Perfil epidemiológico dos recém-nascidos que receberam transfusão de concentrado de hemácias nas unidades de terapia intensiva neonatal na cidade de Juiz de Fora

Abstract

The newborn infant constitutes a unique age group, even on a hematologic point-of-view. The low erythropoiesis and the iatrogenic losses inside the Neonatal Intensive Care Units (NICUs) help explain why these newborns, especially the very-low-birth-weight ones are the most heavily transfused population inside the Intensive Care Units. Over the years, many guidelines tried to reduce the number of erythrocyte transfusions, especially among the premature infants, due to the inherent risks of the procedure. However, the adequate level of haemoglobin and haematocrit that indicates the need to transfuse is not well established. Recent studies have associated some variables as risk factors for submitting these newborn infants to more blood transfusions. The aim of this study is to evaluate the clinical and demographic profiles of this population of transfused newborns in the city of Juiz de Fora, as well as analysing the transfusional practices of these NICUs and trying to establish risk factors that lead to transfusion in this population. This is an epidemiologic observational study that took place within a 12 month-period, through the analysis of the newborn infants‟ health records. It included all neonates that received erythrocyte transfusions and that were admitted to all 5 NICUs of Juiz de Fora. A research protocol was designed and a database was created from it, which was included in the Statistical Package for Social Sciences (SPSS) software, version 15.0. Statistical analysis was performed through descriptive statistics that included the demographic and clinical variables from the study, represented as mean and standard deviation or numbers and percentages. In order to evaluate transfusions under liberal criteria, a special variable was created for that purpose. Univariate analysis identified neonatal clinical conditions associated with the following outcomes: mechanical ventilation, pulmonary bronchodysplasia, necrotizing enterocolitis, retinopathy of prematurity, intra-ventricular haemorrhage and death. Then, a logistic regression model was applied to the significant variables and non-significant ones were removed from the model in a backward stepwise way. From the 949 patients admitted to the NICUs, 133 met the inclusion criteria. The median of transfusions per infant was 2,7 ± 2,16 and 21,07% received at least one transfusion. Most transfusions occurred during the infants‟ first two weeks of life, followed by after its fourth week. The logistic regression model showed the following significant variables correlated to the main studied outcomes: birth weight, gestational age, mechanical ventilation, days of oxygen therapy, late-onset sepsis, delivery room conditions, arterial hypertension on the mother, the differences between multiple birth centres and persistency of Ductus Arteriosus. This study shows that the demographics and clinical characteristics of the transfused newborns admitted in the NICUs of Juiz de Fora are similar to recent literature. The variability on erythrocytes transfusion practices also reflects the difficulty on the decision of when to transfuse, especially on a daily basis.