Abstract: SA-PO1028
Early and Long-Term Mortality After Hemodialysis Initiation for AKI Requiring Dialysis vs. CKD Progression
Session Information
- Hemodialysis: Risk Factors, Epidemiology, and Outcomes
October 24, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 801 Dialysis: Hemodialysis and Frequent Dialysis
Authors
- Okamoto, Keisuke, Nara Kenritsu Ika Daigaku, Kashihara, Nara Prefecture, Japan
- Matsui, Masaru, Nara Kenritsu Ika Daigaku, Kashihara, Nara Prefecture, Japan
- Eriguchi, Masahiro, Nara Kenritsu Ika Daigaku, Kashihara, Nara Prefecture, Japan
- Tsuruya, Kazuhiko, Nara Kenritsu Ika Daigaku, Kashihara, Nara Prefecture, Japan
Background
Patients initiating hemodialysis (HD) for acute kidney injury requiring dialysis (AKI-D) may have outcomes distinct from those initiating HD because of chronic kidney disease (CKD) progression. Early and long-term mortality patterns after HD initiation in these populations remain incompletely described.
Methods
We conducted a single-center retrospective cohort study of 420 adults who initiated HD between 2007 and 2014, including 75 with AKI-D and 345 with CKD progression. Outcomes were all-cause and cause-specific mortality. Analyses were performed for overall follow-up, the first 90 days, and after the 90-day landmark among patients who survived beyond 90 days. Causes of death were classified as infection, cardiovascular/cerebrovascular, malignancy, or other/unknown. Kaplan-Meier methods and Cox regression were used for all-cause mortality, and Fine-Gray models for cause-specific mortality. Models were adjusted for age, sex, and Charlson Comorbidity Index.
Results
Patients with AKI-D were older than those with CKD progression (median age, 72 vs 70 years), with 61.3% versus 66.3% men and a median Charlson Comorbidity Index of 5 versus 6. During a median follow-up of 3.7 years (IQR, 0.7–7.5), 66 of 75 patients with AKI-D and 194 of 345 patients with CKD progression died. Kaplan-Meier curves showed lower survival in the AKI-D group. In adjusted Cox models, AKI-D was associated with higher all-cause mortality during overall follow-up (adjusted hazard ratio [aHR], 3.76; 95% CI, 2.74–5.16), with the strongest association during the first 90 days (aHR, 8.32; 95% CI, 4.36–15.87). This association persisted after the 90-day landmark (aHR, 2.57; 95% CI, 1.71–3.86). In Fine-Gray models, AKI-D was associated with infection-related death during overall follow-up (adjusted subdistribution hazard ratio [asHR], 2.47; 95% CI, 1.38–4.42), particularly during the first 90 days (asHR, 7.90; 95% CI, 3.32–18.80).
Conclusion
Patients initiating HD for AKI-D had higher mortality than those initiating HD for CKD progression, especially within 90 days, with excess risk persisting among 90-day survivors. These findings support separating early and long-term risk periods when evaluating outcomes after HD initiation.