| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Thoracic and Cardiovascular Surgery, Seoul National University Bundang Hospital, Gyeonggi-do, Seoul, Korea
b Department of Thoracic and Cardiovascular Surgery, Sungkyunkwan University School of Medicine, Samsung Medical Center, Seoul, Korea
Accepted for publication June 17, 2008.
* Address correspondence to Dr Park, Department of Thoracic and Cardiovascular Surgery, Seoul National University Bundang Hospital, 300 Gumi-dong, Bundang-gu, Seongnam-si, Gyeonggi-do, 463-707, Korea (Email: drkhpark{at}yahoo.co.kr).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: Preoperative and postoperative computed tomographic (CT) scan images were retrospectively reviewed for 37 patients who underwent surgery for Stanford type A acute IMH. Operative findings were also reviewed from the medical records.
Results: In 18 patients (48.6%), intimal defects were suggested in preoperative computed tomography (CT). During surgery, 27 patients (73.0%) had small intimal defects in the ascending aorta or arch, while 14 of them (51.9%) did not have preoperative CT findings suggestive of intimal defects. In 18 patients, the defects were located in the arch or distal ascending aorta, where they would not have been found if not inspected under total circulatory arrest. In all patients, the identified intimal defects were included in the aortic resection, or locally closed. Follow-up CT done at 4 months or longer after surgery showed that the IMH in the descending aorta disappeared or markedly improved in all patients.
Conclusions: On the basis of our results, we think that a large proportion of IMH may have a similar pathogenic mechanism as classic dissection and the conventional definition of IMH should be changed. For type A lesions treated with surgery, we recommend thorough inspection of the ascending aorta and the arch under hypothermic circulatory arrest.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The presence of an intimal defect has been found to be a significant risk factor for evolution of intramural hematoma to an overt dissection, rupture, or aneurysmal dilatation [16–18]. In addition, many authors reported that an intimal defect was frequently found during surgery or follow-up imaging studies in patients initially diagnosed as intramural hematoma [12–15]. These facts make us reconsider the conventional definition and explanation of pathogenesis. We came to doubt whether it is appropriate to define an intramural hematoma as "a dissection without intimal tear," and whether intramural hematoma develops as a result of rupture of the vasa vasorum, as conventionally described.
Although it is suggested that intimal defect is an important prognostic factor, there have been only a few studies which focused on it's prevalence in the patients with intramural hematoma. Such studies were based on the findings of computed tomographic (CT) scan, and there has been no report about the prevalence and characteristics of intimal defect in patients who underwent surgery. Considering the case reports of finding an intimal defect that had not been detected by preoperative imaging studies, we can expect that the actual prevalence of intimal defect is higher than reported by previous studies based on CT findings [12–14].
This study aimed to investigate the prevalence and features of the intimal defect by reviewing the radiologic and surgical findings of patients who underwent surgery for acute intramural hematoma involving the ascending aorta.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The mean age of the patients was 68.0 years (range, 45 to 81) and there were more female patients; 10 male and 27 female patients. For 18 patients (48.6%), surgery was performed within 24 hours after symptom onset. The remaining patients underwent surgery after a period of medical management and CT follow-up for 1 to 12 days. The indication for surgery in these patients was one of the following findings: (1) aortic rupture or significant leakage of blood evidenced by massive hematoma in the mediastinum or thoracic cavity; (2) presence of intrapericardial hematoma thicker than 1 cm in CT; (3) presence of clinical triads of pericardial tamponade; (4) diameter of the ascending aorta exceeding 5 cm; (5) thickness of ascending aortic intramural hematoma larger than 1 cm; (6) change into overt dissection.
Intramural hematoma extended to the descending aorta in 25 patients (67.6%), and involved only the ascending aorta (with or without the aortic arch) in the remaining patients. In 29 patients (78.4%), rupture of the aorta or significant leakage of blood was evident by clinical or radiologic findings. Seven patients needed high-dose inotropic agents before entering the operating room, and another 3 patients were stabilized by pericardiocentesis before transfer from other hospitals. One patient underwent surgery 3 days after onset due to progression into overt dissection. In another one patient, surgery was needed on 12th day of medical management for initial type B intramural hematoma because pericardial tamponade developed as a result of retrograde propagation of the lesion into the ascending aorta.
The surgery was performed using deep hypothermic circulatory arrest for the distal aortic anastomosis in all patients. Because cardioplegic solution was given only during the rewarming period after completion of the distal anastomosis, the ascending aorta was not cross-clamped in any patient. When the target temperature for circulatory arrest had been reached, the outer wall of the ascending aorta was longitudinally incised while the cardiopulmonary bypass flow was lowered to 1 L/minute. The hematoma in the ascending aorta was removed gently from the false lumen using a vacuum suction tip. If fresh blood gushes out from the false lumen, it was considered a strong sign that an intimal defect is present in the nearby segment of the aorta. After this maneuver, the true lumen of the ascending aorta was incised under total circulatory arrest and the inner wall was thoroughly inspected to find an intimal defect. The aorta was opened from the sinotubular junction to the level of the ostium of the innominate artery in order to inspect the distal aortic arch and the proximal segment of the brachiocephalic branches. Using a dentist's mirror, as much of the descending thoracic aorta was also inspected.
The extent of the replacement was as follows: ascending aorta in 13 patients including 2 who underwent concomitant valve-sparing root replacement; ascending + lower arch (hemiarch) in 15; and ascending + total arch in 9. In 3 patients, intimal defects located in the distal arch were locally closed instead of being treated with total arch replacement [19]. Except for these cases, all grossly identified intimal defects were included in the resected aortic segment. Two patients (in-hospital mortality 5.4%) died postoperatively due to acute myocardial infarction and acute bowel ischemia, respectively. All patients except for 1 patient, who died of acute myocardial infarction on the second postoperative day, underwent at least one follow-up CT study.
In addition to the patients who underwent surgery, we also investigated the prevalence of intimal defect in CT images of the patients who were medically treated for aortic intramural hematoma during the same period. There were a total of 62 patients; 17 patients had type A lesion and 45 patients had type B intramural hematoma.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
|
|
The maximal diameter of the ascending aorta ranged from 4.5 cm to 7.0 cm, with an average of 5.7 ± 0.6 cm. The width of the intramural hematoma in the thickest portion ranged from 4 mm to 20 mm, and the average thickness was 12.8 ± 3.9 mm. There were 6 patients with the ascending aortic diameter less than 5 cm and 9 patients with the hematoma not thicker than 10 mm. All of these patients were complicated by mediastinal or thoracic hemorrhage.
During surgery an intimal defect in the aortic wall was grossly identified in 27 patients (73.0%). Especially, intimal defects were found in all of the 12 patients who had intramural hematoma confined to the ascending aorta with or without the aortic arch (DeBakey type II lesion). The shape of the defect was a round or oval ulcer in 14 patients, and a linear or spiral tear in 13 patients (Fig 1). In all of these patients, communication between the true lumen and the false lumen was confirmed after removal of the intramural hematoma. The size of the ulcers did not exceed 2 cm and the tears were not longer than 3 cm. The location of the defect was the ascending aorta in 15 patients and the aortic arch in 12 patients. In all of these patients, the ascending aorta was grossly free of atherosclerotic change and 5 patients had atheroma and intimal calcification in the aortic arch. There was no correlation between the location or shape of the intimal defects and urgency of the operation. Because 18 patients had intimal defect in the arch or distal portion of the ascending aorta, only 33.3 % (9 of 27) of the defects would have been found if the aorta had been replaced using the cross-clamp technique instead of open anastomosis under circulatory arrest.
In all of 12 patients who had CT findings of intimal defect in the ascending aorta or the arch, intimal defects were found in the corresponding locations during surgery. Aside from these, a short (<1 cm) linear intimal tear was found at the ostium of the left common carotid artery of the patient who had a focal dissection in that artery. In the patient who had a focal dissection in the right innominate artery, an intimal defect was not identified but thought to be present because blood came out from both true and false lumens when the right axillary artery cannula (inserted for cardiopulmonary bypass inflow) was perfused.
If both CT and surgical findings are taken into one count, only 5 patients (13.5%) were free of any finding suggestive of an intimal defect in the aorta or its major branches (Figs 3; 4).
As for the diagnostic accuracy of CT in predicting intimal defects, it identified only 13 of 27 intimal defects (sensitivity 48.1%) that were found during surgery. In other words, intimal defect was found in 14 of 19 patients (73.7%) who had not been thought to have an intimal defect in preoperative CT findings. The sensitivity of CT for predicting presence of intimal defect was different according to the urgency and timing of surgery after symptom onset. The prevalence of intimal defects found during surgery was not different between the 18 patients who underwent surgery within 24 hours after symptom onset and the other 19 patients who underwent delayed surgery; 13 of 18 (72.2%) versus 14 of 19 (73.7%). In contrast, CT scan identified intimal defects only in 4 patients of the immediate surgery group (sensitivity 30.8%) while it identified intimal defects in 9 patients of the delayed surgery group (sensitivity 64.3%).
|
|
In the group of medically treated patients, the prevalence of intimal defects identified in CT images was slightly higher but the location of intimal defects was different. In 17 patients who had type A intramural hematoma and were treated medically, 8 patients (47.1%) had intimal defects identified in at least one of the serial CT images taken during the admission period. The defects were located in the descending aorta in 7 of them and only one patient had an intimal defect in the ascending aorta. In 45 patients with type B intramural hematoma, 26 patients (57.8%) had intimal defects in CT images and all of them were located in the descending aorta.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
How Should We Define It?
Preoperative CT images revealed ulcer-like projections that were confirmed as small intimal tears during surgery in 35.1% of our patients who otherwise had typical findings of intramural hematoma. Including penetrating atherosclerotic ulcers and focal dissection in the branches, the prevalence of intimal defect was as high as 48.6%. Considering the wide variety in quality of CT images taken at different referring hospitals, we think that examination with up-to-date high-quality CT equipment would increase the prevalence.
If we respect the conventional definition of intramural hematoma, we would be at a dilemma how to define such patients. Are they intramural hematoma or dissection? Commonly, aortic intramural hematoma has been defined simply as "a dissection without intimal tear," and the diagnosis is made when there is a concentric nonenhancing space in the aortic wall without evidence of intimal tear in imaging studies. However, considering our result, such definitions may be confusing and even impractical. Confusion exists both in the literature and in clinical practice. While some authors [9] excluded the patients with ulcer-like projections from the study of intramural hematoma, there are authors [16, 17] who included such cases, and most authors [4, 5, 8, 10] did not mention such a criterion. There must be more confusion in clinical practice because intimal defects can be frequently overlooked, especially when CT images are seen by a physician or radiologist who is not specialized in cardiovascular imaging. In our patients, many of initial radiologist's reports or transfer notes did not mention about the intimal defects that were evident for the author who reviewed the images for the purpose of this study.
Another dilemma exists if we stick to "absence of an intimal tear" for definition; should it be radiological absence or absence in pathological and (or) surgical finding? One study showed that the CT findings may differ according to the timing of examination. Moizumi and colleagues [18] reported that "new" intimal defects were found in the follow-up CT in 33% of the patients who had type A or B intramural hematoma but did not have a intimal defect in the first CT. The incidence was much higher (ie, 62%) in the type A patients. In addition, there have been case reports of an intimal defect not diagnosed in preoperative imaging but during surgery [12, 13]. The result of our study showed that such discrepancy is not sporadic but a common finding. It means that it is impractical and inaccurate to differentiate the patients with intimal defect from those without unless examined by surgery or autopsy. Thus, we think that it is no more adequate to define intramural hematoma as a dissection without intimal tear. Instead, it would be better to define it as presence of a false lumen that is entirely thrombosed whether an intimal defect is present or not.
How Does It Develop?
Conventionally, it has been thought that intramural hematoma develops as a result of rupture of the vasa vasorum of the aorta. However, the high prevalence of intimal defect shown in this study may be interpreted as evidence that many, if not all, cases of intramural hematoma develop by a mechanism other than rupture of the vasa vasorum.
One may argue that the intimal defect is not the cause but a secondary lesion that develops during evolution of the intramural hematoma. However, we think that the intimal defect is more likely a causative lesion based on our data showing that the prevalence of surgically identified intimal defects was not different between the patients who underwent immediate surgery and the delayed surgery group. Tittle and colleagues [3] reported several patients who initially presented with penetrating atherosclerotic ulcers and developed intramural hematoma in the corresponding area later during follow-up. Rubinowitz and colleagues [11] reported two cases of intramural hematoma of the ascending aorta that progressed retrogradely from penetrating ulcers of the descending thoracic aorta. A similar case could be found in our series (Fig 2). Such findings also strongly suggest that intimal defect is not a secondary event but an initiating or blood-entering point of the intramural hematoma.
In addition, there have been several articles that reported progression of intramural hematoma into typical dissection. Such progression has been reported to occur in 16% to 45% [4, 5, 10]. In the study by von Kodolitsch and colleagues [5], the incidence was as high as 72% if the patients with type A lesion were counted. These findings can be explained by the presence of an intimal tear that was not detected by initial imaging study. Our study showed that such a "hidden" intimal defect is present in more than half of surgically treated patients for type A intramural hematoma.
Vilacosta and colleagues [2] reported a patient in whom intramural hematoma developed from an iatrogenic intimal tear made during percutaneous coronary intervention. Based on the observation, he proposed that both overt dissection and intramural hematoma may develop from intimal tear. He explained the difference as a result of the size of intimal tear and presence of reentry tears that cause persistence of blood flow through the false lumen. We think that his hypothesis is rational.
One may ask how we can explain the 27% of our patients who did not have intimal defects. Although the conventional idea of rupture of vasa vasorum may explain such cases, we cannot still rule out the presence of hidden intimal defects located in the descending aorta or the major branches of the aorta. We observed in CT images that many patients with medically treated type A intramural hematoma had intimal defects in the descending aorta. Considering that the actual prevalence of intimal defect is higher than identifiable by CT scan, it would be rational to think that some of our surgically treated patients might have intimal defects in the descending aorta. This speculation is strengthened by our observation that all of the patients without grossly identified intimal defects belonged to DeBakey type I intramural hematoma. We could find intimal defects in all of the patients whose intramural hematoma did not extend beyond the arch. In addition, we could find intimal defects in the innominate artery or the left common carotid artery in 2 patients. Such lesions would not have been found if not looked for with a high degree of suspicion or if located distally out of the surgeon's sight.
How Should We Treat It?
There is a controversy regarding the optimal management for type A intramural hematoma. In most articles published from Western countries, surgery is considered to be the choice because the prognosis of intramural hematoma was not found to be different from that of dissection [3–5, 10]. However, some authors from Korea and Japan argued against routine application of surgical treatment based on the relatively good results after medical management of their patients [6–9]. Although our report is from the same area as the latter side, we stand for more active, if not aggressive, attitude toward surgery.
The apparently different prognosis of Western and Far Eastern patients may be explained by genetic difference. However, we think that the difference might also be caused by a difference in the study material of previous studies. In the report by Sohn and colleagues [6] that is from the same country as ours, the average diameter of the ascending aorta and thickness of intramural hematoma were 35 mm (maximal 45 mm) and 5.5 mm (maximal 7.3 mm), respectively. As the values are markedly different from those shown in our result, it is highly probable that the patients who underwent surgery or died early were excluded from the study by Sohn and colleagues. In most of other reports that favored medial treatment, the size of the aorta and thickness of hematoma were not mentioned.
Some authors identified large aortic diameter and thick hematoma as the risk factors for rupture or progression into dissection. Although most of them were those who consider medical or so-called expectant management as the initial treatment of choice, they proposed that surgery is the better treatment for patients with the aortic diameter larger than 5 cm or the hematoma thicker than 10 mm to 12 mm [9, 17, 18]. We do not think that those values should be considered the absolute cutoff criteria. In 6 of our patients who were complicated by significant mediastinal or intrathoracic hemorrhage, the aorta was smaller than 5 cm and intramural hematoma was less than 10-mm thick. In the series by von Kodolitsch and colleagues [5], aortic rupture or progression into dissection occurred in more than 50% of patients whose diameter of the ascending aorta was not larger than 5 cm.
Presence of intimal defect is another factor that has to be taken into consideration. Many studies identified an ulcer-like projection in CT images as a significant risk factor for progression of intramural hematoma into dissection or rupture [16–18]. The high prevalence of intimal defect shown in our study may not represent the whole population of intramural hematoma because we investigated only surgically treated patients. However, we think that our result is meaningful in that it showed the prevalence is much higher than previously thought, and than visible by CT.
Based on the above facts and satisfactory early and follow-up results in our patients, we currently consider surgery as the initial treatment, unless all of the following criteria are met: absence of mediastinal, pericardial, or intrathoracic hemorrhage; diameter of the ascending aorta smaller than 5 cm; thickness of intramural hematoma of the ascending aorta narrower than 10 mm; and absence of intimal defect in the proximal aorta in imaging studies.
As for the surgery, we think that intramural hematoma should be treated with the same principle and technique as overt dissection. We believe that the intimal defect should be included in the resected segment of the aorta as possible because it can cause later problems if left unresected [20]. In our series, routine resection or closure of the defects resulted in excellent follow-up CT results; early regression or marked improvement of the hematoma in the descending aorta of all patients. Because about half of the intimal defects are located in the distal ascending aorta or the arch, those portions are routinely inspected under total circulatory arrest. Fear of intimal disruption and subsequent embolization of hematoma is another reason why we prefer open distal anastomosis to cross-clamping. Because intimal defects are usually small and the inner and outer walls of the aorta are not widely apart in the patients with intramural hematoma, we think that local closure with transfixing sutures can be a good alternative to total arch replacement if the defect is located in the distal arch [19]. All of the 3 patients treated with the technique showed complete closure of the defect and regression of hematoma in the descending aorta.
Conclusion
Our study showed that the prevalence of intimal defect is high in the patients who were diagnosed as intramural hematoma by conventional definition. This finding may be interpreted as evidence that many cases of intramural hematoma develop by a mechanism other than rupture of the vasa vasorum. To avoid confusion, we think that intramural hematoma should be defined other than "a dissection without intimal tear." It would be better to define it as presence of a false lumen that is entirely thrombosed whether an intimal defect is found or not. We stand for an active attitude toward surgical treatment in cases of intramural hematoma involving the ascending aorta. During surgery, we recommend thorough inspection of the aorta under hypothermic circulatory arrest to find and include the intimal defect into the resected segment.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  H. Takagi, H. Manabe, N. Kawai, S.-n. Goto, and T. Umemoto Thrombosed-type acute aortic dissection. Ann. Thorac. Surg., October 1, 2009; 88(4): 1389 - 1389. [Full Text] [PDF]
|
|
|
|
|
|
K.-H. Park Reply. Ann. Thorac. Surg., October 1, 2009; 88(4): 1389 - 1390. [Full Text] [PDF]
|
|
|
|
|
|
O. Reuthebuch Invited Commentary Ann. Thorac. Surg., November 1, 2008; 86(5): 1501 - 1501. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
