Short- and mid-term outcomes of surgical repair of acute type A aortic dissection and concomitant coronary artery bypass grafting or extracorporeal membrane oxygenation support

Abstract

OBJECTIVES

In this study, we aimed to retrospectively analyse the risk factors for failing to wean from CPB and short- and mid-term outcomes of surgical repair of acute type A aortic dissection (TAAD) and concomitant coronary artery bypass grafting (CABG) versus extracorporeal membrane oxygenation (ECMO) support in patients.

METHODS

Two hundred and three patients were enrolled and categorized into the simple TAAD group (n = 106) or the complex TAAD group (n = 97). Following propensity score matching, 68 patients in the complex TAAD group were distributed between the CABG subgroup (n = 34) and the ECMO subgroup (n = 34) for further analysis. Five-year survival was analysed using the Kaplan–Meier method. Multivariable logistic regression analysis was used to identify independent risk factors for failing to wean from cardiopulmonary bypass (CPB).

RESULTS

The complex TAAD group had higher in-hospital mortality than the simple TAAD group (29.9% vs 16.9%, P = 0.049). Multivariable analysis revealed that preoperative high-value cardiac troponin I, operation time, CPB time and circulation assisted time were risk factors for failing to wean from CPB (P = 0.050, 95% CI 1.000–1.105; P = 0.046, 95% CI 1.011–3.143; P = 0.044, 95% CI 1.001–1.039; P < 0.01, 95% CI 1.025–1.092, respectively). There was no significant difference in in-hospital mortality between the CABG and ECMO subgroups (5.9% vs 20.6%, P = 0.15). In contrast, the CABG subgroup demonstrated significantly improved 5-year overall survival compared with the ECMO subgroup, with a statistically significant difference (log-rank P = 0.04).

CONCLUSIONS

Preoperative high-value cardiac troponin I, operation time, CPB time and circulation assisted time were risk factors for failing to wean from CPB. For the patients who failed to wean from CPB, CABG can provide more excellent short- and mid-term outcomes than ECMO support, which was conditional on not being able to wean off CPB.

Type A aortic dissection (TAAD) is one of the most urgent surgical emergencies in cardiac surgical patients.

GRAPHICAL ABSTRACT

INTRODUCTION

Type A aortic dissection (TAAD) is one of the most urgent surgical emergencies in cardiac surgical patients. Despite significant improvements in surgical techniques, cardiopulmonary bypass (CPB) practices, cerebral protection procedures and perioperative management, the extensive TAAD lesions and longer operation time make the surgery very difficult and there is high mortality compared with other conventional cardiac surgery [1]. Data published in the International Registry of Acute Aortic Dissection (IRAD) show that the mortality rate in TAAD surgery is still approximately 20% [2]. However, some patients failed to wean from CPB due to several reasons, such as electrolyte disturbance, coronary artery dissection, coronary artery stenosis, acute myocardial infarction, severe myocardial oedema and so on. Conventional medication may be not helpful for all patients [3]. What should we do next? Increase the dosage of vasoactive drugs, continue the CPB, coronary artery bypass grafting (CABG) or extracorporeal membrane oxygenation (ECMO) support?

ECMO has become a more frequently utilized temporary measure for supporting circulation in instances of persistent cardiac or cardiopulmonary insufficiency. Previous research has shown that the rate of successful weaning from ECMO was 30–60%, while in-hospital mortality ranged from 60% to 80% [4]. Mariscalco also reported that the morbidity of successfully weaning from ECMO and mortality rates for 62 patients who receive ECMO support after TAAD surgery were 37% and 74%, respectively [5].

At present, performing CABG after TAAD surgery is not a rare event and coronary artery involvement in TAAD occurs in 7–20.7% of cases [6]. A study by Wang indicated an early mortality rate of 32.3% among TAAD patients with CABG [7]. Zhang also showed that despite a higher in-hospital mortality rate, patients undergoing total arch replacement (TAR) with CABG had similar conditional survival rates to those without coronary malperfusion if they survived discharge [8].

There are several reports in the literature about clinical outcomes of concomitant CABG or ECMO support in patients with TAAD. However, there is still a lack of comparison of short- and mid-term outcomes between these groups and the risk factors for failing to wean from CPB. The purpose of this study was to analyse the risk factors for failing to wean from CPB and clarify the indication, intraoperative decision, clinical experience and short- and mid-term outcomes of surgical repair of TAAD and concomitant CABG or ECMO support in patients who failed to wean from CPB in a high-volume single-centre experience.

PATIENTS AND METHODS

Ethics

The investigation was conducted in accordance with the Declaration of Helsinki (1975). The collection and storage of data were consistent with requirements outlined in the WMA Declaration of Taipei. The observational study was approved by the Ethics committee of Yangzhou University-affiliated Northern Jiangsu People’s Hospital (IRB approval date: 24 June 2024; approval number: 2024ky196). Written informed consent was obtained from each participant.

Patients

This is a retrospective analysis that enrolled 439 consecutive patients who underwent emergency surgery for TAAD in a single institution between January 2015 and October 2019. Two hundred and three patients were included in this retrospective study. All included patients were divided into the simple TAAD group (n = 106) or the complex TAAD group (n = 97), based on weaning from CPB. The complex TAAD group comprised the CABG subgroup (n = 53) and the ECMO subgroup (n = 44). After 1:1 propensity score matching, 68 patients were distributed between the CABG subgroup (n = 34) and the ECMO subgroup (n = 34) for subsequent analysis. In all patients, the need for aortic dissection was diagnosed with the aid of computerized tomography before surgical treatment. Clinical variables included basic characteristics, perioperative data, in-hospital and follow-up outcome. Follow-up data were obtained through clinical interviews and telephone; the median follow-up time was 57 months for the CABG group and 35 months for the ECMO group.

The inclusion and exclusion criteria were as follows. The inclusion criteria were: (i) the patients were diagnosed as needing TAAD by preoperative aortic computed tomography angiography (CTA) and underwent TAR with frozen elephant trunk (FET) implantation. The exclusion criteria were: (i) advanced age, over 80 years; (ii) patients who had suffered irreversible brain damage prior to the operation; (iii) severe hepatic, renal and other functional disorders who were unable to tolerate the operation; (iv) surgical repair of TAAD during the operation combined with CABG and ECMO; (v) patients with postoperative ECMO support applied due to acute respiratory distress syndrome; and (vi) severe haematological diseases.

Surgical procedure

All procedures were carried out by a median sternotomy after general anaesthesia. We preferred to establish the CPB through the innominate artery and right atrium, and sometimes the right axillary artery or femoral artery. The left heart was vented through the right superior pulmonary vein to avoid ventricular dilation during the operation. Once ventricular fibrillation was identified, a cross-clamp was placed across the distal ascending aorta, which was then incised longitudinally at the distal sinus tubular junction. Myocardial protection was achieved with a cold blood cardioplegic solution. In general, we first trimmed the aorta at 1 cm above the sinotubular junction, and three 5–0 Prolene sutures were used for intermittent fixation of three junctions of the aortic valve to determine the aortic regurgitation, as well as injecting gel into the aortic root entrapment to eliminate the entrapment, and using a 6-mm-wide felt sheet intermittent suture in the proximal periphery of the aorta to facilitate the successive sutures to anastomose the proximal end of the artificial vessel. When the nasal temperature dropped to about 20–22°C, the extracorporeal circulation was stopped, the clamp was opened, and cerebral protection was performed by the innominate artery and the left common carotid artery, with a flow rate of 6–12 ml/kg/min. A FET was placed within the descending aorta. The lower body circulation was returned to normal after completion of the consecutive sutures at the distal end of the aortic arch. We reconstructed the left common carotid artery to achieve bilateral perfusion during the heart rewarming, then reconstructed the innominate and subclavian arteries. Finally, proximal anastomosis of the saphenous vein graft (SVG) was completed. At this point, the main surgical steps were completed. When failing to wean off CPB, we first determined whether there were indications of CABG. If there is severe oedema of the heart and a high dose of vasoactive drugs or not weaning off CPB after CABG, we prefer ECMO support. In brief, if it was not efficient after CABG, we chose ECMO support.

Statistical analysis and follow-up

Categorical variables are represented as frequency distributions and single percentages. Values of continuous variables are expressed as mean ± standard deviation, or median. Normally distributed continuous variables were compared using T-test, non-normally distributed continuous variables using Mann–Whitney U-test, and categorical variables using Fisher’s exact test. The preoperative and surgical variables were first tested using univariable logistic regression analysis. The significant variables in the univariable logistic regression analysis were further analysed via the multivariable logistic regression analysis to identify the independent risk factors for failing to wean from CPB. Propensity score matching was used to match the baseline characteristics between CABG and ECMO subgroups and survival analysis was conducted using the Kaplan–Meier method with the log-rank test. All statistical tests were two-sided. Results were considered statistically significant at a level of P < 0.05. The entire analytical process was executed using IBM SPSS statistical software version 21.0.

Surviving patients were followed up in our outpatient department or by phone to evaluate their clinical status.

RESULTS

Between January 2015 and October 2019, of the 439 TAAD patients, 203 met the inclusion criteria and were assigned to the study. All patients underwent complete protocol-guided follow-up. Patient selection and outcomes are illustrated in Fig. 1.

Figure 1:

The enrolment, allocation and follow-up of TAAD patients who underwent surgical repair of acute TAAD with concomitant CABG or ECMO support.

Baseline and demographic characteristics, comorbidities, laboratory test results and clinical manifestations of patients are shown in Supplementary Table S1. Age, gender, the presence of comorbidities, and laboratory test results were not significantly different in any of the patients. Moreover, in the complex TAAD groups, cardiac troponin I (cTnI) values were generally higher than in the simple TAAD group, and statistically significant (P < 0.01).

Supplementary Table S2 provides detailed procedural characteristics of the two groups. The operation time, CPB time, circulation assisted time, 24-h postoperative traffic diversion, delayed breast closure and continuous renal replacement therapy in the simple TAAD group were lower than in the complex TAAD groups, and were statistically different between the two groups (P < 0.01, P < 0.01, P < 0.01, P < 0.01, P = 0.03, P < 0.01, respectively). Meanwhile, we also found higher morbidity of low cardiac output syndrome and in-hospital mortality in the complex TAAD groups than in the simple TAAD group, which was statistically significant (P = 0.022, P = 0.049) (Supplementary Table S3).

Among preoperative and intraoperative variables, preoperative cTnI, operation time, CPB time and circulation assisted time were significantly different between complex TAAD groups and the simple TAAD group (P < 0.05) in univariable analysis. Using multivariable logistic regression, preoperative cTnI (P = 0.05), operation time (P = 0.046), CPB time (P = 0.044) and circulation assisted time (P < 0.01) were found to be independent predictors of failing to wean from CPB (Supplementary Table S4).

Intraoperatively, we adopted the principle of prioritizing the aortic root. As soon as we found coronary artery involvement and requirement for CABG, we immediately took the saphenous vein as a bypass vessel and performed myocardial arrest fluid perfusion through the saphenous vein. The details of CABG and early patency are shown in Supplementary Table S5. At follow-up, there were five right coronary artery bridge occlusions, one left anterior descending bridge vessel occlusion, and one right coronary artery and left anterior descending bridge vessel occlusion, whereas the surviving patients perhaps survived because of the presence of collateral circulation or other vascular retrograde perfusion.

As shown in Supplementary Table S6, after propensity score matching, 68 patients with concomitant CABG (n = 34) or ECMO support (n = 34) were enrolled. We then conducted further analysis and found that operation time, CPB time, circulatory arrest time, circulatory assisted time, 24-h postoperative drainage, delayed breast closure and continuous renal replacement therapy were significantly different between the CABG and ECMO subgroups (P < 0.01, P < 0.01, P = 0.01, P < 0.01, P < 0.01, P = 0.011, P = 0.011, respectively).

Five-year follow-up mortality and morbidity after discharge are shown in Supplementary Table S7. In the CABG subgroup, there were four sudden cardiac deaths, one rupture of an abdominal aortic aneurysm and one massive cerebral haemorrhage. The ECMO subgroup had two ruptures of an abdominal aortic aneurysm, two cases of non-cardiovascular mortality, one chronic renal failure and one loss to follow-up. Regarding complications, the CABG subgroup included two with severe tricuspid regurgitation, while there was none in the ECMO subgroup. In the CABG subgroup, reintervention was required in two cases due to a left coronary artery ostium tear, one haematuria from excessive graft curvature in the ascending aorta, and one severe tricuspid regurgitation. In comparison, the ECMO subgroup had one case of left coronary ostium tear.

Figure 2 summarizes 5-year follow-up cumulative survival after surgery. Survival curves (including in-hospital deaths) show a statistically significant difference between the CABG subgroup and the ECMO subgroup (log-rank test P = 0.04). Median survival time was 57 months for the CABG subgroup and 35 months for the ECMO subgroup.

Figure 2:

Five-year overall cumulative survival for 68 patients in the CABG and ECMO subgroups with significant different risk profiles: Kaplan–Meier estimation.

DISCUSSION

Owing to the advancements of diagnostic tools, surgical technique and postoperative management, the outcome of aortic repair for TAAD has improved, and hospital mortality in patients with TAAD after aortic repair had decreased to 11% in a recent report [9]. The mortality could be higher for complex TAAD patients who failed to wean from CPB. In our study, we systematically summarized the characteristics, surgical strategies and prognosis of patients with surgical repair of TAAD. We found that preoperative cTnI, operation time, CPB time and circulation assisted time were independent predictors of failing to wean from CPB, by logistic regression. Meanwhile, for complex patients who failed to wean from CPB, the in-hospital mortality was significantly higher than for the simple TAAD patients (P = 0.049). Kazui reported that CPB time of longer than 300 min is an independent determinant of in-hospital mortality in TAR [10]. According to our experience, the main manifestations of failing to wean from CPB were single or double ventricular pulse weakness, low blood pressure, high pulmonary artery pressure and central venous pressure. The main reasons for failing to wean from CPB were: (i) low cardiac output syndrome, the most common, and which has three causes: (a) severe myocardial ischaemia caused by the coronary artery involved and ischaemia-reperfusion injury; (b) inadequate myocardial protection; and (c) long-term chronic coronary atherosclerosis; (ii) severe myocardial oedema, and we often used delayed closure of the chest or opened both sides of the pleura, as well as incising the pericardium in parallel with the phrenic nerve to give the heart a large enough space to beat; if not possible, we could trim the ends of two 20 ml syringes into gear-like grooves, prop the sternum and suture the skin; and (iii) left heart failure caused by long-term aortic regurgitation (Marfan syndrome, aortic root aneurysm, bicuspid aortic valve malformation).

Owing to no preoperative coronary CTA or coronary angiography being performed at our institution, it makes it difficult to accurately determine whether a TAAD patient has concomitant coronary artery disease preoperatively. Therefore, when we encountered that, we needed to quickly and accurately determine whether to choose CABG or ECMO support. According to our surgical experiences in recent years, we have systematically summarized the indications for combining CABG or ECMO support during TAAD aortic repair. Indications for CABG are: (i) weakness of single or double ventricle beat; (ii) severe involvement of the coronary arteries, especially Neri type B and C; (iii) severe stenotic lesions of the coronary arteries themselves or intraoperative detection of significant atheromatous plaques; and (iv) varying epicardium temperatures or purplish colour, and ST-segment changes. Indications for ECMO are: (i) high dosage of vasoactive drugs, severe myocardial oedema with cardiac incompetence; (ii) severe coronary artery involvement or other reasons for being unable to perform CABG; and (iii) unexplained circulatory and respiratory failure.

For patients who failed to wean from CPB, our treatment strategy underwent a shift from intraoperative ECMO support and delayed closure of the chest to a greater preference for prioritizing CABG, owing to it being a relatively routine and simple procedure. The view that concomitant CABG did not increase mortality and morbidity because of its routine nature and simplicity is supported by the study of Takashima [11]. However, the research by Mohammed et al. showed a different result, the mortality in the CABG subgroup being 51.2% among 41 patients who underwent concomitant CABG in their study [12]. In our study, we found that in-hospital mortality was higher in the ECMO subgroup than in the CABG subgroup but without significant differences (20.6% vs 5.9%, P = 0.15). As reported by Lin et al., postoperative ECMO requirement predicted an elevated risk of in-hospital death among TAAD patients, postoperative ECMO being associated with an in-hospital mortality rate exceeding 60% [13]. Loforte also reported that ECMO has been associated with less favourable outcomes, with 1-month survival rates lower than 40% in most studies [14]. The complexity of the cardiac procedure, long cross-clamp time and long CPB time were also associated with higher hospital mortality [15]. In our study, the main causes of premature death in this series were multi-organ failure due to low cardiac output syndrome (12.8%), followed by irreversible brain injury (5.9%) and severe pneumonia (4.4%). For patients who failed to wean from CPB, we adopted intraoperative ECMO support and gauze tamponade haemostasis for delayed closure of the chest previously, but the 24-h postoperative drainage was still higher than in the CABG group (P < 0.01). On the one hand, factors such as preoperative depletion of coagulation factors, along with intraoperative hypothermia, collectively contribute to compromised postoperative haemostatic capabilities [16]. Conversely, anticoagulation during ECMO may further exacerbate bleeding tendencies. Furthermore, most of these patients died within a short period of time because of postoperative low cardiac output syndrome, which explains the lack of statistically significant difference in ICU stay time between the ECMO group and the CABG group (P = 0.492).

ECMO has been widely employed in the treatment of heart and lung failure caused by many kinds of disease [17]. To some extent, it could reduce the burden on the heart, but perhaps it has not addressed the fundamental issue. However, CABG is a relatively routine and simple procedure and the great SVG is the main graft for CABG in TAAD because of its convenience and rapid access. Furthermore, right CABG dominates in this group of patients. This may be related to the specific anatomical location of the right coronary artery, which is located at the anterior wall of the ascending aorta, subjected to greater flow shear, and lacks support from the pulmonary artery. That anatomic relationship may increase the risk of damage or lesions to the right coronary artery. Postoperative follow-up of patients showed that no serious coronary artery stenosis was found in the ECMO group by coronary CTA, which to a certain extent proved the correctness of our choice of intraoperative ECMO support. However, we found that the 5-year survival rate in the CABG subgroup was still higher than in the ECMO subgroup and with significant differences, as shown in Fig. 2. The main cause of death in the CABG subgroup during follow-up was sudden cardiac death. Autopsy of two patients revealed occlusion of the bridge vessel, which may be related to the use of SVGs as the bridge vessel. Chang et al. reported that vein graft onto arteries without atherosclerosis still had a high occlusion rate. The overall freedom from occlusion of vein graft at 1, 5 and 10 years postoperatively was 87.5%, 70.0% and 28.0%, respectively [18]. In the long term, it may be better to apply the internal mammary artery [19].

In terms of postoperative management of TAAD surgical repair combined with CABG or ECMO support, the main concerns are the prevention and management of complications. For CABG, postoperative complications that may be encountered include, but are not limited, to low cardiac output syndrome, renal dysfunction, major neurological events, pneumonia and perioperative myocardial infarction. However, in some patients with very high preoperative troponin values, intraoperative coronary artery involvement was often found to be so severe that combined CABG was required. Although there was a possibility of myocardial ischaemia-reperfusion injury and ECMO support among 382 patients undergoing emergency surgery for TAAD, Mohammed et al. reported that 10.7% also received CABG during TAAD repair, with an in-hospital mortality rate of 51.2% [12]. We should not rely too heavily on the downward trend in troponin levels as a basis for the timing of surgery. On the contrary, it is particularly important to operate as early as possible in order to ameliorate myocardial ischaemia. In a study of 75 acute TAAD patients who had coronary artery dissection, Imoto and colleagues suggested that concomitant CABG may prevent myocardial ischaemia or cardiac-related events during long-term follow-up [20]. In contrast, there are many complications for patients with ECMO support, such as haemorrhage, followed by osteofascial compartment syndrome, renal failure and even multiple organ failure. Vaquer et al. noted that ECMO is associated with a substantial bleeding risk, up to 29%, including a 10% chance of significant bleeding and a 4–10% chance of intracranial haemorrhage [21]. As ECMO patients in our centre receive early enteral nutrition after being removed from a ventilator or tracheotomy, the morbidity of postoperative cerebral haemorrhage is 5.9%. According to our experience, after haemodynamic stabilization, ventilator-assisted respiration is discharged as early as possible, and enteral nutrition is administered, which avoids ventilator-associated pneumonia, reduces the risk of haemorrhage, and determines the recovery of neurological function in the postoperative period. The implications of this complication are devastating as the ELSO registry data report that only 26% of ECMO patients who develop intracranial haemorrhage survive to discharge [22], so early cranial decompression is needed once intracranial haemorrhage is diagnosed.

In our series, the femoral artery served as an arterial cannulation site for the ECMO arterial cannula. Femoral cannulation is easier and faster to perform when the patient is suffering from an unstable haemodynamic condition. The cannula can also have an impact known as downstream compression, restricting blood flow beneath the point of its insertion [23]. The femoral artery should be palpated before intubation to avoid severe calcification. The purse-string should be in the longitudinal direction, and as small as possible, in order to avoid stenosis of the artery after cannula removal. Finally, a 10 Fr sheath tube was inserted below the femoral artery cannula to provide lower limb blood flow. The flow velocity of the posterior tibial artery should be greater than 10 cm/s for adults by Doppler ultrasonography. Antegrade flow of intrathoracic aortic cannulation has been suggested to improve myocardial recovery compared to peripheral cannulation. Schiller et al.’s experiments on healthy pigs revealed that both central and peripheral venoarterial ECMO significantly enlarged left ventricular volumes and consistently impaired left ventricular contractile performance [24]. Currently, we favour inserting arterial cannulas through the axillary artery for antegrade perfusion. However, Frenckner et al. noted that axillary artery cannulation is more technically challenging than femoral artery cannulation and poses a risk of hyperperfusion to the ipsilateral arm and brain [25].

However, it is worth noting that our study is a single-centre-based retrospective study and therefore has some limitations. In particular, the analysis was restricted to the subset of patients not weaning off CPB, which does not allow for assessment of the better method overall, and the relatively small number of patients with TAAD requiring CABG or ECMO support may have resulted in compromised homogeneity of the study results and potential risk of bias. On the one hand, due to the relatively low proportion of TAAD patients requiring CABG or ECMO support, this may lead to a reduction in statistical power and make certain potentially important differences undetectable. On the other hand, some patients may be excluded from the study due to severe disease or other complications, which may cause our sample to not fully reflect the true situation of patients with TAAD. In the future, we will consider that multiple regression, propensity-matched analysis and multicentre controlled studies may provide more evidence to support our data. In addition, as surgical techniques, anaesthesia tools and the level of perioperative care continue to advance, they will undoubtedly have a significant impact on surgical outcomes at different times.

CONCLUSIONS

In conclusion, in our study, preoperative high-value cTnI, operation time, CPB time and circulation assisted time were risk factors for failing to wean off CPB. For the patients who failed to wean off CPB, CABG can provide more excellent short- and mid-term outcomes than ECMO support, which was conditional on not being able to wean off CPB.

Glossary

ABBREVIATIONS

CABG

Coronary artery bypass grafting

CPB

Cardiopulmonary bypass

cTnI

Cardiac troponin I

ECMO

Extracorporeal membrane oxygenation

FET

Frozen elephant trunk

IRAD

International Registry of Acute Aortic Dissection

SVG

Saphenous vein graft

TAAD

Type A aortic dissection

TAR

Total arch replacement

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

FUNDING

None declared.

CONFLICT OF INTEREST

None declared.

DATA AVAILABILITY

Y.S.S. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Author contributions

Dong Zhang: Writing—review & editing. Gui jun Zhu: Writing—review & editing. Ming jun Gao: Supervision. Xiang yang Wei: Conceptualization. Zhe Yan: Supervision. Bin Li: Supervision. Xing peng Chen: Supervision. Xiao lin Wang: Conceptualization. Yu sheng Shu: Writing—review & editing.

Reviewer information

Interdisciplinary CardioVascular and Thoracic Surgery thanks Ahmed Ahmed, Mohammadhosein Akhlaghpasand and the other anonymous reviewer(s) for their contribution to the peer review process of this article.