| Summary: | <p><strong>Introduction:</strong> Malnutrition associated with severe disease represents an important factor that increases mortality in pediatric patients.</p><p><strong>Objective: </strong>To validate a prognostic model of death in malnourished children admitted to intensive care units.</p><p><strong>Material and Methods:</strong> An observational cohort study was carried out in patients with acute malnutrition, admitted to three Intensive Care Units of Pediatric Hospitals in Havana, out of which "Angel Arturo Aballí" Mother-Child Teaching Hospital served as the research center. The work was carried out from January 2011 to May 2018. A total of 234 cases were evaluated. Discrimination and calibration of a prognostic model of death was determined.</p><p><strong>Results:</strong> As general characteristics, there was a greater number of patients among the infants under 7 months of age. The main causes of admission were digestive and respiratory infections and sepsis, this last one had a significant relationship with the probability of death. Likewise, there was a significant association between the longer lengths of stay and death. The validated model showed an excellent discrimination with value of the area under the curve of 0.99 and good calibration with p = 0 .289.</p><p><strong>Conclusions:</strong> The prognostic model of death applied to malnourished children admitted to intensive care units showed a satisfactory performance to be applied in this patient population and has, as an additional characteristic, an easy clinical application given the feasibility of obtaining the variables that constitute it.</p><p class="Sinespaciado1"><strong>Keywords:</strong> Pediatrics, malnutrition, intensive care, prediction, mortality, obtaining of variables.<strong></strong></p>
|
|---|
