Size and Shape

During infancy the neurocranium is larger relative to the face than at any other time during normal growth. Ratios of the respective areas of the neurocranium in lateral projection are roughly 3 : 1 to 4 : 1 at birth, and they decrease to 2 : 1 to 2.5 : 1 by age 6 years. The bones of the calvarium lie in their incompletely mineralized membranous capsule; they are separated by broad strips of connective tissue that form the sutures and by patches of connective tissue, the fontanelles. The six constant, or major, fontanelles are located at the four corners of the parietal bones—two in the midline of the skull and two pairs on each side (Fig. 18-1). Accessory fontanelles may occur in several parts of the cranium but usually are located in the sagittal suture. The sutures and the synchondroses in the base are prominent in newborns but diminish in width during the first 2 to 3 months. Obliteration of the sutures does not begin until the second to third decades. Figures 18-1 through 18-4 and e-Figure 18-5 illustrate sutures, fontanelles, and synchondroses.

The sphenoid bone at birth consists of a single central mass composed of the body and the lesser wings and two symmetric lateral osseous masses, each of which is made up of a greater wing and a pterygoid process. The pituitary fossa in the body of the sphenoid bone tends to be round with smooth margins; the dorsum sella is short and blunt, and the clinoid processes are rudimentary. The angle between the body of the neonatal mandible and the ascending ramus in lateral projection is about 160 degrees; the relatively large bodies are separated in the midline by a prominent cartilaginous symphysis mentalis (see Fig. 18-3). Early calcification of teeth is seen in the fifth fetal month.¹

Components of the individual bones that are not united in infancy may lead to confusion unless they are correctly recognized. The frontal bone is divided in half laterally by the metopic suture (see Figs. 18-1 and 18-3). Apparent discontinuity of the sphenoid bone with the frontal bone superiorly and the occipital bone posteriorly indicates the sites of the sphenoid bone’s synchondroses with these two bones (see Fig. 18-2). The four major components of the occipital bone (e-Fig. 18-5 and Fig. 18-2) likewise may simulate discontinuities of structure.

Normal Variations

Intrasutural, or wormian, bones occur most frequently along the lambdoid sutures (Fig. 18-6 and e-Fig. 18-7; Box 18-2). They occur much less frequently in the fontanelles (see e-Fig. 18-7). The interparietal or Inca bone (Fig. 18-8) results from division of the supraoccipital portion of the occipital bone into two parts by the mendosal suture, with the superior part arising from membranous bone and the inferior part arising from cartilage continuous with that of the supracondylar portions and the basiocciput. A rare synchondrosis or suture line runs vertically through the squamous portion of the occipital bone (Fig. 18-9); persisting superior and inferior portions of the line are known as the superior longitudinal fissure or bi-interparietal suture and the cerebellar synchondrosis or median cerebellar suture. Where the supraoccipital portion of the occipital bone forms the posterior border of the foramen magnum, accessory supraoccipital bones occasionally are found (e-Fig. 18-10). The configuration caused by an outward bulge of the occipital squamosa just above the torcular Herophili in a newborn (Fig. 18-11) is called bathrocephaly. Rarely, a horizontal interparietal suture divides the parietal bones into superior and inferior moieties (e-Fig. 18-12).

Compression of the fetal skull and its molding during passage through the maternal pelvis produce significant radiographic findings that persist after birth (Fig. 18-13).² During the first weeks and months of life, widths of sutures vary so much that caution is required in their evaluation for the diagnosis of increased intracranial pressure, particularly because positioning is difficult and partial superimposition of bilateral sutures can produce spurious widening (Fig. 18-14).

In children older than 2 years, the sutures extend through both tables and the diploic space. The outer table portion of the suture may be deeply serrated when the inner table portion is practically a straight line (e-Fig. 18-15) and may be interpreted erroneously as a “fracture through a suture.” Persistence of the metopic suture may simulate a vertical fracture in the occipital bone in anteroposterior, caudally angulated exposures if extension of the superimposed radiolucent line into the area of the foramen magnum is invisible or if the inferior portion of the suture has been obliterated. The frontal crest on the internal surface of the frontal squamosa in the midsagittal plane may be sufficiently prominent to simulate calcification of the falx cerebri that attaches to it (e-Fig. 18-16).

Normal Variations

The outstanding characteristic of the juvenile skull is its remarkable variability in size, shape, thickness and mineral content, depth of the grooves for the dural sinuses, pattern of the diploic structure and convolutional and vascular markings, degree of pneumatization of the temporal and paranasal bones, and size and shape of the pituitary fossa. These normal variants are so marked that frequently it is difficult to distinguish normal variations from early pathologic changes.

Normal Paranasal Sinuses

The paranasal sinuses are paired pneumatic cavities that communicate with the nasal fossae and are situated in the paranasal bones—maxilla, ethmoid, frontal, and sphenoid. Because of the continuity of their air cell mucosa with that of the nasal cavity via the eustachian tubes, the mastoid cells can be considered an additional component. The size and shape of the cavities vary in different age periods, among persons, and on the two sides of the same person.7,8

The sites of the openings of the sinuses into the nasal cavity are shown in Figures 18-32 and 18-33. The postnatal growth and extension of the sinuses are shown in e-Figure 18-34. The fully developed maxillary, frontal, and ethmoid sinuses are illustrated diagrammatically in e-Figures 18-35 and 18-36. Computed tomography (CT has shown extensions from adjacent sinuses into the orbital roofs and apices of the petrous temporal bone that are not easily recognized on conventional radiographs.

Maxillary Sinuses

Changes in size and configuration of the maxillary sinuses with age are shown in e-Figure 18-36 and Figure 18-37. The maxillary sinuses are present at birth, expand steadily, and are considered mature by the time of puberty.⁹ Variations in development include isolated unilateral hypoplasia and prominent septa that appear to compartmentalize the sinus cavity (Fig. 18-38). The roots of the maxillary molars occasionally impinge on the walls of the sinuses (Fig. 18-39) and sometimes produce folds in the mucous membranes. In oblique ventrodorsal projections of the skull, the roots of the teeth may be superimposed on the sinuses and artifactually appear to project into them. A molar that fails to migrate is found in its fetal position near the posterosuperomedial angle of the maxillary sinus (e-Figs. 18-40 and 18-41). Note should be made of the relative height of the antral floor and structures in the nasal cavity, which can influence surgical approaches.

Frontal Sinuses

Marked variation in size and shape is typical of frontal sinuses, as is asymmetry in development. The frontal sinuses are not sufficiently developed for radiographic identification until they extend into the base of the vertical plate of the frontal bone by 6 years of age. The rate of expansion varies; any time between 6 and 12 years they extend to the level of the orbital roofs. By puberty, most frontal sinuses have reached adult size. Some are extensions of the anterior ethmoid cells. In some cases (especially in instances of hypoplasia of the frontal lobes of the brain), exaggerated extension into the horizontal plates of the frontal bones may occur; this phenomenon is more easily recognized in lateral projections.

Ethmoid Sinuses

The ethmoid sinuses are composed of a series of cells of variable number, forming paired bony labyrinths suspended from the horizontal plate of the ethmoid bone on each side of the vertical plate, with the lateral walls forming the medial walls of the orbits. They are separated by thin osseous septa covered with mucous membrane. They all communicate with the nasal cavities either directly by independent channels or indirectly through cells of the same group. They are present at birth, expand rapidly during the first 5 years, expand less quickly until about 8 years, and usually are complete by age 12 years. They usually form three groups: anterior, middle, and posterior (see Fig. 18-33 and e-Fig. 18-36). The ethmoid cells often extend into the turbinates, crista galli, and neighboring frontal, maxillary, sphenoid, and palate bones. Three anatomic variants of ethmoid cells are found: Haller, Agger nasi, and Onodi cells. These variants are described in Chapter 8. Extension of infection from the sinus to the orbit can occur easily through the lamina papyracea (see Chapter 8).

Sphenoid Sinus

The paired cavities in the body of the sphenoid bone are separated by an osseous partition that may be displaced to one side so that the two cavities vary greatly in size and shape.10–12 The cavities can be visualized when they are superimposed in lateral projections or side by side in submentovertical projections (e-Fig. 18-42). Ridges and septa sometimes divide each single cavity into separate compartments. The air cells are not present at birth; by age 4 years they are 4 to 8 mm in diameter, and they become adult size any time between 7 and 11 years (Fig. 18-43).

Osteomeatal Unit

The components of the osteomeatal unit (OMU) (the central sinus drainage anatomy) are present at birth, albeit crowded (see Figs. 18-32, 18-44, and 18-45). This anatomy is clearly outlined with coronal CT images obtained either directly or reconstructed from axial images.

Plain Radiographs

The traditional, standard, four-view radiographic examination described next is outdated and has limited usefulness. The American Academy of Pediatrics and the American College of Radiology recommend that plain radiographic imaging be used sparingly in the diagnosis of sinusitis in children, and the the American Academy of Pediatrics guidelines state that radiographs are unnecessary for diagnosing sinonasal inflammation in children younger than 6 years.13,14 In an older child in special circumstances, radiographic imaging may prove useful, but it usually should be confined to a single Waters view.

Imaging the sinuses in children is technically demanding. The sinuses develop at varying rates, and scrupulous technique must be used when obtaining multiple views in the sometimes unhappy and moving subject. The radiation dose of a four-view examination of the sinuses is approximately 1.8 mGy to the lens and 0.12 mGy to the entire body.

A standard examination of the sinuses used to include four views (Waters, posteroanterior, Caldwell, and lateral) (see Fig. 18-43). The Waters projection is the most important and is obtained with the patient’s head in just enough extension to place the shadows of the dense petrous pyramids immediately below the maxillary antral floors. It is used primarily for visualization of the maxillary sinuses.

The Caldwell (posteroanterior axial) projection, which uses a central ray angled at 15 degrees below the orbitomeatal line, provides a clear image of the ethmoid and frontal sinuses. A lateral view shows the sphenoid sinus; however, except for visualizing the tonsils and adenoids, it is of limited usefulness in a child younger than 4 years because the sphenoid sinuses are small and not well visualized, and thus they readily can appear to be partially opacified. The lateral film is important for viewing the sella and can reveal intracranial disease masquerading as sinus disease. Plain radiographs cannot adequately evaluate the anterior ethmoid air cells, the upper two thirds of the nasal cavity, the frontal recess, or the sphenoid sinuses.

Technical factors such as patient motion (even normal respirations), incorrect angulation, rotation, and underpenetration almost always overestimate the presence of disease (Box 18-3). Partial ethmoid clouding can occur in the Caldwell view because of superimposition of the ethmoid cells caused by slight rotation or nasal secretion. The normal sloping of the walls of the maxillary antra also can mimic mucosal thickening.

Although the presence of an air-fluid level appears to be evidence for acute inflammatory disease in the absence of trauma, the sensitivity of the plain radiographic finding of mucosal thickening and opacification is questionable at best. However, when radiographs are taken correctly and interpreted in view of the clinical presentation, normal findings do make a significant acute sinus infection unlikely.

Plain radiographs play no role in the investigation of a mass in the sinuses and in assessing suitability for, or complications from, sinus surgery.15,16 Plain radiographs also are inappropriate for imaging complications of sinusitis. The capabilities of CT and magnetic resonance imaging (MRI are far superior. Plain radiographs also have little to offer in children younger than 2 years and certainly in infants younger than 1 year because such a high incidence of false-positive findings occurs.^17–21

Computed Tomography

Coronal CT, that is, imaging in thin sections with low milliamperage (20 to 40 mA) and kilovolts (100 to 120), provides excellent images of the sinuses and can detect changes in thin bone, surrounding soft tissue, and interposed airspaces. The CT dose index “dose” given for such an examination is approximately 6.50 mGy, which makes CT the optimal imaging modality for evaluating the paranasal sinuses and the draining pathway through the ostium and into the nasal cavity. CT is the accepted gold standard for imaging chronic inflammatory changes and their operative management. It also is the best tool for investigating complications of sinusitis and, along with MRI, for examining masses in the sinuses or surrounding soft tissue. These conditions are discussed later in this chapter.

Our current technique on a 64-slice unit is to perform axial scans (which is easier for the patient) at 5 mm reconstructed down to 0.625 mm. We scan at 100 kV with 40 mA for 0.8 seconds (32 mA). Coronal and sagittal reformatted images are obtained.

With the advent of functional endoscopic sinus surgery (FESS) for treating chronic inflammatory disease, the role of CT has expanded further.²² Coronal imaging simulates the view through the endoscope and provides information not only about the extent of inflammation, but also regarding anatomic features that are intimately related to the pathogenesis of the disease process itself. The goal of FESS is to maintain the normal drainage pathway (the OMU) of the frontal, maxillary, and anterior ethmoid sinuses. The OMU, which is in the region of the middle meatus, consists of the ostia, infundibulum, hiatus semilunaris, and middle meatus itself. This anatomy is exquisitely shown with coronal imaging through the nose and sinuses (see Fig. 18-44). Blockage of this pathway by inflammatory tissue or exudate allows mucus and debris to accumulate in the sinus air cells and predisposes to infection. It has been suggested that anatomic variants in the OMU such as a deviated nasal septum, large concha bullosa (aerated middle turbinates), and paradoxically bent middle turbinates (concave medially, rather than the normal configuration, which is bent concave laterally) predispose to blockage of the pathway and disease. Similarly, large Haller cells (ethmoid cells located along the rim of the orbit and protruding into the maxillary antrum), ethmoid bullae (ethmoid air cells above and posterior to the infundibulum), and large Agger nasi cells (the most anterior ethmoid air cells) have been thought to compromise the drainage route and result in disease. These variants are common in asymptomatic persons, however, and have little clinical significance (see Fig. 18-5, A and C, and Fig. 18-45).^23,24

The role of CT before endoscopic surgery also is important to assess anatomic variations that could predispose a person to have complications. The height of the cribriform plate may vary 17 mm on either side of the crista galli. It also is important to note areas of dehiscence in the cribriform plate, which are reported to be present 14% of the time.

No bone is found between the carotid artery and the sphenoid sinus in about 8% to 10% of the population. Also, the relationship of the optic nerve to the posterior ethmoid and sphenoid sinuses should be noted because the posterior ethmoid air cells contact the optic canal in 48% of people, and 78% to 88% of people have a very thin or no bony border between the optic nerve and the sphenoid sinus.

CT also is the study of choice for documenting complications of FESS (e-Fig. 18-46). Orbital hematomas may occur after transection of an ethmoidal artery and could potentially expand, compromising flow in the retinal artery and resulting in ischemia of the optic nerve. Postoperative cerebrospinal fluid leaks are well assessed with nuclear medicine or new cisternographic MRI techniques. CT also may provide ancillary anatomic information in these cases, and CT cisternography is a newer technique being explored. Pseudoaneurysm, is a rare complication of FESS and can be identified with CT angiography, magnetic resonance angiography, or conventional angiography.

CT is considerably more sensitive than plain radiographs in detecting rhinosinusitis. On CT scans obtained for other reasons, 100% of asymptomatic children who had a recent upper respiratory tract infection showed soft tissue changes in their sinus. Seventy percent of pediatric patients show soft tissue changes in CT scans performed for unrelated problems.18–21 The specificity is low, and the problem becomes an even larger issue as the use of CT for the investigation of inflammatory disease increases. Subsequently, CT should be used in cases of acute disease only when the patient is unresponsive to medical treatment, and CT should be used to assess chronic disease only when surgery is being considered.

Magnetic Resonance Imaging

Although MRI of the paranasal sinuses is not widely advocated for assessment of inflammatory disease, it is well accepted for studying masses in this region. Limitations exist when using this modality to image the sinuses, however. Neither bone nor air provides a signal, and thus with MRI, the bony framework and changes that are so important in the imaging of sinus disease are not visualized. It takes longer to obtain the images with MRI than with other modalities, and motion causes profound degradation of images. Most young children require sedation when MRI is performed.

The normal nasal cycle of vasodilation and mucosal edema followed by vasoconstriction and mucosal shrinkage causes signal changes that can result in problems in interpreting findings.9,25 This cycle varies from 50 minutes to 6 hours, and the signal intensity during the edematous phase is indistinguishable from that of inflammatory change. As is the case with CT and plain radiographs, there is a high incidence of findings in asymptomatic persons (13% to 37%), and mucosal thickening less than 3 mm likely is insignificant. MRI usually is not used to image inflammatory disease, but it plays an important role in examining complex or intracranial extension of inflammatory disease or as part of the workup of a sinus or parasinal neoplasm.

Although most acute inflammatory diseases, including polyps, mucoceles, and retention cysts, produce a bright signal on T2-weighted images, 90% to 95% of tumors in the sinuses or nasal cavity exhibit moderately lower signal intensities on T2 weighting. This phenomenon largely occurs because these lesions are histologically cellular and homogeneous. More mature granulation tissue and fibrosis are difficult to distinguish from tumor because these too produce a lower T2 signal. Certain fungal infections, in contrast to other acute inflammatory disease, also exhibit a lower signal on T2 weighting.

Barghouth, G, Prior, JO, Lepori, D, et al. Paranasal sinuses in children: size evaluation of maxillary, sphenoid, and frontal sinuses by magnetic resonance imaging and proposal of volume index percentile curves. Eur Radiol. 2002;12:1451–1458.

Belden, CJ. The skull base and calvaria: adult and pediatric. Neuroimaging Clin N Am. 1998;8:1–20.

Bhattacharyya, N, Jones, DT, Hill, M, et al. The diagnostic accuracy of computed tomography in pediatric chronic rhinosinusitis. Arch Otolaryngol Head Neck Surg. 2004;130:1029–1032.

Huda, W. Assessment of the problem: pediatric doses in screen-film and digital radiography. Pediatr Radiol. 2004;34(suppl 3):S173–S182.

Mann, SS, Naidich, TP, Towbin, RB, et al. Imaging of postnatal maturation of the skull base. Neuroimaging Clin N Am. 2000;10:1–22.

1. Kier, EL, Fetal skull. Radiology of the skull and brain. Newton, TH, Potts, DG, eds. Radiology of the skull and brain, St Louis, Mosby, 1971;Vol 1.

2. Moloy, HG. Studies on head molding during labor. Am J Obstet Gynecol. 1942;44:962–982.

3. Vignaud-Pasquier, J, Lichtenbert, R, Laval-Jeantet, M, et al. Les impressions digitales de la naissance á neuf ans. Biol Neonate. 1964;6:250–276.

4. Greitz, D, Hannerz, J. A proposed model of cerebrospinal fluid circulation: observations with radionuclide cisternography. AJNR Am J Neuroradiol. 1996;17:431–438.

5. Fink, AM, Maixner, W. Enlarged parietal foramina: MR imaging features in the fetus and neonate. AJNR Am J Neuroradiol. 2006;27:1379–1381.

6. Goldsmith, WM. Catlin mark: inheritance of unusual opening in parietal bones. J Hered. 1941;32:301–309.

7. Maresh, MM. Paranasal sinuses from birth to late adolescence. Am J Dis Child. 1940;60:841–861.

8. Maresh, MM. Paranasal sinuses from birth to late adolescence, 1: size of the paranasal sinuses as observed in routine posteroanterior roentgenograms. Am J Dis Child. 1940;60:55–78.

9. Odita, JC, Akamaguna, AI, Ogisi, FO, et al. Pneumatisation of the maxillary sinus in normal and symptomatic children. Pediatr Radiol. 1986;16:365–367.

10. Fujioka, M, Young, LW. The sphenoidal sinuses: radiographic patterns of normal development and abnormal findings in infants and children. Radiology. 1978;129:133–136.

11. Kazkayasi, M, Kardeniz, Y, Arikan, OK. Anatomic variations of the sphenoid sinus on computed tomography. Rhinology. 2005;43:109–114.

12. Reittner, P, Doerfler, O, Goritschnig, T, et al. Magnetic resonance imaging patterns of the development of the sphenoid sinus: a review of 800 patients. Rhinology. 2001;39:121–124.

13. American Academy of Pediatrics, Subcommittee on Management of Sinusitis and Committee on Quality Improvement. Clinical practice guideline: management of sinusitis. Pediatrics. 2001;108:798–808.

14. American College of Radiology. Sinusitis in the pediatric population: ACR appropriateness criteria. Radiology. 2000;215:811–818.

15. McAlister, WH, Kronemer, K. Imaging of sinusitis in children. Pediatr Infect Dis J. 1999;18:1019–1020.

16. Kronemer, KA, McAlister, WH. Sinusitis and its imaging in the pediatric population. Pediatr Radiol. 1997;27:837–846.

17. Diament, MJ, Senac, MO, Gilsanz, V, et al. Prevalence of incidental paranasal sinus opacification in pediatric patients: a CT study. J Comput Assist Tomogr. 1987;11:426–431.

18. Gwaltney, JM, Phillips, CD, Miller, RD, et al. Computed tomographic study of the common cold. N Engl J Med. 1994;330:25–30.

19. Kovatch, AL, Wald, ER, Ledesma-Medina, J, et al. Maxillary sinus radiographs in children with nonrespiratory complaints. Pediatrics. 1984;73:306–308.

20. Kristo, A, Alho, O-P, Luotonen, J, et al. Cross-sectional survey of paranasal sinus magnetic resonance imaging findings in schoolchildren. Acta Paediatr. 2003;92:34–36.

21. Kristo, A, Uhari, M, Luotonen, J, et al. Paranasal sinus findings in children during respiratory infection evaluated with magnetic resonance imaging. Pediatrics. 2003;111:e586–e589.

22. Wolf, G, Wolfgang, A, Kuhn, F. Development of the paranasal sinuses in children: implications for paranasal sinus surgery. Ann Otol Rhinol Laryngol. 1993;102:705–711.

23. Stallman, JS, Lobo, JN, Som, PM. The incidence of concha bullosa and its relationship to nasal septal deviation and paranasal sinus disease. AJNR Am J Neuroradiol. 2004;25:1613–1618.

24. Sivash, E, Sirikci, A, Bayazyt, YA, et al. Anatomic variations of the paranasal sinus area in pediatric patients with chronic sinusitis. Surg Radiol. 2002;24:400–405.

25. Zinrerch, SF, Kennedy, BW, Kuinar, AJ, et al. MRI imaging of the normal nasal cycle: comparison with sinus pathology. J Comput Assist Tomogr. 1988;12:1014–1019.