MACHINE LEARNING ANALYSIS OF THE UK BIOBANK REVEALS PROGNOSTIC AND DIAGNOSTIC IMMUNE BIOMARKERS FOR POLYNEUROPATHY AND NEUROPATHIC PAIN IN DIABETES

Diminished levels of insulin-like growth factor-1 may be a risk factor for peripheral neuropathy in type 2 diabetes patients

AIMS/INTRODUCTION

To investigate risk factors for diabetic peripheral neuropathy (DPN) and to explore the connection between insulin-like growth factor-1 (IGF-1) and DPN in individuals with type 2 diabetes.

MATERIALS AND METHODS

A total of 790 patients with type 2 diabetes participated in a cross-sectional study, divided into two groups: those with DPN (DPN) and those without DPN (non-DPN). Blood samples were taken to measure IGF-1 levels and other biochemical markers. Participants underwent nerve conduction studies and quantitative sensory testing.

RESULTS

Patients with DPN exhibited significantly lower levels of IGF-1 compared with non-DPN patients (P < 0.001). IGF-1 was positively correlated with the average amplitude of both motor (P < 0.05) and sensory nerves (P < 0.05), but negatively correlated with the vibration perception threshold (P < 0.05). No significant difference was observed between IGF-1 and nerve conduction velocity (P > 0.05), or the temperature detection threshold (P > 0.05). Multivariate regression analysis identified diabetes duration, HbA1c, and the low levels of IGF-1 as independent risk factors (P < 0.001). Receiver operating characteristic analysis determined that at 8 years duration of diabetes, 8.5% (69.4 mmol/mol) HbA1c and 120 ng/mL IGF-1, the optimal cut-off points, indicated DPN (P < 0.001).

CONCLUSIONS

A reduction of IGF-1 in patients with DPN suggests a potential protective role against axon injury in large fiber nerves of type 2 diabetes patients.

Interpretable machine learning models for detecting peripheral neuropathy and lower extremity arterial disease in diabetics: an analysis of critical shared and unique risk factors

BACKGROUND

Diabetic peripheral neuropathy (DPN) and lower extremity arterial disease (LEAD) are significant contributors to diabetic foot ulcers (DFUs), which severely affect patients' quality of life. This study aimed to develop machine learning (ML) predictive models for DPN and LEAD and to identify both shared and distinct risk factors.

METHODS

This retrospective study included 479 diabetic inpatients, of whom 215 were diagnosed with DPN and 69 with LEAD. Clinical data and laboratory results were collected for each patient. Feature selection was performed using three methods: mutual information (MI), random forest recursive feature elimination (RF-RFE), and the Boruta algorithm to identify the most important features. Predictive models were developed using logistic regression (LR), random forest (RF), and eXtreme Gradient Boosting (XGBoost), with particle swarm optimization (PSO) used to optimize their hyperparameters. The SHapley Additive exPlanation (SHAP) method was applied to determine the importance of risk factors in the top-performing models.

RESULTS

For diagnosing DPN, the XGBoost model was most effective, achieving a recall of 83.7%, specificity of 86.8%, accuracy of 85.4%, and an F1 score of 83.7%. On the other hand, the RF model excelled in diagnosing LEAD, with a recall of 85.7%, specificity of 92.9%, accuracy of 91.9%, and an F1 score of 82.8%. SHAP analysis revealed top five critical risk factors shared by DPN and LEAD, including increased urinary albumin-to-creatinine ratio (UACR), glycosylated hemoglobin (HbA1c), serum creatinine (Scr), older age, and carotid stenosis. Additionally, distinct risk factors were pinpointed: decreased serum albumin and lower lymphocyte count were linked to DPN, while elevated neutrophil-to-lymphocyte ratio (NLR) and higher D-dimer levels were associated with LEAD.

CONCLUSIONS

This study demonstrated the effectiveness of ML models in predicting DPN and LEAD in diabetic patients and identified significant risk factors. Focusing on shared risk factors may greatly reduce the prevalence of both conditions, thereby mitigating the risk of developing DFUs.

A Prognostic Model Incorporating Relevant Peripheral Blood Inflammation Indicator to Predict Postherpetic Neuralgia in Patients with Acute Herpes Zoster

Objective

To determine the risk of postherpetic neuralgia (PHN) in patients with acute herpes zoster (HZ), this study developed and validated a novel clinical prediction model by incorporating a relevant peripheral blood inflammation indicator.

Methods

Between January 2019 and June 2023, 209 patients with acute HZ were categorized into the PHN group (n = 62) and the non-PHN group (n = 147). Univariate and multivariate logistic regression analyses were conducted to identify risk factors serving as independent predictors of PHN development. Subsequently, a nomogram prediction model was established, and the discriminative ability and calibration were evaluated using the receiver operating characteristic curve, calibration plots, and decision curve analysis (DCA). The nomogram model was internally verified through the bootstrap test method.

Results

According to univariate logistic regression analyses, five variables, namely age, hypertension, acute phase Numeric Rating Scale (NRS-11) score, platelet-to-lymphocyte ratio (PLR), and systemic immune inflammation index, were significantly associated with PHN development. Multifactorial analysis further unveiled that age (odds ratio (OR) [95% confidence interval (CI)]: 2.309 [1.163-4.660]), acute phase NRS-11 score (OR [95% CI]: 2.837 [1.294-6.275]), and PLR (OR [95% CI]: 1.015 [1.010-1.022]) were independent risk factors for PHN. These three predictors were integrated to establish the prediction model and construct the nomogram. The area under the receiver operating characteristic curve (AUC) for predicting the PHN risk was 0.787, and the AUC of internal validation determined using the bootstrap method was 0.776. The DCA and calibration curve also indicated that the predictive performance of the nomogram model was commendable.

Conclusion

In this study, a risk prediction model was developed and validated to accurately forecast the probability of PHN after HZ, thereby demonstrating favorable discrimination, calibration, and clinical applicability.

Increased Systemic Immune-Inflammation Index Was Associated with Type 2 Diabetic Peripheral Neuropathy: A Cross-Sectional Study in the Chinese Population

Background

Systemic immune-inflammation index (SII), a novel inflammatory marker, has been demonstrated to be associated with type 2 diabetes mellitus (T2DM) and its vascular complications, however, the relation between SII and diabetic peripheral neuropathy (DPN) has been never reported. We aimed to explore whether SII is associated with DPN in Chinese population.

Methods

A cross-sectional study was conducted among 1460 hospitalized patients with T2DM. SII was calculated as the platelet count × neutrophil count/lymphocyte count, and its possible association with DPN was investigated by correlation and multivariate logistic regression analysis, and subgroup analyses.

Results

Patients with higher SII quartiles had higher vibration perception threshold and prevalence of DPN (all P<0.01), and SII was independently positively associated with the prevalence of DPN (P<0.01). Multivariate logistic regression analysis showed that the risk of prevalence of DPN increased progressively across SII quartiles (P for trend <0.01), and participants in the highest quartile of SII was at a significantly increased risk of prevalent DPN compared to those in the lowest quartile after adjustment for potential confounding factors (odds rate: 1.211, 95% confidence intervals 1.045-1.404, P<0.05). Stratified analysis revealed positive associations of SII quartiles with risk of prevalent DPN only in men, people less than 65 years old, with body mass index <24 kg/m2, duration of diabetes >5 years, hypertension, dyslipidaemia, poor glycaemic control, and estimated glomerular filtration rate <90 mL/min/1.73 m2 (P for trend <0.01 or P for trend <0.05). The receiver operating characteristic curve analysis revealed that the optimal cut-off point of SII for predicting DPN was 617.67 in patients with T2DM, with a sensitivity of 45.3% and a specificity of 73%.

Conclusion

The present study showed that higher SII is independently associated with increased risk of DPN, and SII might serve as a new risk biomarker for DPN in Chinese population.