doi: 10.1161/01.CIR.0000138219.67028.2a
(Circulation. 2004;110:II-237 – II-242.)
© 2004 American Heart Association, Inc.
Surgery for Aortic and Peripheral Vascular Disease |
Differences in Clinical Presentation, Management, and Outcomes of Acute Type A Aortic Dissection in Patients With and Without Previous Cardiac Surgery
From the University of Michigan, (J.S.C., J.F., J.V.C., D.E.S., K.A.E.), Ann Arbor, Mich; Universitari Vall d’Hebron (A.E.), Barcelona, Spain; University of Rostock (C.A.N.), Germany; National Research Council, Southern Italy (E.B.), Brindisi, Italy; Brigham and Women’s Hospital (P.T.O.), Boston, Mass; Tromsø University Hospital (T.M.), Tromsø, Norway; Hadassah University Hospital (D.G.), Jerusalem, Israel; Massachusetts General Hospital (E.M.I.), Boston, Mass; Cleveland Clinic Foundation (M.P.), Cleveland, Ohio; University of Massachusetts Hospital (L.A.P.), Worchester, Mass; Duke Clinical Research Institute (R.H.M.), Durham, North Carolina.
Correspondence to Rajendra H. Mehta, MD, 2802 Leslie Park Circle, Ann Arbor, MI 48105. E-mail mehta007{at}dcri.duke.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Background— There are less data on the clinical and diagnostic imaging characteristics, management, and outcomes of patients with previous cardiac surgery (PCS) presenting with acute type A aortic dissection (AAD).
Methods and Results— In 617 patients with AAD, we evaluated the differences in the clinical characteristics, management, and in-hospital outcomes of the cohorts with and without PCS. A history of PCS was present in 100 of 617 patients. Patients with PCS were more likely to be males (P=0.02), older (P=0.014), and to have a history of previous aortic dissection (P<0.001) or aneurysms (P<0.001). In contrast, PCS patients were less likely to have presenting chest pain (P<0.001). Cardiac tamponade was less common in PCS patients (P=0.007). Fewer AAD patients with PCS underwent surgical repair (P=0.001). Hospital mortality was not adversely influenced by PCS (odds ratio [OR], 1.46; 95% confidence interval [CI], 0.81 to 2.63), but a trend for increased death was seen in patients with previous aortic valve replacement (AVR) (OR, 2.31; 95% CI, 0.98 to 5.43). Age70 years or older, previous AVR, shock, and renal failure identified PCS patients at risk for death.
Conclusions— Our study highlights differences in clinical characteristics, management, and outcomes of AAD patients with PCS. Importantly, PCS, with the exception of previous AVR, does not adversely influence early outcomes of AAD patients, including those undergoing surgical repair. However, because of otherwise dismal outcomes with medical management of AAD, our data indicate that a history of PCS (even that of previous AVR) should not preclude physicians from recommending surgical correction of type A aortic dissection in appropriate patients.
Key Words: aorta • surgery • risk factors • mortality
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Acute aortic dissection has an incidence of 0.03% to 0.1% after major cardiac surgery.1,2 This incidence is highest for patients undergoing aortic valve replacement (AVR) (0.5% to 1.0%) with other conditions predisposing to subsequent dissection including congenitally deformed valve, overlooked annuloaortic ectasia, root aneurysms, and redissection.3–7 Furthermore, iatrogenic dissection complicating cardiac surgery has been reported at a rate of 0.12% to 0.16%, many of which are missed at primary surgery and are often detected later.8,9 Even in patients undergoing surgery for the repair of type A aortic dissection, re-operation is required in 10% of patients at 5 years and up to 40% at 10 years after initial surgical repair.10–13
Despite the occurrence of aortic dissection in patients with prior cardiac surgery (PCS), less is known about the influence of PCS on the presentation, management, and outcomes in patients with type A aortic dissection. The purpose of this study was to evaluate the differences in clinical features, diagnostic findings, management, and outcomes of patients with and without a history of PCS presenting with acute type A aortic dissection. Additionally, we sought to determine patient characteristics associated with increased risk of in-hospital mortality in the cohort with PCS.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Study Population
We examined data on all patients with acute type A aortic dissection in the International Registry of Acute Aortic Dissection (IRAD) enrolled from January 1, 1996 to December 31, 2001. The rationale and methodology of IRAD have been previously published.14–16 Acute type A dissection was defined as any dissection that involved the ascending aorta with presentation within 14 days of symptom onset.14–17 Patients with acute type A aortic dissection were categorized into 2 groups: those with and without history of PCS.
Data Collection
Data were collected on a standard questionnaire form and included information on patient clinical characteristics, imaging findings, management, and hospital clinical events including mortality. Completed data forms were forwarded to the IRAD coordinating center at the University of Michigan and were entered into an Access database.
Statistical Analysis
Summary statistics of the 2 comparison groups (those with and without PCS) were presented as frequencies and percentages and mean±SD. In all cases, missing data were not defaulted to negative and denominators reflect cases reported. Univariate associations among the 2 groups for nominal variables were compared using the Pearson 
2 test or, when appropriate, 2-sided Fisher exact test, whereas the 2-tailed Student t test or Wilcoxon Mann–Whitney tests was used for continuous variables. Iterative logistic regression modeling was performed to derive independent predictors of hospital mortality and adjusted estimates for the odds ratios (OR) and 95% confidence interval (CI) of in-hospital mortality for the patients with previous cardiac surgery using likelihood ratio tests. Initial modeling used variables marginally suggestive of unadjusted association to in-hospital death (P<0.20). Variables were reviewed for clinical significance before testing. Diagnostic routines (the Hosmer–Lemeshow test for lack of fit and likelihood ratio test) were used for the final model selection. The c-index was calculated to evaluate model discrimination. This model was used to determine the expected mortality of patients with and without PCS treated medically or with surgery. Similarly, multivariate logistic regression analysis was performed to identify features associated with high-risk for in-hospital mortality in patients with type A dissection and PCS. SAS Version 8.2 (SAS Institute) was used for all analyses.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Demographics, Cause, Presenting Features, Imaging Findings, and Management in Patients With Acute Type A Aortic Dissection With and Without PCS
A history of PCS was present in 100 of 617 patients with acute type A dissection (Tables 1 , 2 , and 3
). Patients with PCS were older and more likely to be male. Comorbid conditions varied in PCS patients. Thus, history of previous aortic dissection or aneurysm was almost exclusively prevalent in PCS patients, whereas iatrogenic dissection was 2.4-fold higher in this cohort compared with those without PCS. Diabetes mellitus and atherosclerosis were significantly more common among patients with PCS. Typical presentation with chest or back pain, particularly that with an abrupt or migrating nature, occurred less commonly among PCS patients, and fewer patients in this cohort presented within 6 hours of their symptom onset. Congestive heart failure on presentation was 2.3-fold higher among patients with PCS.
|
|
|
Patients with PCS were 2.5-fold more likely to have coronary angiography before their surgery. The use of other diagnostic imaging studies did not differ between the 2 groups. Findings less frequent in the cohort with PCS included widened mediastinum and involvement of arch vessels and pericardial effusion with evidence of cardiac tamponade on imaging modalities. In contrast, thrombosis of false lumen was 1.6-fold higher in the PCS group.
Patients with type A dissection and PCS were twice as likely to be managed medically and the surgical treatment was more likely to be delayed beyond 24 hours of presentation. The reasons for medical management among patients with PCS included older age (21%), intramural hematoma (21%), refusal of surgery by patient and/or family (25%), and the presence of significant comorbid conditions in the remaining. The circulatory arrest time and frequency of ascending aorta replacement were lower in the PCS group.
Hospital Complications in Patients With Acute Type A Aortic Dissection With and Without PCS
Hospital complications did not differ between the 2 comparison groups with the exception of cardiac tamponade, which occurred less frequently in patients with PCS (Table 4). Furthermore, cardiac tamponade was not reported as a cause of death in any of the 37 PCS patients who died. In-hospital mortality did not differ significantly between the 2 groups (adjusted mortality OR, 1.45; 95% CI, 0.81 to 2.63; P=0.21). Similarly, among patients undergoing surgery for type A dissection, in-hospital mortality did not differ from those who had no history of PCS (OR, 1.64; 95% CI, 0.80 to 3.38; P=0.18; Figure). The mode of death was attributed to rupture (27%), neurologic injury (11%), and visceral ischemia (10.9%), although it was not known in remaining.
|
|
On further evaluation of the impact of the type of PCS on in-hospital death, we found a nonsignificant higher mortality among patients who had previous AVR than in those that did not have previous AVR (42.0% versus 29.0%, P=0.13). In-hospital death occurred in 10 of 25 patients (40%) with previous AVR who underwent surgical repair of their type A dissection. When a history of previous AVR was entered into the logistic regression model instead of PCS to predict death, previous AVR was independently associated with a trend for increased mortality (OR, 2.31; 95% CI, 0.98 to 5.43; P=0.054).
Finally, among patients with PCS, multivariate analysis identified age 70 years or older (OR, 3.70; 95% CI, 1.13 to 12.12; P=0.031), previous AVR (OR, 4.54; 95% CI, 1.10 to 18.73; P=0.036), shock (OR, 4.35; 95% CI, 1.10 to 17.22; P=0.036), and acute renal failure (OR, 13.92; 95% CI, 1.94 to 99.90; P=0.009) as independent predictors of in-hospital mortality (c-statistic=0.81, 2=6.11, degrees of freedom=7, and P=0.53).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
The International Registry of Acute Aortic Dissection (IRAD) provided an opportunity to compile the largest study to date evaluating unselected patients with acute type A aortic dissection with and without a history of PCS. We found that patients with PCS and acute type A aortic dissection represent a unique population with important differences in their cause, diagnostic findings, management, and outcomes. Knowledge of these may be useful to physicians while triaging and treating such patients.
Differences in Clinical Presentation and Diagnostic Findings in Type A Aortic Dissection Patients With and Without PCS
Our study suggests that a history of PCS among patients with acute type A dissection accounts for 1 of every 6 patients with this life-threatening cardiovascular illness. Consistent with findings of previous investigations,10–13,17 our data indicate that aortic aneurysm or dissection and iatrogenic dissection during previous surgical procedures are important predisposing factors for subsequent acute type A dissection. The persistent abnormalities of aortic wall, untreated or recurrent dissection, and injuries originating from aortic cross clamp, intimal tears at the site of suture lines for bypass grafts, or cannulation during PCS account for the increased risk of subsequent ascending aortic tear among such patients.5,12,13,18 Thus, periodic follow-up evaluation, preferably with serial imaging studies, to recognize early occurrence of acute aortic tear among this cohort is extremely important and cannot be overemphasized. In fact, such a strategy is strongly endorsed by the Task Force Report on the Diagnosis and Management of Aortic Dissection of the European Society of Cardiology.17
Although abrupt onset of migratory chest or back pain is recognized as typical symptoms that often herald the onset of acute type A aortic dissection,14,17 these classic symptoms were less frequent among patients with PCS. This is presumably as a result of denervation of cardiac sympathetic nervous system secondary to previous operation. Physicians should be aware of the atypical presentation of acute aortic tear in this cohort, requiring a high index of suspicion for the timely diagnosis of this catastrophic cardiovascular event. These atypical symptoms heralding the onset of acute aortic dissection result in fewer patients with PCS presenting early after their symptom onset or undergoing surgical repair within first 24 hours of their presentation.
The imaging modality of choice did not differ between the 2 comparison groups for diagnosing aortic dissection. However, coronary angiography before surgical repair was performed more frequently among patients with PCS. Coronary angiography is generally considered unnecessary and potentially harmful because it often delays time to operation in acute type A dissection. Additionally, performing an angiogram in patients with type A aortic dissection has not been shown to favorably influence mortality.19,20 However, among patients with PCS, the judicious use of coronary angiography before surgical repair of dissection in selected patients is advocated by some investigators.5,21 Older age and the higher prevalence of atherosclerosis, coupled with these recommendations for its use in selected patient, may have accounted for more PCS patients at IRAD sites undergoing a coronary angiogram before surgical repair.5,21
A previous study has shown that cardiac tamponade is significantly less common in patients with PCS and acute type A aortic dissection.21 Our study findings are consistent with this previous report. The occurrence of a more localized aortic tear and adhesions from previous surgical procedures may result in lower incidence of cardiac rupture and explains, in part, the lower frequency of pericardial effusion and tamponade in this group but also may account for the lower incidence of arch involvement.
Differences in Management and Outcomes in Patients With Acute Type A Aortic Dissection With and Without PCS
Immediate surgery for repair of acute type A aortic dissection has been recommended to prevent rupture and/or cardiac tamponade, improve coronary, cerebral, and visceral circulation, and to treat acute aortic insufficiency.17 However, our data indicate that patients with PCS are less likely to be treated surgically and are more often managed by conservative medical strategy. Several potential explanations may be responsible for this less aggressive approach. Limited data are available on the natural history of acute type A aortic dissection among patients with PCS or the merits of surgical repair in this cohort. This lack of data together with evidence of increased morbidity and mortality for repeat coronary artery bypass or valve surgery (procedures much more commonly performed than aortic repair for dissection)22–24 may have resulted in unfounded perception of increased risk of surgical repair in patients with PCS. In addition, physicians may have opted for a more conservative approach because of greater comorbid conditions in patients with PCS, particularly older age.
Contrary to the widespread perception that PCS adversely influences the outcomes from acute type A dissection, our data suggest no such increased risk in this group of patients. Thus, most in-hospital complications and mortality were similar in patients with and without PCS who had acute type A aortic dissection. In fact, as previously discussed, cardiac tamponade was uncommon in patients with PCS. Although the mortality was somewhat higher in patients with PCS undergoing surgical repair, the risk was neither prohibitive nor statistically different from that observed in patients without PCS who were treated surgically. It is likely that the risk of repeat operation in patients with PCS is partly offset by the fact that rupture with cardiac tamponade is far less likely or that dissection in this cohort is more localized with lower prevalence of arch or aortic valve involvement. In contrast, medical management of type A aortic dissection was associated with uniformly higher mortality in both groups of patients (Figure). In fact, the observed mortality among medically managed patients was significantly higher than their expected mortality estimated based on our previously published model (Figure).16 Similarly, although patients with type A dissection and a history of previous AVR had a higher risk of in-hospital death than those without previous AVR, the 40% mortality observed with surgical management in this group was still less than the 50% to 60% mortality seen among patients with type A dissection managed medically. Thus, our data suggest benefit of surgical correction of acute type A dissection in patients with PCS.
Multivariate analysis identified a subset of patients with PCS and acute type A aortic dissection who are at increased risk for hospital death. These include age 70 or older, history of previous AVR, and the presence of shock or renal failure. These clinical variables explained 81% of the variability in mortality among patients with PCS and are consistent with our previous reports (c-index 0.81).15,16 Physicians should be aware of these high-risk features and use this knowledge to educate patients and families of their perceived risk. This should not be used to deny but to proceed with early surgical treatment of aortic tear in patients in this high-risk cohort, who are most likely to benefit from such aggressive approach.
Limitations
Our study findings should be viewed in light of some inherent limitations. Data were collected retrospectively and subject to incomplete, missing, or inaccurate reporting of events. Most IRAD centers were tertiary referral sites that have significant experience in the management of patients with aortic dissection, thus limiting the applicability to centers that lack such expertise. Third, only patients with acute type A dissection were included and findings should not be extrapolated to patients with chronic stable dissections or type B dissections. Fourth, because of the observational nature of our study and because the treatment allocation was not random, many factors besides those captured in our study may have contributed to the choice of treatment modality. As such, although our data suggest that surgical treatment is better than medical management of acute type A aortic dissection in patients with PCS, this should be regarded as hypothesis-generating rather than definitive evidence to suggest the superiority of surgical repair overconservative management in this cohort. Finally, long-term outcomes were not addressed and follow-up of all patients is underway currently.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Our study highlights important differences in cause, clinical features, diagnostic findings, and management of patients with PCS and acute type A aortic dissection. Importantly, PCS, with the exception of previous AVR, does not adversely influence early outcomes of patients with type A aortic dissection, including those undergoing surgical repair. However, because of otherwise dismal outcomes with medical management of type A dissection, our data indicate that a history of PCS (even that of previous AVR) should not preclude physicians from recommending surgical correction of type A aortic dissection in appropriate patients.
|
Acknowledgments |
|---|
Funding source: University of Michigan Faculty Group Practice and Varbedian Fund for Aortic Research.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
- Murphy DA, Craver JM, Jones EL, et al. Recognition and management of ascending aortic dissection complicating cardiac surgical operations. J Thorac Cardiovasc Surg. 1983; 85: 247–256.[Abstract]
- Suzuki S, Kondo H, Imoto K, et al. Type A acute aortic dissection: late reoperations for dilation of distal false lumen and aortic regurgitation. Kyobu Geka. 1998; 51: 656–660.[Medline]
[Order article via Infotrieve]
- Pieters FA, Widdershoven JW, Gerardy A, et al. Risk of Aortic Dissection After Aortic Valve Replacement. Am J Cardiol. 1993; 72: 1043–1047.[CrossRef][Medline]
[Order article via Infotrieve]
- von Kodolitsch Y, Simic O, Schwartz A, et al. Predictors of Proximal Aortic Dissection at the Time of Aortic Valve Replacement. Circulation. 1999; 100: II287–II294.
- Stanger O, Oberwalder P, Dacar D, et al. Late dissection of the ascending aortia after previous cardiac surgery: risk, presentation and outcome. Eur J Cardiothorac Surg. 2002; 21: 453–458.
[Abstract/Free Full Text]
- Albat B, Thevenet A. Dissecting aneurysms of ascending aorta occurring late after aortic valve replacement. J Cardiovasc Surg. 1992; 33: 272–275.[Medline]
[Order article via Infotrieve]
- Derkac W, Laks H, Cohn LH, et al. Dissecting aneurysms of ascending aorta occurring late after aortic valve replacement. Arch Surg. 1974; 109: 388–390.[Medline]
[Order article via Infotrieve]
- Still R, Hilgenberg AD, Akins CW, et al. Intraoperative Aortic Dissection. Ann Thorac Surg. 1992; 53: 374–380.[Abstract]
- Ruchat P, Hurni M, Stumpe F, et al. Acute ascending aortic dissection complicating open heart surgery: cerebral perfusion defines the outcome. Eur J Cardiothorac Surg. 1998; 14: 449–452.
- Erbel R, Oelert H, Meyer J, et al. Influence of medical and surgical therapy on aortic dissection evaluated by transesophageal echocardiography. Circulation. 1993; 87: 1604–1615.
[Abstract/Free Full Text]
- Glower DD, Speier RH, White WD, et al. Management and longterm outcome of aortic dissection. Ann Surg. 1991; 214: 31–41.[Medline]
[Order article via Infotrieve]
- Crawford ES, Coselli JS. Marfan’s syndrome: combined composite valve graft replacement of aortic root and transaortic mitral valve replacement. Ann Thorac Surg. 1988; 45: 296–302.[Abstract]
- McDonald GR, Schaff HV, Pyeritz RE, et al. Surgical management of patients with Marfan syndrome and dilation of the ascending aorta. J Thorac Cardiovasc Surg. 1981; 81: 180–186.[Abstract]
- Hagan PG, Nienaber CA, Isselbacher et al. The International Registry of Acute Aortic Dissection (IRAD): New Insights Into an Old Disease. JAMA. 2000; 283: 897–903.
[Abstract/Free Full Text]
- Mehta RH, O’Gara PT, Bossone E, et al. Acute Type A Dissection in the Elderly: Clinical Characteristics, Management, and Outcomes in the Current Era. J Am Coll Cardiol. 2002; 40: 685–692.
[Abstract/Free Full Text]
- Mehta RH, Suzuki T, Hagan PG et al. Predicting Death in Patients With Acute Type A Aortic Dissection. Circulation. 2002; 105: 200–206.
[Abstract/Free Full Text]
- Erbel R, Alfonso F, Boileau C, et al. Diagnosis and management of acute aortic dissection: Recommendations of Task Force on Aortic Dissection, European Society of Cardiology. Eur Heart J. 2001; 22: 1642–1681.
[Free Full Text]
- Januzzi JL, Sabatine MS, Eagle KA, et al. Iatrogenic Aortic Dissection. Am J Cardiol. 2002; 89: 623–626.[CrossRef][Medline]
[Order article via Infotrieve]
- Creswell LL, Kouchoukos NT, Cox JL, et al. Coronary artery disease in patients with type A aortic dissection. Ann Thorac Surg. 1995; 59: 585–590.
[Abstract/Free Full Text]
- Penn MS, Smedira N, Lytle B, et al. Does Coronary Angiography Before Emergency Aortic Surgery Affect In-Hospital Mortality? J Am Coll Cardiol. 2000; 35: 889–894.
[Abstract/Free Full Text]
- Gillinov AM, Lytle BW, Kaplon RJ, et al. Dissection of the Ascending Aorta After Previous Cardiac Surgery: Differences in Presentation and Management. J Thorac Cardiovasc Surg. 1999; 117: 252–260.
[Abstract/Free Full Text]
- Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA Guidelines for Coronary Artery Bypass Graft Surgery. J Am Coll Cardiol. 2000; 34: 1262–1338.
- Edwards FH, Petersen ED, Coombs LP, et al. Prediction of operative mortality after valve replacement surgery. J Am Coll Cardiol. 2001; 37: 885–892.
[Abstract/Free Full Text]
- Lytle BW, Cosgrove DM, Taylor PC, et al. Reoperations for Valve Surgery: Perioperative Mortality and Determinants of Risk for 1,000 Patients, 1958–1984. Ann Thorac Surg. 1986; 42: 632–643.[Abstract]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||