Analysis of metabolic alterations as 30 days intensive care mortality predictors for patients undergoing continuous renal replacement therapy

Background

Acute kidney injury patients on continuous renal replacement therapy are subjected to alterations in metabolism, which in turn are associated with worse clinical outcome and mortality. The aim of this study is to determine which metabolism indicators can be used as independent predictors of 30 days intensive care unit (ICU) mortality.

Methods

This was a prospective observational study on critical care patients on renal replacement therapy. Integrated approach of metabolism evaluation was used, combining the energy expenditure measured by indirect calorimetry, bioelectrical impedance provided fat free mass index (FFMI), amino acid and glucose concentrations. ICU mortality was defined as all cause 30 days mortality. Regression analysis was conducted to determine the conventional and metabolism associated predictors of mortality.

Results

The study was conducted between the 2021 March and 2022 October. 60 high mortality risk patients (APACHE II of 22.98 ± 7.87, 97% on vasopressors, 100% on mechanical ventilation) were included during the period of the study. The rate of 30 days ICU mortality was 50% (n = 30). Differences across survivors and non-survivors in metabolic predictors were noted in energy expenditure (kcal/kg/day) (19.79 ± 5.55 vs 10.04 ± 3.97 p = 0.013), amino acid concentrations (mmol/L) (2.40 ± 1.06 vs 1.87 ± 0.90 p = 0.040) and glucose concentrations (mmol/L) (7.89 ± 1.90 vs 10.04 ± 3.97 p = 0.010). No differences were noted in FFMI (23.38 ± 4.25 vs 21.95 ± 3.08 p = 0.158). In the final linear regression analysis model, lower energy expenditure (exp(B) = 0.852 CI95%: 0.741–0.979 p = 0.024) and higher glucose (exp(B) = 1.360 CI95%: 1.013–1.824 p = 0.041) remained as independent predictors of the higher mortality.

Conclusion

The results of the study imply strong association between the metabolic alterations and ICU outcome. Our findings suggest that lower systemic amino acid concentration, lower energy expenditure and higher systemic glucose concentration are predictive of 30 days ICU mortality.