Home Print this page Email this page Small font sizeDefault font sizeIncrease font size
Users Online: 230
About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Advertise Login 

 Table of Contents  
Year : 2021  |  Volume : 10  |  Issue : 3  |  Page : 182-186

A standardized approach to routine cranial ultrasonography in preterm infants: Improved neuromotor outcome predictability at 2 years and communication with parents

1 Department of Obstetrics, Sheffield Teaching Hospitals, Sheffield, England
2 Department of Neonatal Medicine, The James Cook University Hospital, Middlesbrough, England

Date of Submission30-Nov-2020
Date of Decision10-Jun-2021
Date of Acceptance11-Jun-2021
Date of Web Publication28-Jul-2021

Correspondence Address:
Shalabh Garg
Department of Neonatal Medicine, The James Cook University Hospital, Marton Road, Middlesbrough TS4 3BW
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jcn.jcn_192_20

Rights and Permissions

Aim: The aim of this study is to evaluate the predictive value of routine cranial ultrasonography in preterm infants under 32 weeks' gestation with their neuromotor outcome at 2 years. Methods: This was a prospective, single-center, cohort study of preterm infants under 32 weeks' gestation. Each infant had an early (within 1 week of birth) and late (at 6 weeks postnatal age) cranial ultrasound scan performed. Each infant's scan result was independently assessed for the presence of major cranial abnormalities, such as intraventricular hemorrhage Grade 3 or 4, cystic periventricular leukomalacia, and porencephalic cyst. All surviving infants who returned for follow-up at 2 years' corrected age had their neuromotor development assessed using the Bayley Scales of Infant Development. The predictive value of major cranial abnormalities for the neuromotor delay was derived. Results: A total of 134 infants were included over 2 year. Of the 89 children with no major abnormality, 78 did not have a significant neuromotor delay, giving a negative predictive value of 87.6%. Of the six children with major abnormalities, two had a significant neuromotor delay, giving a positive predictive value of 33.3%. Conclusions: The absence of a major abnormality on quality-controlled routine cranial ultrasound scan in preterm infants under 32 weeks' gestation appears to be a good predictor of no significant abnormality in neuromotor development at 2 years of age.

Keywords: Cranial ultrasound scan, intraventricular hemorrhage, neuromotor development, periventricular leukomalacia, prematurity

How to cite this article:
Lal N, Garg S, Lal M. A standardized approach to routine cranial ultrasonography in preterm infants: Improved neuromotor outcome predictability at 2 years and communication with parents. J Clin Neonatol 2021;10:182-6

How to cite this URL:
Lal N, Garg S, Lal M. A standardized approach to routine cranial ultrasonography in preterm infants: Improved neuromotor outcome predictability at 2 years and communication with parents. J Clin Neonatol [serial online] 2021 [cited 2022 Aug 14];10:182-6. Available from: https://www.jcnonweb.com/text.asp?2021/10/3/182/322529

  Introduction Top

The survival of very preterm infants (<32+0 weeks) in general and extremely preterm infants (<28+0 weeks) in particular has continued to improve over the past three to four decades and yet there is little evidence to support similar improvements in their neurodisability and cerebral palsy rates.[1],[2],[3],[4],[5] Up to one-third of preterm infants under 32 weeks' gestation can go on to develop neurodisability during childhood.[6] The burden of neurodisability and cerebral palsy in preterm infants remains a major issue for these children, their parents and families, as well as resource-heavy for healthcare services.[7]

Cranial ultrasound scans (USS) are used routinely and have become the standard of care in preterm infants to screen for and monitor intracranial pathology, including intraventricular hemorrhage (IVH), cystic periventricular leukomalacia (PVL), and porencephalic cysts. It has proven to be a sensitive diagnostic tool without the risks associated with computed tomography, including radiation and sedation.[8] Another key advantage of cranial ultrasonography is that it can be performed at the bedside with relative ease within the neonatal intensive care unit (NICU). The majority of germinal matrix hemorrhage (GMH)-IVH develops within the first few postnatal days and is detected easily with a cranial ultrasound scan. White matter injury (WMI), particularly PVL tends to evolve over 3–4 weeks and hence may only be detected with a late cranial scan. Noncystic PVL or diffuse WMI is not detectable using cranial scan and usually require magnetic resonance imaging (MRI) imaging at term.[9] Brain imaging using MRI with 11.7 teslas allows for the better delineation of WMI of all types and severity but remains a research tool and not available in day-to-day clinical practice.[10] In addition to this, cranial scans are operator-dependent, and there is a significant variation in the timing, frequency, and reporting of these scans.[11]

Previous studies have demonstrated the limitations of cranial scan in predicting neurodevelopmental outcomes at 2 years. Grade 3 and 4 IVH on cranial ultrasound scan have been associated with major neurodevelopmental delay; however, a wide range of neurodevelopmental outcomes are prevalent in babies with periventricular echodensities, and in some cases, even those with severely abnormal scan findings continue to have largely normal neurodevelopmental outcomes.[12] In cases of periventricular hemorrhagic infarction in preterm infants, it has been demonstrated that the likelihood of an abnormality in gross motor function (among other neurodevelopmental outcomes) at a median age of 30 months can be correlated with a cranial ultrasound severity score.[13]

The uncertainty regarding the predictive value of bedside cranial scan poses a major challenge for clinicians during their conversations with parents regarding the overall prognosis and neurodevelopmental outcomes in preterm infants with intracranial pathology.


The aim of this study is to establish the predictive value of high-quality cranial ultrasonography performed and reported in a systematic and structured way with the neuromotor outcomes at 2 years of age for preterm infants under 32 weeks' gestation.

  Methods Top

This study was conducted at a tertiary level NICU in England with an annual delivery rate of approximately 5000 babies. All preterm infants of <32 weeks' gestation receiving intensive care born between January 2011 and December 2012, as recorded in the neonatal database, were eligible, including those booked elsewhere and subsequently transferred to this center in utero or ex utero. As per the unit's protocol, each eligible infant had both an early (within the 1st week of postnatal age) and late cranial scan (6 weeks of postnatal age) performed by one of the two pediatric sonographers, and each scan was reported by one consultant pediatric radiologist.

The results of both the early and late cranial ultrasound scans were collated, and a binary grading of either 0 (no abnormality present on either scan) or 1 (abnormality present on either scan) was allocated to each infant. The predefined criteria for an abnormality present on a scan were: An IVH with ventricular dilatation or parenchymal extension, cystic PVL, or the presence of a porencephalic cyst.[14] The grading was completed independently by two consultant neonatologists, who reviewed each written cranial scan report and were blinded to the 2-year neurodevelopmental outcomes of the infant at the time of scan grading. Each USS grading was compared and any disagreements discussed until a consensus was reached. Inter-rater agreement was determined using a kappa statistical score.[15]

Similarly, a grading of either 0 (no major neuromotor abnormality at 2 years) or 1 (major neuromotor abnormality present at 2 years) was allocated to each subject when the outcome of interest was measured. In most cases, the 2-year neuromotor outcome was measured by the Bayley Scale of Infant and Toddler Development, Third Edition.[16] In a few cases, where this was not possible, the 2-year motor outcome was based on a follow-up clinic report generated from a multidisciplinary outpatient clinic appointment or a parental questionnaire.[17] The criteria for an abnormal 2-year neuromotor outcome were as follows: Bayley Score of Infant Development (BSID) Motor Score <70, BSID Motor Score 70–85 plus not walking/not sitting, >12 months of motor developmental delay if BSID scores not available, or any diagnosis of cerebral palsy. The predefined criteria for cranial scan abnormalities are illustrated in [Table 1]. Similarly, the grading for 2-year neuromotor abnormality is described in [Table 2].
Table 1: Cranial ultrasound grading criteria

Click here to view
Table 2: Two years neuromotor outcome grading criteria

Click here to view

The binary grading of both the cranial ultrasound scan and 2-year neurodevelopmental outcomes was collated in a 2 × 2 contingency table [Table 3]. Sensitivity, specificity, positive, and negative predictive values were calculated from a 2 × 2 contingency table with cranial scan abnormalities and neuromotor outcomes in the study cohort.
Table 3: Correlation of combined cranial USS result (week 1 and week 6) to the 2 years neuromotor outcomes

Click here to view

This study involved analysis of anonymized patient information as recorded prospectively in the neonatal database and hence ethics approval was not required.

  Results Top

There were 8838 births during the 2-year study period and a total of 943 NICU admissions. Of these, 257 infants were under 32 weeks' gestation. All infants alive and with 2-year neurodevelopmental outcomes were included in this analysis [Figure 1]. A total of 134 babies were included over 2-year. Babies who either died or did not have a 2-year outcome assessed were excluded from the analysis. The 2-year outcomes were available for 95 of 134 included children (71% of the study cohort).
Figure 1: Flow diagram of the study population

Click here to view

The mean gestational age was 27 weeks (range, 23–31 weeks; standard deviation ± 1.82) and mean birth weight was 1020 g (range, 490–1690 g; standard deviation ± 265). The male to female ratio was 1.31 (54:41). A total of 13% of included infants were small for gestational age with birth weight <10th centile for gestation and gender. The rates of other comorbidities (bronchopulmonary dysplasia, patent ductus arteriosus, necrotizing enterocolitis, retinopathy of prematurity, etc.,) were comparable to other tertiary neonatal units and the national average. There was no difference in these comorbidities in babies with normal or abnormal scans.

Early and late cranial scans of all 134 babies were rated independently. There was a near-absolute interobserver agreement (κ =0.99). Of the 89 infants with no major abnormality on a routine cranial scan, 78 did not have a significant abnormality in neuromotor development at 2 years, giving a negative predictive value of 87.6%. Of the six infants with an abnormal cranial scan, two had abnormal neuromotor outcomes at 2 years, giving a positive predictive value of 33.3%.

The results of the correlations of cranial scan and the outcome are shown in [Table 3]. The sensitivity of the cranial scan was quite low (15.4%) whereas the specificity was high (95.1%).

  Discussion Top

This study confirms that the absence of a major abnormality in cranial scan in preterm infants under 32 weeks' gestation appears to be a good predictor of the absence of significant neuromotor delay at 2 years of corrected age. It offers clinician's important insight into prognostication based on the cranial scan when offering an explanation to parents.

The key strengths of this study are that it relates to the predictive value of routinely performed cranial scan for preterm infants. After all, a cranial scan is the standard of care in the current clinical practice. Clinicians may have reservations about normal cranial scans mainly due to its inherent limitations as an imaging modality. This study highlights that if performed and reported in a standardized way its predictive value may be high, particularly for neuromotor outcomes. All scans in our study population were performed in a standardized and protocolized way by the two trained pediatric sonographers and were reported by one consultant pediatric radiologist using a standard reporting template. Our previous study highlighted a huge variation in the way, the routine cranial scans are performed in the NICUs in England[11] which is one of the reasons that clinician reliance on these scans is reserved. Hence, in our experience, despite being the most commonly used neuroimaging modality, the clinicians tend not to put much emphasis on the outcome prediction based on results of cranial scans, especially if no major abnormality is found [Table 1]. Having a standardized approach to the scans reducing the subjective variations, the clinicians would feel more confident about the accuracy of the results which is reflected in high inter-rater agreement in our study unlike what is reported previously.[18] There was a high follow-up rate at 2-years corrected age and the majority of children were assessed using the BSID-III tool and we do not feel the addition of babies who were lost to follow-up (mostly out of area babies who were transferred back to their units) would make the difference to the results. Demographic characteristics of the study population are typical of any level 3 NICU in the United Kingdom. The reported rate of cranial scan injury is consistent with the current rates of major cranial scan abnormalities in high-income countries.[19] Hence it is reasonable to expect that the findings of this study relating to outcomes at 2 years of age are likely to be representative of other tertiary neonatal units.

This study indeed has a few limitations. It does not include the whole spectrum of common intracranial injuries in the preterm infants, such as GMH, IVH without ventricular dilatation, transient or more persistent echodensities in the brain parenchyma, or noncystic PVL and are considered to be normal in our study for predictive values. While the benefits of the cranial scan are clear in that these can be performed within the NICU by the bedside and without any need for sedation or general anesthetic or transfer to the radiology department as well as the risk of radiation exposure, these may not identify the whole spectrum of white matter injuries, such as cystic PVL with smaller cysts below 2 mm in size or noncystic PVL or diffuse white matter injuries. Such lesions are better detected using MRI scans at term equivalent age (9). This is by no means commonplace for all preterm infants under 32 weeks' gestation but rather limited to far fewer centers. MRI necessitates sedation and general anesthetic as well as transfer to the radiology department. Even when this is undertaken for the selected few infants at term equivalent age, it may not detect transitional changes, such as diffuse excessive high signal intensity (DEHSI). Even if DEHSI is seen on term equivalent MRI, it does not correlate well to long-term outcomes.[20] High-resolution MRI using 11.7T magnetic fields, although infinitely better at detecting more subtle and diffuse white matter injuries, remains mainly a research tool and not available in day-to-day clinical care.[10] Hence, it is important that a commonly performed investigation like cranial USS should be able to enhance our confidence in giving parents some discrete information. On the contrary, the clinicians generally play down the normal cranial USS findings to avoid parents getting too optimistic. It is, therefore, justifiable to do larger studies to assess the predictive value of high-quality standardized cranial ultrasound scanning with minimal subjective variations in performing and reporting the cranial USS. This will also guide us further in highlighting the neuromotor correlation of some of the benign cranial lesions in preterm babies. The other benefit of such a study would be to study if there are any differences in the neuromotor outcome on the basis of associated comorbidities which may or may not cause any specific cranial USS changes.

Another limitation of this study is that it is mainly restricted to neuromotor development and does not have the scope of all other domains of neurodevelopmental outcomes such as language development, cognition and learning, emotional and behavioral development, and social functioning. This is because neuromotor deficits usually appear in the 1st year of life but cognitive and behavioral impairments evolve slowly from early childhood to adolescence.[5] Sensory outcomes can be more complicated to ascertain at 2 years' and certainly, maybe due to other forms of more subtle brain injuries that are less likely to be detected in the cranial scan. Moreover, sensory outcomes such as deafness or blindness have other likely causations outside of brain injuries, such as intracranial infections including congenital infections, drug toxicities, renal impairment, and retinopathy of prematurity.[5]

  Conclusions Top

Based on the findings of this study in preterm infants under 32 weeks' gestation, the absence of a major abnormality on cranial ultrasound scan performed and reported in a standardized and structured way appears to be a good predictor of no significant abnormality in neuromotor development at 2 years of age corrected for prematurity. The reduction in subjective variations in imaging and reporting will certainly help clinicians during their communication with parents about cranial scan findings.


Prof. Win Tin for his helpful advice and sharing his insight into cranial scan outcome data from the Northern Region in the UK (1990-91 cohorts).

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Santhakumaran S, Statnikov Y, Gray D, Battersby C, Ashby D, Modi N, et al. Survival of very preterm infants admitted to neonatal care in England 2008-2014: Time trends and regional variation. Arch Dis Child Fetal Neonatal Ed 2018;103:F208-15.  Back to cited text no. 1
Draper ES, Manktelow BN, Cuttini M, Maier RF, Fenton AC, Van Reempts P, et al. Variability in very preterm stillbirth and in-hospital mortality across Europe. Pediatrics 2017;139:e20161990.  Back to cited text no. 2
Moore T, Hennessy EM, Myles J, Johnson SJ, Draper ES, Costeloe KL, et al. Neurological and developmental outcome in extremely preterm children born in England in 1995 and 2006: The EPICure studies. BMJ 2012;345:e7961.  Back to cited text no. 3
Draper ES, Zeitlin J, Manktelow BN, Piedvache A, Cuttini M, Edstedt Bonamy AK, et al. EPICE cohort: Two-year neurodevelopmental outcomes after very preterm birth Arch Dis Child Fetal Neonatal Ed 2020;105:350-6.  Back to cited text no. 4
Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008;371:261-9.  Back to cited text no. 5
Trittmann JK, Nelin LD, Klebanoff MA. Bronchopulmonary dysplasia and neurodevelopmental outcome in extremely preterm neonates. Eur J Pediatr 2013;172:1173-80.  Back to cited text no. 6
Colvin M, McGuire W, Fowlie PW. Neurodevelopmental outcomes after preterm birth. BMJ 2004;329:1390-3.  Back to cited text no. 7
Rennie JM. Neonatal Cerebral Ultrasound. Cambridge, UK: Cambridge University Press; 1997.  Back to cited text no. 8
de Vries LS, Volpe JJ. Value of sequential MRI in preterm infants. Neurology 2013;81:2062-3.  Back to cited text no. 9
Riddle A, Dean J, Buser JR, Gong X, Maire J, Chen K, et al. Histopathological correlates of magnetic resonance imaging-defined chronic perinatal white matter injury. Ann Neurol 2011;70:493-507.  Back to cited text no. 10
Hosenie AS, Garg S, Tin W. Routine cranial ultrasound scanning in tertiary neonatal units. Arch Dis Child Fetal Neonatal Ed 2011;96:Fa27.  Back to cited text no. 11
Bennett F, Silver G, Leung E, Mack L. Periventricular echodensities detected by cranial ultrasonography: Usefulness in predicting neurodevelopmental outcome in low-birth-weight preterm Infants. Pediatrics 1990;85:400-4.  Back to cited text no. 12
Bassan H, Limperopoulos C, Visconti K, Mayer DL, Feldman HA, Avery L, et al. Neurodevelopmental outcome in survivors of periventricular hemorrhagic infarction. Pediatrics 2007;120:785-92.  Back to cited text no. 13
Schmidt B, Roberts RS, Davis P, Doyle LW, Barrington KJ, Ohlsson A, et al. Caffeine therapy for apnea of prematurity. N Engl J Med 2006;354:2112-21.  Back to cited text no. 14
Viera AJ, Garrett JM. Understanding interobserver agreement: The kappa statistic. Fam Med 2005;37:360-3.  Back to cited text no. 15
Bayley N. Bayley Scales of Infant and Toddler Development: Administration Manual. 3rd. United States of America: Psychorp; 2006.  Back to cited text no. 16
Lal M, Tin W. Measuring perinatal outcomes – Why, when, and how: A British perspective. Neoreviews 2012;13:e515-26.  Back to cited text no. 17
Hagmann CF, Halbherr M, Koller B, Wintermark P, Huisman T, Bucher HU, et al. Interobserver variability in assessment of cranial ultrasound in very preterm infants. J Neuroradiol 2011;38:291-7.  Back to cited text no. 18
Stoll BJ, Hansen NI, Bell EF, Walsh MC, Carlo WA, Shankaran S, et al. Trends in care practices, morbidity, and mortality of extremely preterm neonates, 1993-2012. JAMA 2015;314:1039-51.  Back to cited text no. 19
Mürner-Lavanchy IM, Kidokoro H, Thompson DK, Doyle LW, Cheong JL, Hunt RW, et al. Thirteen-year outcomes in very preterm children associated with diffuse excessive high signal intensity on neonatal magnetic resonance imaging. J Pediatr 2019;206:66-71.e1.  Back to cited text no. 20


  [Figure 1]

  [Table 1], [Table 2], [Table 3]


Similar in PUBMED
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded80    
    Comments [Add]    

Recommend this journal