Abstract
BACKGROUND:
Perinatal asphyxia is a prominent cause of neonatal mortality in the developing world. Growth in head circumference is associated with improved neurodevelopment. Previous studies found a positive correlation between additional dietary supplementation and growth in head circumference among newborns with perinatal brain injury. This study aims to evaluate the association between anthropometric parameters and developmental outcomes in newborns with hypoxic ischemic encephalopathy (HIE).
METHODS:
Newborns at ≥36 weeks gestation with moderate to severe HIE were included in the study and growth parameters were monitored. Newborns with life-threatening anomalies were excluded. None of the study participants received therapeutic hypothermia (TH). Developmental Assessment Scale for Indian Infants (DASII) was used to evaluate neurodevelopmental outcomes at 1 year of age.
RESULTS:
Of 76 study participants, 46 were followed for 12 months, 28 died, and 2 were lost to follow-up. HIE stage III, Apgar score <5 at 5 minutes of age, pH ≤ 7.1 on first blood gas and base deficit > – 16 was associated with death or disability at 1 year of age. All anthropometric parameters were significantly lower in presence of death or disability. pH ≤ 7.1 at birth (odds ratio: 11.835, 95% CI 2.273–61.629, p = 0.003) and weight gain at one year (odds ratio 1.001, 95% CI 1.000–1.002, p = 0.03) were significantly associated with death and disability.
CONCLUSION:
pH > 7.1 at birth, and weight gain were associated with better neurodevelopmental outcomes at 1 year of age. Thus, in addition to TH, nutritional interventions may potentially improve outcomes among newborns with HIE.
Introduction
In India, 27 million infants are born every year, out of which 1.2 million die within the first month of life. Perinatal asphyxia contributes to 19% of neonatal mortality [1]. Although data corresponding to outcomes of newborns in the developing world is sporadic, they consistently indicate a poor outcome associated with moderate to severe hypoxic-ischemic encephalopathy (HIE) [2–4].
Several international randomized controlled trials have shown that therapeutic hypothermia (TH) improves outcomes among newborns with moderate to severe hypoxia [5–7]. Studies conducted in India have shown that TH is feasible, inexpensive, and has better neuroprotection among asphyxiated newborns with HIE [8–11].
Postnatal nutritional status is an important variable influencing neurodevelopment. Malnutrition is known to negatively affect brain development [12, 13]. In addition, growth in head circumference during infancy has a positive correlation with developmental outcome [14, 15]. Newborns with perinatal brain injury subjected to a diet with 20% higher protein content than the estimated average requirement, had a better head circumference, growth and axonal diameter at 12 months of age [16]. It is, therefore, important to evaluate the association between anthropometric parameters and developmental outcomes in newborns with HIE.
This study compares demographic characteristics and anthropometric parameters in newborns with moderate to severe HIE in presence and absence of death, or moderate to severe disability.
Methods
Selection criteria
Full-term neonates with perinatal asphyxia requiring resuscitation admitted to the neonatal intensive care unit (NICU) at KEM (King Edward Memorial) hospital, Pune, India, within 6 hours of birth and subsequently diagnosed with moderate to severe HIE per Sarnat and Sarnat [17] staging were included in this prospective observational study. Newborns with life-threatening congenital anomalies were excluded. Out of 76 participants matching the study inclusion criteria, 28 died and two were lost to follow-up. The remaining 46 newborns (95.83% of survivors) were followed until 1 year of age. The study enrollment period was November 2007 to November 2008.
Resuscitation for all babies was carried out per Neonatal Resuscitation Protocol guidelines. Apgar scores were recorded at 1, 5 and 10 minutes. Perinatal asphyxia was defined as Apgar score less than 7 at 1 minute and 5 minutes of birth [18]. All neonates were monitored neurologically using the Sarnat and Sarnat staging [17]. A diagnosis of HIE was made and classified as mild, moderate or severe, equivalent to Sarnat stages I, II and III respectively. Detailed maternal history including maternal diseases, obstetric complications, intrapartum monitoring and evidence of fetal distress, along with birth weight, length, and head circumference were noted on a predesigned proforma. Neonates were clinically and biochemically monitored for evidence of multi-organ dysfunction. A trained radiologist performed head ultrasound scans. Involvement of organ systems were recorded according to standard definitions [19]. Morbidities and interventions performed were also recorded. The outcome was noted for all babies to the point of discharge. A detailed neurological examination was done at the time of discharge. In case of death, cause was defined per national neonatal-perinatal data (NNPD) criteria [1].
Assessments after discharge
Outcome was measured by physical growth, neurodevelopmental status, visual and hearing assessments. At each visit, a comprehensive medical, anthropometric, neuromotor and neurodevelopmental assessment was carried out as per methods suggested by Amiel-Tison [20]. Growth assessment was based on the National Centre for Health Statistics (NCHS) standards. Otoacoustic emission (OAE) was performed at discharge from the NICU. If the results were found to be abnormal (Refer) in either or both ears, the test was repeated after one month. Brainstem evoked response audiometry (BERA) was done at 3 months of age.
At 1 year of age, neuromotor and neurodevelopmental assessment was performed by a developmental pediatrician. All infants were assessed by the same pediatrician to avoid confounding factors in assessment. Neuromotor Assessment was done using the Amiel-Tison method and developmental assessment was done using Developmental Assessment Scale for Indian Infants (DASII) [21–23].
The DASII is based on Bayley Scales of Infant development (BSID). This simple test consists of 67 motor and 163 mental measures (Table 1). It assesses the development in infants between birth to 30 months of age and provides a measure of motor and mental development quotients (DQ) as motor DQ and mental DQ. The test was validated against BSID and norms were standardized at the study center (KEM Hospital, Pune, India). The norms of developmental skills are therefore described as the age at which 3%, 50% and 97% of normal infants acquire those skills. 50% pass for age is used to calculate the DQs [24–26].
Developmental assessment scale for Indian infants
Developmental assessment scale for Indian infants
Reference: Spearman’s hypothesis and racial differences on the DAS-II.
The statistical significance of difference between the two groups (Death and Disability) for categorical variables was obtained using Chi-square or Fischer’s exact tests while that for continuous variables was obtained using t-test or Mann Whitney U test, pending distribution. To obtain multivariate independent determinants of outcome measures of mental and motor DQs were used. Binary stepwise logistic regression analysis was performed by using parameters that were significantly different in presence of primary outcome of death or disability at 1 year of age. A p-value less than 0.05 was used to define statistical significance. Statistical analysis was performed using SPSS version 11.5 for Microsoft Windows (IBM, Chicago, IL, USA). Correction for multiple comparisons was not conducted.
Results
A total of 1924 neonates were admitted to the NICU during the study period. Figure 1 describes the distribution of newborns with HIE and the final outcome. The study group had 46 newborns, 43 (93.5%) classified as HIE stage II and three (6.5%) as Stage III. One-year follow-up was obtained for 95.83% of survivors. The baseline characteristics were similar in both groups with outcomes as death or moderate to severe disability, except for Apgar scores at 1, 5 and 10 minutes of age (Table 2).
Distribution of patients during the study enrollment period of November 2007 to November 2008.
Characteristics of the study group
Overall 47 out of 76 newborns (62%) died or had moderate to severe disability at 1 year of age. Table 3 demonstrates univariate analysis of perinatal variables in presence or absence of death, or moderate to severe disability. Parameters included are HIE stages (II and III), Apgar score at 5 min, pH, base deficit, onset of seizures and use of multiple anti-seizure medications (one, two or three). Poor maternal education or socioeconomic status did not affect the developmental outcome.
Correlation of perinatal risk factors with neuromotor development at the age of 1 year
Correlation of perinatal risk factors with neuromotor development at the age of 1 year
Hearing was evaluated with OAE and BERA. It was seen that all neonates with HIE stage III had abnormal OAE and BERA, while 17% of those with moderate HIE had abnormal OAE. Thirteen newborns (33%) with moderate HIE and all those with severe HIE were found to have cerebral palsy (CP) at 1 year of age. With severe HIE, the commonest type of CP was spastic quadriplegia (n = 2) while the other types were choreoathetoid (n = 1) and spastic diplegia (n = 1). There were 9 (23%) CP cases among neonates with moderate HIE and the subtypes were spastic tetraplegia (n = 2), hemiplegia (n = 2), diplegia (n = 2), choreoathetoid (n = 1) and hypotonic (n = 1).
Microcephaly was present among 38% neonates with moderate HIE and 66.7% of those with severe HIE. Early seizures or need of multiple anticonvulsants did not affect mortality but were significantly associated with adverse long-term outcome. Among neonates with early (<12 hours of life) seizures, 70% had a motor DQ < 85 and 65% were associated with mental DQ < 85. Of those on two anticonvulsants, 91% and 82% had a subnormal motor and mental development respectively.
Perinatal course and neuromotor outcome
49% neonates with moderate HIE and all of those with severe HIE had significant motor delays when assessed at 1 year of age, using DASII. Significant factors related to poor neuromotor score (<85th percentile) included arterial blood gas, pH≤7.1, base deficit ≤– 16, early onset of seizures (≤12 hours), use of multiple anti-seizure drugs and the need for artificial ventilation. 55% newborns with an Apgar score <5 at 1 minute and 75% of those with a score <5 at 5 minutes had a subnormal motor outcome.
Perinatal course and cognitive outcome
41% neonates with moderate HIE and all of those with severe HIE had subnormal cognitive outcome (<85th percentile). HIE stages (II and III), Low Apgar score at 5 minutes, pH ≤ 7.1, base deficit ≤ 16, early seizures, use of multiple anticonvulsants and requirement of ventilation were compared with outcome of death or disability using binary stepwise logistic regression. Table 4 shows the distributions of risk factors according to outcome of death or disability at 1 year of age. Risk factors included are HIE stages (II and III), Apgar score at 5 minutes, pH, base deficit, seizure duration, use of multiple anticonvulsant medications and use of ventilation. Arterial blood gas pH, base deficit, use of multiple anticonvulsant medications and use of ventilation demonstrated significant difference in presence or absence of death, or moderate to severe disability.
The distribution of selected risk factors according to groups of mental age (1-year)
The distribution of selected risk factors according to groups of mental age (1-year)
Baseline anthropometric characteristics at birth and at discharge are described in Table 5. Higher weight gain at one year of age was observed in newborns that survived without moderate to severe disability. Anthropometric parameters demonstrated that newborns with moderate to severe disability had significantly lower weight and weight gain at 1 year of age compared to those with no or mild disability. Newborns with moderate to severe disability also had a significantly smaller head circumference and lesser growth of head circumference. Table 6 shows the comparison of anthropometric parameters with no to mild disability versus moderate to severe disability. Weight at 1 year of age, weight difference, head circumference at 1 year of age and head circumference difference were significantly lower in presence of moderate to severe disability. Per capita income was not significantly different between the groups. Table 7 demonstrates bivariate analysis of anthropometric parameters in presence or absence of death, or moderate to severe disability. The Binary logistic regression analysis demonstrates that pH ≤ 7.1 at birth (odds ratio: 11.835, 95% CI 2.273–61.629, p = 0.003) and weight gain at one year (odds ratio 1.001, 95% CI 1.000–1.002, p = 0.03) were significantly associated with death and disability (Table 7).
Composite Outcome for death or neurodevelopmental impairment
Composite Outcome for death or neurodevelopmental impairment
Anthropometric parameters and outcomes
Binary logistic regression analysis of death or moderate to severe disability
Most studies in the developed world have documented outcomes with an emphasis on severe neurological morbidities, such as CP and mental retardation. Neurological function, measured with neurological examination at different ages, has been the most frequently used outcome measure in previous studies of HIE [27, 28]. Unfortunately, the burden of HIE and its morbidity is much more common in developing countries where few studies have been done [29–31]. We have looked at the neurodevelopmental outcomes at 1 year of age in moderate and severe HIE.
Sarnat and Sarnat reported that infants with HIE stage III and those with signs of HIE stage II for >7 days were found to have significant neurological impairments [17]. Finer et al. showed a significant association between HIE staging and outcome at 27 months of age [32]. In a previous study by Aggarwal et al. [33] at the study center, all newborns with severe HIE and 48% of those with moderate HIE had delayed milestones. The severity of HIE was an independent determinant for delayed motor as well as mental development at 1 year of age. Nutrition is a significant predictor of neurodevelopmental outcome in infants with brain injury. Patients with malnutrition are at two to four times increased risk of death and disability [34]. Improving nutrition in early phases improves long-term outcomes in preterm infants [35]. A study has shown that higher calorie and protein content delivered parenterally in patients with neuronal injury can significantly improve the neurological recovery [36]. Therefore, earlier and better nutritional support plays a vital role in the outcome of newborns with HIE.
Newborns with significant brain injury require increased nutritional support in the immediate postpartum period to help with brain growth as depletion of nutrients in these newborns can exacerbate neurological impairments [16]. Brain-injured neonates show a pattern of early growth failure where body mass is lost before length and brain growth is compromised, which indicates that nutritional needs are not sufficiently met. Therefore, these neonates require above-recommended nutrition to not only meet their needs for normal maintenance, but also to catch up with their deficits [37, 38]. Newborns with brain injury have slow growth due to suboptimal nutrition secondary to recurrent aspiration, impaired motor coordination, inadequate intake, increased energy expenditure and deficiency of growth hormones [39–41]. Also, suboptimal nutrition is associated with weakness of respiratory musculature, impaired cough reflex and pneumonia [42, 43]. Several clinical trials have shown that newborns with brain injury require optimal micro-and macro-nutrients to alter the structure and function of the brain in response to injury [44–46]. A small, non-randomized clinical trial showed that neonates with brain injury had improved growth and motor outcome when they received six months of additional nasogastric tube feeds [47]. Many nutritional deficiencies can alter brain growth during an early period of life, but protein malnutrition remains a major cause of low birth weight in developing countries [48]. Growth in the early phase of life is associated with better long-term outcome. Anthropometric parameters such as weight at birth, early neonatal weight gain and post-discharge head circumference are associated with cognitive development [35]. Preterm neonates are at high risk of neurodevelopmental abnormalities and improving early neonatal growth may improve long-term outcome in them [49, 50]. Studies have shown that the neonates with improved weight gain and head circumference growth from birth to discharge had better motor and neurological outcomes later in life and a lower incidence of neurological abnormalities [35, 52]. Intensive early nutritional support and increased energy supply during an early period of life is not only associated with better overall as well as head circumference growth but also with improved neurological outcomes, until six years of life [53, 54]. An adequate diet enriched with protein is essential for brain growth and can better manipulate recovery from brain injury. Optimal protein supplementation is safe and well tolerated in brain-injured neonates and can result in better motor and cognitive outcomes [55–58]. Poor growth is extremely common in brain-injured neonates and is highly associated with poor neurological sequelae. This study re-emphasizes the use of adequate nutrition to improve neurodevelopmental outcome in neonates with HIE.
The most important contribution of this study is the demonstration of significant positive correlation between growth parameters with improved outcomes. Neonates with moderate to severe HIE are at increased risk of poor neurological development. These neonates have poor growth due to poor nutritional intake, secondary to motor problems. Several studies have shown that optimal nutrition is associated with better growth parameters [40, 59–62]. This study re-emphasizes the use of nutrition to help improve neurodevelopmental outcomes. Better nutrition is least controversial and the most time-tested way to improve outcome among newborns at significant risk of developmental delay.
In conclusion, growth parameters are important considerations to be monitored among newborns with HIE. It is further important to study the effect of improved nutritional support on neurodevelopmental outcomes among newborns who received TH.