Information about the energy expenditure assessment of patients with AKI is scarce in the literature. Our study found that the REE estimated by the Harris and Benedict formula [19] was significantly lower than that measured by IC. This finding corroborates the indication not to use this formula in critically-ill patients and in patients with AKI [11, 14, 20, 21] and the need to propose a new equation for AKI on dialysis. This study aimed to develop and validate predictive equations for REE in severe AKI patients using a machine learning approach. It was found that the models developed validly and significantly predict REE in these patients and according to several linear and non-linear algorithms. In the present study, REE was positively correlated with C-reactive protein, minute volume (MV), expiratory positive airway pressure, serum urea, body mass index and inversely with age (attachment 02). The principal variables included in the best model were age, BMI, use of vasopressors, expiratory positive airway pressure, minute volume, C-reactive protein, temperature, and serum urea. The final r-value in the validation set was 0.69. In the literature, there is no consensus regarding the procedures to be used for the validation of predictive models. Some authors do not suggest the use of determination and correlation coefficients for the validation of techniques or estimated variables [15]. Others consider that the Bland Altman plot is likely to show a systematic proportion bias [19]. We used 10-fold cross-validation and select the model with lower RMSE. The performance of the best model was confirmed in a test set of 20% random selected unseen data (internal validation). The linear models had the advantage of simplicity to the model built and better interpretability but do not capture the non-linearity of the data. We confirm this with a very low accuracy of the Harris-Benedict equation in the test set. The use of natural splines to age and BMI predictors improves the linear models but does not reach the same performance as the non-linear models like boost trees or support vector machines [22–23]. The best model had a higher performance but in trade-off lower interpretability. In addition, the traditional model like Harris-Benedict [19] uses linear models that were easy to implement. Otherwise, the implementation of a model like the random forest has a necessity of a calculator. However, in present, we have the facility of computers or apps that may overcome this difficulty.
Another advantage of the machine learning approach is to demonstrate others predictor variables that influence the outcome by finding complex interactions. The principal factor that contributes to the variability of REE was the BMI and age that has been previous describe with traditional models [24–25]. The machine analysis in this study reveals other new contributed variables that predict de REE for example ventilator parameters and biochemical values. Otherwise, a simple linear model with this data set without pre-processing and feature engineering results in a model with lower accuracy and do not shown new interactions (R2 = 0.21, data not shown).
Thus, this study presents the importance of estimating the REE in severe AKI patients, which is a determining factor in the assessment of nutrition status in critically ill patients. From a practical and application point of view, it is worth mentioning the capacity for technical evaluations to reduce the difference in REE between IC and estimated by others conventional formulas as HB, (ii) the use of these predictive equations for the assessment of passive and active trawling, contributing with relevant information for nutritionists and physicians. The present study has as main limitations the predictive models are not valid for no severe AKI patients and the Bland Altman model was not performed. Although we train models with a robust estimator with 10-fold cross-validation using RMSE as metric an approach regular used in machine learning analysis. We test the model in unseen data that may be considered an internal validation.