Acute Type A Intramural Hematoma
Analysis of Current Management Strategy
Background— Management of acute type A intramural hematoma (IMH) remains controversial, varying from immediate surgery to medical management only. Conversion to typical dissection remains a concern. We analyzed our experience managing acute type A IMH.
Methods and Results— Between October 1999 and May 2008, 251 patients with acute type A aortic dissection were treated, including 36 (14.3%) with type A IMH. Seven IMH patients (19%) were repaired immediately, 28 (80%) managed initially with optimal medical management and eventual repair and 1 (3%) with medical management only. End points analyzed were early mortality and conversion to typical dissection (flow in the false lumen of the ascending aorta). Time (hours) from onset of symptoms defined initiation of IMH. Early mortality for acute type A IMH was 8.3% (3/36): 14.3% (1/7) with immediate repair and 7.1% (2/28) when optimal medical management with eventual repair was undertaken (P=0.69). The 1 medically managed Asian patient survived with resolution of the IMH. Conversion to type A IMH to typical dissection occurred in 33% (12/36) of cases. No conversions were observed within 72 hours. Aortic diameter did not predict conversion. In actuarial analysis among the initially medically managed group with eventual repair, the hazard conversion to typical dissection increased significantly at 8 days from the onset of symptoms (P<0.05).
Conclusions— Despite optimal medical management, conversion of type A IMH to typical dissection still remains a concern, with the most significant risk beyond 8 days. In our patient population, timely surgical repair is recommended.
First described by Krukenberg1 in 1920 as aortic dissection without intimal flap, the management of intramural hematoma (IMH) of the aorta remains controversial. Consensus regarding the management of acute IMH involving the descending thoracic aorta (type B) exists, consisting primarily of medical management with anti-impulse therapy.2–6 In contrast, there is less agreement regarding the management of acute type A IMH. Several groups have advocated medical management alone, reporting complete resolution of the IMH,3,6–8 whereas others have recommended immediate surgical repair because nonoperative repair led to excessive mortality and morbidity resulting from conversion to typical dissection and associated rupture and malperfusion.5,6,9,10 Still others have recommended selective repair in type A IMH for those at higher risk for conversion to typical dissection as identified by a larger false channel diameter or aortic diameter.6,11
Although our early experience involveng urgent surgical repair of acute type A IMH was influenced by the Stanford group,5 several of our patients underwent delayed repair due to associated comorbidities. Ultimately, our approach to acute type A IMH has evolved into expeditious but not immediate repair for most patients. Thus, this study set out to analyze our results with the management of acute type A IMH, to compare acute type A IMH with typical acute aortic dissection, and to validate our approach to the management of acute type A IMH.
The Committee for the Protection of Human Subjects at the University of Texas Houston Medical School approved review of the data collected for this study.
Between October 1999 and May 2008, we repaired 251 patients with acute type A aortic dissection. Of these, 36 patients (14%, 36/251) presented with acute type A intramural hematoma. Mean patient age was 62 years (range, 21 to 91). Sixty-four percent (n=161) were men. Ethnicity distribution is depicted in Figure 1.
Patients managed surgically were repaired via a median sternotomy, using full heparinization, cardiopulmonary bypass, and profound hypothermic circulatory arrest with retrograde cerebral perfusion. Arterial cannulation was accessed via the femoral (n=239), axillary, (n=8), or ascending aorta (n=4). Open distal anastomosis was performed under circulatory arrest. Electroencephalograms were used to monitor cerebral function and determine time to initiate circulatory arrest. For typical dissection, obliteration of the false channel was performed with resection of the tear if feasible. For IMH, evacuation of all hematoma from the false channel with subsequent obliteration of the false channel was performed.
Patients were managed with immediate surgical repair (7 patients, 19%), eventual repair with optimal medical management (28 patients, 8%), and medical only (1 patient, 3%). All patients were admitted to the cardiovascular intensive care unit and begun on anti-impulse therapy with maintenance of blood pressure less than 120 mm Hg (systolic).
Follow-up was obtained in all patients. Survival was ascertained by direct patient contact or by the Social Security Death Index. Surveillance protocol included yearly radiographs, either computed tomography or MRI.
Intramural hematoma was defined as aortic dissection without identifiable intimal tear and lack of flow in the false lumen of the aorta.2 This was confirmed radiographically by both computed tomography scan and transesophageal echocardiography. Acute dissection was defined as dissection occurring within 2 weeks of presentation. Conversion to typical dissection was determined by either operative examination or radiographic examination before surgery. Time from the initial onset of pain (in hours) determined the onset of IMH. Time to operative repair was defined as time from onset of pain to operative repair. In addition, time from admission to our hospital to operative repair was determined. Primary end points included early mortality and conversion to typical dissection. Typical, also known as classic, dissection was dissection with free flow into the false channel. This was determined by operative exploration.
Comparisons of descriptive statistics and for short-term outcomes between groups were made by univariate unpaired tests and multivariable logistic regression. In-hospital hazard of conversion from IMH to classic dissection was estimated using hazard function analysis computed by life-table method. Hazard represents the probability that an event will occur within a specified time interval and is interpreted as the instantaneous probability of an event at any given time during follow-up. Long-term survival was estimated by the product-limit method of Kaplan and Meier. All computations were performed using SAS version 9.1.3 (SAS Institute Inc, Cary, NC).
Comparison Between Typical Type A Dissection and IMH
Regarding clinical presentation, patients with IMH presented more frequently with chest pain and less frequently with moderate to severe aortic insufficiency when compared with typical dissection (P=0.04, P=0.002, respectively; Table 1). Approaching significance, patients with IMH were older and in less extremis condition on admission and with less hypotension and tamponade. Mean aortic diameters were similar between typical dissection (5.0±0.8 cm) and IMH (5.2±0.8 cm) (P=0.17). Procedures performed were similar between groups, although transverse arch replacement was performed more frequently in the IMH group and cannulation strategy. Valve replacement, coronary artery bypass, and root replacement were similar between groups (Table 1).
Postoperative mortality and morbidity were similar between groups, with an overall mortality of 8% in the IMH groups and 13% in the typical dissection group (Table 1). Although not significant, there were fewer re-explorations for bleeding in the IMH group (0%) versus the typical dissection group (6%). Although observed mortality was higher in the immediately repaired group versus the purposefully delayed group, at 14% versus 7%, this was not statistically significant (P=0.69) (Table 2).
Conversion to Typical Dissection
In patients who converted from IMH to typical dissection (12 patients), operative exploration revealed the tear in the ascending aorta in 6 patients (50%), in the descending thoracic aorta in 2 patients (17%), and not identified in 2 patients (17%). Within the IMH group, 33% (12/36) were noted to have conversion to typical dissection. No significant difference was noted in aortic diameter between those with and without conversion to typical dissection (5.4±0.74 versus 5.1±0.78 cm, respectively, P=0.29). Moreover, the maximal hematoma thickness of the IMH in the ascending aorta did not significantly differ between the patients who had converted to typical dissection and those who did not convert (1.3±0.6 cm versus 1.1±0.3 cm, P=0.19).
Conversion to typical dissection was observed in 14% (1/7) of the immediately repaired group, which was not significantly different from the purposefully delayed group, in which 38% (11/28) had converted to typical dissection at the time of repair. Among this cohort, no conversions occurred within 3 days of acute presentation. Time periods from symptoms to surgery and from admission to surgery were longer in the purposefully delayed group, at 6.5 days and 5.3 days, respectively (P<0.001). Aortic size did not differ between groups (5.3 cm in the immediately repaired group versus 5.2 cm in the purposefully delayed group, P=0.99).
Reasons for immediate repair included persistent pain, syncope suggestive of cerebral malperfusion, paraparesis suggestive of spinal cord malperfusion, instability/hypotension, signs of coronary ischemia, and altered mental status suggestive of cerebral malperfusion. Reasons for delay included intended (or purposeful) delay, coagulopathy related to coumadin or recent clopidogrel use, initial patient refusal, or comorbidities significantly increasing the risk of open operative repair such as pneumonia, adult respiratory distress syndrome, coma, urosepsis, or leukocytosis from undiagnosed chronic lymphocytic leukemia.
The Medically Managed Patient
The only medically managed patient resolved symptoms during his admission and ultimately declined operative repair, with an ascending aortic diameter of 4.9 cm. Late follow-up at 4 years revealed complete reabsorption of the hematoma with no progression of aortic diameter. Interestingly, this patient’s ethnicity is Japanese (Asian).
Validity of Purposeful Delay
Determining the efficacy of purposeful delay required analyzing the safety of this approach. As a surrogate of outcome, we determined that conversion to typical dissection would be associated with the potential complications linked with acute type A aortic dissection, for example, rupture, pericardial tamponade, valvular insufficiency, and organ malperfusion. For this reason, the hazard plot for risk of conversion to typical dissection was constructed. This plot depicted the instantaneous risk for conversion to typical dissection. Moreover, this plot was compared with the instantaneous risk for those that converted and this was noted to be significantly different (P<0.05). In this plot (Figure 2), a small increase in risk for conversion occurred at 3 days from onset of symptoms but significantly increased up to 8 days, with the greatest risk occurring at 8 days from the onset of symptoms. This suggested that the risk of conversion to typical dissection was lower within 3 days of the onset of pain. This finding, combined with the observation that mortality was no different between the immediately repaired group and the purposefully delayed group, suggested that purposeful delay could be achieved with acceptable outcomes.
Median follow-up was 42 months. Late survival is depicted in Figure 3.
Much controversy regarding the management of acute type A IMH exists. Robbins et al5 first reported an experience of 13 patients with acute aortic IMH, of which 3 patients had IMH involving the ascending and transverse arch. In this series, those with ascending IMH had an associated mortality of 66%, leading the authors to recommend “early graft replacement” for these patients. Several ensuing studies of Asian cohorts with type A IMH, however, reported low early mortality with medical management alone, ranging from 0% to 8%.3,6–8 Moreover, some have reported complete absorption of the IMH in as high as 67% of patients6 with 5-year survival as high as 80% to 85%.3,6,7 In contrast, others have reported poor early outcomes from medical management, with mortalities ranging from 33% to 80%, leading to the recommendation of early surgical repair.2,4,5,10 In an attempt to determine the optimal management for type A IMH, a meta-analysis of 12 studies (9 of which were Asian cohorts) involving 328 patients was performed.12 This analysis ultimately reported no significant difference in early mortality between those medically managed (10.1% in 168 patients) and those surgically repaired (14.4% in 160 patients) (P=0.36). In their conclusion, the authors recommended individualization of treatment, noting that patients with progression of size, enlarged initial aortic diameter, and subadventitial wall thickness >12 mm were candidates for early surgery.12 Despite all available data regarding type A IMH, a general consensus on treatment has not been achieved.
Influenced by early studies reporting both high mortality with medical management of type A IMH but also acknowledging the potential for reabsorption of the hematoma in some cases, we adopted a strategy of initial optimal medical management with eventual or purposefully delayed surgery. Those who were repaired immediately were those in extremis and who clinically demonstrated progression of disease such as conversion to typical dissection leading to malperfusion. Patients who were clinically stable, however, were admitted to the intensive care unit and initiated on anti-impulse therapy to control blood pressure and pain and were scheduled for eventual repair. In most cases, after 2 to 3 days of medical management, surgical repair was performed. In these cases, examination of the aorta revealed clot and organizing thrombus that were completely evacuated, leaving a thickened adventitial aortic wall for repair. An advantage of purposeful delay was the observation that after even a few days, the aortic wall was thicker, allowing for easier reconstruction and ultimately a more expeditious repair. Although not significant, we did observe no reoperations for bleeding in the type A IMH group. Another possible advantage of purposeful delay for acute type A IMH may relate to the avoidance of the extensive systemic inflammatory response encountered during emergent surgery. This, however, is not supported by this study and remains only hypothetical.
Results of this study are comparable to the findings of others when comparing typical type A aortic dissection and IMH. We observed that patients with type A IMH tended to be older (63 years versus 58 years, P=0.06), presenting with more chest pain (100% versus 88%, P=0.04) and less frequently with hypotension (8% versus 22%, P=0.07), tamponade (6% versus 16%, P=0.16), and moderate or greater aortic insufficiency (11% versus 38%, P=0.002). The decreased incidence of malperfusion and tamponade with IMH as compared with typical dissection is not surprising because thrombosis of the false channel should decrease the risk of these events. Conversion to typical dissection was observed in 33% (12/36) of cases in this study, similar to previous studies reporting conversion in up to 40% of cases. Although not demonstrated in our study, previously reported risks factors for conversion have included initial aortic diameter >5.0 cm and adventitial thickness >12 mm.6,11,13
Mean aortic diameters were similar between typical dissection (5.0±0.8 cm) and IMH (5.2±0.8 cm) (P=0.17). Moreover, no significant difference was noted in aortic diameter or maximal hematoma thickness between those with and without conversion to typical dissection.
Despite increasing experience with IMH, much still remains unknown. The influence of ethnicity on IMH and its management is interesting but unclear. Many studies on type A IMH have involved primarily Asian cohorts, with a majority of these series reporting good results with medical management alone. Why Asian cohorts may fare better with medical management remains unclear. It should be noted that the Asian cohort used serial scans to identify complications, which then determined the timing of operative intervention. Although our cohort has a variety of ethnicities, interestingly, the single patient managed medically was Japanese, who refused aortic replacement and whose aorta measured 4.9 cm.
In analyzing these data, before establishing a consistent strategy to the management of acute type A IMH, a number of patients were delayed in their repair for various reasons, including poor medical status or unknown neurological status. As such, this provided an opportunity to analyze the risk for conversion to typical aortic dissection, because almost all patients were eventually repaired and a thorough examination of the aorta could be performed. This resulted in the hazard plot, which provided the instantaneous risk for conversion to typical dissection at any given time during the patient’s clinical course. From this, we observed that no patients converted within 3 days of symptoms when surgery was purposefully delayed, but an increasing risk of conversion up to and beyond 8 days was noted. This strategy, however, assumed that optimal medical management was maintained. These data suggest that in our patient population, a significant risk of conversion to typical dissection ultimately existed and that eventual repair should be performed. Implications of these data suggest that those patients managed medically need close follow-up for at least 2 weeks and a regimen of routine long-term radiographic surveillance.
As with any retrospective study, inherent limitations and biases exist. The series remains small and nonrandomized, thus having these inherent limitations. In addition, the diagnosis of IMH was made radiographically and confirmed using echocardiography, but a small tear not visualized could have been present and only discovered at surgery. Moreover, the hazard plot relied on the definitions of time from onset of symptoms to surgery. The time of conversion was made at surgery; thus, although computed tomography or transesophageal echocardiography may not have revealed a conversion during the hospitalization, the conversion could have occurred any time between the radiographic diagnosis and the time of surgery. The hazard plot analysis assumed that the conversion occurred at the time of operative examination. Although we did not observe any conversions of IMH to typical dissection within 3 days, others have reported such events within 1 day.14 It must be taken into account that all patients in our series were placed on an anti-impulse therapy on admission to the intensive care unit, which is another variable to consider when discussing conversion.
In conclusion, despite optimal medical management, conversion of type A IMH to typical dissection still remains a concern, with the most significant risk beyond 8 days. In our patient population, patients with acute type A IMH may be individualized, allowing for safe purposeful delay. Because the risk of conversion remains significant, however, eventual surgery is still recommended.
We thank G. Ken Goodrick for editing and Chris Akers for the figures.
Sources of Funding
This research was supported by grant 5 P50 HL083794-02 (TAAD-SCCOR) from the National Heart, Lung, and Blood Institute.
Presented in part at American Heart Association Scientific Sessions 2008, November 8–12, 2008, New Orleans, La.
Krukenberg E. Beitrage zur Frage des Aneurysma dissecans. Beitr Patho Anat Allg Pathol. 1920; 67: 329–351.
Evangelista A, Mukherjee D, Mehta RH, O'Gara PT, Fattori R, Cooper JV, Smith DE, Oh JK, Hutchison S, Sechtem U, Isselbacher EM, Nienaber CA, Pape LA, Eagle KA. Acute intramural hematoma of the aorta: a mystery in evolution. Circulation. 2005; 111: 1063–1070.
Nienaber CA, von Kodolitsch Y, Petersen B, Loose R, Helmchen U, Haverich A, Spielmann RP. Intramural hemorrhage of the thoracic aorta: diagnostic and therapeutic implications. Circulation. 1995; 92: 1465–1472.
Robbins RC, McManus RP, Mitchell RS, Latter DR, Moon MR, Olinger GN, Miller DC. Management of patients with intramural hematoma of the thoracic aorta. Circulation. 1993; 88: II-1–II-10.
Song JM, Kim HS, Song JK, Kang DH, Hong MK, Kim JJ, Park SW, Park SJ, Lim TH, Song MG. Usefulness of the initial noninvasive imaging study to predict the adverse outcomes in the medical treatment of acute type A aortic intramural hematoma. Circulation. 2003; 108: II-324–II-328.
Kaji S, Akasaka T, Horibata Y, Nishigami K, Shono H, Katayama M, Yamamuro A, Morioka S, Morita I, Tanemoto K, Honda T, Yoshida K. Long-term prognosis of patients with type a aortic intramural hematoma. Circulation. 2002; 106: I-248–I-252.
Moriyama Y, Yotsumoto G, Kuriwaki K, Watanabe S, Hisatomi K, Shimokawa S, Toyohira H, Taira A. Intramural hematoma of the thoracic aorta. Eur J Cardiothorac Surg. 1998; 13: 230–239.
Ehrlich MP, Ergin MA, McCullough JN, Lansman SL, Galla JD, Bodian CA, Apaydin A, Griepp RB. Results of immediate surgical treatment of all acute type A dissections. Circulation. 2000; 102: III-248–III-252.
von Kodolitsch Y, Csosz SK, Koschyk DH, Schalwat I, Loose R, Karck M, Dieckmann C, Fattori R, Haverich A, Berger J, Meinertz T, Nienaber CA. Intramural hematoma of the aorta: predictors of progression to dissection and rupture. Circulation. 2003; 107: 1158–1163.
Kaji S, Nishigami K, Akasaka T, Hozumi T, Takagi T, Kawamoto T, Okura H, Shono H, Horibata Y, Honda T, Yoshida K. Prediction of progression or regression of type A aortic intramural hematoma by computed tomography. Circulation. 1999; 100: II-281–II-286.
Kan CB, Chang RY, Chang JP. Optimal initial treatment and clinical outcome of type A aortic intramural hematoma: a clinical review. Eur J Cardiothorac Surg. 2008; 33: 1002–1006.
Song JK. Diagnosis of aortic intramural haematoma. Heart. 2004; 90: 368–371.