ABC | Volume 111, Nº5, November 2018

Original Article Lopes et al Mortality for critical congenital heart diseases and newborns Arq Bras Cardiol. 2018; 111(5):666-673 Table 1 – Association between congenital heart disease and factors related to the newborn Variable Factor RR * CI 95% p value AIC Gender Female 1 - Male 0.92 0.66-1.27 0.6 301.2 Weight Above > 2,500 g 1 - Low Weight (< 2,500 g) 2.33 1.26-4.29 0.0068 170.5 Gestational Age > 37 weeks 1 - < 37 weeks 2.89 1.49-5.56 0.0015 157.9 Apgar 1st minute ≥ 7 1 - Less than < 7 2.35 1.25-4.45 0.0084 163.1 Apgar 5 th minute ≥ 7 1 - <7 9.49 1.09-82.85 0.042 43.9 Twinning No 1 - Yes 11.96 1.43-99.85 0.022 48.8 Heart auscultation alteration Normal 1 - Changed 84 11.83-596.21 < 0.0001 112.6 Comorbidities No 1 - Yes 2.27 1.58-3.26 < 0.0001 215.5 (*) Gross RR by Poisson regression; p value - Z statistic. RR: relative risks; 95% CI: 95% confidence interval; AIC: Akaike Information Criterion. twins among the cases (9.9%) (RR: 13.1; 95% IC [1.59-109.1]; p = 0.018) than newborns without heart disease (2.2%), and for this condition, the risk of death was 12 times higher among twin newborns with CHD (Table 1). Clinical data on changed cardiac auscultation were found in 72% of cases and in only 1% of infants without CHD. The difference of this finding was related to the higher risk for CHD (p < 0.0001). Pulse oximetry was recorded even for cases of CHD with intrauterine diagnosis or in those in which another finding was the clinical suspicion and where the diagnosis had been made before 24 hours of life. Figure 1 illustrates the differential distribution density of pulse oximetry measurements among newborns with and without CHD. Some records below the cut-off level are noted for newborns without CHD, for whom the echocardiogram was required, and the possibility of CHD was ruled out. The incidence of death in CHD cases was 81/100 thousand live births. The case fatality rate attributed to CHD was 64.7%, with proportional mortality of 12.0% (17/142). The main cause of death was cardiogenic shock in 41.1% of the cases, followed by sepsis (17.6%) in three newborns with Double Right Ventricular Outflow Tract (DRVOT), and impossibility of therapy for cardiopathy (17.6%) - CHD anatomy was not consistent with any surgical procedure available, progressing to refractory hypoxemia followed by death - in neonates with hypoplastic left heart syndrome and untreatable ill-defined cardiac defects (Table 2). The median hospital stay was 75 days, with an increased risk of death of 0.4 to 0.8 (HR: 0,4->0.8). Still in the neonatal period, 25% of CHD newborns had already died (Figure 2). There was no statistical difference for survival rates when the death event was compared between those who died from CHD and due to other causes (p = 0.076). Although survival in these newborns has declined by more than 50% in the first 10 days of life and within the neonatal period, this survival declined by more than 60% (Figure 3) before newborns achieved 28 days of life. Discussion CHD newborns presented higher morbidity attributed to prematurity, low birth weight, some degree of intrauterine fetal distress, both due to physical examination and changed pulse oximetry. The literature has drawn attention to the greater morbidity, especially of premature newborn infants, who already present a range of other pathologies due to their constitution, which can substantially aggravate these patients progress. 12 For both post-ductal (RR: 46; 95% CI [11.54-184.0]) and pre-ductal (RR: 39; 95% CI [9.72-157.5]) variable oximetry, the differences between groups were well established. This data not only reinforced the validation of the controls, but also confirmed the importance of making this screening test universal. On the other hand, the physical examination had low specificity (40%) and regular sensitivity, a little higher than the POT (89%), but, alone, it was insufficient to rule out the possibility of CHD. The literature states that when the physical examination is performed by a well-trained and experienced pediatrician, there is an increase in the sensitivity of the POT by up to 20%, 18 optimizing the detection capacity when they are appropriately associated. 19,20 The finding of a low Apgar score in the first minute denoted the importance of knowing that some cardiopathies may be active in the uterus, impairing the blood flow that would allow adequate supply of nutrients and oxygen to the fetus, which may affect the morbidity and mortality of this newborn; this 668