|
Correction
to
Smith et al.,
|
| |||||||||||||||||||||||||||||||||||
The authors do not indicate in the most recent paper1 whether these series were consecutive or whether, in fact, these studies all involve the same patients. From the methods described, it would appear that the latter may be the case.
These are important studies, because they suggest that transesophageal echocardiography could replace aortography as the gold-standard investigation for aortic trauma. The conclusions of the papers will obviously be much stronger if the studies involved consecutive series with a total sample of 350 patients. We request that the authors describe more clearly the patients in these three series.
R.J. Young, M.B., B.S.
G.M. Joynt, M.B., B.Ch.
C.D. Gomersall, M.B., B.S.
Chinese University of Hong Kong
Shatin, N.T., Hong Kong
References
We reviewed our experience in patients with blunt chest trauma who underwent aortography because mediastinal blood was apparent on chest radiography. We identified 81 patients who had angiographic evidence of traumatic injury to the thoracic aorta or its branches (75 from motor vehicle accidents, 4 from falls, 1 from being hit by an automobile, and 1 from a crush injury). Among these 81 patients, 66 had only aortic rupture. Fifteen patients (18.5 percent) had injuries of the aortic branch vessels. Three of these patients had not only ruptures of the thoracic aorta but also a total of four injuries to the aortic branches. The other 12 had a total of 19 injuries to the aortic branches but had intact aortas. The 23 ruptured branches were the following arteries: brachiocephalic (5), right subclavian (3), left common carotid (4), left subclavian (7), vertebral (2), and internal thoracic (2). The mediastinal blood seen on chest radiography had an identical appearance whether the injury was to the thoracic aorta, its branches, or both.
Transesophageal echocardiography has completely replaced aortography at our institution in cases of suspected nontraumatic acute aortic dissection.1 However, until the major noncoronary branches of the thoracic aorta can be adequately and consistently evaluated by this method, aortography should remain the primary imaging method used to evaluate patients with mediastinal blood visible on chest radiography after blunt trauma. In this setting, transesophageal echocardiography could be of supplementary benefit in the occasional aortic isthmus in which a "bump" to the ductus arteriosus cannot be distinguished from an aortic rupture.
Douglas C. Smith, M.D.
Ramesh C. Bansal, M.D.
Loma Linda University Medical Center
Loma Linda, CA 92354
References
|
Positive predictive value = true positives/(true positives + false positives) = 10/(10 + 1) = 10/11 = 90.9 percent.
It is not my intention to underestimate the value of this method and the conclusions presented. My concern is that the concept of the positive predictive value is very basic, and the value should not be calculated incorrectly.
José Eduardo R. Azevedo, M.D.
Hospital São Paulo
São Paulo, Brazil
To the Editor: We wish to acknowledge substantial redundancy of patients in our 1995 Journal study1 of 93 patients with suspected aortic injury. Seventy-six of these patients, 8 of whom had positive results on transesophageal echocardiography, had been included in a study of 160 patients with the same condition published in the Journal of Trauma in 1994.2 The 1994 study also included many patients described one year earlier in our initial report of the successful use of transesophageal echocardiography to diagnose thoracic aortic injury.3 Although we cited the 1993 report in our article in the Journal, we neglected to cite the larger 1994 study2 and failed to inform you that that study had been accepted for publication. We sincerely apologize for any confusion about the total number of patients studied and any duplication of data.
In response to Drs. Smith and Bansal, we have reviewed our four-year experience with injuries to arch vessels. Of 6158 patients admitted because of blunt trauma, 251 had suspected aortic injuries, 172 underwent angiography, and none had injuries to the arch vessels. We have evaluated three injuries to the subclavian artery, which involved a pulse deficit in one case, a brachial plexopathy in the second, and both conditions in the third. We have not seen any patients with blunt trauma who had injuries to the innominate or proximal carotid arteries.
The available studies support our conclusions. In an autopsy series of 275 patients4 and a clinical study of 73 patients with blunt thoracic injury,5 there was only 1 patient with an injury to an arch vessel, and that injury was contiguous with an aortic injury. A collective review6 showed that with the exception of the brachiocephalic trunk, other clinical and radiographic signs usually accompany injuries to the arch vessels. Thus, our experience and those reported confirm that injuries to these vessels are "rare and usually apparent clinically."
As pointed out by Dr. Azevedo, the positive predictive value of transesophageal echocardiography was incorrectly reported in our paper. The value is the number of true positive results divided by the number of all positive results7 and is 10/11 or 90.9 percent. Notably, the one patient with a "false positive" result of transesophageal echocardiography did not undergo surgery and may actually have had a false negative angiographic study. Since a patient with negative bedside transesophageal echocardiography could avoid aortography, the most relevant statistic is the negative predictive value: 82 true negative results divided by a total of 82 negative results, or 100 percent.
Mikel D. Smith, M.D.
Paul A. Kearney, M.D.
University of Kentucky College of Medicine
Lexington, KY 40536-0084
References
| |||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
