Author + information
- Received November 3, 2014
- Revision received March 23, 2015
- Accepted April 14, 2015
- Published online June 23, 2015.
- Martin Czerny, MD, MBA∗∗ (, )
- Florian Schoenhoff, MD†,
- Christian Etz, MD‡,
- Lars Englberger, MD†,
- Nawid Khaladj, MD§,
- Andreas Zierer, MD‖,
- Ernst Weigang, MD¶,
- Isabell Hoffmann, MD#,
- Maria Blettner, MD# and
- Thierry P. Carrel, MD†
- ∗University Clinic for Cardiovascular Surgery, University Hospital Zurich, Zurich, Switzerland, and University Heart Center Freiburg-Bad Krozingen, Bad Krozingen, Germany
- †University Clinic for Cardiovascular Surgery, University Hospital Bern, Bern, Switzerland
- ‡Department of Cardiac Surgery, Heart Center Leipzig, Leipzig, Germany
- §Department for Cardiac Surgery, University Hospital Munich, Munich, Germany
- ‖Division of Thoracic and Cardiovascular Surgery, Johann Wolfgang Goethe University, Frankfurt, Germany
- ¶Department of Vascular and Endovascular Surgery, Clinic Hubertus Hospital, Berlin, Germany
- #Institute for Medical Biostatistics, Epidemiology and Informatics, Medical Centre of the Johannes Gutenberg University Mainz, Mainz, Germany
- ↵∗Reprint requests and correspondence:
Dr. Martin Czerny, University Heart Center Freiburg-Bad Krozingen, Hugstetterstrasse 55, Freiburg 79106, Germany.
Background Malperfusion adversely affects outcomes in patients with acute type A aortic dissection, but reliable quantitative data are lacking.
Objectives The aim of this study was to analyze the impact of various forms of malperfusion on early outcome.
Methods A total of 2,137 consecutive patients enrolled in GERAADA (German Registry for Acute Aortic Dissection Type A) who underwent surgery between 2006 and 2010, of whom 717 (33.6%) had any kind of pre-operative malperfusion, were retrospectively analyzed.
Results All-cause 30-day mortality was 16.9% and varied substantially according to the number of organ systems affected by malperfusion (none, 12.6%; 1 system, 21.3%; 2 systems, 30.9%; 3 systems, 43.4%; p < 0.001). Pre-operative cerebral malperfusion, comatose state, peripheral malperfusion, visceral malperfusion, involvement of supra-aortic branches, coronary malperfusion, and renal malperfusion were all independent predictors of developing any post-operative malperfusion syndrome. When survival was considered, age, peripheral malperfusion, involvement of supra-aortic branches, coronary malperfusion, spinal malperfusion, a primary entry in the descending aorta, and pre-operative comatose state were independent predictors, again with increasing significance.
Conclusions Malperfusion remains a severe clinical condition with strong potential for adverse outcomes in patients undergoing surgery for acute type A aortic dissection. The GERAADA registry suggests that the impact of the number of organs involved and the type of malperfusion on outcome differs substantially. Introducing an appropriate classification system, such as “complicated” and uncomplicated” acute type A aortic dissection, might help predict individual risk as well as select a surgical strategy that may quickly resolve malperfusion.
Surgical repair of acute type A aortic dissection (AADA) remains the treatment of choice in the large majority of patients when anticipated benefit outweighs risks (1). Several pre-operative factors are known to adversely affect outcomes (2), among the most decisive of which is malperfusion, defined as compromised blood flow in 1 or more organs resulting in ischemia and organ dysfunction, as a consequence of the dissection process itself (3). Despite major efforts, outcomes in these patients remain suboptimal. Therefore, different strategies to resolve malperfusion have been developed (4), but only limited data exist regarding the impact of malperfusion on outcomes.
The aim of the present study was to determine the impact of various forms of malperfusion on early outcome in patients undergoing surgery for AADA.
Patient selection, data collection, and analysis
GERAADA (German Registry for Acute Aortic Dissection Type A) is a prospective, multicenter registry initiated and set up by the Working Group for Aortic Surgery and Interventional Vascular Surgery of the German Society of Thoracic and Cardiovascular Surgery. We analyzed 90 parameters collected by 50 centers for each patient undergoing surgery for AADA between July 2006 and June 2010. (Participating centers are listed in the Online Appendix.) In this period, a total of 2,137 patients were enrolled in GERAADA. Baseline characteristics, diagnostic information, surgical treatment, post-operative course, and 30-day survival were collected. Data were validated by an independent database-monitoring center, with attention to completeness and plausibility. Data analysis and statistical calculations were performed in collaboration with the Institute for Medical Biostatistics, Epidemiology, and Informatics in Mainz, Germany. The registry itself and all of its studies were approved by a central ethics committee (No. 7293; Landesärztekammer Rheinland-Pfalz, Mainz, Germany).
Primary outcome parameters were any post-operative malperfusion and 30-day mortality.
Summary statistics are presented as frequencies and percents for categorical values and as mean ± SD for continuous values. To compare patients with and without pre-operative malperfusion, chi-square and Student t tests were used. The impact of variables on any post-operative malperfusion was analyzed using multiple logistic regression analysis with forward selection and a selection level for including an effect of 0.05. The following explanatory variables were thus selected: pre-operative coronary malperfusion, cerebral malperfusion, spinal malperfusion, visceral malperfusion, renal malperfusion, peripheral malperfusion, primary entry tear in ascending aorta, primary entry tear in the aortic arch, primary entry tear in descending aorta, sex, hemodynamic status, etiology of aortic dissection, diagnostic modality (echocardiography, computed tomographic scanning, angiography, magnetic resonance angiography), functional status of the aortic valve (aortic insufficiency, aortic stenosis), pre-operative hemiparesis or hemiplegia, pre-operative paraparesis or paraplegia, aphasia, comatose state, propagation of the dissection membrane (into the aortic arch, supra-aortic branches, descending aorta, abdominal aorta, or iliac axis), extent of surgery, cannulation sites, cerebral protection strategy, age, and symptoms. Logistic regression was performed for any post-operative malperfusion as well as for post-operative coronary, cerebral, spinal, visceral, renal, or peripheral malperfusion. Finally, a regression analysis for death during the first 30 days post-surgery was performed.
The descriptive statistical analysis, regression analysis, and tests were conducted using only cases without missing values (complete case analysis). Because this was an exploratory study, it should be noted that the term “statistically significant” must be interpreted with care. There was no formal adjustment for the number of performed tests. For this reason, p values should be considered exploratory.
Of 2,137 patients, 717 (33.6%) had any type of pre-operative malperfusion. The distribution of organs affected by pre-operative malperfusion is shown in Table 1. In the first month, 362 patients died, for a cumulative 30-day mortality rate of 16.9%. There was a substantial difference in mortality depending on the number of malperfused organs before surgery (Table 2).
The primary intimal tear was located in the ascending aorta in 1,613 patients (75%), in the aortic arch in 314 (15%), and in the descending aorta in 41 (2%). The dissection membrane extended into the supra-aortic branches in 798 patients (37%) (Table 1).
The right subclavian and, alternatively, the axillary arteries were used as arterial inflow in 649 (30%) and 255 (12%) patients, respectively. Hemi- and total-arch replacement was performed in 985 (46%) and 346 (16%) subjects, respectively, whereas uni- and bilateral cerebral perfusion techniques were used in 828 (39%) and 654 (31%) patients, respectively (Table 1).
A total of 41 patients (1.9%) experienced post-operative coronary malperfusion. Post-operative cerebral malperfusion was observed in 145 patients (6.8%), and post-operative spinal malperfusion occurred in 23 (1.1%). Post-operative visceral malperfusion was observed in 81 (3.8%) patients, renal malperfusion in 146 (6.8%), and peripheral malperfusion in 71 (3.3%).
Pre- and post-operative risk factors for any malperfusion
Pre-operative cerebral malperfusion (odds ratio [OR]: 1.56; 95% confidence interval [CI]: 1.07 to 2.24; p = 0.019), comatose state (OR: 1.57; 95% CI: 1.08 to 2.25; p = 0.017), peripheral malperfusion (OR: 1.59; 95% CI: 1.13 to 2.24; p = 0.008), visceral malperfusion (OR: 1.68; 95% CI: 1.05 to 2.66; p = 0.028), involvement of supra-aortic branches (OR: 1.71; 95% CI: 1.29 to 2.25; p < 0.001), coronary malperfusion (OR: 1.80; 95% CI: 1.20 to 2.66; p = 0.004), and renal malperfusion (OR: 2.53; 95% CI: 1.71 to 3.71; p < 0.001) were independent predictors of any post-operative malperfusion syndrome in increasing significance (Table 3). The effect of combinations of the binary risk factors can be gauged by multiplying ORs. The combination of pre-operative cerebral and coronary malperfusion correspond to an OR of 2.81; cerebral and visceral malperfusion to an OR of 2.62; cerebral, coronary, and visceral malperfusion to an OR of 4.25; and the combination of all 4 types (including renal malperfusion) of pre-operative malperfusion to an OR of 11.94.
Pre-operative renal malperfusion (OR: 2.60; 95% CI: 1.03 to 5.96; p = 0.031), aphasia (OR: 4.69; 95% CI: 0.72 to 17.73; p = 0.047), and coronary malperfusion (OR: 7.53; 95% CI: 3.56 to 15.73; p < 0.001) as well as aortic stenosis (p < 0.001) were independent predictors of post-operative coronary malperfusion in increasing significance (Table 4).
A pre-operative comatose state (OR: 1.86; 95% CI: 1.14 to 2.97; p = 0.010), involvement of supra-aortic branches by the dissection (OR: 2.18; 95% CI: 1.40 to 3.43; p < 0.001), and pre-operative cerebral malperfusion (OR: 3.94; 95% CI: 2.51 to 6.15; p < 0.001) were independent predictors of post-operative cerebral malperfusion in increasing significance (Table 5).
A primary entry site in the descending aorta with retrograde involvement of the ascending aorta (OR: 2.24; 95% CI 1.25 to 4.19; p = 0.009), pre-operative renal malperfusion (OR: 2.77; 95% CI 1.52 to 4.96; p < 0.001), and pre-operative visceral malperfusion (OR: 9.40; 95% CI: 5.20 to 16.98; p < 0.001) were independent predictors of post-operative visceral malperfusion (Table 6).
Age (OR: 1.02; 95% CI: 1.01 to 1.04; p = 0.006), pre-operative coronary malperfusion (OR: 1.88; 95% CI: 1.07 to 3.17; p < 0.001), and pre-operative renal malperfusion (OR: 10.82; 95% CI: 7.19 to 16.31; p < 0.001) were independent predictors of post-operative renal malperfusion (Table 7).
Pre-operative hemiparesis or hemiplegia (OR: 4.90; 95% CI: 1.36 to 15.59; p = 0.009), pre-operative spinal malperfusion (OR: 6.21; 95% CI: 1.69 to 23.09; p = 0.006), pre-operative paraparesis or paraplegia (OR: 9.11; 95% CI: 2.11 to 34.28; p = 0.002), and aortic stenosis (p = 0.011) were independent predictors of post-operative spinal malperfusion (Table 8).
Pre-operative visceral malperfusion (OR: 1.88; 95% CI: 0.96 to 3.52; p = 0.057), involvement of the supra-aortic branches by the dissection process (OR: 2.23; 95% CI: 1.28 to 3.99; p = 0.006), and pre-operative peripheral malperfusion (OR: 10.32; 95% CI: 6.01 to 18.06; p < 0.001), were independent predictors of post-operative peripheral malperfusion (Table 9).
Risk factors for early mortality
When survival was considered, age (OR: 1.02; 95% CI: 1.01 to 1.03; p < 0.001), peripheral malperfusion (OR: 1.43; 95% CI: 1.01 to 2.01; p = 0.042), involvement of supra-aortic branches (OR: 1.47; 95% CI: 1.13 to 1.89; p = 0.004), coronary malperfusion (OR: 1.61; 95% CI: 1.10 to 2.31; p = 0.012), spinal malperfusion (OR: 2.18; 95% CI: 1.11 to 4.28; p = 0.027), a primary entry in the descending aorta (OR: 2.84; 95% CI: 1.37 to 5.59; p = 0.004), and pre-operative comatose state (OR: 3.42; 95% CI: 2.49 to 4.67; p < 0.001) were independent predictors in increasing significance. Furthermore, post-operative cerebral malperfusion (OR: 2.18; 95% CI: 1.45 to 3.24; p < 0.001), post-operative visceral malperfusion (OR: 3.24; 95% CI: 1.94 to 5.35; p < 0.001), and post-operative coronary malperfusion (OR: 9.54; 95% CI: 4.62 to 20.69; p < 0.001) were independent risk factors for early mortality (Table 10). Just as for the other logistic regression models, the effect of combinations of the binary risk factors can be gauged by multiplying the ORs. For example, post-operative cerebral and coronary malperfusion combined correspond to an OR of 20.80, cerebral and visceral to an OR of 7.06, and the combination of all 3 factors to an OR of 67.38.
Malperfusion remains a severe condition that is frequently associated with adverse outcomes in patients undergoing surgery for AADA. Data from GERAADA suggest that the impact of the number of organs involved and the type of malperfusion on outcome differs substantially. An appropriate classification, such as “complicated” and “uncomplicated” AADA, may help predict individual risk as well as suggest a therapeutic strategy that may resolve malperfusion.
Even at the time of diagnosis, the proportion of patients presenting with malperfusion secondary to AADA was very high. This may be due to the recent interest in the early detection of malperfusion. A decade ago, malperfusion syndrome was not diagnosed as frequently as it is today (5), likely because of the very low threshold for performing computed tomographic scans in emergency units today when patients present with acute chest pain. Because the natural attrition rate of patients with AADA and malperfusion remains very high, early suspicion and early imaging may very well be contributing to the higher prevalence. The present study shows that malperfusion is a common problem in patients with AADA and that the approach to these patients has become more refined.
In this registry, there was a substantial difference in early survival depending on the presence or absence of any type of malperfusion syndrome and also the number of organs involved. Interestingly, a linear correlation was found between the number of malperfusion-affected organs and mortality that increased in steps of 10% per any additional organ involved. This prompted us to suggest a classification for AADA with and without malperfusion. When the results of this analysis are considered, it seems justified to categorize patients with AADA as uncomplicated (no clinical or imaging signs of malperfusion) or complicated (clinical or imaging signs of malperfusion). This simple stratification allows better decision making and may integrate the expected prognosis in the decision, because a direct correlation between malperfusion and early survival was clearly demonstrated (Central Illustration).
With regard to the arterial cannulation site, the operative technique, and the cerebral perfusion strategy, this report does not add new knowledge regarding these parameters and their relationships to outcomes, which have already been extensively reported (6–17). However, it remains remarkable that many surgeons still limit the extent of surgery in patients with AADA to the ascending aorta. Without doubt, limited repair has a higher probability of later need for aorta-related reintervention. However, the primary aim in AADA remains a living patient, and if a center is able to achieve this with limited repair alone, the primary intention is reached (18,19).
Pre-operative cerebral malperfusion, comatose state, peripheral malperfusion, visceral malperfusion, extension of the dissection membrane into the supra-aortic branches, coronary malperfusion, and renal malperfusion were independent predictors of any post-operative malperfusion syndrome with increasing significance. It may be expected that there is a mutual relationship between different types of malperfusion. Cerebral malperfusion, for instance, may well correlate with coma as well as the involvement of the supra-aortic branches by the dissecting membrane. Similarly, visceral and renal malperfusion in the presence of peripheral malperfusion may show a certain common pattern. In any case, a malperfusion-oriented strategy may help improve outcomes. As expected, a prompt diagnosis of malperfusion-induced organ injury remains a crude determinant of outcome (3,4).
Pre-operative coronary malperfusion as well as aortic stenosis were independent predictors of post-operative coronary malperfusion, which may demonstrate different clinical patterns. The most important ones are functional obstruction of the coronary origin by the dissection membrane process and the extension of the dissection into the coronary vessel. Additionally, a recent study has demonstrated that up to 21% of patients with AADA may also present with coronary artery disease if pre-operative angiography is performed systematically (4). It is beyond the design of this database to stratify the extent of myocardial injury caused by pre-operative malperfusion, but it may help to adopt a protective perfusion strategy such as continuous retrograde blood flow through a coronary sinus catheter during cooling. In the majority of cases, surgery will solve the problem by restitution of adequate blood flow into the coronary arteries. It was interesting to observe that aortic stenosis was a risk factor for post-operative coronary malperfusion given that it is unusual to observe patients with calcific degeneration of the aortic valve presenting with AADA. What is addressed here is clearly not aortic stenosis due to calcific degeneration but rather a functional component due to extensive involvement of the aortic root by the dissective process.
Pre-operative coma, involvement of the supra-aortic branches by the dissection process, and pre-operative cerebral malperfusion were independent predictors of post-operative cerebral malperfusion. It is obvious that these 3 entities exist in mutual dependency. Management of cerebral malperfusion is a matter of discussion in the surgical community and there is broad consensus that the primary aim of surgical repair is the quick restoration of antegrade cerebral blood flow by aortic repair. However, other scenarios may be helpful, such as selective antegrade cerebral perfusion during the cooling phase using carotid artery cannulation to quickly reestablish adequate flow (20,21). Nevertheless, if such an approach becomes necessary, the probability of reversing symptoms is already low.
Pre-operative renal malperfusion, pre-operative coronary malperfusion, and a primary entry site in the descending aorta were independent predictors of post-operative visceral malperfusion. Visceral malperfusion is among the most serious and detrimental forms of malperfusion. Consequently, several groups have modified their treatment algorithms in these patients and shifted away from primary surgery to resect the intimal tear to any method of reestablishment of visceral perfusion followed by a period of metabolic and hemodynamic recompensation with delayed surgical repair (4,22,23). If proceeding so, it is obvious that a certain attrition rate due to the dissection itself will occur (3,4). Interestingly, a primary entry tear in the descending aorta was a risk factor for post-operative visceral malperfusion. This is an important finding, because patients with primary entry tears in the descending aorta represent a subgroup at higher risk with additional need for more extensive repair than patients with primary entry tears in the ascending aorta. In particular, some of these patients may benefit from a “frozen elephant trunk” procedure to fix the entire pathology (24). Also in these cases, decision making remains individual. This is clearly a matter of anticipation and planning, as diagnosis may have been made only after a thoracic computed tomographic scan, without any information regarding the abdomen or even merely transthoracic echocardiography. It is still the protocol of many to immediately proceed with surgery without having further diagnostics to confirm or exclude visceral malperfusion as soon as the AADA diagnosis is established.
Age and pre-operative coronary and renal malperfusion were independent predictors of post-operative renal malperfusion, a very heterogeneous entity that might occur alone or in combination with visceral malperfusion. If it occurs alone, the main mechanism is the extension of the membrane into 1 renal artery; surgical repair of the ascending aorta may or may not relieve this problem. Because the second unaffected kidney may compensate, unilateral renal malperfusion is usually diagnosed late and does not need additional intervention. If renal malperfusion occurs in combination, consequences are more serious and a complex proximal repair such as the frozen elephant trunk procedure, with or without any additional intervention (endovascular fenestration) before or after proximal thoracic aortic repair, is warranted.
Pre-operative hemiparesis or hemiplegia, pre-operative spinal malperfusion, pre-operative paraparesis or paraplegia, and aortic stenosis were independent predictors of post-operative spinal malperfusion. It is well known that any kind of pre-operative symptomatic spinal cord injury will persist if the malperfusion-oriented strategy is not adapted. Cerebrospinal fluid drainage may present a highly effective means to attenuate or reverse symptoms (25). It remains an individual decision if the clinical presentation (in particular the presence of cardiac tamponade) allows this additional time-consuming step.
Independent predictors of post-operative peripheral malperfusion included pre-operative visceral malperfusion, involvement of the supra-aortic branches, and pre-operative peripheral malperfusion. Pre-operative peripheral malperfusion might also occur alone or in combination with other organ malperfusion. If occurring by itself, femorofemoral bypass grafting might represent a quick and effective immediate solution that can be performed simultaneously with proximal thoracic aortic repair or during cannulation and cooling (26). If it occurs in combination, the relief of malperfusion in upstream segments is likely to resolve peripheral malperfusion too (27).
When early survival is considered, age, peripheral malperfusion, involvement of the supra-aortic branches, coronary malperfusion, spinal malperfusion, a primary entry site in the descending aorta, and pre-operative comatose state were independent predictors. This finding confirms the need for individual treatment strategies for each malperfusion-affected organ. Furthermore, post-operative cerebral malperfusion, post-operative visceral malperfusion, and post-operative coronary malperfusion were all independent predictors of impaired early survival.
Study limitations, strengths, and future perspectives
This study was limited by the fact that the individual approach to patients with malperfusion differed substantially from institution to institution. Furthermore, the extent of baseline organ injury by malperfusion was not recorded and remains speculative. What we aimed for was to sensitize the community to the substantial difference of presence or absence of malperfusion with regard to outcome as well as the many aspects of such malperfusion and its variable impact. Without a doubt, this leads to the inevitable question of futility of treatment, yet we are hesitant to recommend a cutoff in these patients given that, fortunately, there are outliers with regard to outcomes, and the final decision of accepting or denying treatment should remain an individual one. In addition, the long-term clinical course of these patients are not known, and therefore the effect of pre-operative malperfusion on long-term outcome remains unanswered. Regarding the statistical analyses, regression models for outcomes with only a small number of events might be subject to instability or overfitting. We specifically chose a forward inclusion strategy to ameliorate such problems, but they cannot be completely avoided.
Nevertheless, this study represents the first attempt to stratify malperfusion syndromes and their impact with regard to all involved organs. Finally, this study is the first to suggest a pre-operative stratification of patients with AADA into “complicated” and “uncomplicated” groups on the basis of the presence or absence of malperfusion. This approach may facilitate decision making and better estimate prognosis in patients affected by more complex forms of AADA.
Malperfusion remains a severe condition that is frequently associated with adverse outcome in patients undergoing surgery for AADA. Data from GERAADA suggest that the impact of the number of organs involved and the type of malperfusion on outcomes differs substantially. An appropriate classification, such as “complicated” and uncomplicated” AADA, may help predict the individual risk as well as assist in the choice of a therapeutic strategy that may resolve malperfusion.
COMPETENCY IN MEDICAL KNOWLEDGE: AADA differentially impairs the perfusion of various organ systems, and this has important prognostic implications.
TRANSLATIONAL OUTLOOK: Further studies are needed to expand and validate anatomical patterns of AADA to better inform pre-operative risk stratification and guide clinical decisions.
GERAADA is a project of the Task Force for Aortic Surgery and Interventional Vascular Surgery of the German Society for Thoracic and Cardiovascular Surgery. The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- acute type A aortic dissection
- confidence interval
- odds ratio
- Received November 3, 2014.
- Revision received March 23, 2015.
- Accepted April 14, 2015.
- 2015 American College of Cardiology Foundation
- Bonser R.S.,
- Ranasinghe A.M.,
- Loubani M.,
- et al.
- Czerny M.,
- Krähenbühl E.,
- Reineke D.,
- et al.
- Krüger T.,
- Hoffmann I.,
- Blettner M.,
- Borger M.A.,
- Schlensak C.,
- Weigang E.,
- for the GERAADA Investigators
- Krüger T.,
- Weigang E.,
- Hoffmann I.,
- Blettner M.,
- Aebert H.,
- for the GERAADA Investigators
- Fattori R.,
- Cao P.,
- De Rango P.,
- et al.