Recherche en endocrinologie et diabète

Frequency and Associations of Diabetic Peripheral Sensory Neuropathy: an EMR- Based Retrospective analysis

Khushroo Minhas*, Aziz Fatima, Saira Burney, Khadija Irfan Khawaja, Zobia Zafar, Arsalan Nawaz

The magnitude of burden of diabetic peripheral neuropathy in south-asian population is undetermined, there being a paucity of large-scale studies. Electronic medical records (EMR) of patients offer a promising avenue by which data analysis can be used for clinical research. Objectives: To determine i) the frequency of peripheral sensory neuropathy and ii) its association with various factors using Electronic Medical Record (EMR) database of diabetic patients. Methods: This was a retrospective cross-sectional study conducted at Diabetes Management Centre (DMC), Services hospital, Lahore which undertook a comprehensive review of electronic medical record (EMR) from first visit of 12,485 diabetic patients, over a three-year period. The frequency of peripheral neuropathy was derived from database analysis of specific signs and symptoms including bilaterally symmetrical symptoms of peripheral numbness or paraesthesia, burning / pins and needles sensation. Positive examination included insensitivity to Semmes-Weinstein monofilament (SWM), pinprick sensation, absence of ankle reflexes and vibration perception threshold using biothesiometer. Data was analyzed in SPSS v.25.Chi square and logistic regression analysis was done for association of DPN with body mass index, waist circumference, blood pressure, type and duration of diabetes, age, gender, HbA1c, LDL levels, Total cholesterol levels and eGFR. Results: The frequency of DPN was 84.6% in a sample size of 12485. More females(61.1%)had neuropathy as compared to males(38.9%), similarly greater number of type2 diabetics(91.6%) were found to be suffering from DPN as compared to type1.The mean±SD age of population was 50.85 ±11.13, duration of diabetes 6.70 years ±6.70, BMI 28.26±4.75, HbA1c 9.25 % ±2.31s . Logistic regression analysis predicted 81% of the model correctly. The factors with a significant association with peripheral sensory neuropathy by applying Chi square test included BMI, waist circumference, duration of diabetes, nephropathy, poor glycemic control and hypertension, while serum LDL did not show a positive relation with neuropathy.