Article, Pediatrics

Identifying infant hydrocephalus in the emergency department with transfontanellar POCUS

a b s t r a c t

Hydrocephalus carries significant morbidity in the infant population. Although clinical symptoms are often non- specific, hydrocephalus is easily identified using transfontanellar sonography. In this review, we provide the emergency physician with a succinct overview of infant hydrocephalus and the Point-of-care ultrasound technique for identification of this pathology.

(C) 2018

Introduction

At a rate of roughly 1 in every 1000 live births, infant hydroceph- alus is not uncommon, and is much higher in Developing countries [1]. It is the condition most frequently treated by pediatric neuro- surgeons and costs the United States 2 billion dollars in health ex- penditure every year. Broadly, infant pediatric hydrocephalus is categorized as congenital or acquired. Congenital causes account for the majority of infant hydrocephalus in the US, presenting at birth or developing soon after, while acquired causes are more com- mon globally [2] (Table 1).

Initial signs of hydrocephalus can be subtle including fussiness, poor feeding, vomiting, a full fontanelle and splaying of sutures. Late findings of hydrocephalus in infants can present more dramatically with apnea, bradycardia, hypertension, lethargy, seizures, dilated scalp veins, macrocephaly and the ‘setting-sun’ sign. However, many of these listed signs and symptoms are non-specific and require simultaneous investi- gation of other causes such as sepsis, in tandem with an evaluation for hydrocephalus.

* Corresponding author at: 33 West State Avenue, Phoenix, Arizona 85021, United States of America.

E-mail addresses: [email protected] (S. Subramaniam), [email protected] (A.E. Chen), [email protected] (A. Khwaja), [email protected] (R. Rempell).

In the acute setting, emergency physicians often choose to obtain computed tomography of the head to evaluate for intracranial pathol- ogy, including hydrocephalus. Historically however, radiologists have used ultrasound to screen, identify and monitor hydrocephalus [1,3- 5]. More importantly, if hydrocephalus is identified by ultrasound in the stable infant, an urgent MRI, rather than CT avoids radiation expo- sure, provides more detailed identification of cause and assists with sur- gical planning by neurosurgeons. For infants with Ventricular shunts for hydrocephalus, an open fontanelle can provide a rapid window to eval- uate the ventricles.

With the widespread use of Point-of-care ultrasound , iden- tification of infant hydrocephalus in the ED by emergency physicians is achievable and may help with triage, flow and management decisions. In this article, we review the technique of transfontanellar POCUS, land- marks needed to recognize normal ventricles, and the sonographic fea- tures in infant hydrocephalus.

General scanning technique

There is no specific patient position required when performing transfontanellar POCUS. However, imaging of the brain is optimal when the infant is calm, with minimal head movement. Simple cra- dling by the patient’s guardian, swaddling with a pacifier, warm ul- trasound gel and/or use of sucrose drops as an adjunct may all help the operator keep the infant relaxed and compliant with the POCUS interrogation.

https://doi.org/10.1016/j.ajem.2018.10.012

0735-6757/(C) 2018

Table 1

Congenital and acquired causes of hydrocephalus in infancy. Congenital hydrocephalus Acquired hydrocephalus

Anatomical malformations (primary aqueductal

stenosis, foramen of Munro atresia)

Post infectious hydrocephalus

(ventriculitis, meningitis)

Post hemorrhagic hydrocephalus Neoplastic/non-neoplastic masses (ependymoma, choroid plexus cyst)

Intrauterine infections (TORCH infections) Ventricular Shunt malfunction

Neural tube defects (Chiari malformations, myelomeningocele)

Traumatic injury

The ventricular system of the infant brain can be visualized with transducer frequencies ranging between 5 MHz to 10 MHz. This fre- quency range is achievable with the phased-array, endocavitary, neona- tal curvilinear, and even linear transducer [6-8]. In order to insonate the brain and visualize the ventricles, the footprint of the transducer is placed directly onto the middle of the open anterior fontanelle. The an- terior fontanelle acts as an acoustic window for this POCUS applica- tion and images produced are dependent on the size of the anterior fontanelle and degree of transducer footprint contact over the open window. Although the anterior fontanelle can be open up to 2 years of age, transfontanellar POCUS is more reliable in children up to ages 12-18 months, when the anterior fontanelle is still patent, and more limited thereafter [3]. Preferably, the depth of the ultra- sound beam should be adjusted to allow visualization of the base of the skull.

The infant brain is imaged in 2 planes, a coronal and sagittal view (Fig. 1). In the coronal plane, POCUS convention is to have the indicator marker point towards the right of the patient. The transducer is swept (fanned) in an anterior to posterior direction to visualize the brain in at least 3 coronal sections (anterior, middle and posterior views). In the sagittal plane, the indicator marker points to the anterior aspect of the head (face of the infant) and is swept to visualize the parasagittal left, parasagittal right, and mid- line of the brain.

Fig. 2. Sonographic features of the anterior coronal plane.

POCUS image anatomy acquisition

In the coronal plane, with the indicator marker to the patient’s right, the left side of the infant brain will be on the right side of the monitor. When the brain is imaged proficiently, the normal brain anatomy is virtually symmetric between left and right. Clip 1 demon- strates the images generated by sweeping the ultrasound beam ante- rior to posterior.

The anterior coronal section displays the frontal lobes divided by the interhemispheric fissure in the midline (Fig. 2). Other landmarks are the ‘bull’s horns’ sign of the sphenoid bone and the superior and inferior margins of the orbits. The middle coronal section allows vi- sualization of the body of the lateral ventricles, corpus callosum,

Fig. 1. Transfontanellar POCUS technique demonstrated. The left image demonstrates an ultrasound transducer with indicator marker to right of patient, allowing visualization of brain in the coronal plane. The right image has transducer with indicator marker pointing to the face of patient allowing visualization of brain in the sagittal plane.

Fig. 3. Sonographic features of the middle coronal plane. The left Sylvian fissure indicated on the image separates the parietal and temporal lobe.

Sylvian fissures, tentorium cerebelli, and the brainstem (Fig. 3). Typ- ically, the temporal horns of the lateral ventricles are not visible in the coronal plane. The posterior coronal section demonstrates the re- gion of the trigone or atrium of the lateral ventricles where the cho- roid plexus is most clearly visualized as a linear, oblique, hyperechoic structure bilaterally (Fig. 4).

In the sagittal plane, with the indicator marker facing the in- fants face, the left side of the monitor displays the anterior portion of the brain. Sweeping from left to right will bring the left ventricle, midline portion of the corpus callosum and right ventricle into view (Clip 2). The parasagittal left and right are very similar in a normal brain and usually clearly demonstrate the parietal lobe, lat- eral ventricle, thalamus and hyperechoic choroid plexus (Fig. 5). In infants with a wide-open fontanelle, the acoustic window allows the ultrasound wave to demonstrate the corpus callosum and third ventricle clearly in the midline sagittal plane (Fig. 6).

POCUS identification and interpretation of hydrocephalus

An infant with hydrocephalus is likely to have an anterior fontanelle that will allow ultrasound interrogation and visualization of the dilated ventricular system (Clips 3 and 4). There are several sonographic fea- tures that can be seen on POCUS.

Fig. 4. Sonographic features of the posterior coronal plane.

Fig. 5. Sonographic features of the parasagittal left plane.

An early sign of hydrocephalus is widening or ‘ballooning’ of the anterior horn of the lateral ventricles [9] (Fig. 7). Continued concentric dilation produces enlargement of the frontal horns with a rounded configuration, giving it a ‘Mickey Mouse ears’ ap- pearance (Fig. 8). The mean and maximum diameter of the nor- mal anterior horn of the lateral ventricle is 5 mm and 10 mm respectively. Typically, the 3rd ventricle can be visualized in the sagittal plane, although a diameter larger than 1 mm in the coronal plane is concerning for a dilated ventricular system [6]. Similarly, the temporal horn of the lateral ventricle is not visual- ized in a non-dilated ventricle but can be seen in hydrocephalus (Fig. 8). A temporal horn measurement of more than 3 mm qual- ifies the ventricle as dilated [10]. In the sagittal view the dilated body, anterior and posterior horn of the ventricle is obvious, with the choroid plexus usually appearing to dangle in the atrium of the ventricle (Fig. 9).

The bifrontal index is used in sonography as a ratio of the widest measurement of anterior horns to the widest measurement of the inner table of the skull. A ratio below 0.3 (Fig. 10) is reassuring but above 0.5 is significant and correlates to hydrocephalus [11,12] (Fig. 11). In severe hydrocephalus, significant dilation of the ventricles results in the appearance of thinning of the brain mantle.

Fig. 6. Sonographic features of the midline sagittal plane.

Fig. 7. Left ventriculomegaly found initially on POCUS of a 7-month-old infant with enlarging head circumference, subsequently confirmed with CT head imaging.

Overall, no one sonographic measurement is completely accurate to diagnose hydrocephalus and the constellation of Sonographic findings listed above is used to help with the diagnosis.

Discussion

Untreated hydrocephalus results in significant morbidity. Slow in- crease in intracranial pressure results in neurovascular ischemia, chronic inflammation and neuronal degeneration, impacting infant de- velopment and future Cognitive function [1]. In contrast, with rapidly progressing hydrocephalus, an acute increase in ICP presents an immi- nent risk of intracranial herniation and mortality. Ultrasound is an established modality in the radiology department to detect this entity. Infants with an open anterior fontanelle allow the ultrasound to achieve operating characteristics approaching a sensitivity of 100% and PPV of

Fig. 8. 8-Month-old infant presenting with new onset hydrocephalus. Clear dilation of the ventricular system is present. 1 – Lateral ventricle measures 38 mm, 2 – 3rd ventricle measure 20 mm, 3 – Right temporal horn measures 20 mm.

Fig. 9. – Same 8-month-old infant with dilated left ventricle seen in sagittal.

90% [13-16]. Ultrasound can also be used as a routine modality to mon- itor infants with recently placed ventriculoperitoneal (VP) shunts or when there is concern for shunt malfunction (Fig. 12) [2,17-19].

Point-of-care ultrasound has become a staple adjunct in emergency departments nationwide with both general emergency and pediatric emergency physicians utilizing it for a variety of diagnostic and proce- dural applications, impacting management of our patients. Pediatric emergency ultrasound fellowships have been established, and consen- sus guidelines have been published to provide a framework of POCUS applications in the pediatric population [20-22]. None have thus far commented on the potential scope of transfontanellar POCUS, but as pe- diatric emergency physicians become increasingly facile with POCUS, its application has begun to expand to include evaluation for infant hydro- cephalus [7,23].

This use of transfontanellar POCUS is unique to the infant with spe- cific complaints presenting to the ED – the vomiting child with a large head, the sleepy infant with a bulging fontanelle, the fussy infant with a VP shunt, or the neonate with a brief resolved unexplained event (BRUE). Each raises the question of hydrocephalus within the differen- tial diagnosis, and with a wide-open fontanelle, this condition is no lon- ger veiled to the emergency physician. The entire interrogation of the ventricular system of the infant brain can be performed rapidly, at the bedside, without need for sedation or ionizing radiation and fits well with the flow of the emergency department. Identification of hydro- cephalus at the point-of-care allows quick decision making for both the community and emergency tertiary care provider alike.

Pitfalls

The most obvious disadvantage to transfontanellar POCUS is its reli- ance on the size of the anterior fontanelle. The smaller the anterior fon- tanelle, the smaller the window to visualize the ventricular system reliably. Furthermore, as the infant’s skull size grows with age, the width of ultrasound beam is less likely to insonate the lateral and distal portions of the brain well. This is especially true for the posterior fossa, occipital horns, and third ventricle, resulting in less accurate interpreta- tions of these regions.

Mild ventricular asymmetry is occasionally observed in neonates and is an area of uncertain consequence. It more commonly involves the left ventricle and is more pronounced at the occipital horn. Head po- sition in this age group has some effect, but its influence diminishes with age [9]. The finding of mild asymmetry (difference of up to 5 mm) is of unclear clinical significance at this time. If the usual triangu- lar configuration (in coronal cross section) of the lateral ventricles are

Fig. 10. Bifrontal index in a 3-month-old infant without evidence of hydrocephalus is measured at 0.2. 1 – Widest measurement of the horns of the lateral ventricles (2.47 cm), 2 – Widest measurement between the inner table of inner skull (9.47 cm).

preserved along with no other features of hydrocephalus reviewed in Section 4, non-urgent comprehensive sonography of the brain can be arranged with radiology as follow up for the patient [24].

It is important to remember that when evaluating for VP shunt mal- function with POCUS, clear images of non-dilated or baseline sized lat- eral ventricles (when compared to prior imaging), do not rule out shunt malfunction. Although shunt malfunction is likely if there is inter- val increase in ventricle size, it is still possible in normal sized or slit-like ventricles.

Conclusion

Transfontanellar sonography is an established method of imaging the infant brain and has long been used by radiologists to identify hy- drocephalus. Emergency physicians are beginning to use this applica- tion, and are likely to find it most useful when assessing an infant

with a differential diagnosis that includes hydrocephalus. In this con- text, transfontanellar POCUS has the ability to provide the emergency physician with a binary answer to the condition of the infant ventricular system. More research on this specific POCUS application within the field of emergency medicine is needed to determine its full potential.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2018.10.012.

Sources of support

There was no source of support for this manuscript.

Prior presentations

None.

Fig. 11. Significant hydrocephalus with bifrontal ratio measured at 0.7. 1 – Widest measurement of the horns of the lateral ventricles (8.62 cm), 2 – Widest measurement between the inner table of inner skull (12.48 cm),

Fig. 12. Recently discharged 3-month-old presenting with fussiness 2 months after VP shunt placement. Transfontanellar POCUS revealed significant lateral ventricle dilation compared to previous CT brain. The left proximal shunt indicated in the image is seen as a hyperechoic structure with its tip ending in the right ventricle corresponding to CT brain performed after transfontanellar POCUS.

References

  1. Kahle KT, Kulkarni AV, Limbrick Jr DD, Warf BC. Hydrocephalus in children. Lancet 2016;387(10020):788-99.
  2. Wright Z, Larrew TW, Eskandari R. Pediatric hydrocephalus: current state of diagno- sis and treatment. Pediatr Rev 2016;37(11):478-90.
  3. Dincer A, Ozek MM. Radiologic evaluation of pediatric hydrocephalus. Childs Nerv Syst 2011;27(10):1543-62.
  4. Marchie TT, Ayara CO. Investigation of infant brain with or without hydrocephalous in our environment using anterior transfontanelle ultrasound scan. Niger J Surg 2013;19(1):7-12.
  5. Smith R, Leonidas JC, Maytal J. The value of head ultrasound in infants with macrocephaly. Pediatr Radiol 1998;28(3):143-6.
  6. Garcia R, Lederman H, Brandao J. Sonographic evaluation of cerebral ventricular sys- tem in healthy, full term infants aged 1-6 months. Radiol Bras 2011;44(6):349-54.
  7. Yaeger SK, Marin JR. An infant diagnosed with hydrocephalus by point-of-care ultra- sound. Pediatr Emerg Care 2017;33(4):287-9.
  8. Ecury-Goossen GM, Camfferman FA, Leijser LM, Govaert P, Dudink J. State of the art cranial Ultrasound imaging in neonates. J Vis Exp 2015;96:e52238.
  9. Brouwer MJ, de Vries LS, Pistorius L, Rademaker KJ, Groenendaal F, Benders MJ. Ul- trasound measurements of the lateral ventricles in neonates: why, how and when? A systematic review. Acta Paediatr 2010;99(9):1298-306.
  10. Jacob V, Kumar K. CT assessment of brain ventricular size based on age and sex: a study of 112 cases. J Evol Med Dent Sci 2013;2(50):9842-55.
  11. Bhat V, Bhat V. Neonatal neurosonography: a pictorial essay. Indian J Radiol Imaging 2014;24(4):389-400.
  12. Wilk R, Kluczewska E, Syc B, Bajor G. Normative values for selected linear indices of the intracranial fluid spaces based on CT images of the head in children. Pol J Radiol 2011;76(3):16-25.
  13. Siegel MJ, Patel J, Gado MH, Shackelford GD. cranial computed tomography and real- time sonography in full-term neonates and infants. Radiology 1983;149(1):111-6.
  14. Alves JG, dos Santos Figueira AC, Amaral F, Guimaraes A. Transfontanelle ultrasonog- raphy in infants: comparative study with cranial computed tomography. J Trop Pediatr 1997;43(3):182-3.
  15. Nawab K, Gul H, Al Haq A. Diagnositc accuracy of transfontanelle sonography as compared to computed tomography of brain in the detection of hydrocephalus. Kyber J Med Sci 2017;10(1):101-5.
  16. Al-Hassnaw N. Comparitive study between transfontanelle US and spiral CT scan in the diagnosis of congenital hydrocephalus. AL-Quadisiyah Med J 2008;4(5):143-53.
  17. Bober J, Rochlin J, Marneni S. Ventriculoperitoneal shunt complications in children: an Evidence-based approach to emergency department management. Pediatr Emerg Med Pract 2016;13(2):1-22 (quiz -3).
  18. Steinbok P, Boyd M, Flodmark CO, Cochrane DD. radiographic imaging requirements following ventriculoperitoneal shunt procedures. Pediatr Neurosurg 1995;22(3): 141-6.
  19. Mandiwanza T, Saidlear C, Caird J, Crimmins D. The open fontanelle: a window to less radiation. Childs Nerv Syst 2013;29(7):1177-81.
  20. Vieira RL, Hsu D, Nagler J, Chen L, Gallagher R, Levy JA. Pediatric emergency medicine fellow training in ultrasound: consensus educational guidelines. Acad Emerg Med 2013;20(3):300-6.
  21. Marin JR, Abo AM, Arroyo AC, Doniger SJ, Fischer JW, Rempell R, et al. Pediatric emergency medicine point-of-care ultrasound: summary of the evidence. Crit Ultra- sound J 2016;8(1):16.
  22. Marin JR, Lewiss RE. Point-of-care ultrasonography by pediatric emergency physi- cians. Policy statement. Ann Emerg Med 2015;65(4):472-8.
  23. Doctor K, Cohen JS. Diagnosis of an intraventricular cyst using point-of-care cranial ultrasound in the pediatric emergency department. Pediatr Emerg Care 2017;33 (3):216-8.
  24. Enriquez G, Correa F, Lucaya J, Piqueras J, Aso C, Ortega A. Potential pitfalls in cranial sonography. Pediatr Radiol 2003;33(2):110-7.