Utilidad pronóstica de la hiperglicemia sostenida sobre la mortalidad en niños críticos ingresados en la Unidad de Cuidados Intensivos Pediátricos del Hospital de Niños Dr. Roberto Gilbert durante el período de 1 de enero al 30 de septiembre del 2017.

Abstract

Hyperglycemia is commonly seen in critically ill patients. Many studies have shown the importance of assessing glucose levels in critically ill pediatric patients; it can lead to unfavorable outcomes. Materials and methods: An observational, retrospective, longitudinal, analytical study was carried out, which included patients admitted to the Pediatric Intensive Care Unit (PICU) from January to September 2017, who were measured glucose levels upon admission, taken from blood, during the first days of hospitalization. Hyperglycemia was defined as values above 150 mg/dl. The X2 statistic will be used to identify the association. A LOGIT model was used to determine whether the probability of hyperglycemia is related to higher mortality. Chi-square tests were performed to determine whether the level of hyperglycemia is associated with higher mortality in critically ill children, with an increase in the days of hospital stay, with a higher probability of death measured by the PIM III Scale, with sex and the age of the children with a contrast of proportions. Student's t tests were used to determine if the average number of days of hospital stay, as well as the days of mechanical ventilation for children with hyperglycemia is greater than for those children who do not have it and to know if the average death rate of children with hyperglycemia hyperglycemia is greater than for those who do not present it. Results: 441 patients were registered during January and September 2017, of which 365 met the inclusion criteria considered in the study. The mean age of the patients was 4.53 (± 4.90 years); rank; 1 month-17 years, in the study the children who prevailed with hyperglycemia were younger than 1 year and older than 5 years. The probability of dying on admission, calculated using the PIM III, was on average slightly higher in 33.03% women compared to 30.62% in men. The observed mortality rate in men was 16%, while in women it was 14%. Low weight patients are associated with a higher probability of dying (PIM III) The average length of hospital stay was 23.60 (± 26.93) days (median; 17 days, range 1 to 266 days. Respiratory pathologies had a higher percentage of hospital stay together with neurological pathologies with 34% (23.01 ± (21.09)) and 24% (16.07 ± (18.67)) respectively. 76% of the hospitalized patients had blood glucose values ≤ 150 mg / dl. The age groups that presented hyperglycemia were those younger than 1 year and those older than 5 years. showed that patients with hyperglycemia were complicated by sepsis and left sequelae in many of them; 32% and 42% respectively. The average glycemia upon admission to the PICU was 128.47 (± ABSTRACTXIV 58.75) mg / dl. Only 24% of the patients presented hyperglycemia on admission. There was no significant difference between glucose levels at admission between men and women (p = 0.86 of the Mann-Whitney test). There was a slight positive correlation of 0.0868 between the days of hospital stay for those patients who were admitted with hyperglycemia, they are associated with more days of hospitalization. Glucose values> 160 mg / dl were more frequent in deceased patients. A correlation was observed between each of the glycemic levels with the probability of dying calculated using the PIM III, so as the glycemia levels increase, they are associated with a greater probability of dying. Conclusions: The prevalence of hyperglycemia on admission is low. As blood sugar levels rise in patients, the likelihood of dying increases. The glucose level that best predicted the probability of dying in this study was> 160 mg / dl. Hyperglycemia is associated with the longest hospital stay.