Objective To investigate the association between serum total testosterone (TT ) levels and type 2 diabetes(T2DM )and any gender differences in elderly patients. Methods Based on the Aging Health database built from 2008 to 2012 ,935 elderly individuals over 60 years old with a mean age of 65.8 years receiving physical examinations were included.According to the 1999 WHO criteria for diabetes ,participants were assigned into four groups :a T2DM group(n＝298) ,an impaired fasting glucose(IFG)group(n＝26) ,an impaired glucose tolerance(IGT)group(n＝121) ,and a normal glucose regulation(NGR)group(n ＝ 490).We measured serum TT by ELISA and analyzed the distribution patterns in the groups.Furthermore ,we examined the gender-specific correlation of TT with T2DM and homeostasis model assessment of insulin resistance (HOMA-IR). Results The prevalence of T2DM in the participants was 31.9%(298/935).One-way ANOVA analysis showed that the TT level was higher in the NGR group than in the T2DM and IGT groups (PANOVA＝ 0.001) . Logistic regression analysis indicated a significant protective association between TT and T 2DM in the elderly.Every one unit of increase in the SD of the TT level was accompanied by a 23% reduction in the risk for T2DM (P＝ 0.001).Further gender-stratification analysis suggested that the protective role of TT against T2DM only existed in males (OR＝ 0.55 ,95% CI :0.44-0.68 ;P＜ 0.001).After adjustment for age ,blood pressure ,blood lipids ,and waist circumference ,the protective role of TT against T2DM in males still remained (OR ＝ 0.68 ,95% CI :0.53-0.86 ;P ＝ 0.002 ).Pearson correlation analysis also indicated a significant negative correlation between TT and HOMA-IR in older males(P＝0.002).As for older females ,no significant correlation of TT with T2DM and HOMA-IR was found. Conclusions The serum TT level might be an independent protective factor for T 2DM in older males ,as evidenced by its correlation with improved insulin resistance status ,which is not present in older females.

Objective To investigate the risk factors and appropriate screening methods for diabetic peripheral neuropathy (DPN ). Methods This research is a multicenter ,randomized ,cross-sectional study. Questionnaires ,physical examinations and laboratory tests were performed on 1054 elderly diabetic outpatients at 13 hospitals in urban and rural areas of Beijing. Patients were screened for DPN with the five physical examinations recommended by the Diabetes Branch of the Chinese Medical Association. They were divided into a confirmed DPN group(n=449 ,42.6％ ) ,a suspected DPN group(n=276 ,26.2％ ) ,and a non-DPN group(n=329 ,31.2％ ).Multivariate logistic regression was used to analyze risk factors for DPN.The sensitivity ,specificity ,Youden index ,and area under the receiver operating characteristic (ROC)curve were calculated and used for evaluating each test and their combinations in screening DPN and for choosing the opitmal test combination. Results The differences in age ,family income ,duration of diabetes ,glycated hemoglobin(HbA1c) ,hypoglycemia ,and dyslipidemia among three groups were statistically significant (all P<0.01). The decision whether or not to initiate metformin therapy ,metformin doses and DPN prevalence in patients on long-term metformin therapy showed statistically significant differences among three groups (all P<0.01). The differences in the prevalences of cerebral infarction ,diabetic retinopathy ,and peripheral vascular disease among three groups were statistically significant (all P< 0.05).Multivariate Logistic regression analysis suggested that levels of HbA1c ,hypoglycemia ,LDL cholesterol ,whether or not metformin use , metformin dose ,and duration of metformin use were risk factors for DPN in elderly diabetics (all P<0.05) . The combined methods of ankle reflex ,vibratory sensation and temperature sensation used for screening DPN showed the best results with a sensitivity of 94.1％ ,specificity of 75.3％ ,an area under the ROC curve of 0.847 ,and a Yoden index of 0.6945. Conclusions The prevalence of DPN in elderly diabetic outpatients at third-level referral hospitals in Beijing is high. Poor glycaemic control ,repeated episodes of hypoglycemia , metformin use and its daily dose and duration are risk factors for DPN in elderly diabetics.We should focus on strengthening the DPN screening and management of high-risk population.Symptoms of peripheral neuropathy plus ankle reflex ,vibratory and temperature sensations are simple ,rapid and reliable DPN-screening methods ,and can be promoted in outpatient department and primary hospitals.