Contemporary Results Following Surgical Repair of Acute Type A Aortic Dissection (AAAD): A Single Centre Experience

ORIGINAL ARTICLE

Original Article

Contemporary Results Following Surgical Repair of Acute Type A Aortic Dissection (AAAD): A Single Centre Experience Andrew J.M. Campbell-Lloyd, MBBS a , Julie Mundy, FRACS a , Nigel Pinto, MBBS a , Annabelle Wood, RN a , Elaine Beller b , Stephen Strahan, MBBS a and Pallav Shah, FRACS a,∗ a

Department of Cardiac Surgery, Princess Alexandra Hospital, Woolloongabba, Brisbane, QLD 4120, Australia b University of Queensland, School of Population Health, Brisbane, Australia

Objectives: The study aims to define predictors of neurological dysfunction, 30-day mortality, long-term survival and quality of life following repair of acute type A aortic dissection (AAAD). Methods: Between 2000 and 2008, 65 patients underwent repair of AAAD. Sixty-four pre-, intra- and post-operative variables were studied. Mean follow-up was 26.6 months. Results: The mean age was 61 years; 60% were male and five had Marfan’s syndrome. At presentation, ischaemic ECG changes were seen in 45%, malperfusion syndrome in 59%, moderate–severe aortic regurgitation in 48% and tamponade in 16%. EF was <40% in 17%. There was a delay of >12 hours between diagnosis and operation in 64%. Axillary cannulation was performed in 37%. Cerebral protection was by hypothermic arrest (HCA) alone (19%), HCA with retrograde cerebral perfusion (RCP) (11%), or HCA with antegrade cerebral perfusion (ACP) (46%). The procedure was performed on crossclamp in 24%. Full arch replacement was performed in 14% and concomitant coronary artery grafting was performed in 11%. Post-operative neurological dysfunction was present in 33.8%. The only significant predictor of poor neurological outcome was full arch replacement (p = 0.04) on univariate analysis. In-hospital OR 30 mortality was 23.53%. Significant predictors of mortality were low ejection fraction (p = 0.017) and post-operative renal failure (p = 0.012). Long-term survival was 70% at two years, 50% at five years and 25% at nine years. Functional outcomes and long-term quality of life were assessed in 69% of patients who were alive at last follow-up. Ninety percent of patients reported minimal limitation on functional scores. Quality of life was assessed using the EQ-5D questionnaire. Forty-eight percent of patients recorded full health with an overall mean index of 0.854 (where the best possible score is 1) using the US preference weighted index score. Conclusions: Discharged patients have reasonable long-term survival and good quality of life. (Heart, Lung and Circulation 2010;19:665–672) © 2010 Published by Elsevier Inc on behalf of Australasian Society of Cardiac and Thoracic Surgeons and the Cardiac Society of Australia and New Zealand. Keywords. Aortic aneurysm; Outcome; Mortality; Prognosis

Introduction

S

tanford type A acute aortic dissection (AAAD), encompassing DeBakey types I and II, remains a condition associated with significant morbidity and mortality despite recent advances in surgical treatment and diagnosis. The International Registry of Acute Aortic Dissection (IRAD) recently reported an overall in-hospital surgical mortality of 23.9% based on a combined series of 682 patients [1].

Received 15 September 2009; received in revised form 4 April 2010; accepted 3 May 2010; available online 12 June 2010 ∗

Corresponding author. Tel.: +61 732407675; fax: +61 7 32406954. E-mail address: pallav [email protected] (P. Shah).

The decrease in operative mortality in the last 30 years coincides with the advent of cerebral protection strategies. There is a developing body of literature to suggest that recent changes in technique will improve outcomes, however there is insufficient data to support such assertions at this time. Current data does suggests that a range of pre-operative variables, more so than intra-operative variables [1–6], influence short-term morbidity and mortality, however there remains a paucity of data with respect to longterm survival and outcomes following surgical repair of AAAD. We aim to define the predictors of post-operative neurological dysfunction and mortality, and we aim to improve the understanding of long-term outcomes and quality of life after repair of AAAD.

© 2010 Published by Elsevier Inc on behalf of Australasian Society of Cardiac and Thoracic Surgeons and the Cardiac Society of Australia and New Zealand.

1443-9506/04/$36.00 doi:10.1016/j.hlc.2010.05.009

ORIGINAL ARTICLE

666

Campbell-Lloyd et al. Outcomes of type A dissection

Heart, Lung and Circulation 2010;19:665–672

Methods Patients Sixty-eight patients underwent emergent repair of type A aortic dissection between January 2000 and December 2008. Complete data was available in 65 patients. Data was prospectively collected and retrospectively analysed. Patients were excluded if they presented with chronic type A dissection. The mean age at presentation was 61.15 ± 11.44 years (range 24–80), 60% of patients were male, and five patients had Marfan’s syndrome. There were 12 patients over the age of 70. A delay of greater than 12 hours between diagnosis and operation was present in 35.7% of cases. At presentation,

30% of patients were clinically shocked, 3% required CPR and 59% of patients had documented malperfusion syndromes. Pre-operative aortic regurgitation (AR) was seen in 81% of patients, with moderate to severe AR seen in 48%. Tamponade was seen on echo in 16%. A pre-operative ejection fraction of <40% was seen in 17% of patients.

Variables and Definitions A range of 63 pre-, intra- and post-operative variables were collected for each patient (see Tables 1–3). Malperfusion syndromes were defined by clinical evidence of underperfusion of a particular organ system. Malperfusion syndromes included: neurological dysfunction (e.g. hemiparesis) including blindness; ECG evidence of myocardial ischaemia; acute renal failure as indicated

Table 1. Pre-operative Variables (Variables Not in Table: Diagnosis Modality, Proximal Extent of Dissection, Distal Extent of Dissection). Numbers Represent Percentages, Except Where Indicated. Variable

Overall

30-Day survival

30-Day mortality

No Neurology

Neurology

Mean age at presentation (years) Male Body mass index >25 Time from onset to diagnosis >6 h Shock Cardiopulmonary resuscitation Tamponade New electrocardiogram findings Other malperfusion syndrome Mean ejection fraction (%) Creatinine Previous cardiac operation Atrial fibrillation Previous myocardial infarction Hypertension Marfan’s syndrome Giant cell arteritis Smoker Pre-operative anticoagulation Aortic regurgitation on echo Time to operation >12 h

61 (24–80) 60 68 35 30 3 15 42 57 56 121 6 5 11 61 8 2 56 30 68 36

60 (24–80) 57 66 40 26 2 9 42 53 58 120 4 4 8 63 11 2 59 31 83 41

63 (40–78) 69 73 20 44 6 38 73 75 47 122 13 6 19 56 0 0 50 25 77 20

60 (24–78) 53 62 47 27 3 11 59 57 58 118 0 3 5 63 8 0 55 30 80 44

60 (33–80) 68 78 16 27 0 14 24 55 51 127 9 9 18 59 10 5 60 30 94 25

Table 2. Intra-operative Variables (Variables Not in Table: Freestyle Aortic Root Bioprosthesis [Medtronic, Minneapolis, MN], Intra-operative TOE (Transoesophageal Echocardiogram), Blood Products Administered (Packed Red Cells, Platelets, Cryoprecipitate, Fresh Frozen Plasma, Novoseven [Novo Nordisk, Denmark])). Numbers Represent Percentages, Except Where Indicated. Variable

Overall

30-Day survival

30-Day mortality

No neurology

Neurology

AVR Valve resuspension Root Hemi-arch Full arch Coronary reimplant Coronary artery bypass graft Glue Antigrade cerebral perfusion Retrograde cerebral perfusion Axillary cannulation Femoral cannulation Deep hypothermic circulatory arrest No circulatory arrest Aprotinin Mean pump time (minutes) Mean cross-clamp (minutes) Mean arrest time

34 17 32 63 14 25 11 75 46 11 35 65 52 23 60 206 115 31

31 12 31 63 12 24 8 73 48 13 39 61 49 24 68 187 107 28

44 31 36 60 20 25 19 80 40 7 21 79 63 13 56 261 135 37

26 21 25 62 8 21 8 73 43 5 43 57 39 27 61 192 106 27

41 14 36 64 18 27 14 77 48 19 19 81 64 21 75 216 124 38

Campbell-Lloyd et al. Outcomes of type A dissection

667

Table 3. Post-operative (first 30 post-operative days) variables (variables not included in table: Late Reoperation [After 30 days]). Numbers represent percentages, except where indicated. Variable

Overall

30-Day survival

30-Day mortality

No neurology

Neurology

Early reoperation Length of intensive care unit stay >5 days Ventilation time >24 h Neurological dysfunction Post-op acute renal failure Mean post-op creatinine Post-op other organ malperfusion Post-op atrial fibrillation Post-op cardiogenic shock Post-op myocardial infarction Infection 30-Day mortality Mean total length of stay (days)

42 59 83 37 32 182 12 28 7 3 14 24 19.08

43 63 82 38 21 153 7 33 4 2 17 0 24.15

44 46 85 33 75 281 33 8 17 8 0 100 4.81

50 58 82 0 32 178 5 29 8 3 11 21 17.57

32 65 84 100 30 185 21 26 5 5 20 18 25.89

by raised creatinine and oliguria; limb malperfusion with loss of pulses, isolated limb weakness or numbness; paraplegia; and mesenteric ischaemia with raised lactate, altered liver function tests or an acute surgical abdomen. Post-operatively, in-hospital or 30 day mortality and neurological dysfunction were the two outcomes of interest. We chose to limit post-operative mortality to 30 days to ensure consistency with the majority of studies in the literature and includes patients who died in operating room. Neurological dysfunction was defined as either transient or permanent neurological deficiency as diagnosed by a neurologist and supported by imaging.

Aprotinin (Trasylol, Bayer Pharmaceuticals) was used in 60% to assist with haemostasis following repair.

Follow-up Long-term data was obtained by a combination of chart review and direct patient contact/telephone interview. Mean follow-up was 28.2 months in patients discharged from hospital. Overall, complete follow-up was available to 12 months or death in 88.2%. The EQ-5D questionnaire was used to assess quality of life (QOL) in 69% of patients who were alive at last follow-up.

Statistical Analysis Operative Details Pre-operatively, radial artery and central venous lines are placed, often with placement of a pulmonary artery catheter. Intra-operative transoesophageal echocardiography has been used increasingly since early in this decade cases. BIS (Aspect Medical Systems, Norwood, MA) monitoring is now used routinely for all patients. Operative technique has varied between surgeons, and in some instances, technique for a single surgeon has varied with time. In particular, surgeon preference has meant that both femoral [65%] and axillary cannulation [35%] have been used. Cerebral protection has been by means of deep hypothermic circulatory arrest (DHCA) [19%], isolated cerebral perfusion (both antegrade [46%] and retrograde [11%]) and also by performance of the procedure on cross-clamp [24%]. Mean pump time was 206 min, with mean circulatory arrest time of 39.64 min. In a significant proportion [22.95%] of patients, operative repair was performed without circulatory arrest. Gelatin resorcin formaldehyde (GRF) glue or more recently Bioglue (CryoLife Inc., Kennesaw, GA) was used with or without Teflon strips to reinforce the anastomoses in most instances [75%]. Valve sparing procedures have been in the minority [16.9%], with most surgeons replacing any valve [34%] felt to have been compromised by the dissection. The proximal anastomosis incorporated replacement of the root in 31.7%. The distal anastomosis entailed hemiarch replacement in 62.5%, full arch replacement in 14% of cases and distal ascending aorta in 24% operated on cross-clamp. Concomitant CABG was performed in 11%.

This study involved a retrospective analysis of data which had been collected prospectively. Clinical parameters were entered into a database at the time of the patient’s initial assessment and with follow-up data obtained by accessing patient records and direct patient contact. Results were expressed as frequencies and percentages for categorical variables and mean ± standard deviation for continuous normally distributed variables and median and interquartile range for skewed data. Binary logistic regression was used to model the association between the pre-, intra- and post-operative variables and the two outcomes of interest: 30-day mortality and neurological dysfunction. Logistic regression was performed using backward stepwise elimination and goodness of fit of the model was explored with the Hosmer–Lemeshow test and regression diagnostics were checked. The Kaplan–Meier method was used to display the survival time of patients from operation date to last follow-up or death, and to estimate median survival time. Patients’ data was censored at the date when the patient was last known to be alive. Stata/SE 9.2 (College Station, TX) software was used for all analyses. p-Values less than 0.05 were considered statistically significant and all p values reported are two-sided.

Results Between 2000 and 2008, 68 patients underwent operative repair of AAAD. The number of cases performed per

ORIGINAL ARTICLE

Heart, Lung and Circulation 2010;19:665–672

ORIGINAL ARTICLE

668

Campbell-Lloyd et al. Outcomes of type A dissection

Heart, Lung and Circulation 2010;19:665–672

Table 4. Multivariate Logistic Regression Results of Pre-operative Variables on 30-Day Mortality. Predictor

Odds Ratio

95% CI

p-Value

Tamponade Pre-operative ejection fraction

13.41 0.93a

2.11–85.29 0.88–0.98

0.006 0.011

a

For each % point increase in EF, OR for outcome of mortality = 0.93.

annum has gradually increased in the period examined, with five patients undergoing operation in 2000, and 14 and 13 cases respectively in 2007 and 2008. Complete data was available for 65 patients. The assessed pre-, intra- and post-operative variables are presented in Tables 1–3.

Early Outcomes The in-hospital or 30 day mortality in this series was 23.53%. Between 2000 and 2002, mortality was 37.5%. Mortality reduced to 20% for the period from 2002 to 2005, and has remained at that level in the current period 2005–2008. The mean age of patients has not changed since the start of the series. Of the pre-operative variables, multivariate logistic regression demonstrated that the presence of tamponade (OR 13.41, p = 0.006), and poor ejection fraction (EF) (OR 0.93 [i.e. for each % increase in EF], p = 0.011) were significant independent predictors of mortality (see Table 4). Delay to operation was not a significant predictor of mortality despite an initial hypothesis to that effect. No intra-operative variables were significant as predictors of mortality, however cardiopulmonary bypass time did approach significance (p = 0.089). Similarly, multivariate analysis of post-operative variables failed to yield significant results, although acute renal failure was significant on univariate analysis (p = 0.007). Interestingly, when we considered a model incorporating all 64 variables, multivariate analysis demonstrated that the two variables which were significant predictors of 30-day mortality were pre-operative ejection fraction (OR 0.92, p = 0.017) and post-operative ARF (OR 11.47, p = 0.012) (see Table 5). Of the 12 patients over the age of 70 in this series, three [25%] died. Causes of death were cardiac (50%), neurological (12.5%), multiple organ failure (12.5%), progression of dissection (12.5%), respiratory (6.25%), and intestinal infarction (6.25%). Post-operative neurological dysfunction was present in 36.7% of patients. Based on univariate analysis, poor neurological outcome was predicted by full arch replacement (p = 0.04). Axillary cannulation and deep hypothermic circulatory arrest were found to be of borderline significance as a negative correlate of stroke. No variables remained significant at the 5% level in the multivariate model. Since 2002, there has been an even spread within our unit of various cannulation and cerebroprotective strategies. Axillary cannulation and deep hypothermic circulatory arrest were

Figure 1. Kaplan–Meier survival graph.

found to be of borderline significance as a negative correlate of stroke.

Long-term Outcomes Long-term follow-up focussed on both survival and functional level. Mean follow-up was 28.2 months in patients who were discharged from hospital. In-hospital or 30 day mortality was observed in 23.53% (16). A further 14.7% (10) of patients died after 30 days, with one in-hospital death and nine deaths following discharge. Incomplete follow-up data was available for 14.7% (10) patients. Of these, 7.35% (5) patients had a minimum of 12 months follow-up. The last available data for one patient was at the date of discharge. For 4.41% (three), the only available data was 30-day survival. As at March 2009, 29 patients are confirmed to be alive, with a mean survival 940 days (median 713 days). A Kaplan–Meier survival graph shows survival was 70% at two years, 50% at five years and 25% at nine years (see Fig. 1). Overall mean survival after surgery was 638 days (median 262). For patients who are confirmed to have died following discharge (n = 9), mean survival was 1095 days. Late mortality data was limited, however from that obtained there were no identified instances of mortality directly associated with the initial presentation. Myocardial infarction was the cause of death in 20% (two) of those patients who died following discharge. Follow-up CT scans are available for 23 patients (79.3%). These CT

Table 5. Multivariate Logistic Regression Results for All Significant Variables on the Outcome of 30-Day Mortality. Predictor

Odds ratio

95% CI

p-Value

Pre-operative ejection fraction Post-operative acute renal failure

0.92 11.47

0.86–0.98 1.69–77.56

0.017 0.012

Campbell-Lloyd et al. Outcomes of type A dissection

669

Table 6. EQ-5D Questionnaire Results. 48.28% of Patients Report “Perfect Health”. EQ-5D category

Mobility

Personal cares

Usual activities

Pain/discomfort

Anxiety/depression

No problems (%) Some problems (%) Severe problems (%)

79 21 0

90 10 0

52 48 0

62 35 3

73 17 10

scans were done in the first six months to a year postoperatively. There are no follow-up CT scans at regular intervals planned for them. All these patients had proximal aortic tear, which was completely repaired. We have studied the CT scans at hand, but scans done so shortly after the repair only showed thrombus in the false lumen, no late aneurysms have developed and no distal reoperations have been carried out as yet in this series. We assessed the NYHA class as an indicator of cardiac function in all 29 patients known to be alive. Class I–II symptoms were reported by 26 (89.65%). For the purposes of accurate estimation of day-to-day function, the widely accepted EQ-5D has been used. The EQ-5D is a quality of life assessment based on a composite of patient self-appraisal of mobility, personal cares, usual activities, pain/discomfort and anxiety/depression. We were able to obtain complete data for the EQ-5D from 69% (29 of 42 patients who were alive at last follow-up). “Perfect” health was reported by 48% of these patients, with a mean index of 0.854 (where 1 is best possible health state and 0 is the worst possible) using the US weighted index score (see Table 6).

Discussion DeBakey et al. first described operative repair of dissecting aneurysms of the ascending aorta using homografts in 1954 [7], and subsequently described the use of braided nylon tubes for the same purpose in 1955. Type A dissections involve the ascending aorta, and are a cardiovascular emergency, requiring rapid surgical treatment. Whilst there has been improvement in the outcomes for patients with type A dissections in the last 30 years, mortality remains high even for those patients who reach medical care. Recently, the International Registry of Acute Aortic Dissections (IRAD) reported an overall in-hospital mortality of 23.9% [1]. We have reported an in-hospital mortality of 23.53% in a contemporary series. If the option is taken to treat patients medically (presumably based on the opinion that they would not survive an operation), mortality approaches 60%. Importantly, the IRAD investigators reported the prognostic potential of a range of pre- and intra-operative variables. Age greater than 70 years has been identified as a predictor of mortality [1]. However recent studies in populations greater than 70 years of age have refuted such claims [8–10], with suggestions that age should not limit a patient’s progression to operative repair. In our series, 12 patients were over the age of 70, 3 (25%) of whom died. Hypotension or shock, tamponade and pulse deficit, and ECG evidence of myocardial ischaemia or infarction were found to be important predictors of patient outcome

[1]. Such findings had been suggested in earlier reports [5,11] and more recently have been further supported [3]. It seems that malperfusion syndromes (cardiac, cerebral, iliofemoral, renal, mesenteric, innominate or spinal), indicative as they are of decreased cardiac output, are associated with poor outcome. A recent article has suggested that patients should be stabilised initially by percutaneous techniques with stenting of the true lumen. By facilitating resolution of end-organ malperfusion and allowing recovery from the reperfusion injury, delayed operative repair then results in similar outcomes to those patients who presented initially with uncomplicated dissection [12]. Such findings are distorted however by the fact that one third of patients presenting with malperfusion syndromes died of complications of malperfusion or aortic rupture prior to operation. Multivariate analysis has also supported the role of tamponade (with loss of pulses) as a predictor of short-term mortality, as is pre-operative CPR [2,4,6]. The corollary of the IRAD findings, is that normotensive patients with normal ECG tracings were significantly more likely to survive. We note that shock (43.75 vs. 25.53%), tamponade (37.5 vs. 8.51%), and new ECG findings (73.33 vs. 42.1%) were all seen more commonly in patients who died vs. those who survived. We also found that ejection fraction (EF) was a further potential predictor of mortality. The majority of our patients underwent pre-operative echocardiography, but there is no way to determine how many of these results were indicative of an acute deterioration in EF. Concomitant procedures are widely held to increase risk to patients. Coronary artery bypass grafting (CABG) was found by IRAD to increase mortality, which has been suggested elsewhere [1,13]. One presumes that poor outcome associated with CABG reflects the underlying extent of dissection and potential myocardial ischaemia in the pre-operative period, more so than the procedure itself. From our series, CABG at the time of operative repair of AAAD was performed in more than double the number of patients who died as compared to those who survived (18.75 vs. 8.16%). Interestingly, IRAD found that AVR did not have a significant influence on mortality, however this is disputed [6]. AVR was performed in 43.75% of patients who died, and only 30.61% of those who survived at our institution. The simple conclusion from IRAD was that unstable patients had nearly twice the risk compared with stable patients (30% mortality vs. 15% mortality, p < 0.0001) [1]. Whilst data supporting various risk factors for shortterm outcomes are available, this has proven less so for long-term outcomes. It has been suggested that regardless of aetiology, the long-term survival of all patients undergoing procedures on the proximal or ascending aorta is similar [13]. The IRAD investigators [14] have reported sur-

ORIGINAL ARTICLE

Heart, Lung and Circulation 2010;19:665–672

ORIGINAL ARTICLE

670

Campbell-Lloyd et al. Outcomes of type A dissection

vival of 96.1% and 90.5% at one and three years. Shiono et al. reported long-term survival in an extremely selective consecutive series. In this series, the authors report a 30-day mortality of only 6.1%, with an even more remarkable 3.2% mortality for patients operated on since 2001 [6]. However, of note, of the 185 patients who presented to their institution, only 98 proceeded to operation, suggestive of a strict and limited selection protocol for operating on such patients, which is unlikely to be consistent with practice in most centres. The authors go on to report long-term survival at one, five, and nine years of 85.6, 78.5 and 62.4%. Our long-term survival data extrapolates to a survival of 70% at two years, 50% at five years and 25% at nine years, based on an all-comers treatment policy. Whilst recognising that the all-comers policy leads to worse outcomes, we do understand that better than expected outcomes do occur in delayed presentation patients referred from other institutions, as many of these have already selected themselves as lower risk patients. It has been suggested that the long-term survival of patients with AAAD is largely out of the hands of the surgeon. Whilst operative technique and post-operative factors may influence in-hospital mortality, the only variables to influence long-term outcome appear to be preoperative variables relating to the patients’ underlying health [14,15]. Another long-term series [16] reported that most late complications and mortality associated AAAD were secondary to descending aorta related complications rather than the ascending aorta/arch or operative repair of the AAAD. Reoperation increases the risk of any cardiac surgical procedure. Proximal reoperation is performed after repair of AAAD because of aortic insufficiency (AI) where a valve is resuspended, endocarditis and graft infection, and pseudoaneurysm. Distal reoperation is most often a result of secondary dilatation. Data based on descending thoracic and thoraco-abdominal aneurysms have suggested that chronically dissected aortas pose greater risk of mortality secondary to rupture as compared to non-dissecting aneurysms [17]. It is suggested therefore that secondary dilatation of the dissected aorta with a patent false lumen poses greater risk to the patient. Risk factors for late dilatation include Marfan’s syndrome, initial diameter of the aorta and a patent false lumen. Further, an initial presentation at a younger age, dissection of the supra-aortic branches and/or malperfusion syndrome may favour secondary dilatation [18,19]. The impact of age at presentation seems self-evident, and the latter suggests that the impact of the false lumen at the time of presentation may herald late-term complications also. Most surgeons endeavour to prevent flow in the false lumen, most commonly by use of Bioglue in the current era (where previously GRF glue may have been used). However, it has recently been reported that whilst the average growth of the aorta in patients with a patent false lumen is greater, this growth is slow (<1 mm per year) in nearly 50% of these patients, with no difference in distal reoperation or late survival [20]. Our data is limited, however from that available we have found that late mortality is unrelated to the initial event (AAAD), operative repair, or sequelae (i.e. dilatation). Recent CT

Heart, Lung and Circulation 2010;19:665–672

imaging is available for 80% of our survivors. Because of the early post-operative nature of our CT scan, no appreciable change was noted in the distal pathology. Most importantly however, our long-term follow-up data demonstrates that quality of life in survivors is high. “Perfect” health (indexed score of 1) is reported by 48.28% of those surveyed. “Some” problems in only one category of the EQ-5D questionnaire are reported by 10.34% and 6.89% report some problems in two categories. Severe problems are reported in one category by a further 10.34%, and interestingly nearly all such reports are of debilitating anxiety and/or depression. Technical aspects of the operative repair of AAAD have changed in the last 20 years. In particular, aspects of cerebroprotection by total body cooling or isolated cerebral perfusion, cannulation strategies, and aggressive surgical repair have been advocated. The suggestion that it is these operative techniques [21] that have improved outcomes is contentious, given the wide range of techniques that remain in use. Early post-operative stroke is widely held to be a result of embolic phenomena or hypoperfusion. Consequently, the theoretical advantages of cerebroprotection by deep hypothermia (decreased metabolic demand) or isolated cerebral perfusion have been suggested to be protective against post-operative stroke. The use of hypothermic circulatory arrest alone was the first technique of cerebroprotection described that allowed for exploration of the aortic arch. Despite the frequency of use of this technique, there is no convincing evidence for cerebroprotection by the profound cooling of patients. Lai et al. reported a 33year series comparing open repair of the arch with HCA and operative repair of AAAD using cross-clamp, with no difference between the two groups in terms of mortality, distal aortic reoperation or neurological morbidity [22]. Bachet et al. subsequently revolutionised the concepts of cerebroprotection during surgery on the aortic arch [23,24]. The use of axillary cannulation allows for antegrade perfusion of the brain, thus prolonging the time during which circulatory arrest can be tolerated, and permitting extensive exploration and repair in the arch. Retrograde cerebral perfusion (RCP) by contrast is thought to be less physiological, however it avoids potential complications that may arise from the antegrade route such as embolic phenomena. A recently reported series [25] comparing the use of RCP against hypothermic circulatory arrest (HCA) alone identified protection against transient neurological deficit, however there was no difference in long-term outcome. Circulatory arrest does allow exploration of the arch, however from our series it is worth noting that in 24.49% of survivors, compared with 12.5% of those who died, the procedure was performed on cross-clamp with maintenance of circulation. In contrast, deep HCA was used in 48.98% of survivors vs. 62.5% of those who died. Similarly, cannulation strategies remain contentious. Whilst many studies support the use of axillary cannulation over femoral, there is a paucity of quality data that would allow for either strategy (femoral vs. axillary) to be recommended [26]. The inhomogeneity of these studies

and generally poor quality of study design limits the value of available data. Trends were noted in our data for both mortality and neurological outcome with regards cannulation strategy (seen Table 2). Axillary cannulation was more common in survivors and those with normal neurological recovery.

Campbell-Lloyd et al. Outcomes of type A dissection

[2]

[3]

Conclusion It is accepted that whilst operative mortality after repair of AAAD has fallen over time, it remains both morbid and highly fatal. This is despite a variety of changes in technique which are purported to have theoretical benefits, and also despite earlier and more accurate diagnosis by means of CT angiography. However, it is also widely reported that the number of cases presenting to emergency departments, and the number of cases progressing to operation has increased, owing to earlier diagnosis. This phenomenon is occurring in an aging population with increasing comorbidities. Whilst our study population has not demonstrated any trend towards older age at presentation, this accounts only for the last eight years. It is likely that over the last 10–20 years, such factors have become more significant. Therefore, it can be hypothesised that despite technical advances, surgeons are now operating on older, sicker patients, which may account for a less significant improvement in outcomes than may have been expected. Importantly however, the evidence points to the fact that if patients present to hospital, an attempt at operative repair is warranted. Our slightly higher mortality may be reflective of our “all-comers” policy however in those patients who survive, we have shown that quality of life is maintained, and in fact post-operative function may be as high as that premorbidly. Further we have added support to the prognostic role of cardiac tamponade, and we have suggested that preoperative echocardiography and measurement of ejection fraction may herald worse than expected outcomes. There remains further evidence required for the role of any modifiable (i.e. intra-operative) variable in determining outcome following operative repair of AAAD. Larger studies are required to more closely examine the technical aspects of treatment to come to any conclusion as to the standard of care in this challenging surgical problem.

Conflict of Interest

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

The authors have no conflicts of interest to declare.

Funding No external funding was obtained for this study.

[15]

References [1] Rampoldi V, Trimarchi S, Eagle KA, Nienaber CA, Oh JK, Bossone E, Myrmel T, Sangiorgi GM, De Vincentiis C, Cooper JV, Fang J, Smith D, Tsai T, Raghupathy A, Fattori R, Sechtem U, Deeb MG, Sundt 3rd TM, Isselbacher EM. Simple risk models to predict surgical mortality in acute type A aortic

[16]

[17]

671

dissection: the international registry of acute aortic dissection score. Ann Thorac Surg 2007;83:55–61. Bayegan K, Domanovits H, Schillinger M, Ehrlich M, Sodeck G, Laggner AN. Acute type A aortic dissection: the prognostic impact of preoperative cardiac tamponade. Eur J Cardiothorac Surg 2001;20:1194–8. Geirsson A, Szeto WY, Pochettino A, McGarvey ML, Keane MG, Woo YJ, Augoustides JG, Bavaria JE. Significance of malperfusion syndromes prior to contemporary surgical repair for acute type A dissection: outcomes and need for additional revascularizations. Eur J Cardiothorac Surg 2007;32:255–62. Hata M, Shiono M, Inoue T, Sezai A, Niino T, Funahashi M, Negishi N, Sezai Y. Preoperative cardiopulmonary resuscitation is the only predictor for operative mortality of type A acute aortic dissection: a recent 8-year experience. Ann Thorac Cardiovasc Surg 2004;10:101–5. Kazui T, Washiyama N, Bashar AH, Terada H, Suzuki T, Ohkura K, Yamashita K. Surgical outcome of acute type A aortic dissection: analysis of risk factors. Ann Thorac Surg 2002;74:75–81 [discussion 81–72]. Shiono M, Hata M, Sezai A, Negishi N, Sezai Y. Surgical results in acute type A aortic dissection. Ann Thorac Cardiovasc Surg 2005;11:29–34. Debakey M, Creech Jr O, Cooley D. Occlusive disease of the aorta and its treatment by resection and homograft replacement. Ann Surg 1954;140:290–310. Santini F, Montalbano G, Messina A, D’Onofrio A, Casali G, Viscardi F, Luciani GB, Mazzucco A. Survival and quality of life after repair of acute type A aortic dissection in patients aged 75 years and older justify intervention. Eur J Cardiothorac Surg 2006;29:386–91. Caus T, Frapier JM, Giorgi R, Aymard T, Riberi A, Albat B, Chaptal PA, Mesana T. Clinical outcome after repair of acute type A dissection in patients over 70 years-old. Eur J Cardiothorac Surg 2002;22:211–7. Shiono M, Hata M, Sezai A, Iida M, Yagi S, Negishi N. Emergency surgery for acute type A aortic dissection in octogenarians. Ann Thorac Surg 2006;82:554–9. Long SM, Tribble CG, Raymond DP, Fiser SM, Kaza AK, Kern JA, Kron IL. Preoperative shock determines outcome for acute type A aortic dissection. Ann Thorac Surg 2003;75: 520–4. Patel HJ, Williams DM, Dasika NL, Suzuki Y, Deeb GM. Operative delay for peripheral malperfusion syndrome in acute type A aortic dissection: a long-term analysis. J Thorac Cardiovasc Surg 2008;135:1288–95 [discussion 1295–1286]. Olsson C, Eriksson N, Stahle E, Thelin S. Surgical and longterm mortality in 2634 consecutive patients operated on the proximal thoracic aorta. Eur J Cardiothorac Surg 2007;31:963–9 [discussion 969]. Tsai TT, Evangelista A, Nienaber CA, Trimarchi S, Sechtem U, Fattori R, Myrmel T, Pape L, Cooper JV, Smith DE, Fang J, Isselbacher E, Eagle KA. Long-term survival in patients presenting with type A acute aortic dissection: insights from the international registry of acute aortic dissection (IRAD). Circulation 2006;114:I350–356. Tan ME, Morshuis WJ, Dossche KM, Kelder JC, Waanders FG, Schepens MA. Long-term results after 27 years of surgical treatment of acute type A aortic dissection. Ann Thorac Surg 2005;80:523–9. Suehiro K, Pritzwald-Stegmann P, West T, Kerr AR, Haydock DA. Surgery for acute type A aortic dissection a 37-year experience in Green Lane Hospital. Heart Lung Circ 2006;15:105–12. Griepp RB, Ergin MA, Galla JD, Lansman SL, McCullough JN, Nguyen KH, Klein JJ, Spielvogel D. Natural history of

ORIGINAL ARTICLE

Heart, Lung and Circulation 2010;19:665–672

ORIGINAL ARTICLE

672

[18]

[19]

[20]

[21]

Campbell-Lloyd et al. Outcomes of type A dissection

descending thoracic and thoracoabdominal aneurysms. Ann Thorac Surg 1999;67:1927–30 [discussion 1953–8]. Immer FF, Hagen U, Berdat PA, Eckstein FS, Carrel TP. Risk factors for secondary dilatation of the aorta after acute type A aortic dissection. Eur J Cardiothorac Surg 2005;27: 654–7. Geirsson A, Bavaria JE, Swarr D, Keane MG, Woo YJ, Szeto WY, Pochettino A. Fate of the residual distal and proximal aorta after acute type A dissection repair using a contemporary surgical reconstruction algorithm. Ann Thorac Surg 2007;84:1955–64 [discussion 1955–64]. Kimura N, Tanaka M, Kawahito K, Yamaguchi A, Ino T, Adachi H. Influence of patent false lumen on long-term outcome after surgery for acute type A aortic dissection. J Thorac Cardiovasc Surg 2008;136:1160–6, 1166 e1-3. Krahenbuhl ES, Immer FF, Stalder M, Englberger L, Eckstein FS, Schmidli J, Carrel TP. Technical advances improved outcome in patients undergoing surgery of the ascending aorta and/or aortic arch: Ten years experience. Eur J Cardiothorac Surg 2008;34:595–9.

Heart, Lung and Circulation 2010;19:665–672

[22] Lai DT, Robbins RC, Mitchell RS, Moore KA, Oyer PE, Shumway NE, Reitz BA, Miller DC. Does profound hypothermic circulatory arrest improve survival in patients with acute type A aortic dissection? Circulation 2002;106:I218–28. [23] Bachet J, Guilmet D, Goudot B, Dreyfus GD, Delentdecker P, Brodaty D, Dubois C. Antegrade cerebral perfusion with cold blood: a 13-year experience. Ann Thorac Surg 1999;67:1874–8 [discussion 1891–1874]. [24] Bachet J, Guilmet D, Goudot B, Termignon JL, Teodori G, Dreyfus G, Brodaty D, Dubois C, Delentdecker P. Cold cerebroplegia. A new technique of cerebral protection during operations on the transverse aortic arch. J Thorac Cardiovasc Surg 1991;102:85–93 [discussion 93–84]. [25] Zierer A, Moon MR, Melby SJ, Moazami N, Lawton JS, Kouchoukos NT, Pasque MK, Damiano Jr RJ. Impact of perfusion strategy on neurologic recovery in acute type A aortic dissection. Ann Thorac Surg 2007;83:2122–8 [discussion 2128–2129]. [26] Gulbins H, Pritisanac A, Ennker J. Axillary versus femoral cannulation for aortic surgery: enough evidence for a general recommendation? Ann Thorac Surg 2007;83:1219–24.