Background/Objectives: The co-occurrence of diabetic peripheral neuropathy and depression increases the symptom burden and risk of long-term complications. Methods: This cross-sectional study enrolled 131 patients with type 1 (age: 58.47 years; duration of diabetes: 35.61 years) and type 2 diabetes (age: 63.60 years; duration of diabetes: 11.49 years). All patients underwent assessment of socioeconomic status and evaluation using the Hospital Anxiety and Depression Scale, the Mental Component Score of the Short Form Healthy Survey Questionnaire, neuropathy disability score, nerve conduction studies, corneal confocal microscopy and intraepidermal nerve fibre density (IENFD) assessment. Results: The prevalence of foot pain (45% vs. 23.9%, p = 0.019), tingling (56.7% vs. 32.9%, p = 0.013), weakness (35% vs. 9.9%, p < 0.001), ataxia (40% vs. 16.9%, p = 0.001), and upper limb symptoms (45% vs. 19.7%, p = 0.001) were statistically significantly higher, while cold perception threshold (22.50 ± 8.47 vs. 26.34 ± 3.08, p = 0.007), corneal nerve fibre density (20.49 ± 7.55 vs. 24.16 ± 5.68, p = 0.002) and length (20.06 ± 6.98 vs. 22.95 ± 6.22, p = 0.014) were statistically significantly lower, but no differences in nerve conduction studies or IENFD were observed in patients with depression compared to patients without depression. Furthermore, patients with depression were from a lower socioeconomic class (51.7% vs. 21.1%, p < 0.001), had lower educational attainment (37.9% vs. 12.9%, p < 0.001), had lower income < £37,000 (29.3% vs. 11.4%, p = 0.010) and lived in areas of high deprivation (62.1% vs. 31.4%, p < 0.001). Conclusions: Comorbid depression in people with diabetes was linked to increased socioeconomic deprivation and a greater prevalence of neuropathic symptoms and small fibre pathology.

Diabetic peripheral neuropathy (DPN) is a prevalent and disabling complication of diabetes, yet whether established clinical DPN is reversible remains debated. At the 35th Annual Meeting of NEUROdiab, a formal debate examined arguments 'for' and 'against' the proposition that clinical DPN can be reversed. This review summarises both perspectives and considers their implications for clinical practice and future research. Evidence supporting the reversibility of clinical DPN draws on four main observations. Therapies for hereditary transthyretin amyloidosis demonstrate that substantial improvement in a progressive length-dependent neuropathy is biologically achievable, providing a proof-of-concept in a different disease. Pancreatic transplantation in Type 1 diabetes leads to measurable improvements in large- and small-fibre indices. Structured lifestyle interventions and bariatric surgery improve intraepidermal nerve fibre density, suggesting that metabolic correction can promote neural repair. Finally, emerging agents such as topical oxybutynin show early promise in improving surrogate outcomes, reinforcing the argument that clinical DPN may be partially reversible. Arguments opposing reversibility emphasise that DPN resembles other microvascular complications, in which early abnormalities may regress but established clinical disease is not reversed. Heterogeneity in diagnostic thresholds, phenotype variability and inconsistent outcome measures complicates defining what constitutes true reversal. Long-term studies and major trials, including DCCT/EDIC and BARI 2D, demonstrate slowing of progression rather than restoration of normal nerve function. Moreover, numerous mechanistically promising agents have failed in human trials despite strong pre-clinical results. Collectively, these limitations support the view that established clinical neuropathy remains largely irreversible to date. Overall, the debate underscores that reversibility depends on timing, definitions and outcome thresholds. While early-stage dysfunction is more likely to be modifiable, complete restoration of established clinical DPN (such as full restoration of protective sensation) remains unproven. Future research should prioritise developing validated biomarkers and standardised endpoints in addition to searching for new therapeutics.

Corneal Confocal Microscopy (CCM) is a sensitive tool for assessing small-fiber damage in Diabetic Peripheral Neuropathy (DPN), yet the development of robust, automated deep learning-based diagnostic models is limited by scarce labelled data and fine-grained variability in corneal nerve morphology. Although Artificial Intelligence (AI)-driven foundation generative models excel at natural image synthesis, they often struggle in medical imaging due to limited domain-specific training, compromising the anatomical fidelity required for clinical analysis. To overcome these limitations, we propose a Weight-Decomposed Low-Rank Adaptation (WDLoRA)-based multimodal generative framework for clinically guided CCM image synthesis. WDLoRA is a parameter-efficient fine-tuning (PEFT) mechanism that decouples magnitude and directional weight updates, enabling foundation generative models to independently learn the orientation (nerve topology) and intensity (stromal contrast) required for medical realism. By jointly conditioning on nerve segmentation masks and disease-specific clinical prompts, the model synthesises anatomically coherent images across the DPN spectrum (Control, T1NoDPN, T1DPN). A comprehensive three-pillar evaluation demonstrates that the proposed framework achieves state-of-the-art visual fidelity (Fréchet Inception Distance (FID): 5.18) and structural integrity (Structural Similarity Index Measure (SSIM): 0.630), significantly outperforming GAN and standard diffusion baselines. Crucially, the synthetic images preserve gold-standard clinical biomarkers and are statistically equivalent to real patient data. When used to train automated diagnostic models, the synthetic dataset improves downstream diagnostic accuracy by 2.1

OBJECTIVE:Advanced cardiac autonomic neuropathy (CAN) is associated with increased mortality in people with diabetes. Early identification and reduction of risk factors can limit the progression of CAN. However, the diagnosis of early CAN relies on cardiac autonomic reflex testing (CART's) which is not widely available. We have compared the diagnostic utility of corneal confocal microscopy (CCM) to CART's in diagnosing CAN. RESEARCH DESIGN AND METHODS:Two-hundred and thirty eight individuals with type 1 and type 2 diabetes and thirty seven healthy controls were assessed using CARTs and CCM. RESULTS:There was a progressive and significant reduction in DB-HRV, E:I ratio, 30:15 ratio, corneal nerve fibre density (CNFD), corneal nerve branch density (CNBD) and corneal nerve fibre length (CNFL) with increasing severity of CAN. The receiver operating characteristic (ROC) area under the curve (AUC) and sensitivity/specificity of CCM were comparable to those of CARTs for identifying early and definite CAN. CONCLUSION:CCM is a rapid, non-invasive ophthalmic test which could be used to detect early and established CAN.

OBJECTIVE:Peripheral neuropathies contribute to patient disability but may be diagnosed late or missed altogether due to late referral, limitation of current diagnostic methods and lack of specialized testing facilities. To address this clinical gap, we developed NeuropathAI, an interpretable deep learning-based multiclass classification system for rapid, automated diagnosis and differentiation of 88 patients with diabetic peripheral neuropathy (DPN), chemotherapy-induced peripheral neuropathy (CIPN), chronic inflammatory demyelinating polyneuropathy (CIDP), and human immunodeficiency virus-associated sensory neuropathy (HIV-SN). METHODS:A deep learning-based multiclass system was developed to analyze corneal nerve images. These images were preprocessed to train and validate the proposed model and the diagnostic utility was evaluated from the accuracy, F1-score and area under the curve to derive sensitivity, specificity and precision. RESULTS:NeuropathAI achieved excellent results: AUC-96.75%, sensitivity-83.87%, specificity-95.07%, and demonstrated excellent discrimination for CIDP, CIPN, HIV-SN and DPN with one-vs-all AUC scores of 97%, 93.1%, 99.7% and 96.9%, respectively. Explainability visualization through heatmaps demonstrated that regions of decision making by the model localized to areas with nerve fiber loss, enhancing interpretability. INTERPRETATION:NeuropathAI achieved rapid and accurate diagnosis of four of the most prevalent peripheral neuropathies globally, underscoring the potential of artificial intelligence-driven corneal image analysis for the rapid diagnosis and differentiation of peripheral neuropathies.

BACKGROUND & OBJECTIVES:Hypertriglyceridaemia (HTG) is an independent risk factor for small fibre neuropathy in patients with type 1, type 2 diabetes, and obesity. This study assessed for evidence of small nerve fibre damage and cardiac autonomic dysfunction in individuals with HTG, without diabetes. METHODS:Participants with HTG (n=60) and age and sex-matched controls (n=31) underwent assessment of the lipid profile, neuropathic symptoms and disability, corneal confocal microscopy (CCM) and cardiac autonomic nerve function tests. RESULTS:Triglyceride (TG) concentration was significantly higher [6.0 (3.6-8.7) vs 1.1 (0.7-1.4) mmol/L, p < 0.001], and HDL-C [0.9 (0.7-1.2) vs 1.5 (1.3-1.8) mmol/L, p < 0.001) was lower in the HTG group. The Neuropathy Symptom Profile score [2 (0-5) vs 0, p < 0.001], Neuropathy Disability Score [2 (0-3) vs 0, p < 0.001] and vibration perception threshold [8.9 (5.5-10.5) vs 3.0 (2.0-4.0) volts, p < 0.001] were higher, whilst corneal nerve fibre density (CNFD) [28.9 (5.9) vs 35.1 (7.2) no./mm2, p < 0.001], corneal nerve branch density (CNBD) [50.0 (30.4-66.5) vs 76.1 (56.2-112.5) no./mm2, p < 0.001], corneal nerve fibre length (CNFL) [19.8 (4.8) vs 26.0 (6.2) mm/mm2, p < 0.001], deep breathing heart rate variability (DB-HRV) [18 (13-20) vs 25 (20-30) beats/min, p < 0.001], E-I ratio [1.13 (1.09-1.20) vs 1.25 (1.18-1.31), p < 0.001], valsalva ratio [1.29 (1.18-1.49) vs. 1.47 (1.28-1.59), p = 0.01] and 30-15 ratio [1.24 (1.16-1.36) vs. 1.40 (1.27-1.49), p = 0.002] were lower in the HTG group compared to controls. Serum TG concentration correlated negatively with CCM parameters and cardiac autonomic function tests. CONCLUSION:HTG, independent of diabetes, is associated with signs and symptoms of neuropathy, small nerve fibre damage and cardiac autonomic dysfunction.

Purpose:Familial chylomicronemia syndrome (FCS) is a rare autosomal recessive disorder. This study aimed to analyze the genotype distribution of FCS-causing genes in the United Kingdom. Methods:Data were anonymously collated from 2 genetic testing laboratories providing national genetic diagnosis services for severe hypertriglyceridemia in the United Kingdom. Results:As of December 2023, 880 individuals underwent genetic testing for FCS. The mean (SD) age at the time of genetic testing was 42.5 (15.3) years. After genotyping, 12.9% of the individuals (n = 114) received a genetic diagnosis of FCS. The detection rate of variant-positive multifactorial chylomicronemia syndrome, ie, heterozygous for pathogenic/likely pathogenic (P/LP) variants in 1 of the 5 canonical genes was 11.4% (n = 100).Among 114 genetically proven FCS individuals, 52.6% (n = 60) had biallelic P/LP LPL variants (ie, LPL-FCS), 45.6% (n = 52) had biallelic non-LPL P/LP variants (ie, non-LPL-FCS) and 1.7% (n = 2) individuals were digenic. Among non-LPL-FCS (n = 52), the most common gene implicated was GPIHBP1 (42.3%, n = 22), followed by APOA5 (32.7%, n = 17), LMF1 (13.5%, n = 7) and APOC2 (11.5%, n = 6). Most variant-positive multifactorial chylomicronemia syndrome harbored P/LP variants in LPL (61%) or APOA5 (37%).The geographical distribution of FCS demonstrated regional variability, where the Northwest of England had the highest number of FCS cases per million population. Individuals of European geographic ancestry predominantly had LPL-FCS (60.9%); however, genotype was more diverse in individuals of non-European origin (LPL 47.1%, GPIHBP1 30.9%, APOA5 8.8%, LMF1 7.4%, and APOC2 4.4%). Variants in specific causal genes, GPIHBP1 and LMF1, were predominantly observed in non-European FCS individuals. Conclusion:The genetic architecture of FCS in the United Kingdom is complex, with a substantial proportion affected by non-LPL FCS-causing genes. It also displays a significant regional and ethnic variations.

Diabetic Peripheral Neuropathy (DPN) affects nearly half of diabetes patients, requiring early detection. Corneal Confocal Microscopy (CCM) enables non-invasive diagnosis, but automated methods suffer from inefficient feature extraction, reliance on handcrafted priors, and data limitations. We propose HMSViT, a novel Hierarchical Masked Self-Supervised Vision Transformer (HMSViT) designed for corneal nerve segmentation and DPN diagnosis. Unlike existing methods, HMSViT employs pooling-based hierarchical and dual attention mechanisms with absolute positional encoding, enabling efficient multi-scale feature extraction by capturing fine-grained local details in early layers and integrating global context in deeper layers, all at a lower computational cost. A block-masked self supervised learning framework is designed for the HMSViT that reduces reliance on labelled data, enhancing feature robustness, while a multi-scale decoder is used for segmentation and classification by fusing hierarchical features. Experiments on clinical CCM datasets showed HMSViT achieves state-of-the-art performance, with 61.34

Background: Diabetic neuropathy is a dying-back neuropathy with evidence of greater distal corneal nerve loss at the inferior whorl. Individuals with obesity exhibit neuropathy and central corneal nerve loss. We have assessed whether there was evidence of greater corneal nerve loss at the inferior whorl indicative of a dying back neuropathy in subjects with obesity with and without diabetes. Methods: Fifty-seven individuals with obesity, with diabetes (OBDM+), without diabetes (OBDM-) (n = 30 and n = 27 respectively) and age and sex-matched normal weight controls (n = 21) underwent venous blood sampling and assessment of the neuropathy symptom profile (NSP), neuropathy disability score (NDS), vibration, cold and warm threshold testing, cardiac autonomic function and corneal confocal microscopy (CCM). Results: The Neuropathy Symptom Profile and Neuropathy Disability Score were significantly higher in OBDM- (NSP: p G 0.0001; NDS p G 0.01) and OBDM+ (NSP p G 0.0001; NDS p G 0.001) subjects compared to controls. Vibration perception threshold was significantly higher (p = 0.001) and deep breathing heart rate variability was significantly lower (p = 0.01) in OBDM+ compared to controls. Corneal nerve fibre density (CNFD) [26.8 f 6.22 vs 26.8 f 6.01 vs 35.3 f 7.41, p G 0.00011, branch density [55.4 f 28.2 vs 58.4 f 28.5 vs 88.2 f 31.1, p G 0.0011, fibre length (CNFL) [17.6 f 4.43 vs 19.9 f 5.43 vs 26.7 f 5.31, p G 0.00011, inferior whorl length (IWL) [17.9 f 6.10 vs 18.6 f 7.42 vs 35.3 f 9.70, p G 0.00011 and total nerve fibre length (TNFL) [35.5 f 9.58 vs 38.5 f 11.0 vs 62.0 f 12.3, p G 0.00011 were significantly lower in OBDM+ and OBDM- compared to control but comparable between the two groups of individuals with obesity. The inferior whorl showed a disproportionately greater loss of nerve fibres compared to the central cornea in OBDM+ and OBDM- patients. The mean % difference (between patients and controls) for IWL was 65.4 % and 62.8 % for OBDM- and OBDM+ respectively. This difference was numerically greater than the mean % difference (between patients and controls) for CNFL which was 41.1 % and 29.2 % for OBDM- and OBDM+ respectively. Conclusion: We provide evidence of a distal dying back small fibre damage by demonstrating greater corneal nerve loss at inferior whorl in obesity regardless of diabetes status.

BACKGROUND AND OBJECTIVES:Familial chylomicronemia syndrome (FCS) is a rare autosomal recessive disorder associated with markedly elevated triglyceride concentration and acute pancreatitis, but neuropathic pain and cognitive dysfunction are also increasingly recognized. This study undertook detailed quantification of somatic and autonomic neuropathy in participants with FCS. METHODS:Sixteen individuals with FCS and 16 age and sex-matched controls underwent assessment of the lipid profile, neuropathic symptoms and disability, vibration perception, corneal confocal microscopy (CCM), and cardiac autonomic reflex testing. RESULTS:Age (36.4 [26.5-47.2] vs 36.7 [31.3-46.9] years, P = .7), gender distribution (males 44% [n = 7] vs 50% [n = 8], P = .7), body mass index (23.7 [20.3-27.1] vs 24.7 [22.3-27.2] kg/m2, P = .5), and glycosylated hemoglobin (36.0 [32.5-39.2] vs 36.8 [33.0-38.0] mmol/mol, P = .9) were comparable between participants with FCS and controls. Triglycerides were significantly higher (23.7 [17.4-34.8] vs 1.0 [0.7-1.3] mmol/L, P < .001), whilst low-density lipoprotein cholesterol (0.9 [0.7-1.2] vs 2.7 [2.3-3.1] mmol/L, P < .001) and high-density lipoprotein cholesterol (0.4 [0.3-0.6] vs 1.5 [1.3-1.9] mmol/L, P < .001) were lower in participants with FCS. The Neuropathy Symptom Profile Score (4 [0-14] vs 0, P = .003), Neuropathy Disability Score (1 [0-3.5] vs 0, P = .01), and vibration perception threshold (6.5 [4.7-8.1] vs 3.0 [2.0-3.5] volts, P < .001) were higher, whilst corneal nerve fiber density (30.2 [27.3-32.3] vs 37.0 [32.5-38.5] no./mm2, P < .001), corneal nerve branch density (54.1 [42.0-76.6] vs 100.5 [67.0-147.1] no./mm2, P < .001), corneal nerve fiber length (20.0 [18.3-25.5] vs 27.5 [23.9-34.3] mm/mm2, P < .001), deep breathing heart rate variability (17.0 [14.2-28.0] vs 29.5 [25.2-34.7] beats/min, P = .002), and expiration to inspiration ratio (1.18 [1.14-1.29] vs 1.36 [1.25-1.41], P = .001) were lower in participants with FCS compared to controls. CONCLUSION:FCS is associated with neuropathic symptoms, elevated vibration perception, small nerve fiber damage, and cardiac autonomic dysfunction.

AIMS:Dyslipidaemia contributes to the pathogenesis of diabetic peripheral neuropathy (DPN). While statins improve cardiovascular outcomes in diabetes, their potential neurotoxic effects remain debated. This study examined the impact of statin use on neuropathy in type 1 diabetes mellitus (T1DM). MATERIALS AND METHODS:Participants with T1DM (n = 160) and healthy controls (n = 64) underwent symptom and clinical evaluation of DPN, cardiac autonomic neuropathy (CAN) and corneal confocal microscopy (CCM). T1DM participants were stratified by statin use (non-statin: n = 68; statin treated: n = 92). RESULTS:There were significant differences between the non-statin and statin patients with T1DM and healthy controls for the diabetic neuropathy symptom score (DNS) (0.49 ± 0.14 vs. 0.90 ± 0.13 vs. 0.15 ± 0.13, p < 0.001), neuropathy disability score (NDS) (2.55 ± 0.29 vs. 3.67 ± 0.26 vs. 0.44 ± 0.28, p < 0.001), vibration perception threshold (13.68 ± 1.16 vs. 16.03 ± 1.03 vs. 6.05 ± 1.08, p < 0.001), corneal nerve fibre density (19.61 ± 1.04 vs. 19.02 ± 0.92 vs. 28.48 ± 0.97, p < 0.001), branch density (20.40 ± 2.21 vs. 21.39 ± 1.94 vs. 37.31 ± 2.05, p < 0.001), fibre length (11.97 ± 0.51 vs. 11.51 ± 0.45 vs. 16.55 ± 0.47, p < 0.001), DB-HRV (26.27 ± 1.76 vs. 24.21 ± 1.51 vs. 30.18 ± 1.67, p = 0.033) and 30:15 ratio (1.32 ± 0.04 vs. 1.21 ± 0.03 vs. 1.15 ± 0.07, p = 0.033). Despite the statin group being significantly older (p < 0.001) with a higher BMI (p = 0.001) and longer duration of diabetes (p < 0.001), statin-treated patients showed no significant differences in most neuropathy measures, except DNS (p = 0.04), NDS (p = 0.009) and 30:15 ratio (p = 0.04). CONCLUSIONS:This study demonstrates that individuals with T1DM exhibit neuropathic symptoms and disability, increased vibration perception thresholds, corneal nerve fibre loss and evidence of CAN. However, statin therapy was associated with comparable measures of DPN and CAN, despite statin-treated patients having a longer duration of diabetes and a higher BMI.