A study of neurosonogram abnormalities, clinical correlation with neurosonogram findings, and immediate outcome of high-risk neonates in Neonatal Intensive Care Unit.

BACKGROUND: Neonatal sonography of the brain is now an essential part of newborn care, particularly in high risk and unstable premature infants. Cranial ultrasound is the most available and easily repeatable imaging technique for the neonatal brain showing brain development and the most frequently occurring forms of cerebral injury in the preterm and terms. This study aims to assess the importance of cranial ultrasound as an investigatory modality for high-risk neonates and to find out the morphology of various cerebral lesions and correlate clinically.
METHODOLOGY: An observational correlation clinical study was conducted at Sardar Patel Medical College, Bikaner involving 100 high-risk neonates admitted to Neonatal Intensive Care Unit (NICU) who was subjected to neurosonography on selected days as per protocol. Perinatal details were recorded, and clinical examination with appropriate investigations was done. The cranial ultrasound was done, and morphology of various findings was studied and recorded. Clinical correlation with cranial ultrasound findings and follow-up was done.
RESULTS: On cranial ultrasound, 38% of neonates had abnormal findings. Twelve percent of these had evidence of intracranial bleed, 13% periventricular echogenicity, 7% had ventriculomegaly, 2% had cerebral edema, and 1% had leukomalacia. Three neonates had findings suggestive of simple cyst in middle cranial fossa, agenesis of corpus callosum, and choroid plexus cyst.
CONCLUSIONS: Cranial ultrasonography is the best point of care neuroimaging method available for high-risk neonates. It is critical as an investigatory modality in NICU and effectively documents morphology of cerebral damage.

Introduction

Cranial ultrasonography (CUS) has become an essential diagnostic tool in modern neonatology for depicting normal anatomy and pathological changes in the neonatal brain. In the neonate, many sutures and fontanelles are still open, and these can be used as acoustic windows to “look” into the brain.[1] Any neonate, regardless of birth weight, size or gestational age, who has a greater than average chance of morbidity or mortality, due to fetal, maternal or placental anomalies or an otherwise compromised pregnancy, especially within the first 28 days of life is categorized as high-risk neonate.[2] Cranial ultrasound plays an important role in assessing neurological prognosis of these high-risk infants.

It is cheap, easy to perform, noninvasive, and can be initiated at a very early stage, even immediately after birth. It can be repeated as often as necessary, and thereby enables visualization of ongoing brain maturation and the evolution of brain lesions. In addition, it can be used to assess the timing of brain damage.[1] As a result of ongoing development in ultrasonography, image quality is high nowadays, provided optimal settings and techniques are applied. Using additional acoustic windows can significantly augment the diagnostic power of cranial ultrasound. Scanning through the posterior and mastoid fontanelles, can help to detect lesions and structural malformations in cerebellum, brainstem, and posterior subcortical white matter? Imaging through the temporal window allows good views of the mesencephalon and brainstem.[3]

Cranial ultrasound (CUS) provides bedside imaging access to the neonatal brain. It detects most of the hemorrhagic, ischemic, and cystic brain lesions as well as calcifications, cerebral infections, and major structural abnormalities in preterm and full-term infants.[1] In seriously, ill neonates and in neonates with serious cerebral abnormalities, either congenital or acquired, it plays a role in decisions on continuation or withdrawal of intensive treatment. In neonates surviving with cerebral injury, it may help to optimize the treatment of the infant both during the neonatal period and thereafter.[4] If the quality of cranial ultrasound is good, timing is carefully chosen, proper transducers are used, and in the case of preterm birth, serial examinations are continued until term age, most diagnoses will not remain undetected, and the reliability and prognostic value of cranial ultrasound can be high. Serial cranial ultrasound examinations enable assessment of the onset of injury and the evolution of lesions.[1]

While early cranial ultrasound imaging concentrated on intraventricular hemorrhage (IVH) and cystic periventricular leukomalacia (PVL), the incidence of these pathologies has diminished considerably. Current studies aim more at detecting subtle white matter disease, assessing brain growth and maturation, and predicting neurodevelopment outcome from the cranial ultrasound.

Methodology

This study was conducted in the Department of Pediatrics, Sardar Patel Medical College, Pediatric PBM Hospital, Bikaner, Rajasthan, from September 2014 to September 2015. One hundred high-risk neonates admitted to NICU were selected as per the inclusion criteria on nonrandomized purposive sampling basis and were subjected to neurosonography on the selected days. If CUS revealed various findings, repeat neurosonogram were done to follow-up squeal if any. This study includes high-risk neonates with any of the following: Neonatal convulsions, birth asphyxia and hypoxic ischemic encephalopathy (HIE), respiratory distress, neonatal sepsis, preterm neonates, neonates born out of traumatic/instrumental labor, metabolic disturbances with convulsions, congenital malformation of central nervous system, and neural tube defects and excludes transient tachypnea of newborn, babies with only hyperbilirubinemia,babies >28 days.

Informed consent was obtained from the parents/guardian regarding the inclusion of the neonate in the study. Assessment of factors placing the neonate in a high-risk category was done taking detailed maternal history reviewing antenatal records. All perinatal details were recorded, and detailed clinical examination was done including anthropometric measurements. Vital parameters were recorded within 24–48 h of admission, and complete neurological examination was done during baby's stay in NICU. Gestational age was assessed as per modified Ballard's scoring method for all preterm neonates. Evaluation with baseline routine investigations (septic and metabolic workup) and lumbar puncture in case of neonatal convulsions and neonatal sepsis, chest X-ray in all respiratory distress cases was done. Cranial ultrasound of the high-risk neonate fulfilling the inclusion criteria was performed. Follow-up cranial ultrasound was done in case of findings revealed and for preterm neonates. Morphology of cranial ultrasound findings was studied and recorded, and clinical correlation with various findings on cranial ultrasound was done. Neonates were followed until recovery and discharge from NICU. Statistical analysis was done by using SPSS software (SPSS 15, Windows, Linux on Z systems, IBM Cororation, United States) methods.

Results

The incidence of cranial ultrasound abnormalities in high-risk neonates is 38% in the present study. There were 63% of male and 37% of female neonates, 62% preterm and 38% term high-risk neonates enrolled in the study. There was no significant correlation of incidence of abnormal cranial ultrasound findings in male and female. Correlation of gestational age with cranial ultrasound findings was statistically significant. Of these, 17% preterm were <32 weeks, 45% preterm were between 33 and 36 weeks, and rest 38% were term neonates. Sixty-five percent neonates were born via normal labor, and 35% via lower segment cesarean section for various reasons. In the correlation of perinatal risk factors with abnormal cranial ultrasound findings, there was no significant correlation with pregnancy-induced hypertension, antepartum hemorrhage, premature rupture of membranes, multiple births, and birth trauma.

In this study, 62% of high-risk neonates were preterm, 23% had evidence of perinatal asphyxia, 4% had birth trauma, 39% had evidence of neonatal sepsis, 42% had neonatal seizures in NICU stay, 42% clinically had respiratory distress, 8% had documentation of hypoglycemia, and 11% had hypocalcemia on investigatory record. There were no neonates with congenital malformations or neural tube defects. On cranial ultrasound, 38% of neonates had abnormal findings [Figure 1]. About 12% of these had evidence of intracranial bleed, 13% periventricular echogenicity, 7% had ventriculomegaly, 2% had cerebral edema, and 1% had leukomalacia three neonates had findings suggestive of simple cyst in middle cranial fossa, agenesis of corpus callosum, choroid plexus cyst.

Figure 1

Periventricular leucomalacia

There was statistically significant correlation between the presence of icterus (P = 0.038) and abnormal respiratory system (P = 0.013) on clinical examination and the presence of abnormalities on cranial ultrasound. There was statistically significant correlation between the incidence of positive C-reactive protein (CRP) (P = 0.000) and platelet (P = 0.028) on the investigation and the presence of abnormalities on cranial ultrasound.

Of all the high-risk neonates 75.4% had normal Apgar score, and 24.6% of these had abnormal findings on cranial ultra sound [Tables 1 and 2]. Of the 24.6% neonates who had abnormal Apgar scores, 52.9% with mild asphyxia, 50% with moderate asphyxia, and 80% with severe asphyxia had abnormal findings on cranial ultrasound of which evidence of intracranial bleed and periventricular echogenicity were most common [Figures 2 and 3]. Correlation between cranial ultrasound findings of neonates with perinatal asphyxia was statistically significant (P = 0.003) [Table 3]. All neonates in the present study who were found to have cerebral edema on cranial ultrasound had no evidence of perinatal asphyxia.

Table 1

Incidence of different cranial ultrasonography abnormalities in high-risk neonates

Table 2

Incidence of abnormal investigations in high-risk neonates with abnormal cranial ultrasonography findings and their correlation

Figure 2

(a) Grade 3 intraventricular hemorrhage. (b) Grade 4 intraventricular hemorrhage

Figure 3

(a) Right germinal matrix hemorrhage Grade 1. (b) Grade 2 intraventricular hemorrhage

Table 3

Incidence of abnormal findings on clinical examination in high-risk neonates with cranial ultrasonography findings and their correlation

Of the high-risk neonates with term gestation, 52.6% had normal and 47.4% had abnormal cranial ultrasound. Of the high-risk neonates with preterm gestation, 67.7% had normal and 32.3% had abnormal cranial ultrasound. Correlation between cranial ultrasound findings of neonates with sepsis, birth trauma, seizures, and prematurity was statistically significant (P = 0.015). There was no significant correlation with abnormal findings on cranial ultrasound and day of life it was done.

There was statistically significant correlation between gestational age of high-risk neonate and day of life CUS was done (P = 0.000) [Graph 1]. Of the neonates with gestational age <32 weeks having abnormal findings on cranial ultrasound, 41.2% had germinal matrix hemorrhage (GMH). Fifty-six percent of neonates enrolled had good outcome at the time of NICU discharge, 10% died, and 33% of neonates were relieved at the time of NICU discharge, 11% discharged from NICU for various reasons before clinical recovery (discharge against medical advice) [Graph 2].

Graph 1

Bar chart showing cranial ultrasonography findings correlating with and days of life and gestational age

Graph 2

Bar chart showing correlation of clinical outcome with various cranial ultrasound findings

Discussion

Daneman et al.[5] proved that cranial ultrasound remains an extremely useful modality for the evaluation of the full-term neonatal brain. The present study aims at proving the same. The incidence of cranial ultrasound abnormalities in high-risk neonates is 38% in the present study. Mercuri et al.[6] reported an incidence 20% of ultrasound abnormalities in apparently well neonates. De Vries et al.[7] assessed sequential high-resolution cranial ultrasound in high-risk preterm infants to predict cerebral palsy (CP) and found 79% of CP cases had major cranial ultrasound abnormalities. Badrawy et al.[8] showed in their study that 37% preterm had abnormal cranial ultrasound findings. Their study had 64% of male and 36% of female neonates. Of these, 9% were <30 weeks, 33% between 30 and 33 weeks, and 58% were between 34 and 37 weeks. In this study, 17% neonates were <32 weeks, 45% between 32 and 36 weeks, and the rest 38% term gestation. Chowdhury et al.[9] in their study, on fifty preterm neonates of which 58% were males and 42% of females, detected intracranial pathology in 12% of preterm and 6% of these had intracranial hemorrhage (ICH). In this study, 62% of high-risk neonates were preterm of which 32.3% had the intracranial pathology of which 11.3% had GMH and 3.2% other ICH. Soni et al.[10] in their study suggested that cranial ultrasound is sensitive and specific for the detection of various types of ICH. One hundred and eleven high-risk neonates were subjected to cranial ultrasound, one-quarter of these neonates developed ICH within 120 h of birth. Trounce et al.[11] reported that ultrasonic evidence of hemorrhage was evident within the first 7 days of life in 78% of neonates and 2nd week in 15% neonates. In this study, 12% high-risk neonates have evidence of intracranial bleed (GMH, IVH) all of which were picked up between 24 and 72 h of life. In this study, 32.3% preterm had abnormal findings on cranial ultrasound of which 11.3% had GMH. The maximum incidence of GMH 41.2% was found in preterm <32 weeks. Rehan et al.[12] concluded that frequency of IVH was found in 47.5% preterm neonates. In this study, 12% of high-risk neonates had intracranial bleeds of which 41.7% had other hemorrhages, and 58.3% had Grade 1 hemorrhages.

The overall incidence of abnormalities on cranial ultrasound in low birth weight neonates in the present study was 73.6%. Maria et al.[13] concluded cranial ultrasound remains an important bedside diagnostic tool for PVL. In the present study, one preterm neonate on regular follow-up cranial ultrasound developed findings suggestive of cystic PVL. In this study, 13% high-risk neonates had periventricular echogenicity findings by a neurosonogram. Of the neonates having periventricular echogenicity, 5.9% were early preterm, 4.4% were late preterm, and 26.3% were term neonates. Eken et al.[14] reported 26.4% full-term neonates with HIE had areas of increased echogenicity in the periventricular and/or subcortical white matter.

In this study, of all high-risk neonates 13% had periventricular echogenicity, and 2% had cerebral edema. Glass et al.[15] in their study reported that 3.8% preterm neonates had clinical seizures. The cranial ultra sound was abnormal in all these infants and was accurate for detecting IVH. In this study, of all high-risk neonates presenting with seizures, 47.6% had normal, and 52.4% had abnormal cranial ultrasound of which thalamic periventricular echogenicity was most commonly detected (30.9%) by cranial ultrasound. Van Houten et al.[16] reported that 59% of infants with congenital heart disease had a higher incidence of cranial ultrasound abnormalities. In this study, 23.7% of high-risk neonates with abnormal findings on cranial ultra sound had congenital heart diseases. Of the 38%, neonates which had abnormal cranial ultra sound, 5.2% had positive CRP, which was statistically significant correlation. In this study, there was statistically significant correlation with neonates having positive CRP, low platelet with abnormal cranial ultrasound findings. There was no correlation of hemoglobin, packed cell volume, and total leukocyte count, retic and positive culture, serum electrolytes, serum bilirubin, and cerebro spinal fluid analysis with abnormal cranial ultrasound findings.

Amess et al.[17] proved that neonatal cranial ultrasound revealed a subcategory of infants that were at high risk of developing epilepsy. Canadian Pediatric Society Statement in (2001),[18] concluded that it is important to consider the value of determining the need for routine screening cranial ultrasound examinations in NICUs. Follow-up of these high-risk neonates for the neurodevelopmental outcome was not a part of this study.

Conclusions

Neonatal care in India is advancing at an impressive phase at the level of the community as well as in tertiary care units. The concept of “survival” of the newborn has predictably given way to the importance of “intact survival” of the high-risk infant, prompting initiation of strategies to identify neurological subnormality at the earliest. The cranial ultrasound is an ideal tool for the primary screening of the neonatal brain. Despite the wide availability of ultrasound machines in the hospitals, the penetration of cranial ultrasound in Indian NICU's is still very little.

This study highlights the convenience and diagnostic efficiency of cranial ultrasound in high-risk neonates in NICU. It also emphasizes its use as a screening modality for preterm neonates influencing their neurodevelopmental outcome. High efficacy of cranial ultrasound in detecting the presence of brain damage and its evolution on regular follow-up guides clinical decisions and prognosis. This is particularly important in the anticipation of potential preventive, protective, and rehabilitative strategies for the management of critically ill newborn infants. The study concludes cranial ultrasound is critical as an investigatory modality in NICU and effectively documents morphology of brain damage. Cranial ultrasound mandatory screening for only preterm and low birth weight babies. In developing countries like India, cranial ultrasound screening available only advanced NICU centers. Our study was done in a peripheral tertiary care center. Our studies show abnormalities in term babies also. Cranial ultrasound should be mandatory screening for all NICU high-risk babies.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.