Original Article
Intraventricular hemorrhage risk factors in very low birth weight newborns: a case-control study Fatores de risco para hemorragia intraventricular em recém-nascidos de muito baixo peso: estudo caso-controle Laura Martins1, Maria João Borges2, Sónia Melo Gomes3, Israel Macedo4, José Nona5, Filomena Pinto6, Conceição Faria7, Odília Nascimento8, António Marques Valido9
ABSTRACT Objective: Identification of variables that affect the risk of severe intraventricular hemorrhage (IVH) in very low birth weight (VLBW) newborns. Methods: Analytic case-control study, in a population consisting of all VLBW newborns admitted to the Neonatal Intensive Care Unit of a maternity hospital, between January 2002 and December 2007. The authors considered as cases all VLBW newborns with severe IVH (grade ≥ 3), and control all VLBW newborns without IVH. Independent variables included obstetric, perinatal and neonatal diagnosis and therapy. Bivariate analysis and multivariate logistic regression analysis were performed. Results: During this period, of the 864 VLBW newborns admitted to the Neonatal Intensive Care Unit, 9.7% had severe IVH. With bivariate analysis an association between severe IVH, gestational age and birth weight was found. Prenatal care and pre-eclampsia were associated with a decrease in the incidence of severe IVH. Amnionitis, being outborn, vaginal delivery, male gender, intubation in the delivery room, surfactant, hyaline membrane disease, pneumothorax, necrotizing enterocolitis (NEC) perforation and oscillatory high frequency ventilation were associated with an increased incidence of severe IVH. By multivariate logistic regression, the variables associated with increased risk of severe IVH were: pneumothorax (OR = 3.8; 95%CI = 1.7-8.3), NEC with perforation (OR = 8.8; 95%CI = 1.7-45.0), vaginal delivery (OR = 2.0; 95%CI = 1.0-4.1) and high frequency ventilation (OR = 4.8; 95%CI = 1.3-17.3). The following were protective of severe IVH: gestational age (OR = 0.61; 95%CI = 0.52-0.72), patent ductus arteriosus treatment with indomethacin (OR = 0.26; 95%CI = 0.11-0.6) and fertility treatment (OR = 0.24;
95%CI = 0.06-0.94). Conclusion: These data outline the importance of improvement of pre and neonatal care to reduce severe IVH. Keywords: Intracranial hemorrhages; Risk factors; Infant, very low birth weight; Infant, newborn, Infant, premature
RESUMO Objetivo: Identificação de variáveis que influenciem o risco de hemorragia intraventricular (HIV) grave em recém-nascidos de muito baixo peso (RNMBP). Métodos: Efetuou-se um estudo analítico, caso-controle, em uma população constituída por todos os RNMBP admitidos em uma Unidade de Cuidados Intensivos Neonatal (UCIN), no período compreendido entre Janeiro de 2002 e Dezembro de 2007. Consideraram-se casos todos os RNMBP com HIV grave (grau ≥ 3) e controle todos os RNMBP sem HIV. As variáveis independentes foram dados obstétricos, perinatais, diagnóstico e terapêutica neonatal. Realizou-se análise bivariada e análise de regressão logística multivariada. Resultados: Foram admitidos na Unidade de Cuidados Intensivos Neonatal, neste período, 864 RNMBP, dos quais 9,7% apresentaram HIV grave. Na análise bivariada, verificou-se uma associação entre HIV grave, idade gestacional e peso ao nascer. A atenção pré-natal e pré-eclampsia foram associadas a uma menor incidência de HIV grave. Amnionite, nascimento no exterior, parto vaginal, sexo masculino, intubação na sala de parto, surfactante, doença da membrana hialina, pneumotórax, enterocolite necrotizante (EN) com perfuração e a ventilação de alta frequência oscilatória
Study carried out at Neonatology Department, Neonatal Intensive Care Units of the Maternidade Dr. Alfredo da Costa, Lisboa, Portugal. 1
Medical student, Complementary Internship in Pediatrics at Hospital do Espírito Santo de Évora, EPE, Portugal.
2
Pediatrician; Hospital assistant at Hospital Central do Funchal, Portugal.
3
Medical student, Complementary Internship in Pediatrics at Centro Hospitalar das Caldas da Rainha, Portugal.
4
Pediatrician; Hospital assistant at Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
5
Pediatrician; Hospital assistant at Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
6
Pediatrician; Hospital assistant at Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
7
Pediatrician; Hospital assistant at Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
8
Head of Pediatrics Department of Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
9
Director of Pediatrics Department of Maternidade Dr. Alfredo da Costa, Lisboa, Portugal.
No financial support was given to this study. All authors declare they took part in the study, are responsible for it, and there is no conflict of interest regarding the statements made in the article. Corresponding author: Laura Martins – Rua do Raimundo, 72 – CEP 7000-661 – Évora, Portugal – e-mail:
[email protected] Received on May 19, 2009 – Accepted on Jul 1, 2009
einstein. 2009; 7(3 Pt 1):279-83
280
Martins L, Borges MJ, Gomes SM, Macedo I, Nona J, Pinto F, Faria C, Nascimento O, Valido AM
foram associados a uma maior incidência de HIV grave. No modelo de regressão logística multivariada, as variáveis associadas a um maior risco de HIV grave foram pneumotórax (OR = 3,8; IC95% = 1,7-8,3), EN com perfuração (OR = 8,8; IC95% = 1,7-45,0), parto vaginal (OR = 2,0; IC95% = 1,0-4,1) e ventilação de alta frequência oscilatória (OR = 4,8; IC95% = 1,3-17,3). Foram fatores protetores para HIV grave: idade gestacional (OR = 0,61; IC95% = 0,52-0,72), tratamento da persistência do ducto arterioso com indometacina (OR = 0,26; IC95% = 0,11-0,60) e tratamento de fertilidade (OR = 0,24; IC95% = 0,06-0,94). Conclusão: Os resultados obtidos neste estudo realçam a importância da melhoria da prestação de cuidados pré e neonatais na redução da HIV grave. Descritores: Hemorragias intracranianas; Fatores de risco; Recémnascido de muito baixo peso; Recém-nascido, prematuro
INTRODUCTION Intraventricular hemorrhage (IVH) is a major cause of severe cognitive and motor development impairment in very low birth weight (VLBW) newborns, and is associated with high mortality rate(1-3). Although the incidence of IVH has decreased, due to better pre and neonatal care, it still remains a problem in VLBW newborns, with an incidence of approximately 20%(1). Identifying risk factors would allow the development of effective strategies to reduce the incidence of IVH and, subsequent, improve neurological outcomes in VLBW newborns(3). Several risk factors have been implicated in the pathogenesis of IVH: low birth weight, low gestational age, Apgar score, maternal smoking, breech presentation, sex, premature rupture of membranes, intrauterine infection, mode of delivery, prolonged labor, postnatal resuscitation and intubation, transfer from one hospital to another, early onset sepsis, development of the respiratory distress syndrome or pneumothorax, recurrent endotracheal suctioning, metabolic acidosis and rapid bicarbonate infusion, high frequency ventilation, low mean arterial pressure, surfactant application and patent ductus arteriosus (PDA). Factors that have been considered to reduce the risk of IVH are: pregnancy induced hypertension, antenatal steroids, prenatal tocolytic therapy, postnatal administration of low indomethacin dose(2). Many of the studies performed until now did not use multivariave analysis to identify independent risk factors to IVH. The purpose of this study was to identify variables that may affect the risk of severe IVH in VLBW newborns. METHODS The Neonatology Department of Maternidade Dr. Alfredo da Costa collects data on all VLBW newborns. These data einstein. 2009; 7(3 Pt 1):279-83
include prenatal history, details of delivery, neonatal care and treatments, diagnoses, procedures, complications and outcome at discharge. These data are included in the Portuguese national database of VLBW newborns. Based on this database, an analytic case-control study was made, in the period between January 1st 2002 and December 31st 2007. The authors considered all VLBW newborns with severe IVH (grade ≥ 3), and as controls all VLBW newborns without IVH. Other data from obstetric history, perinatal events and neonatal strategies including therapy and pathology in both groups were collected from the above-mentioned database (Chart 1). Chart 1. Variables analyzed Obstetric data Maternal age Fertility treatment Prenatal care Twins Antenatal steroids Amnionitis Preeclampsia Gestational age Delivery method In/outborn
Perinatal data Apgar score at 5’ Birth weight Sex CRIB and SNAPPE II
Neonatal data Intubation in delivery room Surfactant Hyaline membrane disease High frequency ventilation Pneumothorax Early onset of sepsis NEC perforation Patent ductus arteriosus Indomethacin IVH and grade
At this Neonatal Intensive Care Unit (NICU), IVH was diagnosed according to the routine protocol by brain ultrasound, which recommends a first evaluation in the first day of life and, then, subsequent scans 24 to 48 hours after, and on postnatal days 4, 5, 7, 14, 21 and at 40 weeks of conception age or more often if clinically indicated. When present, the grade of IVH was determined according to Papile et al.(4). All images were acquired using a Siemens Sequoia Ultrasonograph. Statistical analysis was performed with the Statistical Package for Social Sciences Software, version 15.0, for Windows (SPSS, Inc; Chicago, EUA). Bivariate analysis was performed to identify differences between Case and Control Groups, using the Pearson χ2 test, Fisher’s exact test, and Mann-Whitney non-parametric test, as appropriate. Results are expressed as median, percentage, Odds Ratio (OR) with 95% Confidence Interval (CI). Statistical significance was defined as p ≤ 0.05. The variables that in a bivariate analysis had an association at a significance level of 0.1 or less were entered into a stepwise logistic regression model.
RESULTS A total of 864 VLBW preterm infants were born at Maternidade Dr. Alfredo da Costa during a six-year period between January 1st 2002 and December 31st 2007. Forty-three deaths occurred during the first 48 hours of life, and these newborns were excluded from
Intraventricular hemorrhage risk factors in very low birth weight newborns: a case-control study
Table 1. Obstetric data Factors Maternal age # Fertility treatment ° Prenatal care * Twins * Antenatal steroids * Amnionitis * Preeclampsia * Vaginal delivery * Outborn *
Case (n = 84) 28.8 3.6% 85.9% 26.2% 79.8% 11.9% 8.3% 47% 20.2%
Control (n = 620) 29.7 10.3% 93.5% 35.6% 80.5% 6% 19.8% 19.6% 7.1%
p-value 0.182 0.048 0.015 0.087 0.876 0.041 0.011 < 0.001 < 0.001
OR (95% CI)
P
