A clinical screening score for diabetic polyneuropathy: KORA F4 and AusDiab Studies

AimsSince screening for distal sensorimotor polyneuropathy (DSPN) in individuals with diabetes is being underused, our aim was to develop a clinical screening score for identifying individuals with DSPN.MethodsAll participants with type 2 diabetes and aged 61-82 years from the German population-based KORA F4 Study (n = 177) and the Australian population-based AusDiab Study (n = 244) were combined into one study sample. Risk indicators of DSPN were identified and entered into a stepwise model-selection procedure, constructing two consecutive scores with increasing complexity (a base and clinical model).ResultsThe prevalence of DSPN was 18.2% (95% confidence interval (CI): 14.7-22.3). The base model comprised age (years), height (cm), weight (kg), pain or discomfort in the feet and/or legs (yes/no), and duration of diabetes (years), yielding an area under the receiver operating characteristics curve (AUC) of 0.80 (95% CI: 0.76-0.85). The clinical model additionally included diastolic blood pressure (mmHg) and serum creatinine levels (mmol/l). The AUC increased only marginally to 0.82 (0.77-0.87) (p for AUC comparison = 0.188). The internal validation of the scores produced similar AUCs.ConclusionsThe screening scores developed in this study are a simple tool to differentiate between a high and low likelihood of having DSPN among individuals with type 2 diabetes.