frontal sinus fractures: evaluation of ct scans in 132 ... sinus fractures: evaluation of ct scans...
TRANSCRIPT
Frontal Sinus Fractures: Evaluation of CT Scans in 132 Patients
Erik M. Olson , 1 Darin L. Wright, 1 Henry T. Hoffman,2 David B . Hoyt,3 .4 and Robert D. Tien 1
Purpose: To determine the frequency of detection of frontal sinus fractures on initial CT scans of
patients with intracranial injuries, and to characterize associated injuires. Methods: The initial head
CT scans in 132 patients with clinical or radiographic evidence of a frontal sinus fracture were
retrospectively reviewed to further characterize the fracture. Additional radiographic studies and
medical records were reviewed to determine associated injuries, therapy , clinical outcome, and
complications. Results: In 90% (124) of the patients, the frontal sinus fractures were visualized
on initial head CT scans· that were obtained to evaluate suspected intracranial injury. Complex
fractures involving both the anterior and posterior wall of the sinus accounted for 65% of cases
(86 patients), whereas fractures of the anterior wall only or posterior wall only occurred in 24%
(32) and 11 % (14) of patients, respectively . Significant intracranial hemorrhage occurred in over
90% of patients with fractures involving the posterior wall. Conclusions: In general , fractures that
involved the posterior wall had more complications and a worse clinical outcome than fractures
that only involved the anterior wall ; nearly all frontal sinus fractures can be detected on head CT
studies in patients with intracranial injuries.
Index terms: Sinus, computed tomography ; Sinus, frontal ; Head, trauma
AJNR 13:897-902, May/ June 1992
Frontal sinus fractures account for 5% to 15% of all facial fractures (1-5). Frontal sinus fractures are frequently first detected on the initial head computed tomography (CT) scan of trauma patients being evaluated for suspected intracranial injury. The evaluation of frontal sinus fractures has previously received little attention in the radiology literature; to our knowledge, no study has fully described their CT scan appearance and their associated injuries in a large series of patients. This study was undertaken to determine how frequently frontal sinus fractures could be detected on the initial head CT scan, and to
Received May 13, 1991 ; accepted and revision requested August 14; revision received September 26.
1 University of California San Diego Medical Center, Department of
Radiology, San Diego, CA 92103. 2 University of Iowa Medical Center, Department of Otolaryngology ,
Iowa City , Iowa 52242. 3 University of California San Diego Medical Center, Department of
Surgery, Division of Trauma, San Diego, CA 92103 and The Trauma
Research and Education Foundation, San Diego, CA. 4 Address reprint requests to David B. Hoy t, MD, Assoc iate Professor
of Surgery, UCSD Medical Center, Department of Surgery , 225 Dickinson
Street, 8896, San Diego, CA 92103.
AJNR 13:897- 902, May/ June 1992 0195-6108/ 92/ 1303-0897 ·9 American Society of Neuroradiology
897
characterize and define associated injuries and complications.
Materials and Methods
San Diego County trauma registry records of patients entering the t rauma units at University of California San Diego Medical Center, Mercy Hospital and Medical Center, Palomar Medical Center, Scripps Memorial Hospita l, and Sharp Memorial Hospital between January 1, 1985 and May 1, 1990 were reviewed to identify patients with a discharge diagnosis of frontal sinus fracture. Data collected for the trauma registry during the initial hospitalization included age, sex, mechanism of injury, associated injuries, therapy , and immediate complications. The complete medical record of each patient was retrospectively reviewed at the time of this study to confirm the trauma registry data and obtain clinical follow-up.
Of the 204 patients with frontal sinus fractures recorded in the trauma registry, 132 patients were included in the study. Patients were only included in this study if the initial CT scans were available, and included 1 0-mm thick , contiguous slices with brain, bone, and subdural window settings. Patients with missing or incomplete head CT scans were not included in the study. The initial head CT scans of these 132 patients (obtained shortly after admission to evaluate suspected intracranial t rauma) were retrospectively reviewed to determine if a frontal sinus fracture was imaged on the initial scan.
898
CT scans were performed on either a GE 9800 or GE 8800 (General Electric Medical, Milwaukee, WI) , a Picker International 1200SX (Picker International, Cleveland, OH) or a Somaton (Siemens Medical Systems, Iselin, NJ) CT scanner. All scans were interpreted at the time the patient presented by at least one radiologist. At the time of this study, at least two of the authors reviewed the scans without knowledge of the initial interpretation. If these two interpretations of initial head CT scan differed , an additional radiologist reviewed the scan to arrive at a final interpretation.
After the final interpretation of the initial scan was determined, thin-cut facial CT scans, follow-up head CT scans, and plain radiographs of the face was retrospectively reviewed to further classify the frontal sinus fractures, determine additional facial and skull fractures, and evaluate soft-tissue injury to the brain and face. The frontal sinus fractures were characterized as anterior wall nondisplaced, anterior wall displaced, posterior wall nondisplaced, posterior wall displaced, or no fracture demonstrated. No attempt was made to evaluate the integrity of the nasofrontal duct. Additional facial and skull fractures, the presence of abnormal soft-tissue or air-fluid levels within the paranasal sinuses, orbital hematomas, and intracranial hemorrhage, hematomas, and contusions were also recorded.
Results
The average age of the 132 patients was 32 years, with an age range of 14 to 83 years. There was a 9:1 male to female predominance, with fractures in 119 (90%) males and 13 (1 0%) females. The majority ( 119, 90%) of fractures were the result of blunt trauma, with penetrating trauma accounting for 13 (10%). Motor vehicle accidents accounted for 84 (64%) cases. The second and third most frequent causes were falls from heights 16 (14%), and gunshot injuries 12 (9 % ). Miscellaneous blunt trauma, usually assaults, accounted for 17 (13%) of the fractures. Each type of frontal sinus fracture had a similar distribution of etiologic mechanisms.
A frontal sinus fracture was detectable on the initial head CT scan in 124 (94%) patients. Of the eight (6 %) patients whose fractures were not detected on the initial scan, fractures were discovered in six patients on thin-slice axial and coronal facial CT scans performed to evaluate other facial injuries. Fractures in two of the patients were discovered on physical examination but were not seen on the head CT scan. One was a small , minimally displaced , open anterior wall fracture and the other was a nondisplaced posterior wall fracture discovered incidently at craniotomy for evacuation of a hemorrhagic frontal lobe contusion.
AJNR: 13, May/ June 1992
Most of the patients, 86 (65 % ), had complex fractures involving both the anterior and the posterior wall. Fractures of the anterior wall only were present in 32 (24%) patients, while isolated posterior wall fractures were present in 14 ( 11 %) patients. Fractures were displaced for a distance greater than the width of the wall in 59% of the anterior wall fractures and 54% of the posterior wall fractures .
Figure 1 shows typical examples of displaced and nondisplaced fractures. Most (86%) of the posterior wall fractures were associated with either nasoethmoid or orbital roof fractures. Linear fractures through the anterior and posterior wall without involvement of the floor or ethmoid air cells were uncommon, occurring in only four (3%) of the patients.
In general, the patients with frontal sinus fractures had significant associated facial or skull fractures or intracranial injury (Table 1). Abnormal soft-tissue findings within the sinuses, orbits, and brain parenchyma were also relatively common (Table 2 , Fig. 2). Periorbital hematoma or intraorbital abnormalities (hematoma, gas, bone fragments, or foreign bodies) were seen in 36 (27 %) of the patients. Pneumocephalus was present in 4 7 (37% ), over one third of the patients. Frontal lobe contusion (43 %), frontal subdural hematoma (14%), and subarachnoid hemorrhage (7 %) were the most common intracranial abnormalities (Table 3). Over 90% of the intracranial injuries were associated with fracture of the posterior wall of the frontal sinus. Of the 100 patients whose injuries involved the posterior wall, 73% had radiographic evidence of intracranial hemorrhage or contusion. Although, the more severely comminuted or displaced fractures had a higher incidence of associated injuries, involvement of the posterior wall was the best predictor of associated injuries.
Therapy was variable and often dependent on associated injuries. Observation alone was the treatment in 38 (29%) of the patients. Seventeen (13%) patients died from associated injuries before receiving treatment. Surgical therapy, usually a combination of sinus obliteration, elevation of bony fragments , or sinus cranialization, was performed in 59 (45%) cases. Eighteen (14%) patients were lost to follow-up before treatment could be completed.
Frontal sinus fractures frequently have significant complications. Because of the short duration of the study and the poor clinical follow-up , only the complications apparent during the initial hos-
AJNR: 13, May/ June 1992 899
Fig. 1. Frontal sinus fractures seen on initial head CT scans in three separate patients involved in automobile accidents. A, CT scan showing a displaced fracture of the anterior wall with the fragment (arrow) adjacent to the posterior wall and a large
overlying soft-tissue defect. B, A relatively subtle nondisplaced posterior wall fracture (arrows) is well shown on this initial head CT scan. C, CT scan demonstrating a minimally displaced anterior wall fracture on the right (arrow) and severely displaced anterior and
posterior wall fracture on the left (open arrows). Note the abnormal soft tissue within much of the frontal sinus.
TABLE 1: Frequency of associated fractures with frontal sinus
fractures
Fracture Site
Orbit
Nasoethmoid
Cranium
Maxillary sinus
Zygomatic arch
Sphenoid sinus
LeFort type
Frequency(%)
110 (83)
90 (68) 77 (58) 72 (54)
28 (21) 12 (9)
11 (8)
TABLE 2: Frequency of abnormal soft tissue findings associated with
frontal sinus fractures
Site
Ethmoid sinus
Frontal sinus
Maxillary sinus
Sphenoid sinus
Orbit (hematoma)
Frontal lobe (contusion)
Other brain injuries
Frequency(%)
122 (92)
120 (91) 62 (47)
51 (39)
54 (41) 57 (43)
31 (24)
pitalization were documented in this series. There were 15 ( 11%) cases of significant ocular complications, such as blindness, extraocular muscle disorders, or diplopia. Meningitis occurred in six (5%) patients. Persistent cerebrospinal fluid leaks,
requiring surgery for repair, occurred in three (2%) patients.
Discussion
The frontal sinuses, air-filled cavities within the frontal bone, arise as small invaginations from an anterior ethmoidal cell tract (6). These cells may belong to the uncinate process, middle meatus, or ethmoidal bullae, which accounts for the variability in the drainage of the frontal sinus (7). The adult frontal sinus, a pyramid-shaped structure with its base inferior and apex superior, has a thick, strong anterior wall and a thin, relatively fragile posterior wall and floor. A boney septum separates the sinus into two asymmetric cavities, each with its own drainage system via a nasafrontal duct into the anterior part of the middle meatus or the anterior portion of the ethmoid infundibulum (8-1 0). The ducts originate near the intrasinus septum in the anterior portion of the floor of the sinus. They run posteriorly and caudally, through the anterior ethmoidal air cells, into the meatus or infundibulum (7, 11 ).
Frontal sinus fractures have traditionally been characterized as anterior wall, posterior wall, or nasofrontal duct fractures (1, 5, 6, 9, 12). Some
900
Fig. 2 . An example of a frontal lobe hemorrhagic contusion compl icating a severe posterior wall frontal sinus fracture.
A, Initial head CT scan showing a nondisplaced fracture through the anterior and posterior wall of the left frontal sinus (small arrows). There is also a fracture through the posterior wall of the right frontal sinus (large arrow) , which was subsequently shown to be comminuted and displaced on a facial CT scan (data not shown). There are also fractures through both sphenoid bones (open arrows) and through the lateral aspect of the left frontal bone (arrowhead).
8 , Follow-up head CT scan on the same patient demonstrates a huge left frontal lobe hematoma (arrows).
TABLE 3: Intracranial hemorrhage associated with frontal sinus fractures
No. of Percent with
Brain Injury Patients
Posterior Wall Fractures
Frontal lobe contusion 57 (43) 91 Frontal subdural hematoma 18 (14) 100
Temporal lobe contusion 9 (7) 100 Temporal subdural hematoma 4 (3) 100
Other contusion/ hematoma 3 (2) 100
Other subdural hematoma 3 (2) 100 Subarachnoid hematoma 10 (8) 80 Shear injury 6 (5) 67 Epidural hematoma 4 (3) 75 Intraventricular hematoma 4 (3) 75
authors include through-and-through fractures, involving both the anterior and posterior wall, as a separate category (9). The fractures are further categorized as linear or comminuted and depressed or nondepressed. High-resolution CT is the method of choice for defining and characterizing anterior and posterior wall fractures ( 13-18). Nasofrontal duct fractures can seldom be imaged directly even with high-resolution CT scans (11). However, fractures through the ethmoid air cells and the orbital roof are easy to visualize and have a high association with nasofrontal duct involvement (11, 12, 19-23). In our series, posterior wall fractures that did not involve the ethmoid air cells or orbital roof were very uncommon. Extension of a fracture from the orbital roof into the posterior wall accounted for most of the fractures of the posterior wall that did not also involve the anterior wall.
Proper characterization of the fractures is essential for planning treatment. Nondisplaced fractures of the anterior wall require no treatment. Displaced anterior wall fractures require cosmetic
AJNR: 13, May/June 1992
elevation. Posterior wall fractures and fractures that may involve the nasofrontal duct usually require surgical exploration with repair of the fracture and duct, or obliteration or cranialization of the sinus (3- 6, 9). If the frontal sinus is functionally eliminated, involvement of the nasofrontal duct is irrelevant, as the sinus will no longer produce secretions.
No large imaging study of frontal sinus fractures has been conducted since CT scanning has become the standard method of evaluating facial trauma. Associated facial fractures are common, but the exact frequency is not known (9). One series of 21 patients described facial fractures in 14 patients, usually nasoethmoid or LeFort fractures (24). In another series (25), nasoethmoidal, maxillary imd zygomatic, and mandibular fractures were reported in 64%, 25%, and 11% of patients, respectively, with frontal sinus fractures. The incidence of associated facial or skull fractures (94%) in our study was very high. This is due partly to improved detection with CT scanning and partly to a more seriously injured patient population. Orbital fractures, nasoethmoid fractures, and maxillary sinus fractures were the most common associated facial fractures. Interestingly, LeFort fractures were present in only 8% of the patients. Fractures of the posterior wall had the highest incidence of associated facial fractures. This high incidence is not surprising, as it has been shown that two to three times greater force is required to fracture the frontal sinus than any other facial bone (26). Accordingly, most (64%) of the fractures in this series were caused by high-speed motor vehicle accidents.
The strong forces required to fracture the frontal sinus also account for the extremely high incidence of associated brain injuries. The peste-
AJNR: 13, May/ June 1992
rior wall of the frontal sinus is the anteroinferior wall of the anterior cranial fossa. Over 70% of the patients with fractures of the posterior wall developed detectable brain contusion or hemorrhage, and over 90% of the patients with serious brain injury had fractures involving the posterior wall (Fig. 2). The brain injury was not always apparent on the initial head CT scan. The high frequency of brain injury with posterior wall fractures suggests that patients with posterior wall fractures should have follow-up CT scans or MR scans to exclude brain injury if the initial CT scan of the brain is normal.
Frontal sinus fractures are frequently associated with ocular complications. One study of 727 patients receiving formal ophthalmologic consultation for facial fractures revealed that 89% of the patients with frontal sinus fractures had associated ocular injuries (27) . Although our study had a much different patient selection bias, which should have had a lower frequency of associated eye injuries, the incidence of ocular trauma was still high: 83% of patients had orbital fractures and 11 % had significant ocular complications. Most of these complications occurred in patients with severely comminuted fractures involving the posterior wall and floor of the frontal sinus.
The incidence of cerebrospinal fluid leaks and meningitis were similar but slightly lower than that reported in previous series (28, 29). No mucoceles were reported in our study; however, long-term follow-up was poor and posttraumatic mucoceles can take years to develop (30, 31). Twenty-seven percent of the patients either died or were lost to follow-up before evaluation and therapy could be completed.
Our study demonstrates that nearly all frontal sinus fractures can be detected on head CT scans obtained to evaluate intracranial injury. The patients in this series were not randomly selected, but chosen from trauma unit registries, which biases the group toward more serious injuries. The 94% detection rate would likely be lower in less critically injured or randomly selected patients. Even so, the detection rate on head CT scans is undoubtedly quite high. Detecting a frontal sinus fracture on an initial head CT scan may lead the patient, if hemodynamically and neurologically stable, to receive a thin-cut facial CT scan during initial evaluation without moving the
atient from the scanner. Frontal sinus fractures are relatively uncom
mon but serious manifestations of facial trauma. igh-resolution, thin-slice axial and coronal facial
901
CT scans are required to define details of the fractures and associated injuries. All frontal sinus fractures , but particularly posterior wall fractures , have a : high incidence of associated facial fractures and intracranial hemorrhage and contusion.
Acknowledgment
We gratefully thank the Department of Surgery, UCSD Medica l Center and the Trauma Research and Education Foundation of San Diego for assistance in collecting the clinical material.
References
1. Calvert CA, Cairns H. Discussion on injuries of fronta l and ethmoidal
sin u se~. Proc R Soc Med 1942;35:805-8 1 0
2. Schultz RC. Frontal sinus and supraorbital fractures from vehicle
accidents. C/in P/ast Surg 1975;2:93-1 06
3. Luce EA. Frontal sinus fractures: guidelines to management. Plast Reconstr Surg 1987;80:500- 51 0
4. Sm ith WW, Yanagisawa E. Paranasal sinus trauma: In: Blitzer A,
Lawson W, Friedman WH, eds. Surgery of the paranasa/ sinuses. Philadelphia: Saunders, 1985:299- 315
5. Helmy ES, Koh ML, Bays RA. Management of frontal sinus fractures:
review of the li tera ture and cl inical update. Oral Surg Oral Med Oral Patho/1 990;69:137- 148
6. Ri tter F. The paranasa/ sinuses: anatomy and surgical technique. 2nd
ed. St Louis: Mosby, 1978:34
7. Terrer E, Weber W, Ruefinacht D, Poralin in B. Anatomy of the
ethmoid: CT , endoscopic , and macroscopic. AJNR 1985;6:77-84
8. O'Rahi)ly R. Nose and paranasal sinuses. In : Anatomy, a regional study of human structure. 4th ed. Gardner E, ed. Philadelphia:
Saunders, 1975:732-740
9. Donald PJ . Frontal sinus fractures. In : Cummings CW, ed. Otolaryngology-head and neck surgery. Vol 1. St. Louis: Mosby , 1986:901-
922
10. Graney DO. A natomy of the paranasal sinuses. In: Cum m ings CW,
ed. Otolaryngology-head and neck surgery, Vol 1. St. Louis: Mosby,
1986:845- 850
11 . Harris L, Marano GD, McCorkle D. Nasofronta l duct: CT in fronta l
sinus trauma. Radiology 1987; 165:195- 198
12. Pollak K, Payne EE. Fractures of the fronta l sinus. Oto/aryngol C/in North Am 1976;9:5 17- 522
13. Rowe LD, Miller E, Brandl-Zawadzki M . Computed tomography in
maxi llofac ial trauma. Laryngoscope 1981;91:745- 757
14. Zikkha A. Computed tomography in facial trauma. Radiology 1982; 144:545-548
15. Gentry LR, Manor WF, Turski PA, Strother CM. High resolution CT
analysis of facial struts in trauma. I. Normal anatomy. AJR 1983; 140:523- 532
16. Gentry LR , Manor WF, T urski PA, Strother CM. High resolution CT
analysis of fac ial struts in trauma. II . Osseous and soft-tissue compli
cations. AJR 1983; 140:533- 541
17. Kassel EE, Noyek AM , Cooper PW. CT in facia l trauma. J Otolaryngol 1983; 12:2- 15
18. Johnson DH. CT of maxi llofac ial trauma. Radio/ C/in North Am
1984;22:107- 118
19. Schenck NL, Rauchback E, Ogura JH. Fronta l sinus disease. II.
Development of the fronta l sinus model: occlusion of the nasofronta l
duct. Laryngoscope 1974 ;84: 1233- 1247
902
20. Hybels RL, Newman MH. Posterior table fractures of the fronta l sinus.
I. An experimenta l study . Lary ngoscope 1977 ;87:171-1 79
2 1. May M , Ogura JH, Schramm V. Nasofronta l duct in frontal sinus
fractures. A rch Otolary ngol 1970;92:523- 532
22. Stanley RB, Becker TS. Injuries of the nasofrontal orifices in frontal
sinus fractures. Lary ngoscope 1987;97:728-731
23. Newman MH, T ravis LW. Fron tal sinus fractures. Lary ngoscope
1973;83: 128 1- 1292
24. Levine SB, Weiser P, Gillen J . Evaluation and treatment of fron tal
sinus fractures. Oto/ary ngol Head Neck Surg 1986;95: 19- 22
25. Schul tz RC. Supraorbi ta l and glabellar fractures. Plast Reconstr Surg
1970;45:227-233
AJNR: 13, May/ June 1992
26. Nahum AM. The biomechanics of maxillofacial trauma. C/in P/ast
Surg 1975;2:59-64
27. Holt GR, Holt JE. Incidence of eye injuries in facial fractures: an
analysis of 727 cases. Oto/ary ngol Head Neck Surg 1983;9 1 :276-
279
28. Lewin W. Cerebrospina l fluid rhinorrhea in closed head injuries. Br J
Surg 1954;42: 1- 18
29. Frenckner P, Richter NG. Operative treatment of skull fractures
through the frontal sinus. A cta Otolary ngol 1960;51 :63-72
30. Evans C. Aetiology in treatment of frontoethmoid mucocele. J Lar
y ngol Oto/1 981;95:36 1-375
3 1. Bordley JE, Bosley WR. Mucoceles of the frontal sinus: causes and
treatment. Ann Otol 1973;82:696-702