Back
Frailty is a syndrome that is often overlooked but common in severely ill patients, characterized by a state of increased susceptibility to systemic adversities due to the loss of physiological and cognitive reserves. While not diagnostic, some tools are helpful in screening for frailty, such as the Clinical Frailty Scale, the Frailty Index, and the Modified Frailty Index (MFI). This analysis aimed to assess patient frailty, measured by the MFI, as an independent prognostic factor in the incidence of Acute Kidney Injury (AKI) in critically ill patients and its association with mortality.
A retrospective cohort analysis was conducted on critically ill patients admitted to the intensive care units (ICU) of two large hospitals in Rio de Janeiro, Brazil, between 2020 and 2022. Exclusion criteria included end-stage renal disease patients and patients with fewer than two creatinine measurements. Frailty was assessed using the MFI, which assigns 1 point for each present frailty component out of 11 items, including comorbidities, functional capacity, and previous complications. Based on the sum of these points, patients were categorized as non-frail (MFI < 2) or frail (MFI ≥ 3). The frailty score was defined by the sum of individual points divided by 11, and frailty corresponds to a value greater than or equal to 0.27. Logistic regression analysis was conducted to account for relevant confounders in outcome analyses. Primary outcomes analyzed include the incidence of AKI and mortality.
After applying the exclusion criteria, 9291 patients admitted to the ICU were analyzed. Of these, 2237 (24.1%) were considered as frail (MFI ≥ 3). There was no difference between frail and non-frail patients concerning ethnic groups, sex, and median days spent in the ICU (4 days, IQR 2.00 – 9.00). Frail patients were older (74 years, IQR 66.00 – 82.00) compared to non-frail patients (62 years, IQR 49.00 – 73.00) and presented with a higher Charlson Comorbidity Index (2.00, IQR 1.00 – 3.00 vs. 0.00, IQR 0.00 – 2.00, p<0.001). The severity of presentation, as measured by the SAPS 3 index, was higher for frail patients (49.00, IQR 41.00 –56.00) compared to non-frail patients (42.00, IQR 34.00 – 51.00). The overall incidence of AKI was 59.8% and was higher in frail patients compared to non-frail patients (67.8% vs. 57.3%, p<0.001). After adjustments, in the multivariate model, including variables such as comorbidities and SAPS3 index, the MFI score was independently associated with AKI (OR 1.95, CI 95% 1.17 - 3.24, p<0.001). Mortality was higher in frail compared to non-frail (22.1 vs. 19.7%, p<0.014). The univariate regression analysis showed a significant association between the MFI score and mortality (OR 2.89, 95% CI 1.92 – 4.36, p < 0.001). However, after the multivariate logistic regression for mortality, in adjustment to other relevant variables, the MFI was marginally non-significant (OR 1.70, 95% CI, 0.99 – 2.89, p = 0.053).
The MFI serves as a robust independent prognostic factor for the incidence of AKI but does not independently predict mortality. Despite higher rates of AKI and mortality among frail patients, the MFI's predictive power for mortality remains inconclusive. These findings indicate the need for further research to refine frailty indices and their role in predicting AKI and mortality.