Abstract Background Early rehabilitation in the intensive care unit (ICU), including early mobility therapy, is known to improve the clinical outcomes in patients with critically ill. However, the effectiveness of acute-phase cardiovascular rehabilitation (CR) during ICU treatment have not been thoroughly evaluated in patients with cardiovascular disease (CVD). Purpose We aimed to investigate the trends and outcomes of acute-phase CR in the ICU for patients with CVD, including in-hospital and long-term clinical outcomes. Methods We reviewed 1948 consecutive patients with CVD admitted to tertiary academic ICU at a university hospital. Patients were arbitrarily assessed by an ICU team consisting of medical and surgical doctors, nurses and physiotherapists within 24 hours after admission to ICU to discover whether their rehabilitation could be initiated according to the specific clinical trial and statement. As clinical characteristics, disease aetiology, comorbid conditions, and ICU treatment were obtained from an electronic database. We evaluated the probability of return to walking independence and return to home as in-hospital clinical outcomes. All patients were followed for five years and investigated all-cause and cardiovascular events after hospital discharge as long-term clinical outcomes. The associations between the implementation of CR during ICU treatment (ICU-CR) and clinical outcomes were evaluated using propensity score-matched analysis with adjustment for clinical characteristics in all matched patients and various subgroups, including aged >65 years, surgical patients, emergency, and length of ICU stay ≥48 hours. Results Out of studied patients, 1092 patients received ICU-CR, the number of which positively correlated with year-trend (r=0.986, P<0.001). After propensity score matching with adjustment for clinical characteristics including calendar years, 758 patients were included for analysis (pairs of n=379 ICU-CR and non-ICU-CR). The ICU-CR was significantly associated with a higher probability of return to walking independence (incident rate ratio [IRR], 2.04; 95% confidence interval [CI], 1.77–2.36) and return to home (IRR, 1.22 95% CI, 1.05–1.41). These associations were consistently observed in various subgroups regarding CVD conditions (Figure 1). During the median follow-up periods of 2.6 years, all-cause clinical events and cardiovascular events occurred in 289 patients (38.1%) and 153 patients (20.2%), respectively. The ICU-CR showed significantly lower rates of five-year all-cause and cardiovascular events than non-ICU-CR (hazard ratio [95% CI] for all-cause events and cardiovascular events, 0.71 [0.56–0.89] and 0.69 [0.50–0.95], respectively, Figure 2). Conclusions The implementation of acute-phase CR in the ICU increased with year-trend, considered beneficial to improve in-hospital and long-term clinical outcomes in patients with CVD and various subgroups of relatively severe disease conditions. Funding Acknowledgement Type of funding sources: None.