Abstract: TH-PO083
Incidence of Hypophosphatemia and Clinical Outcomes in Patients Exposed to Kidney Replacement Therapy: A Post Hoc Analysis of the STARRT-AKI Trial
Session Information
- AKI: Clinical, Outcomes, and Trials - Epidemiology and Pathophysiology
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- Neyra, Javier A., University of Alabama at Birmingham, Birmingham, Alabama, United States
- Ghamarian, Ehsan, University of Toronto, Toronto, Ontario, Canada
- Bagshaw, Sean M., University of Alberta, Edmonton, Alberta, Canada
- Wald, Ron, University of Toronto, Toronto, Ontario, Canada
Background
Patients receiving renal replacement therapy (RRT) are at increased risk of hypophosphatemia, which can predispose to functional muscle weakness and prolonged respiratory failure. The impact of hypophosphatemia on clinical outcomes in patients receiving RRT has been described in small retrospective studies, yet has not been evaluated in a contemporary, large prospective study. We conducted a secondary analysis of the STARRT-AKI trial to identify clinical parameters associated with incident hypophosphatemia during RRT and to examine the association of hypophosphatemia with clinical outcomes.
Methods
This was a post hoc study of the STARRT-AKI trial, a multinational RCT comparing 2 RRT initiation strategies in critically ill adult patients with AKI. Eligibility for this analysis included 1) receipt of RRT, 2) CRRT >24 hours or IHD/SLED >1 session, 3) normal serum phosphate on the day of RRT initiation, and 4) >1 serum phosphate measurement during RRT. A truncated hurdle model was built to determine which baseline characteristics were associated with any hypophosphatemia (<0.7 mmol/L) and severe hypophosphatemia (<0.5 mmol/L). The primary outcome of ventilator-free days (VFD) at 28 days was assessed with an inverse probability weighted zero-inflated negative binomial model and a win-ratio analysis (mortality as first hierarchy). Secondary outcomes of RRT dependence and mortality at 90 days were assessed with logistic regression.
Results
Of 1,940 patients included, hypophosphatemia occurred in 33%. Male sex, higher weight, standard-strategy arm, and higher baseline phosphate were associated with the occurrence of hypophosphatemia during RRT. Duration of hypophosphatemia was associated with lower first-day serum phosphate and the use of CRRT. Patients without severe hypophosphatemia during RRT had more VFD at 28 days (β 1.16 [95% CI 1.06- 1.27], p<0.001. A similar trend of VFD was found among those without any hypophosphatemia during RRT (β 1.06 [95% CI 1.0 - 1.1], p=0.05. These associations were consistent in the win-ratio analysis. There was no association between incident hypophosphatemia and RRT dependence or mortality at 90 days.
Conclusion
Among patients in the STARRT-AKI trial who received RRT, incident hypophosphatemia following RRT initiation was associated with fewer ventilator-free days.