[Clinical effect of acupuncture combined with medication on diabetic peripheral neuropathy and its influence on serum lipid metabolism].

OBJECTIVE: To observe the clinical effect on diabetic peripheral neuropathy (DPN) treated with acupuncture combined with medication and explore its effect mechanism. METHODS: Sixty-two patients of DPN were randomly divided into a combined therapy group (31 cases) and a medication group (31 cases, 2 cases dropped out); besides, 20 healthy subjects were recruited as a normal group. On the base of routine intervention, in the medication group, thioctic acid capsules were administrated orally, 0.2 g each time, 3 times a day. In the combined therapy group, besides the medication as the medication group, acupuncture was performed on bilateral Quchi (LI 11), Waiguan (TE 5), Hegu (LI 4), Tianshu (ST 25), Zusanli (ST 36), Sanyinjiao (SP 6) and Taichong (LR 3) and the needles were retained for 30 min, acupuncture was delivered once daily, 6 times a week. The duration of treatment was 4 weeks in the two groups. The score of Toronto clinical scoring system (TCSS), the nerve conduction velocity of median nerve (MN) and common peroneal nerve (CPN) were observed before and after treatment in the two intervention groups; and the serum lipid metabolism was detected before and after treatment in the two intervention groups and the normal group. RESULTS: Compared with that before treatment, the scores of TCSS were reduced in the combined therapy group and the medication group (P<0.05) after treatment, and the score decrease in the combined therapy group was larger than that of the medication group (P<0.001). The motor nerve conduction velocity and the sensory nerve conductive velocity of MN and CPN after treatment all increased in the combined therapy group and the medication group compared with those before treatment (P<0.05), and the improvements in the combined therapy group were larger than those of the medication group (P<0.001). Before treatment DPN patients had 365 differential lipid metabolites, including sphingosine (SPH, d18:0), involved in the inositol phosphate metabolism, compared with the subjects of the normal group. There were 103 differential lipid metabolites in the medication group before and after treatment, including lysophosphatidyl ethanolamine (LPE, 18:1/0:0), participated in glycerophospholipid metabolism. In the combined therapy group, before and after treatment, there were 99 differential lipid metabolites, including lysophosphatidylcholine (LPC, 18:0/0:0), participated in the neuroactive ligand-receptor interaction. Acupuncture greatly affected 50 lipid metabolites such as lysophosphatidic acid (LPA, 0:0/22:6), LPA(0:0/18:2) and LPC(O-18:0), which was mainly involved in glycerophospholipid metabolism. CONCLUSION: Acupuncture combined with medication ameliorates the symptoms and the nerve conduction velocity in DPN patients, which may be related to the regulation of serum lipid metabolism.

Study on risk factors of diabetic peripheral neuropathy and establishment of a prediction model by machine learning.

BACKGROUND: Diabetic peripheral neuropathy (DPN) is a common complication of diabetes. Predicting the risk of developing DPN is important for clinical decision-making and designing clinical trials. METHODS: We retrospectively reviewed the data of 1278 patients with diabetes treated in two central hospitals from 2020 to 2022. The data included medical history, physical examination, and biochemical index test results. After feature selection and data balancing, the cohort was divided into training and internal validation datasets at a 7:3 ratio. Training was made in logistic regression, k-nearest neighbor, decision tree, naive bayes, random forest, and extreme gradient boosting (XGBoost) based on machine learning. The k-fold cross-validation was used for model assessment, and the accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (AUC) were adopted to validate the models' discrimination and clinical practicality. The SHapley Additive exPlanation (SHAP) was used to interpret the best-performing model. RESULTS: The XGBoost model outperformed other models, which had an accuracy of 0·746, precision of 0·765, recall of 0·711, F1-score of 0·736, and AUC of 0·813. The SHAP results indicated that age, disease duration, glycated hemoglobin, insulin resistance index, 24-h urine protein quantification, and urine protein concentration were risk factors for DPN, while the ratio between 2-h postprandial C-peptide and fasting C-peptide(C2/C0), total cholesterol, activated partial thromboplastin time, and creatinine were protective factors. CONCLUSIONS: The machine learning approach helped established a DPN risk prediction model with good performance. The model identified the factors most closely related to DPN.