[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.167.142.229. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Download PDF
Figure 1.
Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 2-year-old skull.

Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 2-year-old skull.

Figure 2.
Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 5-year-old skull.

Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 5-year-old skull.

Figure 3.
Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 8-year-old skull.

Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in the 8-year-old skull.

Figure 4.
Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in adult skull No. 14.

Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in adult skull No. 14.

Figure 5.
Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in adult skull No. 20.

Resting state and resultant interferograms at the infraorbital foramen (A and B), malar eminence (C and D), nasofrontal suture (E and F), and supraorbital notch (G and H) in adult skull No. 20.

Figure 6.
Fringe density measured at periorbital loci. IOF indicates infraorbital foramen; SON, supraorbital notch; NOF, nasofrontal suture; and ME, malar eminence.

Fringe density measured at periorbital loci. IOF indicates infraorbital foramen; SON, supraorbital notch; NOF, nasofrontal suture; and ME, malar eminence.

1.
Koltai  PJ Maxillofacial injuries in children. Smith  JDBumsted  Reds.Pediatric Facial Plastic & Reconstructive Surgery New York, NY Raven Press1993;283- 316
2.
Koltai  PJRabkin  D Pediatric maxillofacial trauma. Curr Opinion Otolaryngol Head Neck Surg. 1994;2482- 486Article
3.
Koltai  PJRabkin  D Rigid fixation for facial fractures in children. J Craniomaxillofac Trauma. 1995;132- 42
4.
Koltai  PJAmjad  IMeyer  DFeustel  P Orbital fractures in children. Arch Otolaryngol Head Neck Surg. 1995;1211375- 1379Article
5.
Conerty  MCastracane  JCacace  AJ  et al.  Preliminary results of real-time in vitro ESPI measurements in otolaryngology. SPIE. 1995;2395124- 134
6.
Castracane  JConerty  MCacace  AJ  et al.  Applications of ESPI in otolaryngology.  Paper presented at: 17th Midwinter Meeting of the Association for Research in Otolaryngology February 6-10, 1994 St Petersberg, Fla
7.
Castracane  JConerty  MCacace  AJ  et al.  Advanced ESPI-based medical instruments for otolaryngology. SPIE. 1993;189333- 42
8.
Ovryn  B Holographic interferometry. Crit Rev Biomed Eng. 1989;16269- 322
9.
Ovryn  BManley  MTStern  LS Holographic interferometry: a critique of the technique and its potential for biomedical measurements. Ann Biomed Eng. 1987;1567- 78Article
10.
Gardner  GCastracane  JConerty  M  et al.  Laser holography and interferometry of the larynx: preliminary studies with a cadaver model. Trans Am Laryngol Assoc. 1992;113104- 109
11.
Spetzler  RFSpetzler  H Holographic interferometry applied to the study of the human skull. J Neurosurg. 1980;52825- 828Article
12.
Hoyer  HEDorheide  J A study of human head vibrations using time-averaged holography. J Neurosurg. 1983;58729- 733Article
13.
Anderson  RLPanje  WRGross  CE Optic nerve blindness following blunt forehead trauma. Ophthalmology. 1982;89445- 455Article
14.
Raflo  GT Blow-in and blow-out fractures of the orbit: clinical correlations and proposed mechanisms. Ophthalmic Surg. 1984;15114- 119
15.
Antonyshyn  OGruss  JSKassel  EE Blow in fractures of the orbit. Plast Reconstr Surg. 1989;8410- 20Article
16.
Gruss  JS Orbital roof fractures in the pediatric population. Plast Reconstr Surg. 1989;84217- 218Article
17.
Messinger  ARadkowski  MAGreenwald  MJPensler  JM Orbital roof fractures in the pediatric population. Plast Reconstr Surg. 1989;84213- 216Article
18.
Converse  JM Facial injuries in children. Kazanjian  VHConverse  JMeds.The Surgical Treatment of Facial Injuries 3rd ed. Baltimore, Md Williams & Wilkins1974;
19.
Enlow  DH Facial Growth. 3rd ed. Philadelphia, Pa WB Saunders; Co1990;
20.
Morin  JDHill  JCAnderson  JEGrainger  RM A study of growth of the interorbital region. Am J Ophthalmol. 1963;56895- 901
21.
Onodi  A Accessory Sinuses of the Nose of Children.  New York, NY William Wood & Co1911;plates 4-90
22.
Moore  MHDavid  DJCooter  RD Oblique craniofacial fractures in children. J Craniofac Surg. 1990;14- 7Article
23.
Hirano  ATsuneda  KNisimura  G Unusual fronto-orbital fractures in children. J Craniomaxillofac Surg. 1991;1981- 86Article
Original Article
July 1999

Evaluation of Orbital Stress Dissipation in Pediatric and Adult Skulls Using Electronic Speckle Pattern Interferometry

Author Affiliations

From the Section of Pediatric Otolaryngology, Department of Surgery, Albany Medical College, Albany, NY (Dr Mouzakes); the Section of Pediatric Otolaryngology, The Cleveland Clinic Foundation, Cleveland, Ohio (Dr Koltai); Interscience Inc, Troy, NY (Ms Simkulet); and the Center for Environmental Sciences and Technology Management, State University of New York at Albany (Dr Castracane).

Arch Otolaryngol Head Neck Surg. 1999;125(7):765-773. doi:10.1001/archotol.125.7.765
Abstract

Objectives  To measure and quantitatively compare the degree of force dissipation in pediatric and adult skulls subjected to similar dynamic forces.

Design  An anatomical study using electronic speckle pattern interferometry, which allows generation of displacement vectors after application of a force.

Subjects  Five human skulls (3 pediatric and 2 adult).

Intervention  Each skull was subjected to a reproducible and quantifiable force created by a steel ball pendulum striking a precise periorbital focus: (1) infraorbital foramen, (2) supraorbital notch, (3) malar eminence, and (4) nasofrontal suture. Electronic speckle pattern interferometry was used to construct interferogram fringe patterns to determine skull regions with the greatest degree of displacement.

Results  Interferogram analysis revealed that the adult skull has a tendency to dissipate force with minimal resultant displacement. In contrast, the pediatric skulls demonstrated greater displacements (ie, increased fringe density) at the same periorbital foci.

Conclusions  The pediatric skull dissipates periorbital stress differently than the adult skull, as illustrated by quantitative interferogram analysis. This finding parallels clinical data that demonstrate a varying pattern of fractures in pediatric and adult skulls related to craniofacial development.

PEDIATRIC maxillofacial trauma generally receives attention with regard to etiology, epidemiology, treatment, and management.14 Consistent with other developmental processes, the face of a child represents a changing mosaic of growth and structural alteration. This dynamic state is illustrated by the changing fractures in and around the orbit.4 For example, orbital floor fractures usually do not appear before the age of 5 years, since the maxillary sinus lacks complete pneumatization. In contrast, orbital roof fractures are primarily seen in the younger pediatric age group owing to the greater craniofacial ratio in the younger child and the absence of frontal sinus pneumatization. These developmental differences provided a basis for our study, in which we analyzed the structural and mechanical characteristics of both pediatric and adult skulls as they relate to stress dissipation. The assessment of stress applied to the orbital region, and the resultant skeletal displacement was measured by electronic speckle pattern interferometry (ESPI).

Interferometry is an investigative tool that has traditionally been used in the field of optical engineering but that has recently evolved into a modality with broader application; eg, it is now used in various biomedical fields.510 This noncontact technique determines the 3-dimensional surface relationships of an object for the calculation of displacement. After a stress force or vibratory stimulus is applied, the object's surface pattern is compared with the surface geometry in the resting state. The resulting changes in the surface are displayed as a series of interference fringes superimposed on the image of the object. As the degree of displacement from the resting state increases, the number and proximity of the fringes will increase. Traditional holographic interferometry (HI) uses photographic film to record the interference patterns. This is a time-consuming process and has limited applicability for facial structural analysis. We have adopted the method of ESPI, in which the interference patterns are combined on the face of a digital high-resolution camera for processing and storing information. We have the capability of measuring the displacement both quantitatively and qualitatively. By counting the number of fringes generated by ESPI, a displacement vector can be calculated. This technique lends itself to qualitative analysis, because strain pattern recognition can be used.

Interferometry was first described more than 3 decades ago, and researchers have since found that it can be useful in the field of clinical medicine.510 Clinical studies on human skulls have used conventional HI. Spetzler and Spetzler11 conducted an early study involving the application of HI to the human skull. They described the basic laboratory technique for HI on skulls using a helium-neon laser to generate the interferograms, which provided predictive information about the patterns of fracture. They concluded that HI is more sensitive than conventional methods that use multiple strain gauges. They noted that the test object (the skull) had to be completely immobile for 5 seconds during the holographic recording. Even subtle movements rendered the image useless, thus limiting the method to static forces. By using ESPI, we can eliminate this problem, since the exposure time is reduced to as little as 1125 of a second.

Hoyer and Dorheide12 postulated that the use of dry skulls results in a misinterpretation of deformation patterns, since the sutural connections of the specimens lack cartilage and fibrous tissue. Using 2 preserved human heads, they found displacement patterns that were different from those obtained with dry skulls, and speculated that the preserved heads had minimal shearing forces across the sutures. In contrast, the relative movements of the component bones of the dry skulls changed the dissipation of forces.

Anderson et al13 described 7 cases of sudden monocular blindness that occurred after blunt frontal head trauma, suggesting that indirect optic nerve injury is a result of

. . . a stretching, tearing, torsion, or contusion of the nerve caused not only from the momentum of the eyeball and orbital contents being absorbed by the fixed canalicular portion of the optic nerve but also by skeletal distortion caused by forces remote from the initial impact.

Double-exposure holograms were used to assess stress distribution in the region of the optic foramen. Supraorbital ridge pressure resulted in maximal fringe densities in the orbital roof, approximately 5 to 8 mm from the optic foramen. Malar eminence pressure resulted in maximal fringe densities in the orbital floor. Anderson and colleagues concluded that the forces from frontal trauma may be transmitted to the optic nerve directly and that fractures are not necessary to transmit these forces.

Raflo14 reported on blow-out fractures of the orbit and speculated on their etiology. He described one patient who experienced a blow to the supraorbital rim that resulted in an orbital floor blow-out fracture and another patient who sustained a blow-in fracture of the orbital roof secondary to blunt orbital trauma. To account for these fractures, he proposed a mechanism that differed from the traditional idea of increased intraorbital or intracranial hydraulic pressure causing a floor or roof fracture. His argument for the concept of buckling forces on the orbital walls as causative factors in blow-in and blow-out fractures was based on the holographic studies of Anderson et al.13

This article discusses the clinically relevant applications of ESPI for craniomaxillofacial trauma. Hoyer and Dorheide12 suggested that skull composition may influence interferogram patterns, and their theory is likely to apply to the different stages of growth and development as well.

MATERIALS AND METHODS

Institutional review board approval was obtained before we began our investigation. Five human skulls (3 aged 2, 5, and 8 years, and 2 adult skulls) were used. Each skull was mounted in an identical fashion to a vibration isolation table. The skulls were then subjected to a similar impact at each of 4 identical periorbital foci: (1) infraorbital foramen, (2) supraorbital notch, (3) malar eminence, and (4) nasofrontal suture. The dynamic force used in this experiment was created by a steel ball pendulum impact system. A ball bearing, measuring 5 mm in diameter and weighing 3 g, was threaded to free-fall in pendulum motion. The initial position was secured by an electromagnet. Resultant impact forces at each skull locus were reproducible by manipulation of the height of initial release and/or fulcrum length to create similar stress forces. Data were collected at each skull locus to achieve at least 3 consistent interferograms. Because of the large amount of acquired data and computer storage constraints, 1 representative interferogram was chosen (from each locus of each skull) for the purposes of analysis and comparison.

Interferogram patterns were created using a helium-neon laser in the ESPI setup. The laser source was introduced into a beam splitter that sends 90% of the laser power to the object leg and 10% of the laser power to the reference leg. The object leg provided the illumination of the skull positioned along the optical axis. The object was then imaged by a lens and the adjustable aperture through the wedged beam splitter onto a digital camera (Polaroid Digital Microscope Camera; 800×600 pixels; shutter speed range, 1125-1 second). A wedged beam splitter was used to minimize the back reflections picked up by the camera. The beam splitter coaxially recombines the object image and a reference wave front on the camera. A smooth reference wave front is achieved by spatially filtering the reference wave by a second lens and the pinhole aperture. The interference pattern achieved by the coaxial recombination of the wavefronts was imaged by the digital camera in high-resolution mode. Static images captured at a specific time point from impact were then processed and reviewed on a computerized format. In the current static mode, the interferograms are created by subtracting the displaced image from the reference image in software. The resultant difference image is a static interferogram. Analysis of the representative interferograms from each skull locus consisted of a quantitative assay of fringe density. Fringe density represents the counted number of fringes per square centimeter (fringes/cm2) of skull surface.

RESULTS

Figure 1, Figure 2, Figure 3, Figure 4, and Figure 5 are representative interferograms generated from the 2-, 5-, and 8-year-old skulls and the 2 adult skulls (Nos. 14 and 20), respectively. The left side of the interferogram pairs represents the resting state of the skull before impact, and the adjacent image depicts the fringe pattern after it is struck by the steel ball pendulum.

At the infraorbital foramen, the 2-year-old skull demonstrated the greatest displacement (0.71 fringes/cm2), with uniform densities noted at the 5- and 8-year-old skulls (0.53 fringes/cm2). In contrast, the adult fringe densities recorded were only 0.09 and 0.18 fringes/cm2 for the adult skulls (Nos. 14 and 20, respectively). At the supraorbital notch, the 2-year-old skull produced a fringe density of 1.68 fringes/cm2. Less displacement was demonstrated by the 5- and 8-year-old skulls (0.44 and 0.18 fringes/cm2, respectively). Both adult skulls manifested densities of 0.26 fringes/cm2.

At the nasofrontal suture, both the 2- and 5-year-old skulls exhibited increased displacements, with densities of 1.94 and 0.88 fringes/cm2, respectively. These densities were unlike those of the 8-year-old skull and both adult skulls (Nos. 14 and 20), which produced displacements of 0.26, 0.35, and 0.26 fringes/cm2, respectively.

Similarly, at the malar eminence, both the 2- and 5-year-old skulls demonstrated increased densities of 1.23 and 0.88 fringes/cm2. The 8-year-old skull produced a slightly less displacement of 0.62 fringes/cm2, with the adult skulls demonstrating the least density (0.26 and 0.35 fringes/cm2). Overall fringe density trends are illustrated in Figure 6.

COMMENT

The developmentally specific anatomy of the pediatric skull, and the forces that result in injury, are the determinants of pediatric maxillofacial fractures. Specifically, we have observed a clinical variation among children with orbital injuries that appears to be age related, which suggests that the orbit—the boundary between the face and cranium—exhibits fracture patterns that are influenced by the changing craniofacial geometry of a growing child. Koltai et al4 reviewed the records of 40 children (aged 1-16 years) with orbital fractures. Of the 40 children, 14 had orbital roof fractures, 10 had orbital floor fractures, 14 had mixed orbital fractures, and 2 had medial wall fractures. The age distribution of the children with orbital roof fractures was significantly different from that of the children with other types of orbital fractures (P<.01). The age at which the probability of lower orbital fractures exceeds the probability of roof fractures was estimated to be 7.1±1.0 (mean±SD) years by logistic regression. Overall, orbital fractures in the pediatric population differ from those in adults, with a higher incidence of roof fractures noted.1517 This finding suggests that there is developmental specificity in the changing pattern of fractures. Orbital roof fractures are more likely to occur in children aged 7 years or younger, whereas other orbital fractures tend to occur in older children. The biomechanics of orbital fractures and the development of the face best explain this observation.

There are 2 distinct, but related, factors of facial growth that have an impact on the varied pattern of pediatric orbital fractures: (1) craniofacial geometry and (2) expansion and growth of the paranasal sinuses. At birth, the craniofacial ratio is 8:1, with the face located in a recessed position relative to a large skull. By the age of 2 years, the cranium has achieved 80% of its mature size. Gradually, facial growth continues into the second decade of life, with the final resultant craniofacial ratio approaching 2:1.18,19 Additionally, pneumatization of the paranasal sinuses plays a significant role in the architectural changes of facial enlargement. In infancy, the sinuses are present as buds of mucosa within a matrix of cancellous bone covered by thin cortical plates. The ethmoid cells steadily expand, ultimately yielding 70% of the interorbital width by the age of 7 years.20 At birth, the maxillary sinus is medial to the globe, with the alveolar margin in close proximity to the inferior orbital rim. By the age of 7 years, the maxillary sinus has widened past the midpupillary line to occupy half of the lateral nasal wall. Also, the slowly expanding frontal sinuses begin to reside above the superior orbital rim at that age.21 Thus, before the age of 7 years, the large cranium is more vulnerable to trauma than is the proportionately smaller face. Concomitantly, the unfolding of the face beneath the protective overhanging cranium (a consequence of continued growth of the maxilla) results in the exposure of the face to trauma.

Facial maturation is the result not only of a change in size, but also of an evolution in the structural physiology of the face—a dynamic transformation in the mechanism and direction of stress dissipation. Moore et al22 described a series of 12 children who sustained high-velocity injuries that had a pattern of obliquity that traversed the frontal bone, orbit, and midline nasal structures. There was radiation to the contralateral aspect of the midface area and relative sparing of the mandible, a pattern at variance with fractures observed in adults. They note that this pattern of fracturing is attributable to the higher craniofacial ratio, as well as to the lack of complete buttress development. Their description of these fractures suggests a "tearing" or "shearing" rather than the "egg shelling" of the facial bones, as is often seen in adults. Hirano et al23 reported 2 unusual cases of fronto-orbital fractures in which bone fragments containing orbital roof were detached in 1 piece from the wound, and injury was thought to occur by the tangential forces to the forehead. While the authors speculate that this type of injury is characteristic in children as a result of the higher craniofacial ratio, they also note a shearing quality to the injury of the softer, more elastic pediatric bone. This attribute is most likely the cause of a high incidence of greenstick fractures in the pediatric population.

It is these clinically apparent differences in stress dissipation in pediatric and adult skulls that prompted our investigation. To our knowledge, no studies have quantitatively compared the degree and characteristics of force dissipation in pediatric and adult skulls subject to similar dynamic forces. Our experimentation has demonstrated age-related differences in displacements resulting from impacts created at 4 separate periorbital foci. Interferogram fringe-pattern analysis reveals a tendency of the adult skull to dissipate the impact force rapidly, with a minimal displacement from the resting state. Hence, this displacement manifested as broad, widely spaced fringes (Figure 4 and Figure 5). On the contrary, the series of pediatric skulls subjected to a similar force consistently formed fine- to medium-sized fringes with increased density (thus indicating a greater displacement from the resting state) (Figure 1, Figure 2, and Figure 3). This trend was most evident with the 2- and 5-year-old skulls, with less displacement noted at the 8-year-old skull (Figure 6).

Given these stress dissipation results, it becomes necessary to account for the contrasts that were observed. Specifically, it is important to consider the composition and thickness of the adult and pediatric craniofacial skeleton. The adult skull clearly possesses a dense osseous framework that allows rapid dissipation of force and minimal bone movement. In comparison, the thin, more elastic bone that composes the pediatric skeleton may result in greater displacements from a resting state, as noted in the generated interferograms.

The limitation of our study is related to the composition of the test objects, all of which were dry skulls. The lack of fibrocartilaginous interfaces in dry skulls may result in artifactual stress dissipation. This issue may be relevant to the pediatric skull, which consists of plates of bone separated by well-defined fibrocartilaginous sutures. Also, the small number of subjects makes it difficult to relate the results to a larger population, and the uniqueness of each individual skull in this small series does not easily apply to tests of statistical significance.

Although this phase of investigation has identified a distinct difference in periorbital stress dissipation in pediatric vs adult skulls, further analysis is necessary. A larger series of skulls would favor statistical analysis. Most significantly, the current interferograms represent a summative moment during stress dissipation. Future research will incorporate multiple time loci after a single impact so that the complex stress development and eventual decay over several points in time can be more closely examined.

Back to top
Article Information

Accepted for publication February 26, 1999.

This study was supported by grant 96-K73 from the Arbeitsgemeinschaft für Osteosynthesfragen/Association for the Study of Internal Fixation Research Foundation, Davos, Switzerland.

Presented in part at the 13th annual meeting of the American Society of Pediatric Otolaryngology, Palm Beach, Fla, May 12, 1998.

Reprints: Jason Mouzakes, MD, Division of Otolaryngology–Head and Neck Surgery, A-41, Albany Medical College, Albany, NY 12208.

References
1.
Koltai  PJ Maxillofacial injuries in children. Smith  JDBumsted  Reds.Pediatric Facial Plastic & Reconstructive Surgery New York, NY Raven Press1993;283- 316
2.
Koltai  PJRabkin  D Pediatric maxillofacial trauma. Curr Opinion Otolaryngol Head Neck Surg. 1994;2482- 486Article
3.
Koltai  PJRabkin  D Rigid fixation for facial fractures in children. J Craniomaxillofac Trauma. 1995;132- 42
4.
Koltai  PJAmjad  IMeyer  DFeustel  P Orbital fractures in children. Arch Otolaryngol Head Neck Surg. 1995;1211375- 1379Article
5.
Conerty  MCastracane  JCacace  AJ  et al.  Preliminary results of real-time in vitro ESPI measurements in otolaryngology. SPIE. 1995;2395124- 134
6.
Castracane  JConerty  MCacace  AJ  et al.  Applications of ESPI in otolaryngology.  Paper presented at: 17th Midwinter Meeting of the Association for Research in Otolaryngology February 6-10, 1994 St Petersberg, Fla
7.
Castracane  JConerty  MCacace  AJ  et al.  Advanced ESPI-based medical instruments for otolaryngology. SPIE. 1993;189333- 42
8.
Ovryn  B Holographic interferometry. Crit Rev Biomed Eng. 1989;16269- 322
9.
Ovryn  BManley  MTStern  LS Holographic interferometry: a critique of the technique and its potential for biomedical measurements. Ann Biomed Eng. 1987;1567- 78Article
10.
Gardner  GCastracane  JConerty  M  et al.  Laser holography and interferometry of the larynx: preliminary studies with a cadaver model. Trans Am Laryngol Assoc. 1992;113104- 109
11.
Spetzler  RFSpetzler  H Holographic interferometry applied to the study of the human skull. J Neurosurg. 1980;52825- 828Article
12.
Hoyer  HEDorheide  J A study of human head vibrations using time-averaged holography. J Neurosurg. 1983;58729- 733Article
13.
Anderson  RLPanje  WRGross  CE Optic nerve blindness following blunt forehead trauma. Ophthalmology. 1982;89445- 455Article
14.
Raflo  GT Blow-in and blow-out fractures of the orbit: clinical correlations and proposed mechanisms. Ophthalmic Surg. 1984;15114- 119
15.
Antonyshyn  OGruss  JSKassel  EE Blow in fractures of the orbit. Plast Reconstr Surg. 1989;8410- 20Article
16.
Gruss  JS Orbital roof fractures in the pediatric population. Plast Reconstr Surg. 1989;84217- 218Article
17.
Messinger  ARadkowski  MAGreenwald  MJPensler  JM Orbital roof fractures in the pediatric population. Plast Reconstr Surg. 1989;84213- 216Article
18.
Converse  JM Facial injuries in children. Kazanjian  VHConverse  JMeds.The Surgical Treatment of Facial Injuries 3rd ed. Baltimore, Md Williams & Wilkins1974;
19.
Enlow  DH Facial Growth. 3rd ed. Philadelphia, Pa WB Saunders; Co1990;
20.
Morin  JDHill  JCAnderson  JEGrainger  RM A study of growth of the interorbital region. Am J Ophthalmol. 1963;56895- 901
21.
Onodi  A Accessory Sinuses of the Nose of Children.  New York, NY William Wood & Co1911;plates 4-90
22.
Moore  MHDavid  DJCooter  RD Oblique craniofacial fractures in children. J Craniofac Surg. 1990;14- 7Article
23.
Hirano  ATsuneda  KNisimura  G Unusual fronto-orbital fractures in children. J Craniomaxillofac Surg. 1991;1981- 86Article
×