〔Abstract〕 Objective To investigate the effects of aerobic exercise intervention at low temperature on renal metabolism and short-term prognosis in diabetic nephropathy. Methods (1) 32 mice with diabetic nephropathy were randomly divided into a model group, a low-temperature group, an exercise group, and a low-temperature exercise group, and another 8 mice were selected as members of the normal group. The exercise group and the low-temperature exercise group were given an 8-week running table exercise intervention at room temperature and low temperature, respectively, while the low-temperature group was only given the low temperature intervention. Fasting blood glucose(FBG), serum creatinine(Scr), blood urea nitrogen(BUN), and 24 h urine protein levels were measured by a biochemical analyzer; renal lesions were observed by the hematoxylin-eosin(HE) method; and nuclear factor E2-related factor 2(NRF2) and heme oxygenase 1(HO-1) levels in renal tissues were measured by Western blot.(2) 194 patients with diabetic nephropathy were prospectively enrolled and divided into the exercise group(1 time/week,n=35), the exercise group(≥2 times/week,n=47), and the control group(0 time/week,n=112) according to the frequency of aerobic exercise intervention at low temperature. Scr, BUN, uric acid(UA), urinary albumin/creatinine ratio(UACR), and estimated glomerular filtration rate(eGFR) levels were compared in the 3 groups. Kaplan-Meier and COX regression were performed to analyze the relationship between aerobic exercise interventions at low temperature and short-term prognosis. Results (1) Compared with the normal group, the levels of FBG, Scr, BUN, 24 h urine protein, and the kidney injury score in the model group increased, while the levels of NRF2 and HO-1 in the model group decreased(P<0.05). Compared with the model group, the levels of FBG, Scr, BUN, 24 h urine protein, and the kidney injury score in the exercise group and the low-temperature exercise group decreased, while the levels of NRF2 and HO-1 in the exercise group and the low-temperature exercise group increased(P<0.05). Compared with the low-temperature group and the exercise group, the levels of FBG, Scr, BUN, 24 h urine protein, and the kidney injury score in the low-temperature exercise group decreased, while the levels of NRF2 and HO-1 in the low-temperature exercise group increased(P<0.05).(2) After the aerobic exercise intervention at low temperature, the levels of Scr, BUN, and UACR in the exercise group(1 time/week) and the exercise group(≥2 times/week) were lower than those in the control group, and eGFR were higher than that in the control group(P<0.05). The cumulative incidence of poor prognosis in the exercise group(≥2 times/week) was 2.13%(1/47), which was lower than in the exercise group[1 time/week, 11.43%(4/35)] and the control group[16.07%(18/112),P<0.05]. COX regression analysis showed that exercise intervention ≥2 times/week(HR=0.123, 95%CI: 0.016-0.925) reduced the risk of poor short-term prognosis in patients with diabetic nephropathy compared with no intervention. Conclusion Aerobic exercise intervention at low temperature can correct renal metabolic abnormalities in diabetic nephropathy, protect renal tissues, and reduce the risk of poor short-term prognosis by a mechanism that may be related to upregulation of NRF2/HO-1 axis expression.