Ann Thorac Surg 2005;79:1957-1960
© 2005 The Society of Thoracic Surgeons
Original article: Cardiovascular
Transferring Diagnosis Versus Actual Diagnosis at a Center for Thoracic Aortic Disease
Thomas M. Beaver, MD*,
Francis N. Herrbold, MD,
Philip J. Hess, Jr, MD,
Charles T. Klodell, MD,
Tomas D. Martin, MD
Division of Thoracic and Cardiovascular Surgery, University of Florida, Gainesville, Florida
Accepted for publication December 20, 2004.
* Address reprint requests to Dr Beaver, Division of Thoracic and Cardiovascular Surgery, University of Florida College of Medicine, P.O. Box 100286 JHMHSC, Gainesville, FL 32610-0286 (E-mail: beavetm{at}surgery.ufl.edu).
 |
Abstract
|
|---|
BACKGROUND: Lack of physician awareness of thoracic aortic disease has received increased media attention. As a referral center for thoracic aortic disease our institutional experience confirms discrepancies between the transferring diagnosis and the actual pathologic diagnosis. A retrospective review was undertaken to identify the incidence and sources for disparate diagnoses.
METHODS: Medical records from 100 consecutive patients transferred to The University of Florida—Shands Hospital between April 2002 and October 2003 were reviewed. To identify sources for error, the charts of 24 patients with diagnostic discrepancies were examined in detail with attention to outside radiologic reports, level of physician experience, and additional diagnostic testing required.
RESULTS: The transferring diagnosis of 24 patients was different from the final aortic pathologic disease. The most common discrepancies were misclassifications of dissections and aneurysms. Seven patients had either no leak or no dissection. Two patients had misleading "pulsation artifacts" on their computed tomographic scans. In half of the patients diagnostic differences were secondary to initial misinterpretation by the referring radiologist. Seventeen of 24 patients underwent additional diagnostic testing. Misdiagnoses were more common when the referring physician was not a surgeon (15 of 24). The diagnosis of 5 patients was confirmed only in the operating room.
CONCLUSIONS: A significant incidence of disparate diagnosis was identified between transferring facilities and our referral center. Discrepancies were secondary to initial radiographic misinterpretation and the complexity of thoracic aortic pathologic disease. Medical schools and continuing medical education programs should place increased emphasis on thoracic aortic disease in their curricula.
|
Introduction
|
|---|
Public awareness of thoracic aortic disease was heightened last year by the death of a popular Hollywood celebrity from an ascending aortic dissection [1]. Shortly thereafter, The Wall Street Journal published a report on the "education gap among physicians" regarding thoracic aortic disease [2]. As a referral center for thoracic aortic disease in the Southeastern United States, our institution has noted discrepancies between the transferring diagnosis from outside hospitals and the final thoracic aortic pathologic diagnosis. The purpose of this review was to quantify the incidence of diagnostic discrepancies and to identify the reasons they occur.
|
Material and Methods
|
|---|
After institutional review board approval, the records from 100 consecutive patients with thoracic aortic disease transferred to The University of Florida between April 2002 and October 2003 were reviewed. Transferring diagnoses included dissections, aneurysms, pseudoaneurysms, aortic leak, and traumatic rupture. All available transferring records were reviewed, including referring physician’s notes, nurses’ notes, radiology reports, and the emergency room record to identify the "transferring diagnosis." The "actual diagnosis" was defined after complete review of all diagnostic testing by an in-house attending cardiothoracic surgeon with experience in thoracic aortic disease. Discrepancies between the transferring and actual diagnosis were grouped into six categories: aneurysm instead of dissection, dissection instead of aneurysm, no leak, no dissection, "pulsation artifact," and error in dissection classification (Debakey type III versus I).
Of the 100 transferred patients, 76 patients were identified as having an accurate diagnosis of their thoracic aortic disease. The medical records of 24 patients with disparate diagnosis were further reviewed to identify sources for discrepancies with attention to initial radiographic findings at the outside facility, requirement for further diagnostic evaluation, intraoperative findings, and the experience level of the transferring physician (emergency room or primary care physician versus a cardiothoracic or vascular surgeon). Importantly, there remain an unknown number of patients with unrecognized thoracic aortic disease, which are not included in this report.
|
Results
|
|---|
Twenty-four patients (24%) were transferred with a disparate diagnosis (Table 1). There were 63 men and 37 women. The mean patient age was 63 years (range, 11 to 87 years). The most common discrepancies were secondary to errors in discrimination between dissections and aneurysms. Nine patients had no aortic disease, including 7 patients with either no leak or no dissection and 2 other patients who presented with computed tomographic (CT) scans showing pulsation artifacts that mimic aortic dissection (Fig 1). A revised diagnosis in 7 patients was made after review of their existing studies, and they required no further diagnostic testing. In total, 17 of 24 patients underwent additional CT scans, magnetic resonance imaging (MRI), or cardiac catheterization and transesophageal echocardiography (TEE; Table 2).
View larger version (61K):
[in this window]
[in a new window]
|
Fig 1. Computed tomographic scan with a pulsation artifact mimicking an aortic dissection in the ascending aorta.
|
|
Twelve of the 24 differences in diagnosis were secondary to initial misinterpretation by the outside radiologist (Table 2). In addition, a higher number of discrepancies were noted when the transferring physicians had less experience in thoracic aortic disease (n = 15) than cardiothoracic or vascular surgeons (n = 9; Table 2).
The actual diagnosis of 5 patients was confirmed only in the operating room (Table 3). The first patient was a 16-year-old boy in a motor vehicle accident who was believed to have an aortic tear. After review of the outside CT scan by our in-house radiologists, injury could not be ruled out; an intraoperative TEE was performed but was still not diagnostic. At the time of exploration an insignificant adventitial hematoma was found. Two patients with the diagnosis "ruptured thoracic aortic aneurysm" were found to have stable type III dissections after thoracotomy. An additional patient with the diagnosis of "ruptured thoracic aortic aneurysm" had a ruptured type III dissection that required graft replacement of his descending thoracic aorta. The fifth patient had a transferring diagnosis of acute ascending dissection and was taken directly to the operating room as described by others [3]. Surgery was avoided after the surgical team raised concern for a pulsation artifact on the outside CT scan, which was confirmed by a TEE in the operating room.
|
Comment
|
|---|
The diagnosis of thoracic aortic disease can be challenging for physicians who do not have frequent exposure to thoracic aortic disease [2]. Given the recent heightened public scrutiny of physicians’ knowledge in this area and our institutional experience as a referral center for thoracic aortic disease, we elected to examine disparities in diagnoses from transferring facilities.
Indeed, 24% of patients with thoracic aortic disease transferred to our facility had a different diagnosis than the final pathologic diagnosis. In the diagnostic discrepancies in this series, more often the referring physician was not a surgeon. Most commonly, dissections were labeled as aneurysms and vice versa. Increased emphasis on thoracic aortic disease in continuing medical education should be able to impact on this finding. Half of the diagnostic discrepancies could be traced back to the outside radiologist’s initial misinterpretation of CT scans. At our center we have seen several radiology reports from centers with limited thoracic aortic experience reporting "leaking" aneurysms when there was no leak identified by our hospital radiologists or surgical team.
Referring physicians should have a low threshold to transfer patients to centers for thoracic aortic disease as complicated aortic pathologic disease can require additional diagnostic testing. The sensitivity of CT, MRI, and TEE in aortic disease exceeds 93% [4, 5]. The sensitivity of MRI has been noted as high as 100%; however, its use has been limited secondary to procedure time and patient monitoring issues [4, 6]. Transesophageal echocardiography is operator dependent, and errors in diagnosis can occur at centers that do not frequently encounter thoracic aortic disease [7]. Accordingly, the availability and speed of CT scanning makes it the diagnostic modality of choice at most centers [8].
Nonetheless, despite the high sensitivities of spiral CT, TEE, and MRI, definitive diagnoses are sometimes made only at the time of operative intervention secondary to the complexity of thoracic aortic disease [9]. The diagnoses of 5 patients in this series were confirmed only in the operating room. Increasing numbers of patients with acute dissection are being taken straight to the operating room [3]. Outside radiologic reports and intraoperative TEE should be interpreted judiciously, as 1 of our patients was found to have no disease, but rather a pulsation artifact.
Pulsation artifact is a recently reported imaging phenomenon that is secondary to the pulsatile movement of the heart and aorta during scanning [8, 10]. It is a product of advanced rapid spiral CT technology, which captures images so fast that the thoracic aorta is imaged in two planes, falsely simulating an aortic dissection. This artifact is usually limited to three axial images and can be duplicated when the scan time is comparable to the cycling time of the aortic motion [10, 11]. Clinicians should be aware that radiologists with limited exposure to thoracic aortic disease could misinterpret pulsation artifacts as aortic dissections.
This retrospective review confirmed that a significant incidence of diagnostic discrepancies occur with thoracic aortic disease. Twenty-four percent of patients were transferred to our center with a different diagnosis than the actual pathologic diagnosis. Patients with thoracic aortic disease should receive consultation from a cardiothoracic or vascular surgeon. Efforts should be undertaken to incorporate thoracic aortic disease into continuing medical education programs for emergency room, radiology, and cardiovascular specialists. In addition, increased emphasis on thoracic aortic disease should be placed in medical school curricula.
|
References
|
|---|
- Gorman C. An uncommon death. What to know about the vascular condition that killed John Ritter Time 2003;162(12):78(Sept. 22).[Medline]
- Helliker K, Burton TM. Knowledge gap: medical ignorance contributes to toll from aortic illness; many doctors don’t realize aneurysms are treatable; a paucity of specialists; ‘I could have saved him.’ The Wall Street Journal (Nov. 4) 2003, section A, page 1..
- Bavaria JE, Brinster DR, Gorman RC, Woo YJ, Gleason T, Pochettino A. Advances in the treatment of acute type A dissectionan integrated approach. Ann Thorac Surg 2002;74:S1848-S1852discussion S1857–63.[Abstract/Free Full Text]
- Nienaber CA, von Kodolitsch Y, Nicolas V, et al. The diagnosis of thoracic aortic dissection by noninvasive imaging procedures N Engl J Med 1993;328:1-9.[Abstract/Free Full Text]
- Sommer T, Fehske W, Holzknecht N, et al. Aortic dissectiona comparative study of diagnosis with spiral CT, multiplanar transesophageal echocardiography, and MR imaging. Radiology 1996;199:347-352.[Abstract/Free Full Text]
- Laissy JP, Blanc F, Soyer P, et al. Thoracic aortic dissectiondiagnosis with transesophageal echocardiography versus MR imaging. Radiology 1995;194:331-336.[Abstract/Free Full Text]
- Vignon P, Spencer KT, Rambaud G, et al. Differential transesophageal echocardiographic diagnosis between linear artifacts and intraluminal flap of aortic dissection or disruption Chest 2001;119:1778-1790.[Abstract/Free Full Text]
- Ledbetter S, Stuk JL, Kaufman JA. Helical (spiral) CT in the evaluation of emergent thoracic aortic syndromes. Traumatic aortic rupture, aortic aneurysm, aortic dissection, intramural hematoma, and penetrating atherosclerotic ulcer Radiol Clin North Am 1999;37:575-589.[Medline]
- Sherwood JT, Gill IS. Missed acute ascending aortic dissection J Card Surg 2001;16:86-88.[Medline]
- Qanadli SD, El Hajjam M, Mesurolle B, et al. Motion artifacts of the aorta simulating aortic dissection on spiral CT J Comput Assist Tomogr 1999;23:1-6.[Medline]
- Parry CK, Rajagopalan B. Characterization of artifact simulating aortic dissection in computed tomography imaging J Digit Imaging 2001;14:220-221.[Medline]
This article has been cited by other articles:
|
|
|
|
 
R Meenakshisundaram and P Thirumalaikolundusubramanian
Resident's knowledge on thoracic aortic dissection
Postgrad. Med. J.,
May 1, 2008;
84(991):
278 - 278.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
