Clinical Impact of the Use of Warfarin in Patients with Atrial Fibrillation Undergoing Maintenance Hemodialysis

Background: We evaluated the impact of warfarin use on the clinical outcomes of patients with atrial fibrillation who were undergoing hemodialysis (HD). Methods: A retrospective analysis was conducted utilizing data from patients undergoing maintenance HD who participated in HD quality assessment programs. Patients who were assigned the diagnostic code for atrial fibrillation (n = 4829) were included and divided into two groups based on the use of warfarin: No group (no warfarin prescriptions (n = 4009)), and Warfarin group (warfarin prescriptions (n = 820)). Results: Cox regression analyses revealed that the hazard ratio for all-cause mortality in the Warfarin group was 1.15 (p = 0.005) in univariate analysis and 1.11 (p = 0.047) in multivariable analysis compared to that of the No group. Hemorrhagic stroke was significantly associated with warfarin use, but no significant association between the use of warfarin and ischemic stroke or cardiovascular events was observed. The subgroup results demonstrated similar trends. Conclusions: Warfarin use is associated with a higher risk of all-cause mortality and hemorrhagic stroke, and has a neutral effect on ischemic stroke and cardiovascular events in patients with atrial fibrillation who are undergoing HD, compared to those who are not using warfarin.

Prediction of osteoporosis in patients with rheumatoid arthritis using machine learning

Osteoporosis is a serious health concern in patients with rheumatoid arthritis (RA). Machine learning (ML) models have been increasingly incorporated into various clinical practices, including disease classification, risk prediction, and treatment response. However, only a few studies have focused on predicting osteoporosis using ML in patients with RA. We aimed to develop an ML model to predict osteoporosis using a representative Korean RA cohort database. The KORean Observational study Network for Arthritis (KORONA) database, established by the Clinical Research Center for RA in Korea, was used in this study. Among the 5077 patients registered in KORONA, 2374 patients were included in this study. Four representative ML algorithms were used for the prediction: logistic regression (LR), random forest, XGBoost (XGB), and LightGBM. The accuracy, F1 score, and area under the curve (AUC) of each model were measured. The LR model achieved the highest AUC value at 0.750, while the XGB model achieved the highest accuracy at 0.682. Body mass index, age, menopause, waist and hip circumferences, RA surgery, and monthly income were risk factors of osteoporosis. In conclusion, ML algorithms are a useful option for screening for osteoporosis in patients with RA.

Phase Angle as a Risk Factor for Mortality in Patients Undergoing Peritoneal Dialysis

Phase angle (PhA) is measured using bioimpedance analysis and calculated using body reactance and resistance in the waveform at 50 kHz. Further studies are necessary to clarify the predictive efficacy of PhA in the mortality of peritoneal dialysis (PD) patients. The objective of this study was to assess the utility of PhA for predicting patient mortality and technique failure and compare the predictability of PhA with other risk factors. Our study had a retrospective cohort design. Our center routinely evaluates bioimpedance measurements for all prevalent PD patients (n = 199). The PhA was measured using multifrequency bioimpedance analysis. Our study evaluated patient and technique survival. There were 66, 68, and 65 patients in the low, middle, and high tertiles of PhA, respectively. The PhA values of the low, middle, and high tertiles were 3.6° (3.4-3.9), 4.4° (4.2-4.7), and 5.5° (5.2-6.0), respectively. The 5-year patient survival rates for the high, middle, and low tertiles were 100%, 81.7%, 69.9%, respectively (p < 0.001). The 5 year technique survival rates for the high, middle, and low tertiles were 91.9%, 74.8%, 63.7%, respectively (p = 0.004). Patient and technique survival increased as the PhA tertiles increased. Both univariate and multivariate Cox regression analyses demonstrated a consistent pattern. The prediction of patient or technique survival was better in PhA than in the other classical indicators. The present study demonstrated that PhA may be an effective indicator for predicting patient or technique survival in PD patients. Furthermore, it suggests that routine measurement of PhA and pre-emptive intervention to recover PhA according to causes of low PhA may help improve patient or technique survival in PD patients.

Correlation between Statin Solubility and Mortality in Patients on Chronic Hemodialysis

This study aimed to evaluate the effect of statin solubility on the survival of patients undergoing hemodialysis (HD). This retrospective study used laboratory and clinical data from a national HD quality assessment program and claims data (n = 53,345). The use of statins was defined as prescription ≥30 days during 6 months of each HD quality assessment period. We divided the patients into three groups based on the use and solubility of statins: No group, patients without a prescription of statins (n = 37,944); Hydro group, patients with a prescription of hydrophilic statins (n = 2823); and Lipo group, patients with a prescription of lipophilic statins (n = 12,578). The 5-year survival rates in the No, Hydro, and Lipo groups were 69.6%, 67.9%, and 67.9%, respectively (p < 0.001 for the trend). Multivariable Cox regression analyses showed that the Lipo group had better patient survival than the No group. However, multivariable analyses did not show statistical significance between the Hydro and No or Lipo groups. In all subgroups based on sex, age, presence of diabetes mellitus, and heart disease, the Lipo group had better patient survival than the No group. We identified no significant association between hydrophilic and lipophilic statins and patient survival. However, patients taking lipophilic statins had a modest survival benefit compared with those who did not receive statins.

Influence of Different Types of β-Blockers on Mortality in Patients on Hemodialysis

Previous results regarding the association between types of β-blockers and outcomes in patients on hemodialysis (HD) were inconsistent. Our study aimed to evaluate patient survival according to the type of β-blockers administered using a large sample of patients with maintenance HD. Our study included patients on maintenance HD patients from a national HD quality assessment program (n = 54,132). We divided included patients into four groups based on their use and type; Group 1 included patients without a prescription of β-blockers, Group 2 included patients with a prescription of dialyzable and cardioselective β-blockers, Group 3 included patients with a prescription of non-dialyzable and non-cardioselective β-blockers, and Group 4 included patients with prescription of non-dialyzable and cardioselective β-blockers. The number of patients in Groups 1, 2, 3, and 4 were 34,514, 2789, 15,808, and 1021, respectively. The 5-year survival rates in Groups 1, 2, 3, and 4 were 69.3%, 66.0%, 68.8%, and 69.2%, respectively. Univariate Cox regression analyses showed the hazard ratios to be 1.10 (95% CI, 1.04-1.17) in Group 2 and 1.05 (95% CI, 1.02-1.09) in Group 3 compared to Group 1. However, multivariate Cox regression analyses did not show statistical significance among the four groups. Our study showed that there was no significant difference in patient survival based on the use or types of β-blockers.