vitamin B12 deficiency is a common but often underdiagnosed complication in patients with type 2 diabetes (T2D) undergoing long-term metformin therapy. Accurate early prediction could enable targeted screening and timely intervention. This study aimed to develop and interpret a machine learning model for predicting vitamin B12 deficiency in metformin-treated patients with T2D, using eXtreme Gradient Boosting (XGBoost). Methods: A retrospective cross-sectional study was conducted at a single endocrinology centre (La Rabta University Hospital, Tunis, Tunisia). Patients with T2D treated with metformin for at least three years were included (n = 257); those with conditions independently affecting vitamin B12 metabolism were excluded. Vitamin B12 deficiency was defined as a serum B12 level below 150 pmol/L or a borderline level (150–221 pmol/L) with concurrent hyperhomocysteinemia (>15 μmol/L). XGBoost was selected after comparison with Logistic Regression (L2), Random Forest, and Support Vector Machine on the same 5-fold stratified cross-validated pipeline. Hyperparameters were optimized via Bayesian search (100 iterations × 5-fold stratified cross-validation), with the Matthews correlation coefficient (MCC) as the primary optimization metric to account for class imbalance. Model interpretability was achieved using SHapley Additive exPlanations (SHAP). Discrimination and calibration were assessed on an independent test set using bootstrap 95% confidence intervals (2000 resamples). Results: Of 257 patients, 95 (37.0%) presented with vitamin B12 deficiency. On the independent test set (n = 52), the optimized XGBoost model achieved an ROC-AUC of 0.671 [95% CI: 0.514–0.818], sensitivity of 0.737 [95% CI: 0.533–0.938], specificity of 0.545 [95% CI: 0.375–0.710], MCC of 0.273 [95% CI: 0.018–0.517], and a Brier Score of 0.259. SHAP analysis identified HbA1c, microalbuminuria, autonomic neuropathy, BMI, DN4 score, and fasting glucose as the most influential predictors. Nonlinear SHAP interaction plots revealed an increased predicted risk in patients with low HbA1c combined with a high cumulative metformin dose. Conclusions: The XGBoost–SHAP framework provided interpretable predictions of vitamin B12 deficiency in patients with T2D on metformin, identifying key clinical profiles for targeted screening. External multi-centre validation is required before clinical deployment. Full article