Abstract: TH-PO064
Persistent AKI Risk Index to Predict Risk of Persistent AKI in Critically Ill Adult Patients: A Validation Cohort Study
Session Information
- AKI: Biomarkers, Risk Factors, Treatments, Outcomes
November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical‚ Outcomes‚ and Trials
Authors
- Pelayo, Blessa Marie Hernandez, St. Luke's Medical Center - Quezon City, Quezon City, NCR, Philippines
- Alolod, Maria kristina, St.Luke's Medical Center - Quezon City, Quezon City, NCR, Philippines
Background
The persistent AKI risk index(PARI) scoring tool is designed as a simple calculator using changes in serum creatinine and only three clinical risk factors such as hyperbilirubinemia, sepsis, ventilation or use of inotropic support. It is an effective bedside tool to exclude low risk patients and detect high risk patients to develop persistent AKI. This risk assessment tool can provide real-time clinical decision to prevent persistent AKI and help clinicians decide on aggressive management such as kidney replacement therapy. The PARI has not been externally validated to detect the potential population of patients at high risk for persistent AKI.
Methods
A total of 2560 critically ill adult patients age ≥19 years were screened for the study from 2015 to 2019 at the critical care units of St. Luke’s Medical Center – Quezon City. The mean age of participants were 63.8 years and 59.3% of which were male. Primary outcome is to predict risk of developing persistent AKI using PARI were score of 8 is considered positive. Secondary outcomes includes risk for kidney replacement therapy(KRT), mortality and length of hospital stay. Discrimination using logistic regression C testing and calibration using Hosmer & Lemeshow test were used to evaluate the predictive ability of the scoring system.
Results
A total of 386 patients were eligible after exclusion evaluation. The PARI was statistically significant in predicting risk of persistent AKI and need for KRT with an OR of 15.13 (CI 9.1 to 25.0), p value 0.0001 and OR 11.35 (CI 6.7 to 19.2), p value 0.0001, respectively and provides acceptable discrimination with C statistics of 0.79 and 0.70. Patients with positive PARI score showed HR of 3.94 (CI 2.6 to 6.1), p value 0.0001 for mortality with acceptable performance in discriminating patients risk for mortality. No statistical significance in length of hospital stay, p value 0.83. The calibration study showed p values of <0.05 indicating poor goodness to fit for all outcomes.
Conclusion
PARI is a convenient and cost-effective bedside tool that can be used to predict persistent AKI with an impact in management of critically ill adult patients. This will provide as an add on tool to biomarkers for early intervention and goal-directed therapies for better prognosis.