Objective: We analyzed 1,529 patients with CKD stages 1 to 5(pre-dialysis). The predictor was serum tCO2 level measured at baseline. tCO2 was analyzed either as a continuous or a categorical variable of 3 levels (low<22, normal 22-29.9, high≥30mmol/L). The outcome was osteoporosis at baseline. Osteoporosis was defined as a Z score≤-2.0 for those under 50 years of age and a T-score≤-2.5 for over 50 years of age based on the lowest BMD at three measurement points. Among the patients who measured BMD both at baseline and after 4 years(N=885), the reduction of BMD over four years in lumbar vertebra L1, femur neck(FN), and total hip(TH)(defined as BMD at 4 years – BMD at baseline<0) was defined as a longitudinal outcome. The associations between continuous level and low group of tCO2 and outcomes were assessed using multivariable logistic regression with adjustment for the age and sex.  Methods: The low tCO2 as a continuous variable (adjusted odds ratio[OR] 0.92, confidence interval[CI] 0.88-0.97, P<0.001) and the low tCO2 group were related to osteoporosis (adjusted OR 1.95, CI 1.02-3.87, P=0.05). For subgroup analysis according to CKD stage, this tendency remained significant in CKD stage 3a to 5 as continuous tCO2(adjusted OR 0.93, CI 0.87-0.99, P=0.046) and the lowest group of tCO2(adjusted OR 1.24-9.64, P=0.026), but not in the early CKD patients. There was a significant relationship between the low tCO2 group and TH BMD reduction at 4 years(adjusted OR 1.18, CI 1.02-1.37, P=0.025), but not in the L1 and FN BMD reduction. Results: In CKD patients, low serum tCO2 was associated with osteoporosis, and decrease of BMD at 4 years might be related to low serum tCO2. Conclusions: Objective: Metabolic acidosis leads to lower bone mineral density(BMD), which is a risk of osteoporosis. We evaluated the relationship between serum total CO2(tCO2) and BMD in the chronic kidney disease(CKD) population.  Methods: We analyzed 1,529 patients with CKD stages 1 to 5(pre-dialysis). The predictor was serum tCO2 level measured at baseline. tCO2 was analyzed either as a continuous or a categorical variable of 3 levels (low<22, normal 22-29.9, high≥30mmol/L). The outcome was osteoporosis at baseline. Osteoporosis was defined as a Z score≤-2.0 for those under 50 years of age and a T-score≤-2.5 for over 50 years of age based on the lowest BMD at three measurement points. Among the patients who measured BMD both at baseline and after 4 years(N=885), the reduction of BMD over four years in lumbar vertebra L1, femur neck(FN), and total hip(TH)(defined as BMD at 4 years – BMD at baseline<0) was defined as a longitudinal outcome. The associations between continuous level and low group of tCO2 and outcomes were assessed using multivariable logistic regression with adjustment for the age and sex.  Results: The low tCO2 as a continuous variable (adjusted odds ratio[OR] 0.92, confidence interval[CI] 0.88-0.97, P<0.001) and the low tCO2 group were related to osteoporosis (adjusted OR 1.95, CI 1.02-3.87, P=0.05). For subgroup analysis according to CKD stage, this tendency remained significant in CKD stage 3a to 5 as continuous tCO2(adjusted OR 0.93, CI 0.87-0.99, P=0.046) and the lowest group of tCO2(adjusted OR 1.24-9.64, P=0.026), but not in the early CKD patients. There was a significant relationship between the low tCO2 group and TH BMD reduction at 4 years(adjusted OR 1.18, CI 1.02-1.37, P=0.025), but not in the L1 and FN BMD reduction. Conclusions: In CKD patients, low serum tCO2 was associated with osteoporosis, and decrease of BMD at 4 years might be related to low serum tCO2.