Kanazawa N1, Iijima H2, Fushimi K3
1National Hospital Organization Headquarters, Clinical Research Center, Tokyo, Japan, 2Hokkaido University, Graduate School of Medicine, Department of Biostatistics, Hokkaido, Japan, 3Tokyo Medical and Dental University Graduate School, Department of Health Policy and Informatics, Tokyo, Japan
Background: Cardiac rehabilitation (CR) is strongly recommended for patients with acute myocardial infarction (AMI) because of its effectiveness in improving clinical outcomes. CR is offered to outpatients several weeks after discharge. In Japan, CR is started early after admission. Although the safety of early CR has already been reported, its effectiveness is unclear.
Purpose: To investigate the effectiveness of early CR for AMI on a nationwide scale.
Methods:
Data source: Japanese administrative database (The Diagnosis Procedure Combination; the Japanese national case-mix system database)
Study population: Patients aged 18 years and older hospitalized due to AMI and who underwent percutaneous coronary intervention (PCI) between April 1 2012 and March 31 2014 were enrolled in this study. The target hospitals were limited to hospitals authorized to provide CR in accordance with Japanese standards.
Treatment: The treatment group was defined as patients who received early CR. A patient was included in the treatment group when he or she participated in a CR program for at least 1 session within 30 days from admission.
Endpoints: We evaluated revascularization, all-cause re-hospitalization, and cardiac re-hospitalization as primary endpoints. All-cause mortality and recurrence of AMI were evaluated as secondary endpoints.
Follow-up: For follow-up, we used all records, including inpatient and outpatient records, which were generated after index admissions, and evaluated the status of usage of medical services after admissions. As the database was composed of hospital-based data, we limited the cohort to patients who regularly visited the hospitals where they underwent PCI in order to improve the precision of follow-up.
Statistics: To examine the relationship between early CR and outcomes, we constructed a Kaplan-Meier survival curve, and performed a log-rank test for matched pairs based on propensity scores. To confirm the robustness of results, cox proportional hazard model was applied to the whole cohort with an estimated adjusted hazard ratio.
Results: Eventually, 19,064 records of patients with AMI were extracted from the database. Men comprised 79.1% of the cohort. The average age was 66.3 years (SD±12.3). The percentage who attended CR based on the study population was 65.8 % (n=12,541). Based on propensity score, 2,494 matched-pairs were generated. As result of the Kaplan-Meier method and log-rank test for these matched-pairs, the risk of revascularization in CR users was significantly lower than that in non-CR users (p 0.001). Similarly, the risk of all-cause re-hospitalization and cardiac re-hospitalization were also lower in CR users than that in non-CR users (p 0.001). In addition, as a result of cox regression analysis, the adjusted hazard ratios for each outcome were 0.66, 0.74, and 0.81, respectively.
Conclusion(s): It was shown that early CR for patients with AMI reduced the risk of revascularization, all-cause re-hospitalization, and cardiac re-hospitalization.
Implications: This study showed that early CR could reduce the consumption of medical resources after AMI and also reduce physical and economic burdens on patients. Therefore, CR should be started earlier, even in countries other than Japan.
Keywords: Cardiac rehabilitation, myocardial infarction, secondary prevention
Funding acknowledgements: This work is supported by JSPS KAKENHI Grant Number JP16K21697.
Purpose: To investigate the effectiveness of early CR for AMI on a nationwide scale.
Methods:
Data source: Japanese administrative database (The Diagnosis Procedure Combination; the Japanese national case-mix system database)
Study population: Patients aged 18 years and older hospitalized due to AMI and who underwent percutaneous coronary intervention (PCI) between April 1 2012 and March 31 2014 were enrolled in this study. The target hospitals were limited to hospitals authorized to provide CR in accordance with Japanese standards.
Treatment: The treatment group was defined as patients who received early CR. A patient was included in the treatment group when he or she participated in a CR program for at least 1 session within 30 days from admission.
Endpoints: We evaluated revascularization, all-cause re-hospitalization, and cardiac re-hospitalization as primary endpoints. All-cause mortality and recurrence of AMI were evaluated as secondary endpoints.
Follow-up: For follow-up, we used all records, including inpatient and outpatient records, which were generated after index admissions, and evaluated the status of usage of medical services after admissions. As the database was composed of hospital-based data, we limited the cohort to patients who regularly visited the hospitals where they underwent PCI in order to improve the precision of follow-up.
Statistics: To examine the relationship between early CR and outcomes, we constructed a Kaplan-Meier survival curve, and performed a log-rank test for matched pairs based on propensity scores. To confirm the robustness of results, cox proportional hazard model was applied to the whole cohort with an estimated adjusted hazard ratio.
Results: Eventually, 19,064 records of patients with AMI were extracted from the database. Men comprised 79.1% of the cohort. The average age was 66.3 years (SD±12.3). The percentage who attended CR based on the study population was 65.8 % (n=12,541). Based on propensity score, 2,494 matched-pairs were generated. As result of the Kaplan-Meier method and log-rank test for these matched-pairs, the risk of revascularization in CR users was significantly lower than that in non-CR users (p 0.001). Similarly, the risk of all-cause re-hospitalization and cardiac re-hospitalization were also lower in CR users than that in non-CR users (p 0.001). In addition, as a result of cox regression analysis, the adjusted hazard ratios for each outcome were 0.66, 0.74, and 0.81, respectively.
Conclusion(s): It was shown that early CR for patients with AMI reduced the risk of revascularization, all-cause re-hospitalization, and cardiac re-hospitalization.
Implications: This study showed that early CR could reduce the consumption of medical resources after AMI and also reduce physical and economic burdens on patients. Therefore, CR should be started earlier, even in countries other than Japan.
Keywords: Cardiac rehabilitation, myocardial infarction, secondary prevention
Funding acknowledgements: This work is supported by JSPS KAKENHI Grant Number JP16K21697.
Topic: Cardiorespiratory
Ethics approval required: Yes
Institution: National Hospital Organization
Ethics committee: Clinical Research Center Ethics Review Committee
Ethics number: H29-1106007
All authors, affiliations and abstracts have been published as submitted.
