| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Hospital of Saint Raphael Yale University School of Medicine New Haven, CT 06511
Daffner [1] reported on determining the time needed to perform the helical CT examination of the entire cervical spine in trauma patients and compared it with the average time for performing a primary radiographic examination of the cervical spine. He concluded that evidence now exists to support the replacement of radiography by helical CT for this examination. Convincing evidence exists that this practice can be undertaken [2, 3]. I do not believe, however, that Daffner has shown in his study that the time gained by using cervical CT is the ultimate factor to be considered in deciding to replace radiography. To make such an assumption, one needs to measure the time involved in obtaining end products with comparable diagnostic information.
A radiographic examination is completed when the images are processed and reviewed by the technologist or radiologist. Inherent information of cervical spine radiography includes assessment of alignment. Beyond consideration of whether the trauma team physician should be the one judging on completion of an imaging examination, a helical CT examination of the cervical spine is completed when not only the axial but also the multiplanar reformatted images are available for viewing. The patient may come off the table after the acquisition of the axial images, but the study is not yet completed. Typically, depending on the reviewing workstation, a variable period of time (sometimes considerable) elapses before the reformations can be seen. If we are to replace radiography with helical CT, the sagittal and coronal reformations need to be an integral part of the study. A coronal reformation provides similar information to the open-mouth odontoid view, and the sagittal reformation compares to the lateral radiograph.
I also wonder what percentage of his consecutive trauma patients underwent CT examinations of other body segments. Frequently, studies of the chest, abdomen, and pelvis are also indicated in trauma patients, and they need to be performed without delay. Even if the cervical spine is scanned first, viewing the reconstructed images is frequently deferred while the scanning of other body segments continues. This practice is common procedure, and it adds variable delay to the true completion of cervical helical CT in patients who undergo scanning of multiple body segments. This is also the case even when the faster multidetector CT units are available.
Furthermore, one can argue that most emergency centers rely on the traditional three-view radiographic trauma series and not on a routine six-view examination. The oblique and swimmer's views, beyond frequently providing redundant information for alignment, are more technically demanding and likely take most of the time required to complete the examination at Daffner's institution. If we were to add the time to generate and review the reformatted multiplanar images to the total CT acquisition time, and if we were to subtract the time needed to produce acceptable oblique and swimmer's radiographs from the overall radiography time, we would have a more realistic comparison and very different results.
It has always been my contention—and I agree with Daffner on the concept—that CT can replace radiography in the subset of patients with multiple trauma. However, cervical spine helical CT should replace radiography in examinations of such patients because of its higher sensitivity [2] and cost-effectiveness [3] in uncooperative patients, not because it is intrinsically faster than the basic radiographic examination. This practice is additionally justified by the fact that helical CT is being performed to screen for other body injuries and because obtaining radiographs after CT is redundant when we routinely include the coronal and sagittal reformatted images [4].
References
Allegheny General Hospital Pittsburgh, PA 15212
I thank Dr. Nuñez for his comments on my article on helical CT of the cervical spine in the trauma setting [1]. As a colleague and friend, he knows that we agree to occasionally disagree on controversial issues. In this instance, I agree with his comments, but I think that some clarification is in order because of the confusion this study created. Often, through author oversight, reviewer request, or editorial polishing, information is omitted from the paper regarding the methodology. I hope I can provide this information to satisfy not only Nuñez but also many other readers of the American Journal of Roentgenology.
First, this study was designed as a follow-up to an earlier one that determined the time to perform a six-view radiographic examination on trauma patients [2]. That study showed that the average radiographic examination took 22 min to perform. Furthermore, 79% of patients required one or more radiographs to be repeated [2]. The time for the CT examination was approximately half that of radiography [1]. The reviewers of both manuscripts questioned exactly how the times for starting and ending the examinations were recorded. There are a large number of variables, as Nuñez and others involved in the care of trauma patients know. These include patient size, severity of injuries, and the number of medical devices and life support elements to be transferred. Often, the mere transfer of a patient from the stretcher to the table takes much longer than the examination itself. In an effort to get reproducible data, I chose the actual start time of the examination. For radiography, that time was when the patient was satisfactorily on the table and the technologist had positioned the tube for the first exposure. For the CT examination, that time was when the patient had been placed on the table and positioned in the gantry. The end point for the radiographic study was when the final radiograph was determined to be satisfactory; for the CT study, the end point was when the last axial image appeared on the technologist's console. We did not add the times for sagittal or coronal tomographic reconstruction, because these reconstructions are often performed after the patient is removed from the CT suite. In other words, once the cervical examination was determined to be satisfactory, the patient could be either repositioned for additional body CT or removed from the CT suite. To this end, I did not record the number of patients who also underwent additional body (thorax, abdomen, pelvis) imaging, although most of our patients do.
Second, it is apparent that there is some confusion as to when a CT or radiographic study is "completed." I agree with Nuñez that, ideally, the study is completed when all images (axial and reformatted) are available for study. However, the nature of trauma departments dictates that, on occasion, there will be a delay in "completing" the examination. The axial images are, for all intents and purposes, adequate for determining the presence or absence of a fracture or of a dislocation [3, 4]. Thus, for practical purposes, I chose the end points that were used in the studies.
Third, and perhaps most important, is the fact that both studies were performed to provide additional "ammunition" in the on-going discussion between radiologists, emergency physicians, and trauma surgeons on what type of imaging (if any) should be performed on trauma patients. We, as radiologists, have written and spoken extensively about the merits of helical CT over radiography for evaluating patients with suspected cervical injury. Unfortunately, most of that discourse has been in the radiology literature and in front of audiences that are predominantly, if not all, radiologists. Murray Dalinka made the most appropriate comment after I gave a presentation, "Controversies in Cervical Spine Imaging for Trauma," at the most recent meeting of the International Skeletal Society. He said, "Dick, that's a great presentation. However, you are preaching to the converted." We need to be getting our message to our clinical colleagues.
Finally, one must consider the issue of whether helical CT can completely replace radiography. When this idea was first proposed, in the early part of the last decade, I was among the skeptics [5] because at that time there was not enough evidence to support the concept. Now, through the work of Nuñez [3, 4], Blackmore [6], Hanson [7], and their colleagues, we know that helical CT is not only accurate but also cost-effective [8]. Rybicki et al. [9] make a convincing case for abandoning all but the lateral radiograph in favor of helical CT. However, I believe that we need additional studies to confirm their results.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
