Background/Aims: Traditional cardiovascular (CV) risk factors (RFs) and their management targets may not be applicable to specific medical subpopulations, particularly dialysis patients. This study aimed to evaluate the dose-response association between measurements of RFs, cardiovascular mortality, and potential metabolic targets among Chinese patients initializing peritoneal dialysis (PD). Methods: Risk-set sampling was applied to two population based 1: 10 case-control studies of incident PD patients, matched by age, sex and the year of initialisation of PD: a main sample (204 cases and 2,040 controls) and a replication sample (81 cases and 810 controls). The dose-response association between continuous measurements of CV RFs (blood pressure, fasting glucose, body mass index, total cholesterol, phosphate and ejection fraction) at baseline and the 2-year CV mortality were analyzed using conditional Logistic regression. The final threshold was chosen based upon a significant break in the regression coefficients and achievement of the minimum Bayesian information criterion (BIC). Results: A linear relationship was identified between fasting glucose and CV mortality. Non-linear associations between other measurements and CV mortality suggested potential metabolic treatment intensification thresholds as < 145/92mmHg for blood pressure, < 1.70mmol/L for phosphate, 24 kg/m2 for body mass index, 4.6mmol/L for total cholesterol, and > 60% for ejection fraction respectively. Conclusion: Our findings highlight the potential importance of more intensive glucose management, anti-hypertensive treatment and dietary management among PD patients. We recommend that the clinical relevance of these epidemiological associations be tested using randomized controlled trials of multifaceted interventions.