Study population and design
This retrospective cohort study obtained data from the Korean National Health Insurance Service (KNHIS) database, which provides comprehensive coverage to approximately 97% of the Korean population. The remaining 3% of individuals who are unable to afford national insurance are covered by the Medical Aid Program44. Therefore, the KNHIS database effectively represents the entirety of the South Korean population and contains the national records of all covered inpatient and outpatient visits, prescriptions, and procedures. The KNHIS data includes modules on insurance eligibility and medical treatment. The insurance eligibility module contains the information on age, sex, residential area, and income level. The medical treatment module contains information related to treatments, including conditions and prescriptions45. This study was conducted with the approval of the KNHIS official review committee (protocol number: NHIS-2022-1-240), and access to the data was granted per approval.
We aimed to evaluate the long-term consequences of anemia in relation to cardiovascular events and mortality after hospital discharge. For this, we included adult patients > 18 years who were admitted to the ICU and received CRRT for ≥ 3 days between January 1, 2010 and December 31, 2019 (N = 119,421). We excluded participants with preexisting end-stage kidney disease (ESKD) (N = 19,605); history of acute MI, stroke, heart failure, or revascularization (N = 43,039); and history of other cardiovascular diseases (N = 28,479). We further excluded patients who received major surgery, endoscopic hemostasis, or vascular embolization (N = 52,946) and those who died during hospitalization (N = 19,294). Thus, a total of 10,923 participants were included in the final analyses.
Ethics approval and consent to participate
The study was approved by the Institutional Review Board of Samsung Medical Center in compliance with the Declaration of Helsinki (IRB No. 2021-01-052, December 31, 2020, title: “Investigation of Evidence-Based Optimal Management Strategies for Continuous Renal Replacement Therapy”). The study procedures were followed in accordance with the ethical standards of the responsible institutional committee on human experimentation and with the Helsinki Declaration of 1975. The Institutional Review Board of Samsung Medical Center waived the requirement for informed consent due to the retrospective nature of the study and deidentified data collection13.
Exposure
Anemia was defined as the requirement for RBC transfusion or the prescription of ESAs. ESA prescription was defined as having at least one prescription of darbepoetin or methoxy polyethylene glycol-epoetin, or a minimum of three prescriptions of short-acting erythropoietin during hospitalization.
Study variables
The KNHIS records for inpatient and outpatient visits, prescriptions, and procedures were coded using the International Classification of Diseases (ICD), 10th Revision46. These data are highly reliable due to routine audits by the KNHIS and have been used in numerous peer-reviewed publications47,48. In a validation study, the diagnostic accuracy for MI in KNHIS data was reported to be 93%49.
We collected claim codes that encompassed information regarding comorbidities, management procedures during ICU admission, prescriptions, and demographic characteristics. Comorbidities, such as chronic liver disease, diabetes mellitus, chronic kidney disease, and cancer diagnosed within the year preceding hospitalization, were summarized using the Charlson index50,51. Additionally, we included in our analysis hypertension (ICD-10 codes: I10, I13, I15) and septic shock (defined as ≥ 2 days of vasopressor and ≥ 1 week of antibiotics administration), despite not being covered by the Charlson index. Management procedures included CRRT (Korean NHI procedure codes O7031-O7035, O7051-O7055) and mechanical ventilation (Korean NHI procedure codes M5857, M5858, or M5860).
Outcomes
The primary outcomes were cardiovascular events and all-cause mortality after hospital discharge. Cardiovascular events were defined as hospitalization for heart failure (ICD-10 codes I110, I130, I132, I255, I420, I425-I429, I43, I50, and I971), acute MI (ICD-10 codes I21-I23, and I252), the presence of coronary revascularization codes (percutaneous coronary intervention, Korean NHI codes M6551-M6554, M6561-M6567, and M6571-M6572; coronary artery bypass graft surgery [CABG], O1640-O1642, O1647-O1649, OA640-OA642, and OA647-OA649), or stroke (ICD-10 codes I63, I64, and G45). The data regarding all-cause mortality was obtained from death certificates collected by Statistics Korea, part of the Ministry of Strategy and Finance of South Korea47.
Subgroup analysis
We conducted subgroup analysis to evaluate the association of anemia with post-discharge cardiovascular events based on various preexisting factors. Subgroup analyses were stratified by age (age < 65 vs. ≥ 65 years old), sex, and specific comorbidities, including chronic liver disease, diabetes mellitus, chronic kidney disease, cancer, septic shock, and severe respiratory failure requiring mechanical ventilation.
Sensitivity analysis
We performed additional analysis based on different definitions of anemia, categorizing patients into three groups: those who received only RBC transfusions, those prescribed with ESA alone, and those who received both RBC transfusions and ESA prescriptions.
Statistical analysis
We compared the control and anemia groups using the following methods: continuous variables were presented as mean and standard deviation or as median and interquartile range, and we performed comparisons using the t-test or Mann–Whitney Test. Categorical variables were presented as numbers and proportions and compared using the chi-square test or Fisher’s exact test. Person-time was calculated from the date of hospital discharge to the event date, death, or the last follow-up date. Survival curves were generated by the Kaplan–Meier product-limit method and compared using log-rank tests. We used Cox proportional hazards regression models to estimate hazard ratio (HR) with 95% confidence interval (CI) for cardiovascular events and all-cause mortality, adjusting for age, sex, tertiary hospital, comorbidities, septic shock, CRRT duration, and mechanical ventilation. The proportional hazards assumption was assessed through log–log plots survival function and Schoenfeld residuals. To account for competing risks due to mortality, a proportional sub-distribution hazards regression model was employed for the incidence of cardiovascular events, considering death as a competing event. All analyses were two-sided and P-values < 0.05 were considered statistically significant. Statistical analyses were performed using SAS version 9.2 (SAS Institute, Inc, Cary, NC, USA) and R software version 3.3.2 (Free Software Foundation, Inc., Boston, MA, USA).
- The Renal Warrior Project. Join Now
- Source: https://www.nature.com/articles/s41598-024-56772-1