Respiratory dysfunction index as a predictor of mechanical ventilation in patients with acute renal failure

Background: Acute kidney injury (AKI) can cause respiratory dysfunction, increasing the risk of mechanical ventilation (MV). Objective: To develop and evaluate the respiratory dysfunction index in acute kidney failure (RII-AKI) as a predictor of the need for MV in the first 14 days of hospitalization. Material and method: An analytical study based on a database of 5060 patients hospitalized with AKI, of whom 2679 had information on MV. The RII-AKI was developed using factor analysis, and its predictive capacity was evaluated using the area under the curve (AUC), Kaplan-Meier analysis, and Cox regression. Results: The RII-AKI showed good discriminatory capacity for predicting MV in 14 days (AUC: 0.756), superior to other parameters such as anion gap (AUC: 0.627) and respiratory rate (AUC: 0.625). A high RDI-ARI was associated with a shorter time to MV (median: 25.4 vs. 58.5 days; p < 0.001). In Cox regression, it was the main predictor (HR: 2.199; p < 0.001), followed by creatinine at admission (HR: 1.511; p = 0.031) and anion gap (HR: 1.040; p = 0.004). Its sensitivity was 76.75%, specificity 61.97%, positive predictive value 23.70%, and negative predictive value 94.54%. Conclusions: The RDI-ARI is a useful tool for predicting MV in patients with ARI. External validation studies are required.