| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Head and Neck Imaging |
1 All authors: Department of Radiology (FND 210), Massachusetts General Hospital, 32 Fruit St., PO Box 9657, Boston, MA 02114-2698.
Received December 1, 2003; accepted after revision April 15, 2004.
Address correspondence to R. A. Novelline.
OBJECTIVE. CT has replaced conventional radiography of the face in many trauma centers. Concern exists that increased costs are associated with increased use of CT. Our goal was to compare the amount of CT and radiography performed for facial trauma at a level 1 trauma center in 1992 and in 2002 and to determine hospital costs for the imaging of these patients.
MATERIALS AND METHODS. The changes in volume and types of facial imaging examinations were determined comparing 1992 and 2002. Hospital costs of different imaging examinations were determined for 2002. Current costs of imaging facial trauma were compared with what 2002 costs would have been if the practice pattern in 1992 had continued.
RESULTS. In 1992, 890 patients were evaluated for facial trauma. Six hundred seventy-one had only radiography, 153 only CT of the face, and 66 both CT and radiography. In 2002, 828 patients were evaluated. Five hundred eighty-four patients had only CT of the face; 228, only radiography; and 16, both CT and radiography. The number of facial imaging examinations per patient in 1992 and 2002 was 1.23 and 1.03, respectively. The 2002 hospital cost of a facial CT examination was $121 and of a facial radiography series was $154. Using CT instead of radiography for evaluating facial injury resulted in an overall cost savings of 22% per patient in 2002.
CONCLUSION. The availability of CT has not resulted in increased use of facial imaging. The increased use of CT from 1992 to 2002 results in decreased current costs for the hospital.
Each year in the United States, an estimated 175,000 individuals sustain severe facial injury [1]. Thousands more suffer moderate and minor facial injuries. Most commonly, facial trauma results from automobile collisions and assault-related injuries. In fact, 45,550 facial injuries classified as moderate or severe occurred as a direct result of motor vehicle collisions [2]. A substantial number of the remaining injuries are due to falls, sports-related injuries, and, less often, penetrating trauma such as gunshot or knife wounds [3].
In the past, radiography was widely used for screening patients with suspected facial injury. Today, MDCT has revolutionized the imaging of suspected facial fracture and has improved diagnosis. MDCT now permits accurate and complete facial evaluation in a single 35-sec axial scan. By postprocessing this single data acquisition, we can obtain high-quality axial and reformatted coronal, sagittal, curved plane, oblique, and 3D images of the face to show bone and soft-tissue injury [4, 5].
In the past decade, we have observed a change in practice pattern, specifically an increase in the number of CT examinations and a decrease in radiography examinations for evaluating facial trauma. The ready availability of CT, its speed, and its relative ease of performance are raising concerns among insurance companies about the potential overuse of CT in the assessment of facial trauma. Our purpose was to determine the extent of the increase in use of CT and decrease in use of radiography for screening facial trauma by comparing data for 1992 and 2002. In addition, we wanted to compare the costs actually incurred in 2002 for imaging facial trauma with the costs in 2002 that would have been incurred if the practice pattern in 1992 had not changed.
Materials and Methods
A retrospective review of imaging methods was performed for all patients assessed at a level 1 trauma center for suspected facial trauma during the years 1992 and 2002. In both 1992 and 2002, CT and radiography were available and were performed on patients with facial trauma.
In 1992, radiography examinations of the face were as follows. The nasal series consisted of three views: Waters, lateral nasal, and occlusal. The mandibular series consisted of five views: Towne, true lateral, left oblique, right oblique, and base. The orbital series consisted of five views, including Waters, lateral, bilateral optic foramen, and Caldwell. The facial series consisted of six views: Waters, Caldwell, Towne, left lateral, right lateral, and base. In 2002, radiography views of the face were the same as those described for 1992.
CT of the face in 1992 was performed on a 9800 CT scanner (GE Healthcare). The 1992 CT protocol required two unenhanced acquisitions: direct axial and direct coronal with hyperextension of the neck. Direct coronal scans were not possible in patients with neck injuries. Axial scanning and coronal scanning were performed using 3-mm-thick slices at 3-mm intervals. Slices obtained in the axial plane were taken from below the mandible to above the frontal sinuses. Coronal slices were obtained from in front of the nasal bones to behind the sphenoid sinuses. No PACS workstations were available in 1992, and interpretations of both CT scans and radiographs were made from hard-copy films. All slices were photographed using bone and soft-tissue windows settings.
CT of the face in 2002 was performed on a Light-Speed MDCT scanner (GE Healthcare). The following CT protocol was used during 2002: Axial scanning was performed from above the frontal sinuses to below the mandible with a detector configuration of 4 x 1.25 mm producing 1.25-mm images at 1.25-mm spacing. Scanning was performed in the HQ (high-quality) mode, representing a pitch of 3.
For coronal, sagittal, and other 2D reformations, the 1.25-mm axial slices were reformatted in the other imaging planes at 1.25-mm slices with 1.25-mm spacing. The 3D reformations were performed on a separate Vitrea workstation (Vital Images). All images were reviewed on PACS using workstations (AGFA). Routine coronal and sagittal reformations were obtained for all patients. Three-dimensional reformations were obtained only in patients with displaced fractures. In 2002, a PACS system was in place, and CT and radiography images were viewed electronically rather than on hard copy.
The patient cohorts from 1992 and 2002 were identified from the radiology computer database Views 4.2 (Folio, a division of Open Market) using keywords for CT and radiography of the face. From the hospital database Clinical Application Suite (Partners Healthcare Systems), additional imaging examinations of the 2002 cohort performed on or within 7 days of the injury were also retrieved for analysis. These examinations included CT images of the head, cervical spine, chest, and abdomen. In refining the data, Panorex views and skull films were excluded.
To establish approximate examination times for CT of the face, we performed the following analysis: A total of 50 CT examinations of the face were analyzed. The times were retrieved from the radiology database IdxRad (IDX Systems) and represent the times the patient enters and exits the scanner room. An average examination time per patient was calculated.
The number of patients identified in 2002 as undergoing CT or radiography of the face was compared with the number of facial imaging examinations performed in 1992. Trends in the techniques used and in the 2002 hospital costs of producing the examinations were analyzed (charges were not analyzed). Cost data for 2002 were retrieved from the hospital cost database, Sunrise Decision Support Manager, version 5.1.02 (Eclipsys, formerly Transition Systems). These costs represent total costs and include departmental fixed and variable costs and hospital overhead. In addition, for 2002 the categorization of facial fractures, associated injuries, and positivity rate was determined.
Results
Data were collected retrospectively from January 1, 1992, through December 31, 1992, and from January 1, 2002, through December 31, 2002. In 1992, 67,004 patients were evaluated in the emergency department. Of this number, 890 (1.3%) were examined for suspected facial injury; 671 (75.4%) had only radiography, 153 (17.2%) patients had only CT of the face, and 66 (7.4%) had both CT and radiography examinations. The 890 patients (1992 cohort) with suspected facial injury included 587 males (66.0%) and 303 females (34%) with ages ranging between 2.6 and 99.6 years (mean, 36.0 years). Of those examined with CT, 30.1% were examined with at least one or more radiography examinations, including 63 facial series, 14 mandibular series, and five nasal series. For patients who underwent only radiography of the face, examinations included 405 facial series, 250 nasal series, 132 mandible series, and five orbital series. The total number of initial examinations for assessing facial trauma was greater than the total number of patients in the cohort, indicating that on occasion more than one type of examination was performed on the same patient.
In 2002, at the same level 1 trauma center, 75,770 patients were seen. Of this number, 828 (1.1%) were examined for suspected facial injury; 584 patients (70.5%) had only CT of the face, 228 (27.5%) had only radiography, and 16 (1.9%) had both CT and radiography. The 828 patients (2002 cohort) with suspected facial injury included 518 males (62.6%) and 310 females (37.4%) with ages ranging between 2 months and 98 years (mean, 39.6 years). The most common causes of trauma were falls, assaults, and motor vehicle collisions.
In 2002, 600 patients were examined with CT. Additional radiography was performed as follows: 10 mandible series, four nasal series, and two facial series. In the 228 patients undergoing only radiography, examinations included 107 nasal series, 66 mandible series, 63 facial series, and two orbital series. Again, the total number of initial examinations for assessing facial trauma was greater than the total number of patients in the cohort, indicating that on occasion more than one type of examination was performed on the same patient.
Also in 2002, 484 patients (80.7%) under-going facial CT had another type of CT of another injured body part at or around the time of trauma. There were 458 CT examinations (76.3%) of the head, 205 (34.2%) of the cervical spine, 123 (20.5%) of the abdomen, and 93 (15.5%) of the chest.
The CT examinations in 2002 were interpreted as positive for facial fracture in 340 patients (56.7%). In three cases (0.05%), suspected fractures were age-indeterminate. A total of 663 individual fractures were identified in the 340 patients. Nasal and nasoorbitalethmoid fractures accounted for 175 (26.4%), maxillary and Le Fort's fractures for 169 (25.5%), orbital fractures for 168 (25.3%), zygoma fractures (including isolated arch and zygoma complex fractures) for 74 (11.2%), mandibular fractures for 44 (6.6%), and frontal fractures for 33 (5.0%).
Frequently, injury to other parts of the body was associated with facial fractures. Of those interpreted as positive for fracture, 271 patients (79.7%) were examined with head CT; traumatic brain injury was seen in 58 (17.1%) of these patients. Cervical spine CT was performed in 199 patients (58.5%), with 12 examinations (6.0%) positive for injury. Abdominal CT was performed in 116 patients (34.1%), and injury was observed in 21 cases (18.1%). Finally, chest CT was performed in 83 patients (24.4%), with 37 (44.6%) sustaining intrathoracic injury.
The number of facial imaging examinations per patient declined between 1992 and 2002. During 1992, a total of 1,093 facial examinations were performed on 890 patients. Thus, on average in 1992, 1.23 imaging examinations were performed per patient. During 2002, a total of 854 facial examinations were performed on 828 patients, for an average of 1.03 imaging examinations per patient (Table 1).
|
In 2002, the actual hospital cost of a facial series per the Sunrise Decision Support Manager cost database was $154, an orbital series was $145, a maxillofacial CT series was $121, a mandibular series was $120, and a nasal series was $103. These costs represent the true cost to the hospital, which is different from the charges the hospital may have submitted to third-party payers.
The costs in 2002 that would have been incurred if the practice pattern experienced in 1992 had continued, were as follows: The total costs for all facial radiography series would have been $71,960, for maxillofacial CT would have been $26,501, for nasal series would have been $26,349, for mandibular series would have been $17,494, and for orbital series would have been $726. These costs represents a total cost of $143,030 or an average of $161 per patient for all imaging examinations. In 2002, the total cost for maxillofacial CT was $72,606, for nasal series was $11,469, for facial series was $9,994, for mandibular series was $9,106, and for orbital series was $291. This represents a total cost of $103,467, an average of $125 per patient for all imaging examinations (Table 1). When CT examinations instead of radiography examinations are used as the predominant technique to evaluate facial trauma, there is an overall cost savings in 2002 dollars of $36 (22.2%) per patient evaluated in the emergency department for facial fracture.
In 1992, CT took approximately 40 min for both axial and coronal scans. In 2002, scanning took 35 sec for axial scans, and coronal reformatted images were produced with no additional scanning time. The average on-and-off-the-table times for both years was 18 min. This time was shorter in relatively healthy patients, whereas less stable multitrauma patients required more time for proper positioning and setup.
Discussion
A fast and accurate imaging technique is essential for assessing facial trauma. Injury to the face can be devastating and can lead to neurologic deficits or blindness [6]. In addition, substantial facial asymmetry or disfigurement can result in emotional and cosmetic concerns.
Diagnosis and treatment of these patients are costly and time-intensive. In our prior investigation, the average time of a radiography skull series, facial series, or cervical spine series was 28 min (range, 10–60 min) [7]. MDCT is faster, having a scanning time of 35 sec and average on-and-off-the-table time of 18 min.
MDCT is not only faster than radiography, it is also safer and more comfortable for the patient. Manipulation of the head to position the patient is necessary for a high-quality radiography series, but today is unnecessary with CT. Other authors have observed that manipulation of the head in the unconscious multitrauma patient is not advisable [8]. With MDCT scanners, only one axial acquisition is necessary compared with non-MDCT scanners, which require hyperextension of the neck to obtain direct axial and coronal acquisitions.
In 1992, facial radiography was used for screening in many cases and resulted in patients having to undergo two techniques of facial imaging. Our investigation shows that in 1992, 30% of patients had both radiography and CT during the initial evaluation of facial trauma. Today, a single rapid scan of the face produces detailed diagnostic images and thus greatly reduces radiation exposure compared with that of 1992, when patients underwent direct axial and direct coronal CT acquisitions. The high positivity rate (56.7%) of facial CT in this investigation suggests that CT is predominantly being used in patients in whom facial fracture is a strong clinical suspicion.
With respect to the use of hospital resources and costs, our results show a significant overall cost savings because CT has become less expensive than radiography, and the use of CT has increased relative to radiography. In 2002 alone, $29,601 was saved using CT as the predominant imaging method when compared with what 2002 costs would have been if radiography had been the predominant imaging technique. CT was much more expensive 10 years ago because of longer scanning times and lower throughput. The speed with which MDCT can be performed today results in the relatively high fixed costs of CT being spread out over a large number of examinations, thus decreasing the overall cost per examination. The speed with which a radiography series can be performed has changed little during the past 10 years.
In addition, in 2002 both CT and radiography were viewed and interpreted on a PACS. The PACS costs were part of the departmental fixed costs in 2002 and did not exist in 1992. Because of this change in the method of handling images, the actual 1992 costs were not analyzed. Rather, the 2002 examination costs were used with the 1992 volumes of examinations in order to reflect how 2002 costs would have been different if the 1992 practice pattern had persisted.
In the past decade, a clear change has occurred in the method of imaging facial trauma. The use of radiography has decreased while the use of CT has increased. In the past, radiography with positive or suspicious results had to be followed by CT for better definition of the injuries. Today, a single MDCT examination is sufficient and is more accurate than radiography alone.
In conclusion, in the past decade CT has largely replaced radiography in our emergency department as the technique of choice to evaluate facial trauma. This change has resulted in a decrease in the total number of imaging examinations per patient evaluated. The high positivity rate of all examinations used in evaluating facial trauma in 2002 does not suggest overuse of imaging.
The increased use of CT and decreased use of radiography over the decade have had a favorable impact on hospital costs because CT of the face has become less expensive than radiography of the face. Costs in 2002 would have been higher if the old practice pattern of reliance on radiography had persisted.
Acknowledgments
We thank Kathy Oullette for data retrieval regarding CT scanning times and Julia Sinclair for providing the epidemiologic data for our emergency department.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. T. Rhea and R. A. Novelline How to Simplify the CT Diagnosis of Le Fort Fractures Am. J. Roentgenol., May 1, 2005; 184(5): 1700 - 1705. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
