[Joint diagnostic value of four temperature sensation tests in elderly patients with type 2 diabetic peripheral neuropathy].

Objective: To explore the joint diagnostic value of four temperature sensation tests in elderly patients with type 2 diabetic peripheral neuropathy.

Methods: Thermal sensory analyzer-II were applied to measure cool sensation (CS), warm sensation (WS), cold pain sensation (CP)and heat pain sensation (HP) of 308 elderly patients with type 2 diabetes. Logistic regression model was adopted to create the new variable Temp4 from four temperature sensation tests to diagnose type 2 diabetic peripheral neuropathy. The ROC curve analysis was used to determine the best cut-off points of the four temperature sensation and Temp4, and the diagnostic value of it was evaluated.

Results: The means of temperature sensation tests of the diabetic peripheral neuropathy (DPN) group were significantly different from those of the non-DPN group (P < 0.05). According to the current reference intervals of the four temperature sensation tests to diagnose the DPN, the sensitivity of WS test was the highest, and the value was 0.710; but the specificity, positive predictive value, negative predictive value, Youden index, diagnostic accuracy and Kappa value of cold sensation test were the highest, and the values were 0.842, 0.746, 0.799, 0.528, 77.92% and 0.535, respectively; the Kappa values of the other three temperature sensation tests were all greater than 0.4 (P < 0.05). The area under the ROC curve of the new variable Temp4 was 0.93 (95% CI 0.91-0.96), and was larger than the four temperature sensation tests (P < 0.05). The sensitivity, specificity, Youden index and diagnostic accuracy of Temp4 were 0.823, 0.897, 0.719 and 86.69%, respectively. The new best cut-off points of the CS test, WS test, CP test, HP test and Temp4 was 27.5 degrees C, 34.7 degrees C, 20.5 degrees C, 43.5 degrees C and 0.416, respectively.

Conclusion: The results of the four temperature sensation quantitative tests were in good agreementand could be applied to diagnose DPN; the new variable Temp4 could be used for diagnosis of DPN with a higher diagnostic accuracy.