- Email: [email protected]
A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension
+
MODEL
Journal of Plastic, Reconstructive & Aesthetic Surgery (2015) xx, 1e9
A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension Ravi K. Garg a, Ahmed M. Afifi a,d,*, Jennifer Gassner a, Michael J. Hartman b, Glen Leverson c, Timothy W. King a, Michael L. Bentz a, Lindell R. Gentry b a Division of Plastic and Reconstructive Surgery, Department of Surgery, University of Wisconsin, Madison, WI, USA b Section of Head and Neck Imaging, Department of Radiology, University of Wisconsin, Madison, WI, USA c Department of Surgery, University of Wisconsin, Madison, WI, USA d Division of Plastic Surgery, Cairo University, Cairo, Egypt
Received 5 November 2014; accepted 2 February 2015
KEYWORDS Frontal bone fracture; Craniomaxillofacial trauma; Orbital fracture; Frontal sinus fracture; Skull base trauma
Summary Purpose: The broad spectrum of frontal bone fractures, including those with orbital and skull base extension, is poorly understood. We propose a novel classification scheme for frontal bone fractures. Methods: Maxillofacial CT scans of trauma patients were reviewed over a five year period, and frontal bone fractures were classified: Type 1: Frontal sinus fracture without vertical extension. Type 2: Vertical fracture through the orbit without frontal sinus involvement. Type 3: Vertical fracture through the frontal sinus without orbit involvement. Type 4: Vertical fracture through the frontal sinus and ipsilateral orbit. Type 5: Vertical fracture through the frontal sinus and contralateral or bilateral orbits. We also identified the depth of skull base extension, and performed a chart review to identify associated complications. Results: 149 frontal bone fractures, including 51 non-vertical frontal sinus (Type 1, 34.2%) and 98 vertical (Types 2e5, 65.8%) fractures were identified. Vertical fractures penetrated the middle or posterior cranial fossa significantly more often than non-vertical fractures (62.2 v. 15.7%, p Z 0.0001) and had a significantly higher mortality rate (18.4 v. 0%, p < 0.05). Vertical fractures with frontal sinus and orbital extension, and fractures that penetrated the middle or posterior cranial fossa had the strongest association with intracranial injuries, optic neuropathy, disability, and death (p < 0.05).
* Corresponding author. Division of Plastic and Reconstructive Surgery, University of Wisconsin, Madison 608-265-2535, USA. E-mail address: [email protected] (A.M. Afifi). http://dx.doi.org/10.1016/j.bjps.2015.02.021 1748-6815/ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
2
R.K. Garg et al. Conclusions: Vertical frontal bone fractures carry a worse prognosis than frontal bone fractures without a vertical pattern. In addition, vertical fractures with extension into the frontal sinus and orbit, or with extension into the middle or posterior cranial fossa have the highest complication rate and mortality. ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Introduction Although frontal sinus fractures and their associated complications have been described extensively,1e6 fracture patterns of the frontal bone as a broad unit of the craniofacial skeleton have not been rigorously characterized. Considering that the frontal bone is the most frequently fractured cranial bone in craniofacial trauma patients and accounts for 37 percent of cranial fractures,7 it is important to understand frontal bone fracture patterns and related complications in greater detail. The important elements to include in a thorough characterization of frontal bone fractures are the status of adjacent bony structures, including the frontal sinus, orbit, and skull base.8 Previous studies have shown that frontal bone fractures may acquire an “oblique” or vertical trajectory through the frontal bone and extend to adjacent bones. In particular, vertical fractures that begin in the frontal bone and penetrate the midface have been described in children as a common fracture pattern related to their prominent foreheads and underdeveloped paranasal sinuses.9,10 The occurrence of vertical fractures has not been made as explicit in adults, although “linear” fractures were recently described in an adult population as originating in the frontal bone and extending to the skull base.11 The literature therefore supports the possibility of vertically oriented fractures in multiple contexts, but a comprehensive description of frontal bone fractures that accounts for involvement of adjacent bony structures is needed. To our knowledge, no study has examined the spectrum of frontal bone fractures with consideration of frontal sinus, orbital and skull base involvement in both children and adults. An improved understanding of different frontal bone fracture types and their associated morbidity and mortality may enhance our ability to distinguish injury severity among patients with frontal bone trauma and provide prognostic information that can assist with patient counseling and treatment planning. Here we classify different frontal bone fracture patterns and their associated complications.
Methods After receiving approval from the University of Wisconsin Health Sciences IRB, we used ICD-9 codes for calvarial, facial, mandibular and cervical spine trauma to retrospectively identify all patients with craniofacial trauma over a five year interval. Subsequently, all CT scans were reviewed
by a neuroradiology attending to identify those individuals with frontal bone fractures. Frontal bone fractures were primarily distinguished as having a non-vertical or vertical trajectory. Type 1 fractures were defined as comminuted fractures of the frontal sinus without a vertical trajectory. Vertical fractures were broken down into 4 distinct types, Types 2e5, as it was clear from radiologic analysis that these fractures represented a spectrum of injury patterns (Figure 1). Type 2 fractures are vertical fractures involving the orbit but not the frontal sinus. Type 3 fractures are vertical fractures involving the frontal bone and sinus but not the orbit. Type 4 fractures involve both the frontal sinus and the ipsilateral orbit. Type 5 fractures cross the midline of the face, involving the frontal sinus and the contralateral or bilateral orbits. The term non-vertical fracture is therefore used to describe Type 1 fractures while the term vertical fracture is used to describe Type 2e5 fractures. In the case of children without a frontal sinus present, the classification scheme still applies with a few modifications. A Type 1 fracture is a non-vertical fracture through the frontal bone. A Type 2 fracture is a vertical fracture through the orbit only. A Type 3 fracture is a vertical fracture through the frontal bone that does not involve the orbit. There is no Type 4 fracture since there is no frontal sinus. A Type 5 fracture is a vertical fracture through the frontal bone that involves the contralateral or bilateral orbits. If a frontal sinus is present in some phase of development, then the adult classification scheme is used. The depth of skull base extension was also classified for all fractures (Figure 2). Depth A is defined as involvement of the frontal bone without extension into the skull base. Depth B is characterized by extension into the anterior cranial fossa (orbital roof, fovea ethmoidalis, cribriform plate). Depth C fractures extend into the middle cranial fossa (sella, sphenoid body, carotid canal, optic chiasm sulcus). Depth D fractures involve the posterior cranial fossa (clivus, petromastoid temporal bone, petrosal segment of the carotid canal). CT images were also reviewed by a neuroradiology attending to characterize intracranial injuries for all patients with frontal bone fractures. Injuries of interest included subarachnoid hemorrhage, subdural hematoma, epidural hematoma, stroke, subfalcine herniation, and transtentorial herniation. Additionally, the presence of midface fractures that fit accepted definitions of zygomaticomaxillary complex, nasoorbitoethmoid, and Le Fort 1, 2 and 3 fractures was described. Medical records were reviewed to characterize patient demographics and outcomes. The outcomes we measured
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures
3
Figure 2 Skull base penetration depths. Depth A fractures involve the anterior table of the frontal bone with or without posterior table involvement and do not extend into the anterior cranial fossa (purple). Depth B fractures involve the floor of the anterior cranial fossa (blue). Depth C fractures involve the middle cranial fossa (yellow). Depth D fractures extend into the posterior cranial fossa (red).
Figure 1 Frontal bone fracture types. Type 1 fractures are isolated to the frontal sinus without a vertical trajectory (purple). Type 2 fractures are vertically oriented and extend into the orbit but not the frontal sinus (blue). Type 3 fractures are vertically oriented and extend into the frontal sinus but not the orbit (yellow). Type 4 fractures are vertically oriented and extend into ipsilateral frontal sinus and orbit (green). Type 5 fractures extend into the frontal sinus and extend into the orbit on both sides of the face or the contralateral side of the face (red).
version 9.1 was used to apply the test. For data in Tables 3 and 4, the Fisher exact test was applied to two by five tables and two by four tables respectively. The two rows represent whether or not a particular injury or finding was present and the columns represent fracture Type or depth. A two-tailed p-value <0.05 was determined to be significant and indicates a significant relationship between injury/finding and fracture Type or depth.
Results include optic neuropathy, neurologic disability, and mortality. With the exception of mortality, each of these outcomes was measured for patients with at least six months of follow-up in order to minimize the chance of missing a complication. The presence of optic neuropathy was identified in the medical record as the occurrence of a decrease in vision or permanent blindness. Neurologic disability was measured using the Glasgow Outcomes Scale (GOS).12,13 A score of 5 indicated low disability, 4 indicated moderate disability, 3 indicated severe disability, 2 indicated a vegetative state, and 1 indicated death. To simplify this, patients with a GOS score of 2 or 3 were considered disabled. A statistician in the Department of Surgery (G.L.) was consulted who recommended the Fisher exact test as the most appropriate statistical tool for our data. SAS software
Patient demographics and complications Among 1980 patients identified to have craniofacial fractures, 149 (7.5%) patients had frontal bone fractures. There were 121 male patients (81.2%). Average patient age was 32.5 (range 1e83). Motor vehicle accident was the most common mechanism of injury, occurring in 97 patients (65.1 percent, Table 1). Average follow-up for patients who survived was 4.2 years (range 2 dayse12.6 years). Seventy-five patients (50.3%) had a minimum of six months of follow-up. Intracranial injuries occurred in 98 patients (65.8%). Eighteen patients (12.1%) developed optic neuropathy. Neurologic disability occurred in 16 patients (10.7%). No patients developed meningitis, mucocele, or intracranial
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
4
R.K. Garg et al. Table 1 Patient characteristics.
demographics
and
fracture No. (%)
Gender Male Female Injury mechanism MVA Fall Sports Fight Work related Other Fracture type 1 2 3 4 5 Fracture depth A B C D
121 (81.2) 28 (18.8) 97 28 7 9 2 6
(65.1) (18.8) (4.7) (6.04) (1.34) (4.0)
51 23 9 28 38
(34.2) (15.4) (6.0) (18.8) (25.5)
14 66 34 35
(9.4) (44.3) (22.82) (23.49)
abscesses. Eighteen patients (12.1%) died soon after arrival in the emergency room or during their hospital admission.
Frontal bone fracture types Firstly, a comparison was performed of non-vertical and vertical frontal bone fractures, which was the primary
distinction of interest. Imaging analysis revealed a total of 51 non-vertical (Type 1) fractures (34.2%, Figure 3) and 98 vertical (Type 2e5) fractures (65.8%, Figure 4). Motor vehicle accident was the most common mechanism of injury for both non-vertical (70.6%) and vertical (62.2%) fracture types, and there was no significant difference in cause of injury between these two groups (p Z 0.3670). There was no significant difference in optic neuropathy (p Z 0.1139) between vertical and non-vertical fractures. Although vertical fracture types 2, 4 and 5 extended into the orbit, 86.3% of non-vertical fractures isolated to the frontal sinus had associated orbital fractures, which may explain the similar incidence of optic neuropathy. There was also no significant difference in disability (p Z 0.0567) between patients with vertical and non-vertical fractures, but vertical fractures had a significantly higher rate of intracranial injuries (p < 0.05, Table 2). In addition, mortality was significantly higher in patients with vertical fractures (n Z 18, 12.1%) compared to patients with nonvertical fractures (n Z 0, 0%, p Z 0.0004). Given the significant association of vertical fractures with intracranial injuries and mortality, further analysis was performed to identify differences among vertical fractures. A significant relationship was noted between fracture type and the occurrence of intracranial injuries as identified by CT, with 75% of Type 4 and 95% of Type 5 fractures having an associated intracranial injury compared to 65 percent of Type 2 and 67 percent of Type 3 fractures (p < 0.0001, Table 3). Similarly there was a significant relationship between fracture type and outcomes of optic neuropathy, neurologic disability, and mortality, with Types 4 and 5 fractures having the worst outcomes (p < 0.05). Type 4 and 5 fractures were further compared given their strong association with adverse outcomes. Although there was a significant increase in intracranial injuries in patients with Type 5 compared to Type 4 fractures (p Z 0.03), there was no significant difference in optic neuropathy (p Z 0.13), disability (p Z 0.62), and mortality (p Z 0.36).
Influence of age, midface fracture and skull base extension on outcome
Figure 3 Type 1 frontal bone fracture isolated to the bone immediately surrounding the frontal sinus without a vertical fracture trajectory.
There were 131 adults (87.9 percent) and 18 children (12.1 percent) age 12 or younger in this study. All children except five had frontal sinuses in some phase of development, enabling application of the adult classification scheme to this group. For the five without frontal sinuses, three children had Type 2 vertical fractures only involving the orbit. The remaining 2 children had Type 5 fractures in which the vertical frontal bone fracture crossed the midline and involved bilateral orbits. Subsequently, a comparison of children and adults with frontal bone fractures showed no significant difference in fracture type (p Z 0.59) or depth (p Z 0.66). Additionally, there was no significant difference in intracranial injuries (p Z 0.79), optic neuropathy (p Z 0.13), disability (p Z 0.25), and mortality (p Z 1.00). Midface fractures including zygomaticomaxillary complex, nasoorbitoethmoid, and Le Fort 1, 2, and 3 fractures occurred in a total of 83 patients (55.7%). These fractures
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+ A novel classification of frontal bone fractures
MODEL
5
Figure 4 Patients with vertical frontal bone fractures. Type 2 fracture shown with involvement of the orbit (arrow) but not the frontal sinus. Type 3 fracture depicted with involvement of the frontal sinus (arrow) but not the orbit. Type 4 fracture involves the ipsilateral right frontal sinus and orbit (arrows). Type 5 fracture involves the right frontal sinus and extends inferiorly into the bilateral orbits (arrows).
Table 2 Comparison of vertical and non-vertical frontal bone fractures. Vertical Non-vertical pa Fx No. (%) Fx No. (%) Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
35 (35.71)
6 (11.76)
0.0019
38 34 10 20 22
5 4 0 1 1
0.0001 0.0003 0.0159 0.0012 0.0006
(38.78) (34.69) (10.20) (20.41) (22.45)
(9.8) (7.84) (0.00) (1.96) (1.96)
78 (79.59) 20 (39.22)
11 (23.40) 14 (29.79) 18 (18.37)
2 (7.14) 2 (7.14) 0 (0.00)
<0.0001
0.1139 0.0567 0.0004
P values < 0.05 are written in bold. a Fisher’s exact test was used for statistical analysis. b Data reported for patients with a minimum of 6 months of follow-up.
as a group occurred significantly more often in patients with non-vertical (Type 1) fractures (n Z 35, 68.6%) compared to patients with vertical (Type 2e5) fractures (n Z 48, 49.0%, p Z 0.02). However, there was no significant relationship between the presence of midface fractures and the development of intracranial injuries (p Z 0.30), optic neuropathy (p Z 1.00), disability (p Z 0.14), or mortality (p Z 0.07). The relationship between depth of fracture extension into the skull base and complications was also assessed. Type 2e5 vertical fractures penetrated the middle (depth C) or posterior (depth D) cranial fossa (n Z 61, 62.2%) significantly more often than Type 1 fractures (n Z 8, 15.7%; p Z 0.0001, Figure 5). Among vertical fractures, Types 4 and 5 fractures penetrated the middle or posterior cranial fossa significantly more often than Types 2 and 3 fractures (n Z 52, 78.8% v. n Z 9, 28.1%; p < 0.0001). There was a significant trend towards fractures with deeper extension into the skull base having higher intracranial injury rates (Table 4). The intracranial complication rate was 14.3% for depth A fractures, 57.6% for depth B fractures, 76.5% for depth C fractures, and 91.4% for depth D fractures (p < 0.0001). There was also a trend towards significantly higher rates of optic neuropathy, disability and mortality as fractures extended deeper into the skull base (p < 0.05).
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
6
R.K. Garg et al. Table 3
Relationship between fracture type and intracranial injuries, optic neuropathy, disability, and mortality. pa
Fracture type
1
2
3
4
5
Total number of fractures
51
23
9
28
38
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
6 5 4 0 1 1 20
6 5 3 1 2 2 15
2 2 1 1 1 1 6
9 9 8 2 4 5 21
18 22 22 6 13 14 36
Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
(11.76) (9.80) (7.84) (0.00) (1.96) (1.96) (39.22)
2 (7.14) 2 (7.14) 0 (0.00)
(26.09) (21.74) (13.04) (4.35) (8.70) (8.70) (65.22)
0 (0.00) 1 (16.67) 2 (8.70)
(22.22) (22.22) (11.11) (11.11) (11.11) (11.11) (66.67)
0 (0.00) 0 (0.00) 2 (22.22)
(32.14) (32.14) (28.57) (7.14) (14.29) (17.86) (75.00)
7 (50.0) 6 (42.86) 4 (14.29)
(47.37) (57.89) (57.89) (15.79) (34.21) (36.84) (94.74)
4 (20.0) 7 (35.00) 10 (26.32)
0.0049 <0.0001 <0.0001 0.0203 0.0004 0.0001 <0.0001 0.0085 0.0416 0.0006
a
Fisher’s exact test was used for statistical analysis, and the p value reflects whether the trend demonstrated between variables of fracture type and intracranial injuries or outcomes is significant. b Data reported for patients with a minimum of 6 months of follow-up.
Comparison of frontal bone fracture location and skull base extension depth Given the significant association of Type 4 and 5 frontal bone fractures and skull base extension depths C and D with complications ranging from intracranial hemorrhage to blindness, disability and death, a comparison was performed of complication rates between these two groups. There was no significant difference in rates of optic neuropathy (p Z 0.66), disability (p Z 0.82) and mortality (p Z 0.84). A crossmatch was also performed between patients with Type 4 or 5 fractures and skull base extension depth C or D in order to identify whether combining fracture type and depth would capture a greater number of complications than either parameter alone. There was no significant difference in complication rates among patients with Type 4 or 5 and depth C or D fractures compared to either Type or depth alone.
Discussion To our knowledge, no previous study has described the broad spectrum of frontal bone fractures and defined which injury patterns carry the worst prognosis. Although vertical craniofacial fractures have been alluded to in the literature,9,11,14e16 this study was designed to characterize the broad range of frontal bone fracture patterns and their association with intracranial injuries, optic neuropathy, disability, and mortality. We found that vertical frontal bone fractures have a significantly higher complication rate and mortality than non-vertical fractures, and that vertical fractures transecting both the orbit and frontal sinus or with extension into the middle or posterior cranial fossa carry the worst prognosis. The importance of analyzing both skull base and midface involvement in patients with frontal bone fractures is evident from the work of Manson et al.11 In their cadaveric
and retrospective study of patients with fractures involving the frontal bone and skull base, they found that both vertical linear and comminuted fractures of the frontal bone may extend into the skull base. Their study identified the highest rates of CSF leak and infection in patients with comminuted frontal bone fractures with skull base extension and associated midface fractures. Our study builds off this work by demonstrating that the specific fracture trajectory through the frontal bone and depth of skull base penetration are significantly related to the severity of intracranial injuries and complications. Furthermore, we demonstrate that extension of vertical fractures into both the frontal sinus and orbit is associated with a poorer prognosis. We were unable to identify a previous study that makes such a correlation between location of frontal bone injury, depth of fracture extension into the skull base, and patient outcome. Some important differences between our study and Manson’s include the absence of central nervous system infection as a complication of trauma even though the average follow up time was over 4 years. Also, in our cohort, midface fractures were more common in patients with non-vertical fractures than vertical fractures and not significantly associated with intracranial injuries, optic neuropathy, disability or mortality. When assessing a patient with frontal bone trauma, our work demonstrates that an important determination is whether the frontal bone fracture has a vertical trajectory. Vertical fractures are hypothesized to occur because the initial trauma to the frontal bone occurs outside the sturdy buttress surrounding the frontal sinus, enabling the fracture to extend into adjacent bony structures. Non-vertical frontal sinus fractures, on the other hand, are hypothesized to occur because the bone immediately surrounding the sinus forms a sturdy buttress resistant to fracture.11,17,18 All 18 patients in our study who died had vertical (Type 2e5) fractures, indicating that this fracture pattern is highly associated with mortality. Of note, there was no significant difference in injury mechanism between vertical and non-
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures
7
Figure 5 Patients with frontal bone fractures and skull base extension. Depth A fracture demonstrated with fracture of anterior and posterior frontal bone tables (arrow) but no skull base involvement. Depth B fracture shown with involvement of the fovea ethmoidalis (arrows). Depth C fracture shown extending to the fovea ethmoidalis, sphenoid sinus, greater sphenoid wing, and the pituitary fossa (arrow). Depth D fracture depicted with extension beyond the clivus (arrow) into the foramen magnum.
Table 4
Relationship between fracture depth and intracranial injuries, optic neuropathy, disability, and mortality. pa
Fracture depth
A
B
C
D
Total number of fractures
14
66
34
35
No. (%)
No. (%)
No. (%)
No. (%)
1 0 1 0 1 1 2
16 10 9 0 4 2 38
7 13 9 1 3 6 26
17 20 19 9 13 14 32
Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
(7.14) (0) (7.14) (0.0) (7.14) (7.14) (14.29)
0 (0) 0 (0) 1 (7.14)
(24.24) (15.15) (13.64) (0.0) (6.06) (3.03) (57.58)
4 (6.06) 5 (15.63) 1 (1.52)
(20.59) (38.24) (26.47) (2.94) (8.82) (17.65) (76.47)
7 (20.59) 2 (11.11) 3 (8.82)
(48.57) (57.14) (54.29) (25.71) (37.14) (40.0) (91.43)
0.0100 <0.0001 0.0001 <0.0001 0.0004 <0.0001 <0.0001
7 (20.0) 9 (52.94) 13 (37.14)
0.0352 0.0052 <0.0001
a Fisher’s exact test was used for statistical analysis, and the p value reflects whether the trend demonstrated between variables of fracture depth and intracranial injuries or outcomes is significant. b Data reported for patients with a minimum of 6 months of follow-up.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
8 vertical fracture types, suggesting that differences in outcomes between vertical and non-vertical fracture groups was not a mere consequence of injury mechanism. Although the primary distinction in this study was between vertical and non-vertical fractures, a sub-analysis of vertical fractures was performed to better understand differences in patient outcomes within this heterogenous group of injuries. We found that vertical fractures with extension through both the frontal sinus and orbit (Types 4 and 5) had the strongest association with intracranial injuries, optic neuropathy, disability and mortality. We did not identify a significant difference in outcomes between Types 4 and 5 fractures, possibly due to limited sample size of each of these subgroups. Nonetheless, when considering vertical fracture types, we conclude that Types 4 and 5 carry the worst prognosis. Based on this analysis, the clinical practitioner need not memorize all vertical fracture subtypes. It is enough to take away from this study that vertical fractures have a significantly worse prognosis than non-vertical fractures, especially those with extension into both the frontal sinus and orbit. Fracture extension into the skull base was also demonstrated to be significantly associated with intracranial injuries, optic neuropathy, disability and mortality. Complication rates were highest for patients with fracture extension into the middle (depth C) or posterior (depth D) cranial fossa. Although patients with Type 4 and 5 fractures were most likely to have fracture extension into the middle or posterior cranial fossa, we were able to demonstrate similar adverse patient outcomes between Type 4e5 frontal bone fractures and fractures with middle or posterior cranial fossa extension without regard for fracture type. Therefore, assessment of either fracture type or skull base extension depth may be useful in predicting the likelihood of intracranial injuries and complications in patients with frontal bone fractures. Although this study is limited by its retrospective approach, it does indicate that there is a wide spectrum of frontal bone injuries and that characterizing fracture trajectory, location, and skull base involvement using high resolution CT may be helpful in considering patient prognosis. Furthermore, our study demonstrates that vertical fractures with extension from the frontal bone into the midface occur in both children and adults. Although previous work has described “vertical” fractures beginning in the frontal bone and extending into the midface as a unique feature of pediatric facial trauma,9,19 these injuries were clearly identified in adults as well in our study. Based on our findings, we conclude that vertical frontal bone fractures with frontal sinus and orbital involvement or fractures with middle or posterior cranial fossa extension carry the worst prognosis.
Conclusions Frontal bone fractures occur along a continuum ranging from frontal sinus fractures to vertical frontal bone fractures with sinus, orbital and skull base extension. Here we have presented a novel classification system of frontal bone fractures that accounts for involvement of the frontal sinus, orbits, and skull base. Frontal sinus fractures without
R.K. Garg et al. a vertical trajectory have the best prognosis. On the other hand, vertical fractures traversing the frontal sinus and orbit, and fractures penetrating the middle or posterior cranial fossa have the worst prognosis. Awareness of these findings may help surgeons and families stratify the severity of a patient’s frontal bone injury and anticipate potential complications.
Ethical approval The University of Wisconsin Health Sciences IRB provided approval for this retrospective chart review under protocol number 2013-0220.
Funding None.
Conflict of interest None.
References 1. Follmar KE, Debruijn M, Baccarani A, et al. Concomitant injuries in patients with panfacial fractures. J Trauma 2007;63: 831e5. 2. Alvi A, Doherty T, Lewen G. Facial fractures and concomitant injuries in trauma patients. Laryngoscope 2003;113:102e6. 3. Doonquah L, Brown P, Mullings W. Management of frontal sinus fractures. Oral Maxillofac Surg Clin North Am 2012;24:265e74. ix. 4. Gassner R, Tuli T, Ha ¨chl O, et al. Cranio-maxillofacial trauma: a 10 year review of 9543 cases with 21,067 injuries. J Craniomaxillofac Surg 2003;31:51e61. 5. Stanwix MG, Nam AJ, Manson PN, et al. Critical computed tomographic diagnostic criteria for frontal sinus fractures. J Oral Maxillofac Surg 2010;68:2714e22. 6. Echo A, Troy JS, Hollier Jr LH. Frontal sinus fractures. Semin Plast Surg 2010;24:375e82. 7. Pappachan B, Alexander M. Correlating facial fractures and cranial injuries. J Oral Maxillofac Surg 2006;64:1023e9. 8. Gentry LR. Facial trauma and associated brain damage. Radiol Clin North Am 1989;27:435e46. 9. Moore MH, David DJ, Cooter RD. Oblique craniofacial fractures in children. J Craniofac Surg 1990;1:4e7. 10. Losee JE, Afifi A, Jiang S, et al. Pediatric orbital fractures: classification, management, and early follow-up. Plast Reconstr Surg 2008;122:886e97. 11. Manson PN, Stanwix MG, Yaremchuk MJ, et al. Frontobasal fractures: anatomical classification and clinical significance. Plast Reconstr Surg 2009;124:2096e106. 12. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480e4. 13. Jennett B, Snoek J, Bond MR, et al. Disability after severe head injury: observations on the use of the Glasgow outcome scale. J Neurol Neurosurg Psychiatry 1981;44:285e93. 14. Kelamis JA, Mundinger GS, Feiner JM, et al. Isolated bilateral zygomatic arch fractures of the facial skeleton are associated with skull base fractures. Plast Reconstr Surg 2011;128: 962e70.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures 15. Nelson EL, Melton 3rd LJ, Annegers JF, et al. Incidence of skull fractures in Olmsted County, Minnesota. Neurosurgery 1984; 15:318e24. 16. Samii M, Tatagiba M. Skull base trauma: diagnosis and management. Neurol Res 2002;24:147e56. 17. Luce EA, Tubb TD, Moore AM. Review of 1,000 major facial fractures and associated injuries. Plast Reconstr Surg 1979;63: 26e30.
9 18. Schultz RC. Supraorbital and glabellar fractures. Plast Reconstr Surg 1970;45:227e33. 19. Rubinstein R, Afifi A, Jiang S, et al. Oblique pediatric craniofacial fractures. In: Paper presented at: 65th anniversary meeting American Cleft Palate-Craniofacial Association; 2008 [Pittsburgh].
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
MODEL
Journal of Plastic, Reconstructive & Aesthetic Surgery (2015) xx, 1e9
A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension Ravi K. Garg a, Ahmed M. Afifi a,d,*, Jennifer Gassner a, Michael J. Hartman b, Glen Leverson c, Timothy W. King a, Michael L. Bentz a, Lindell R. Gentry b a Division of Plastic and Reconstructive Surgery, Department of Surgery, University of Wisconsin, Madison, WI, USA b Section of Head and Neck Imaging, Department of Radiology, University of Wisconsin, Madison, WI, USA c Department of Surgery, University of Wisconsin, Madison, WI, USA d Division of Plastic Surgery, Cairo University, Cairo, Egypt
Received 5 November 2014; accepted 2 February 2015
KEYWORDS Frontal bone fracture; Craniomaxillofacial trauma; Orbital fracture; Frontal sinus fracture; Skull base trauma
Summary Purpose: The broad spectrum of frontal bone fractures, including those with orbital and skull base extension, is poorly understood. We propose a novel classification scheme for frontal bone fractures. Methods: Maxillofacial CT scans of trauma patients were reviewed over a five year period, and frontal bone fractures were classified: Type 1: Frontal sinus fracture without vertical extension. Type 2: Vertical fracture through the orbit without frontal sinus involvement. Type 3: Vertical fracture through the frontal sinus without orbit involvement. Type 4: Vertical fracture through the frontal sinus and ipsilateral orbit. Type 5: Vertical fracture through the frontal sinus and contralateral or bilateral orbits. We also identified the depth of skull base extension, and performed a chart review to identify associated complications. Results: 149 frontal bone fractures, including 51 non-vertical frontal sinus (Type 1, 34.2%) and 98 vertical (Types 2e5, 65.8%) fractures were identified. Vertical fractures penetrated the middle or posterior cranial fossa significantly more often than non-vertical fractures (62.2 v. 15.7%, p Z 0.0001) and had a significantly higher mortality rate (18.4 v. 0%, p < 0.05). Vertical fractures with frontal sinus and orbital extension, and fractures that penetrated the middle or posterior cranial fossa had the strongest association with intracranial injuries, optic neuropathy, disability, and death (p < 0.05).
* Corresponding author. Division of Plastic and Reconstructive Surgery, University of Wisconsin, Madison 608-265-2535, USA. E-mail address: [email protected] (A.M. Afifi). http://dx.doi.org/10.1016/j.bjps.2015.02.021 1748-6815/ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
2
R.K. Garg et al. Conclusions: Vertical frontal bone fractures carry a worse prognosis than frontal bone fractures without a vertical pattern. In addition, vertical fractures with extension into the frontal sinus and orbit, or with extension into the middle or posterior cranial fossa have the highest complication rate and mortality. ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Introduction Although frontal sinus fractures and their associated complications have been described extensively,1e6 fracture patterns of the frontal bone as a broad unit of the craniofacial skeleton have not been rigorously characterized. Considering that the frontal bone is the most frequently fractured cranial bone in craniofacial trauma patients and accounts for 37 percent of cranial fractures,7 it is important to understand frontal bone fracture patterns and related complications in greater detail. The important elements to include in a thorough characterization of frontal bone fractures are the status of adjacent bony structures, including the frontal sinus, orbit, and skull base.8 Previous studies have shown that frontal bone fractures may acquire an “oblique” or vertical trajectory through the frontal bone and extend to adjacent bones. In particular, vertical fractures that begin in the frontal bone and penetrate the midface have been described in children as a common fracture pattern related to their prominent foreheads and underdeveloped paranasal sinuses.9,10 The occurrence of vertical fractures has not been made as explicit in adults, although “linear” fractures were recently described in an adult population as originating in the frontal bone and extending to the skull base.11 The literature therefore supports the possibility of vertically oriented fractures in multiple contexts, but a comprehensive description of frontal bone fractures that accounts for involvement of adjacent bony structures is needed. To our knowledge, no study has examined the spectrum of frontal bone fractures with consideration of frontal sinus, orbital and skull base involvement in both children and adults. An improved understanding of different frontal bone fracture types and their associated morbidity and mortality may enhance our ability to distinguish injury severity among patients with frontal bone trauma and provide prognostic information that can assist with patient counseling and treatment planning. Here we classify different frontal bone fracture patterns and their associated complications.
Methods After receiving approval from the University of Wisconsin Health Sciences IRB, we used ICD-9 codes for calvarial, facial, mandibular and cervical spine trauma to retrospectively identify all patients with craniofacial trauma over a five year interval. Subsequently, all CT scans were reviewed
by a neuroradiology attending to identify those individuals with frontal bone fractures. Frontal bone fractures were primarily distinguished as having a non-vertical or vertical trajectory. Type 1 fractures were defined as comminuted fractures of the frontal sinus without a vertical trajectory. Vertical fractures were broken down into 4 distinct types, Types 2e5, as it was clear from radiologic analysis that these fractures represented a spectrum of injury patterns (Figure 1). Type 2 fractures are vertical fractures involving the orbit but not the frontal sinus. Type 3 fractures are vertical fractures involving the frontal bone and sinus but not the orbit. Type 4 fractures involve both the frontal sinus and the ipsilateral orbit. Type 5 fractures cross the midline of the face, involving the frontal sinus and the contralateral or bilateral orbits. The term non-vertical fracture is therefore used to describe Type 1 fractures while the term vertical fracture is used to describe Type 2e5 fractures. In the case of children without a frontal sinus present, the classification scheme still applies with a few modifications. A Type 1 fracture is a non-vertical fracture through the frontal bone. A Type 2 fracture is a vertical fracture through the orbit only. A Type 3 fracture is a vertical fracture through the frontal bone that does not involve the orbit. There is no Type 4 fracture since there is no frontal sinus. A Type 5 fracture is a vertical fracture through the frontal bone that involves the contralateral or bilateral orbits. If a frontal sinus is present in some phase of development, then the adult classification scheme is used. The depth of skull base extension was also classified for all fractures (Figure 2). Depth A is defined as involvement of the frontal bone without extension into the skull base. Depth B is characterized by extension into the anterior cranial fossa (orbital roof, fovea ethmoidalis, cribriform plate). Depth C fractures extend into the middle cranial fossa (sella, sphenoid body, carotid canal, optic chiasm sulcus). Depth D fractures involve the posterior cranial fossa (clivus, petromastoid temporal bone, petrosal segment of the carotid canal). CT images were also reviewed by a neuroradiology attending to characterize intracranial injuries for all patients with frontal bone fractures. Injuries of interest included subarachnoid hemorrhage, subdural hematoma, epidural hematoma, stroke, subfalcine herniation, and transtentorial herniation. Additionally, the presence of midface fractures that fit accepted definitions of zygomaticomaxillary complex, nasoorbitoethmoid, and Le Fort 1, 2 and 3 fractures was described. Medical records were reviewed to characterize patient demographics and outcomes. The outcomes we measured
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures
3
Figure 2 Skull base penetration depths. Depth A fractures involve the anterior table of the frontal bone with or without posterior table involvement and do not extend into the anterior cranial fossa (purple). Depth B fractures involve the floor of the anterior cranial fossa (blue). Depth C fractures involve the middle cranial fossa (yellow). Depth D fractures extend into the posterior cranial fossa (red).
Figure 1 Frontal bone fracture types. Type 1 fractures are isolated to the frontal sinus without a vertical trajectory (purple). Type 2 fractures are vertically oriented and extend into the orbit but not the frontal sinus (blue). Type 3 fractures are vertically oriented and extend into the frontal sinus but not the orbit (yellow). Type 4 fractures are vertically oriented and extend into ipsilateral frontal sinus and orbit (green). Type 5 fractures extend into the frontal sinus and extend into the orbit on both sides of the face or the contralateral side of the face (red).
version 9.1 was used to apply the test. For data in Tables 3 and 4, the Fisher exact test was applied to two by five tables and two by four tables respectively. The two rows represent whether or not a particular injury or finding was present and the columns represent fracture Type or depth. A two-tailed p-value <0.05 was determined to be significant and indicates a significant relationship between injury/finding and fracture Type or depth.
Results include optic neuropathy, neurologic disability, and mortality. With the exception of mortality, each of these outcomes was measured for patients with at least six months of follow-up in order to minimize the chance of missing a complication. The presence of optic neuropathy was identified in the medical record as the occurrence of a decrease in vision or permanent blindness. Neurologic disability was measured using the Glasgow Outcomes Scale (GOS).12,13 A score of 5 indicated low disability, 4 indicated moderate disability, 3 indicated severe disability, 2 indicated a vegetative state, and 1 indicated death. To simplify this, patients with a GOS score of 2 or 3 were considered disabled. A statistician in the Department of Surgery (G.L.) was consulted who recommended the Fisher exact test as the most appropriate statistical tool for our data. SAS software
Patient demographics and complications Among 1980 patients identified to have craniofacial fractures, 149 (7.5%) patients had frontal bone fractures. There were 121 male patients (81.2%). Average patient age was 32.5 (range 1e83). Motor vehicle accident was the most common mechanism of injury, occurring in 97 patients (65.1 percent, Table 1). Average follow-up for patients who survived was 4.2 years (range 2 dayse12.6 years). Seventy-five patients (50.3%) had a minimum of six months of follow-up. Intracranial injuries occurred in 98 patients (65.8%). Eighteen patients (12.1%) developed optic neuropathy. Neurologic disability occurred in 16 patients (10.7%). No patients developed meningitis, mucocele, or intracranial
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
4
R.K. Garg et al. Table 1 Patient characteristics.
demographics
and
fracture No. (%)
Gender Male Female Injury mechanism MVA Fall Sports Fight Work related Other Fracture type 1 2 3 4 5 Fracture depth A B C D
121 (81.2) 28 (18.8) 97 28 7 9 2 6
(65.1) (18.8) (4.7) (6.04) (1.34) (4.0)
51 23 9 28 38
(34.2) (15.4) (6.0) (18.8) (25.5)
14 66 34 35
(9.4) (44.3) (22.82) (23.49)
abscesses. Eighteen patients (12.1%) died soon after arrival in the emergency room or during their hospital admission.
Frontal bone fracture types Firstly, a comparison was performed of non-vertical and vertical frontal bone fractures, which was the primary
distinction of interest. Imaging analysis revealed a total of 51 non-vertical (Type 1) fractures (34.2%, Figure 3) and 98 vertical (Type 2e5) fractures (65.8%, Figure 4). Motor vehicle accident was the most common mechanism of injury for both non-vertical (70.6%) and vertical (62.2%) fracture types, and there was no significant difference in cause of injury between these two groups (p Z 0.3670). There was no significant difference in optic neuropathy (p Z 0.1139) between vertical and non-vertical fractures. Although vertical fracture types 2, 4 and 5 extended into the orbit, 86.3% of non-vertical fractures isolated to the frontal sinus had associated orbital fractures, which may explain the similar incidence of optic neuropathy. There was also no significant difference in disability (p Z 0.0567) between patients with vertical and non-vertical fractures, but vertical fractures had a significantly higher rate of intracranial injuries (p < 0.05, Table 2). In addition, mortality was significantly higher in patients with vertical fractures (n Z 18, 12.1%) compared to patients with nonvertical fractures (n Z 0, 0%, p Z 0.0004). Given the significant association of vertical fractures with intracranial injuries and mortality, further analysis was performed to identify differences among vertical fractures. A significant relationship was noted between fracture type and the occurrence of intracranial injuries as identified by CT, with 75% of Type 4 and 95% of Type 5 fractures having an associated intracranial injury compared to 65 percent of Type 2 and 67 percent of Type 3 fractures (p < 0.0001, Table 3). Similarly there was a significant relationship between fracture type and outcomes of optic neuropathy, neurologic disability, and mortality, with Types 4 and 5 fractures having the worst outcomes (p < 0.05). Type 4 and 5 fractures were further compared given their strong association with adverse outcomes. Although there was a significant increase in intracranial injuries in patients with Type 5 compared to Type 4 fractures (p Z 0.03), there was no significant difference in optic neuropathy (p Z 0.13), disability (p Z 0.62), and mortality (p Z 0.36).
Influence of age, midface fracture and skull base extension on outcome
Figure 3 Type 1 frontal bone fracture isolated to the bone immediately surrounding the frontal sinus without a vertical fracture trajectory.
There were 131 adults (87.9 percent) and 18 children (12.1 percent) age 12 or younger in this study. All children except five had frontal sinuses in some phase of development, enabling application of the adult classification scheme to this group. For the five without frontal sinuses, three children had Type 2 vertical fractures only involving the orbit. The remaining 2 children had Type 5 fractures in which the vertical frontal bone fracture crossed the midline and involved bilateral orbits. Subsequently, a comparison of children and adults with frontal bone fractures showed no significant difference in fracture type (p Z 0.59) or depth (p Z 0.66). Additionally, there was no significant difference in intracranial injuries (p Z 0.79), optic neuropathy (p Z 0.13), disability (p Z 0.25), and mortality (p Z 1.00). Midface fractures including zygomaticomaxillary complex, nasoorbitoethmoid, and Le Fort 1, 2, and 3 fractures occurred in a total of 83 patients (55.7%). These fractures
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+ A novel classification of frontal bone fractures
MODEL
5
Figure 4 Patients with vertical frontal bone fractures. Type 2 fracture shown with involvement of the orbit (arrow) but not the frontal sinus. Type 3 fracture depicted with involvement of the frontal sinus (arrow) but not the orbit. Type 4 fracture involves the ipsilateral right frontal sinus and orbit (arrows). Type 5 fracture involves the right frontal sinus and extends inferiorly into the bilateral orbits (arrows).
Table 2 Comparison of vertical and non-vertical frontal bone fractures. Vertical Non-vertical pa Fx No. (%) Fx No. (%) Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
35 (35.71)
6 (11.76)
0.0019
38 34 10 20 22
5 4 0 1 1
0.0001 0.0003 0.0159 0.0012 0.0006
(38.78) (34.69) (10.20) (20.41) (22.45)
(9.8) (7.84) (0.00) (1.96) (1.96)
78 (79.59) 20 (39.22)
11 (23.40) 14 (29.79) 18 (18.37)
2 (7.14) 2 (7.14) 0 (0.00)
<0.0001
0.1139 0.0567 0.0004
P values < 0.05 are written in bold. a Fisher’s exact test was used for statistical analysis. b Data reported for patients with a minimum of 6 months of follow-up.
as a group occurred significantly more often in patients with non-vertical (Type 1) fractures (n Z 35, 68.6%) compared to patients with vertical (Type 2e5) fractures (n Z 48, 49.0%, p Z 0.02). However, there was no significant relationship between the presence of midface fractures and the development of intracranial injuries (p Z 0.30), optic neuropathy (p Z 1.00), disability (p Z 0.14), or mortality (p Z 0.07). The relationship between depth of fracture extension into the skull base and complications was also assessed. Type 2e5 vertical fractures penetrated the middle (depth C) or posterior (depth D) cranial fossa (n Z 61, 62.2%) significantly more often than Type 1 fractures (n Z 8, 15.7%; p Z 0.0001, Figure 5). Among vertical fractures, Types 4 and 5 fractures penetrated the middle or posterior cranial fossa significantly more often than Types 2 and 3 fractures (n Z 52, 78.8% v. n Z 9, 28.1%; p < 0.0001). There was a significant trend towards fractures with deeper extension into the skull base having higher intracranial injury rates (Table 4). The intracranial complication rate was 14.3% for depth A fractures, 57.6% for depth B fractures, 76.5% for depth C fractures, and 91.4% for depth D fractures (p < 0.0001). There was also a trend towards significantly higher rates of optic neuropathy, disability and mortality as fractures extended deeper into the skull base (p < 0.05).
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
6
R.K. Garg et al. Table 3
Relationship between fracture type and intracranial injuries, optic neuropathy, disability, and mortality. pa
Fracture type
1
2
3
4
5
Total number of fractures
51
23
9
28
38
No. (%)
No. (%)
No. (%)
No. (%)
No. (%)
6 5 4 0 1 1 20
6 5 3 1 2 2 15
2 2 1 1 1 1 6
9 9 8 2 4 5 21
18 22 22 6 13 14 36
Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
(11.76) (9.80) (7.84) (0.00) (1.96) (1.96) (39.22)
2 (7.14) 2 (7.14) 0 (0.00)
(26.09) (21.74) (13.04) (4.35) (8.70) (8.70) (65.22)
0 (0.00) 1 (16.67) 2 (8.70)
(22.22) (22.22) (11.11) (11.11) (11.11) (11.11) (66.67)
0 (0.00) 0 (0.00) 2 (22.22)
(32.14) (32.14) (28.57) (7.14) (14.29) (17.86) (75.00)
7 (50.0) 6 (42.86) 4 (14.29)
(47.37) (57.89) (57.89) (15.79) (34.21) (36.84) (94.74)
4 (20.0) 7 (35.00) 10 (26.32)
0.0049 <0.0001 <0.0001 0.0203 0.0004 0.0001 <0.0001 0.0085 0.0416 0.0006
a
Fisher’s exact test was used for statistical analysis, and the p value reflects whether the trend demonstrated between variables of fracture type and intracranial injuries or outcomes is significant. b Data reported for patients with a minimum of 6 months of follow-up.
Comparison of frontal bone fracture location and skull base extension depth Given the significant association of Type 4 and 5 frontal bone fractures and skull base extension depths C and D with complications ranging from intracranial hemorrhage to blindness, disability and death, a comparison was performed of complication rates between these two groups. There was no significant difference in rates of optic neuropathy (p Z 0.66), disability (p Z 0.82) and mortality (p Z 0.84). A crossmatch was also performed between patients with Type 4 or 5 fractures and skull base extension depth C or D in order to identify whether combining fracture type and depth would capture a greater number of complications than either parameter alone. There was no significant difference in complication rates among patients with Type 4 or 5 and depth C or D fractures compared to either Type or depth alone.
Discussion To our knowledge, no previous study has described the broad spectrum of frontal bone fractures and defined which injury patterns carry the worst prognosis. Although vertical craniofacial fractures have been alluded to in the literature,9,11,14e16 this study was designed to characterize the broad range of frontal bone fracture patterns and their association with intracranial injuries, optic neuropathy, disability, and mortality. We found that vertical frontal bone fractures have a significantly higher complication rate and mortality than non-vertical fractures, and that vertical fractures transecting both the orbit and frontal sinus or with extension into the middle or posterior cranial fossa carry the worst prognosis. The importance of analyzing both skull base and midface involvement in patients with frontal bone fractures is evident from the work of Manson et al.11 In their cadaveric
and retrospective study of patients with fractures involving the frontal bone and skull base, they found that both vertical linear and comminuted fractures of the frontal bone may extend into the skull base. Their study identified the highest rates of CSF leak and infection in patients with comminuted frontal bone fractures with skull base extension and associated midface fractures. Our study builds off this work by demonstrating that the specific fracture trajectory through the frontal bone and depth of skull base penetration are significantly related to the severity of intracranial injuries and complications. Furthermore, we demonstrate that extension of vertical fractures into both the frontal sinus and orbit is associated with a poorer prognosis. We were unable to identify a previous study that makes such a correlation between location of frontal bone injury, depth of fracture extension into the skull base, and patient outcome. Some important differences between our study and Manson’s include the absence of central nervous system infection as a complication of trauma even though the average follow up time was over 4 years. Also, in our cohort, midface fractures were more common in patients with non-vertical fractures than vertical fractures and not significantly associated with intracranial injuries, optic neuropathy, disability or mortality. When assessing a patient with frontal bone trauma, our work demonstrates that an important determination is whether the frontal bone fracture has a vertical trajectory. Vertical fractures are hypothesized to occur because the initial trauma to the frontal bone occurs outside the sturdy buttress surrounding the frontal sinus, enabling the fracture to extend into adjacent bony structures. Non-vertical frontal sinus fractures, on the other hand, are hypothesized to occur because the bone immediately surrounding the sinus forms a sturdy buttress resistant to fracture.11,17,18 All 18 patients in our study who died had vertical (Type 2e5) fractures, indicating that this fracture pattern is highly associated with mortality. Of note, there was no significant difference in injury mechanism between vertical and non-
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures
7
Figure 5 Patients with frontal bone fractures and skull base extension. Depth A fracture demonstrated with fracture of anterior and posterior frontal bone tables (arrow) but no skull base involvement. Depth B fracture shown with involvement of the fovea ethmoidalis (arrows). Depth C fracture shown extending to the fovea ethmoidalis, sphenoid sinus, greater sphenoid wing, and the pituitary fossa (arrow). Depth D fracture depicted with extension beyond the clivus (arrow) into the foramen magnum.
Table 4
Relationship between fracture depth and intracranial injuries, optic neuropathy, disability, and mortality. pa
Fracture depth
A
B
C
D
Total number of fractures
14
66
34
35
No. (%)
No. (%)
No. (%)
No. (%)
1 0 1 0 1 1 2
16 10 9 0 4 2 38
7 13 9 1 3 6 26
17 20 19 9 13 14 32
Intracranial injuries Subarachnoid hemorrhage Epidural hematoma Subdural hematoma Stroke Subfalcine herniation Transtentorial herniation Any intracranial injury Outcomesb Optic neuropathy Disability Mortality
(7.14) (0) (7.14) (0.0) (7.14) (7.14) (14.29)
0 (0) 0 (0) 1 (7.14)
(24.24) (15.15) (13.64) (0.0) (6.06) (3.03) (57.58)
4 (6.06) 5 (15.63) 1 (1.52)
(20.59) (38.24) (26.47) (2.94) (8.82) (17.65) (76.47)
7 (20.59) 2 (11.11) 3 (8.82)
(48.57) (57.14) (54.29) (25.71) (37.14) (40.0) (91.43)
0.0100 <0.0001 0.0001 <0.0001 0.0004 <0.0001 <0.0001
7 (20.0) 9 (52.94) 13 (37.14)
0.0352 0.0052 <0.0001
a Fisher’s exact test was used for statistical analysis, and the p value reflects whether the trend demonstrated between variables of fracture depth and intracranial injuries or outcomes is significant. b Data reported for patients with a minimum of 6 months of follow-up.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
8 vertical fracture types, suggesting that differences in outcomes between vertical and non-vertical fracture groups was not a mere consequence of injury mechanism. Although the primary distinction in this study was between vertical and non-vertical fractures, a sub-analysis of vertical fractures was performed to better understand differences in patient outcomes within this heterogenous group of injuries. We found that vertical fractures with extension through both the frontal sinus and orbit (Types 4 and 5) had the strongest association with intracranial injuries, optic neuropathy, disability and mortality. We did not identify a significant difference in outcomes between Types 4 and 5 fractures, possibly due to limited sample size of each of these subgroups. Nonetheless, when considering vertical fracture types, we conclude that Types 4 and 5 carry the worst prognosis. Based on this analysis, the clinical practitioner need not memorize all vertical fracture subtypes. It is enough to take away from this study that vertical fractures have a significantly worse prognosis than non-vertical fractures, especially those with extension into both the frontal sinus and orbit. Fracture extension into the skull base was also demonstrated to be significantly associated with intracranial injuries, optic neuropathy, disability and mortality. Complication rates were highest for patients with fracture extension into the middle (depth C) or posterior (depth D) cranial fossa. Although patients with Type 4 and 5 fractures were most likely to have fracture extension into the middle or posterior cranial fossa, we were able to demonstrate similar adverse patient outcomes between Type 4e5 frontal bone fractures and fractures with middle or posterior cranial fossa extension without regard for fracture type. Therefore, assessment of either fracture type or skull base extension depth may be useful in predicting the likelihood of intracranial injuries and complications in patients with frontal bone fractures. Although this study is limited by its retrospective approach, it does indicate that there is a wide spectrum of frontal bone injuries and that characterizing fracture trajectory, location, and skull base involvement using high resolution CT may be helpful in considering patient prognosis. Furthermore, our study demonstrates that vertical fractures with extension from the frontal bone into the midface occur in both children and adults. Although previous work has described “vertical” fractures beginning in the frontal bone and extending into the midface as a unique feature of pediatric facial trauma,9,19 these injuries were clearly identified in adults as well in our study. Based on our findings, we conclude that vertical frontal bone fractures with frontal sinus and orbital involvement or fractures with middle or posterior cranial fossa extension carry the worst prognosis.
Conclusions Frontal bone fractures occur along a continuum ranging from frontal sinus fractures to vertical frontal bone fractures with sinus, orbital and skull base extension. Here we have presented a novel classification system of frontal bone fractures that accounts for involvement of the frontal sinus, orbits, and skull base. Frontal sinus fractures without
R.K. Garg et al. a vertical trajectory have the best prognosis. On the other hand, vertical fractures traversing the frontal sinus and orbit, and fractures penetrating the middle or posterior cranial fossa have the worst prognosis. Awareness of these findings may help surgeons and families stratify the severity of a patient’s frontal bone injury and anticipate potential complications.
Ethical approval The University of Wisconsin Health Sciences IRB provided approval for this retrospective chart review under protocol number 2013-0220.
Funding None.
Conflict of interest None.
References 1. Follmar KE, Debruijn M, Baccarani A, et al. Concomitant injuries in patients with panfacial fractures. J Trauma 2007;63: 831e5. 2. Alvi A, Doherty T, Lewen G. Facial fractures and concomitant injuries in trauma patients. Laryngoscope 2003;113:102e6. 3. Doonquah L, Brown P, Mullings W. Management of frontal sinus fractures. Oral Maxillofac Surg Clin North Am 2012;24:265e74. ix. 4. Gassner R, Tuli T, Ha ¨chl O, et al. Cranio-maxillofacial trauma: a 10 year review of 9543 cases with 21,067 injuries. J Craniomaxillofac Surg 2003;31:51e61. 5. Stanwix MG, Nam AJ, Manson PN, et al. Critical computed tomographic diagnostic criteria for frontal sinus fractures. J Oral Maxillofac Surg 2010;68:2714e22. 6. Echo A, Troy JS, Hollier Jr LH. Frontal sinus fractures. Semin Plast Surg 2010;24:375e82. 7. Pappachan B, Alexander M. Correlating facial fractures and cranial injuries. J Oral Maxillofac Surg 2006;64:1023e9. 8. Gentry LR. Facial trauma and associated brain damage. Radiol Clin North Am 1989;27:435e46. 9. Moore MH, David DJ, Cooter RD. Oblique craniofacial fractures in children. J Craniofac Surg 1990;1:4e7. 10. Losee JE, Afifi A, Jiang S, et al. Pediatric orbital fractures: classification, management, and early follow-up. Plast Reconstr Surg 2008;122:886e97. 11. Manson PN, Stanwix MG, Yaremchuk MJ, et al. Frontobasal fractures: anatomical classification and clinical significance. Plast Reconstr Surg 2009;124:2096e106. 12. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480e4. 13. Jennett B, Snoek J, Bond MR, et al. Disability after severe head injury: observations on the use of the Glasgow outcome scale. J Neurol Neurosurg Psychiatry 1981;44:285e93. 14. Kelamis JA, Mundinger GS, Feiner JM, et al. Isolated bilateral zygomatic arch fractures of the facial skeleton are associated with skull base fractures. Plast Reconstr Surg 2011;128: 962e70.
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021
+
MODEL
A novel classification of frontal bone fractures 15. Nelson EL, Melton 3rd LJ, Annegers JF, et al. Incidence of skull fractures in Olmsted County, Minnesota. Neurosurgery 1984; 15:318e24. 16. Samii M, Tatagiba M. Skull base trauma: diagnosis and management. Neurol Res 2002;24:147e56. 17. Luce EA, Tubb TD, Moore AM. Review of 1,000 major facial fractures and associated injuries. Plast Reconstr Surg 1979;63: 26e30.
9 18. Schultz RC. Supraorbital and glabellar fractures. Plast Reconstr Surg 1970;45:227e33. 19. Rubinstein R, Afifi A, Jiang S, et al. Oblique pediatric craniofacial fractures. In: Paper presented at: 65th anniversary meeting American Cleft Palate-Craniofacial Association; 2008 [Pittsburgh].
Please cite this article in press as: Garg RK, et al., A novel classification of frontal bone fractures: The prognostic significance of vertical fracture trajectory and skull base extension, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/10.1016/ j.bjps.2015.02.021