Background: Bioimpedance analysis (BIA) has been reported as helpful in identifying hypervolemia. Observation data showed that hypervolemic maintenance hemodialysis (MHD) patients identified using BIA methods have higher mortality risk. However, it is not known if BIA-guided fluid management can improve MHD patients' survival. The objectives of the BOCOMO study are to evaluate the outcome of BIA guided fluid management compared with standard care. Methods: This is a multicenter, prospective, randomized, controlled trial. More than 1300 participants from 16 clinical sites will be included in the study. The enrolment period will last 6 months, and minimum length of follow-up will be 36 months. MHD patients aged between 18 years and 80 years who have been on MHD for at least 3 months and meet eligibility criteria will be invited to participate in the study. Participants will be randomized to BIA arm or control arm in a 1: 1 ratio. A portable whole body bioimpedance spectroscopy device (BCM-Fresenius Medical Care D GmbH) will be used for BIA measurement at baseline for both arms of the study. In the BIA arm, additional BCM measurements will be performed every 2 months. The primary intent-to-treat analysis will compare outcomes for a composite endpoint of death, acute myocardial infarction, stroke or incident peripheral arterial occlusive disease between groups. Secondary endpoints will include left ventricular wall thickness, blood pressure, medications, and incidence and length of hospitalization. Discussions: Previous results regarding the benefit of strict fluid control are conflicting due to small sample sizes and unstable dry weight estimating methods. To our knowledge this is the first large-scale, multicentre, prospective, randomized controlled trial to assess whether BIS-guided volume management improves outcomes of MHD patients. The endpoints of the BOCOMO study are of utmost importance to health care providers. In order to obtain that aim, the study was designed with very careful important considerations related to the endpoints, sample size, inclusion criteria, exclusion criteria and so on. For example, annual mortality of Beijing MHD patients was around 10%. To reach statistical significance, the sample size will be very large. By using composite endpoint, the sample size becomes reasonable and feasible. Limiting inclusion to patients with urine volume less than 800 ml/day the day before dialysis session will limit confounding due to residual renal function effects on the measured parameters. Patients who had received BIS measurement within 3 months prior to enrolment are excluded as data from such measurements might lead to protocol violation. Although not all patients enrolled will be incident patients, we will record the vintage of dialysis in the multivariable analysis.