Objective To investigate the value of cardiovascular autonomic reflex test (CART) and heart rate variability (HRV) for the diagnosis of diabetic cardiac autonomic neuropathy (DCAN). Methods A total of 154 hospitalized patients with type 2 diabetes mellitus (T2DM) in Shanghai Traditional Chinese Medicine-Integrated Hospital, Shanghai University of Traditional Chinese Medicine from Feb. to Dec. 2020 were enrolled. All patients underwent CART and 24-h Holter monitoring, and HRV indexes were analyzed. The differences in diagnostic value of DCAN between CART and HRV were analyzed. Results Among the 154 T2DM patients, 101 in the DCAN group and 53 in the non-DCAN group were diagnosed using CART. The incidence of DCAN was 65.58% (101/154). The differences in age, diabetes mellitus duration and incidence of diabetic peripheral neuropathy were statistically significant between DCAN and non-DCAN groups (allP＜0.05), while the differences in gender, hypertension, smoking or drinking habits, body mass index, waist-to-hip ratio, glycated albumin or brain natriuretic peptide were not statistically significant (all P＞0.05). There were significant differences in expiration/inspiration difference, 30/15 ratio, Valsalva test and blood pressure differences after postural change between the 2 groups (all P＜0.05). There were significant differences in HRV indexes (standard deviation of all normal N-N interval ［SDNN］, mean standard deviation of the normal N-N interval every 5 min ［SDANN］, root mean square of successive differences between adjacent N-N intervals ［RMSSD］, percentage of the number of pairs of adjacent N-N intervals differing by more than 50 ms ［pNN50］, total power ［TP］, very low frequency ［VLF］, low frequency ［LF］, and high frequency ［HF］) (all P＜0.05). The incidence of DCAN was 59.74% (92/154) according to the HRV indexes of 24-h Holter. The total coincidence rate was 61.68% between HRV and CART, and Kappa value was 0.184. The sensitivity and specificity of HRV in diagnosing DCAN were 66.34% and 52.83%, respectively. According to the CART score, T2DM patients were divided into 4 grades: non-, sub-clinical, confirmed and severe DCAN. The degree of DCAN was negatively correlated with SDNN, SDANN, RMSSD, pNN50, TP, VLF, LF, and HF (r＝－0.252, －0.234, －0.260, －0.219, －0.315, －0.280, －0.316 and－0.311, all P＜0.01). The correlation analysis of HRV indexes in the DCAN patients diagnosed by CART found that SDNN and SDANN had the strongest correlation (r＝0.977, P＜0.001), suggesting that sympathetic nerve damage was obvious; correlations between LF and VLF (r＝0.761, P＜0.001) and HF (r＝0.829, P＜0.001) were strong, suggesting that sympathetic and parasympathetic nerves were involved in cardiac autonomic damage. CART combined with HRV diagnosed 134 cases in the DCAN group and 20 cases in the non-DCAN group, and the incidence of DCAN was 87.01% (134/154). The sensitivity and specificity of CART combined with HRV in diagnosing DCAN were 100.00% and 37.74%, respectively. Conclusion CART combined with HRV has a higher detection rate in the diagnosis of DCAN. HRV analysis can be further performed on T2DM patients with low CART score at the early stage of DCAN to reduce the incidence of missed diagnosis.