Anterolateral thoracotomy with partial sternotomy: a feasible approach for treating the complex pathology of the aortic arch

Abstract

OBJECTIVES

Our goal was to review our surgical experiences in patients with complex pathologies of the aortic arch who have undergone anterolateral thoracotomy with a partial sternotomy (ALPS).

METHODS

From October 2019 to November 2023, a total of 23 patients underwent one-stage repairs of complex pathologies of the aortic arch through the ALPS approach. The mean age was 61.9 ± 16.7 years old. The aortic pathologies were as follows: aorta-related infection in 11 (aorto-oesophageal fistula: 4, graft infection: 6, native aortic infection: 1); aortic dissection in 9 including shaggy aorta in 2, non-dissecting aneurysm in 1, and coarctation of the aorta (CoA) in 2.

RESULTS

Eighteen patients underwent aortic replacement from either the sinotubular junction or the ascending aorta to the descending aorta; 1 patient underwent it from the aortic root to the descending aorta (redo Bentall procedure and extensive aortic arch replacement); 3 patients underwent it from the aortic arch between the left carotid artery and left subclavian artery to the descending aorta; and 1 patient underwent a descending aortic replacement. Ten patients underwent omentopexy, latissimus dorsi muscle flap installation or both procedures. The hospital mortality rate was 13.0% (3/23). The overall survival and freedom from aortic events were 73.3%±10.2% and 74.1%±10.2%, respectively, at the 3-year follow-up. There was an absence of aorta-related deaths, and no recurrent infections were identified.

CONCLUSIONS

The short-term outcomes using the ALPS approach for the treatment of complex pathologies of the aortic arch were acceptable. Further studies will be required to determine the long-term results.

Open in new tabDownload slide

INTRODUCTION

With the widespread use of thoracic endovascular aortic repair (TEVAR) for aortic arch repair, the number of TEVAR-related complications, including endoleaks, migrations, ruptures and infections, has risen. In such cases, most patients have complex pathologies of the aortic arch, including extensive thoracic arch aneurysms and aorta-related infection. The implementation of surgical interventions in patients with complex pathologies remains a challenge. A one-stage approach is highly effective for the treatment of complex pathologies of the aortic arch, although the presence of perioperative comorbidities may be a contraindication for some patients. On the other hand, a staged approach is often preferable, with safety as its priority [1–3]. However, this strategy has the disadvantages of a high cumulative mortality for the 2 procedures and the risk of rupture after the initial operation [4, 5].

In the one-stage approach, good exposure is essential for patient management for the protection of multiple organs. Several options for a one-stage approach are available, such as a median sternotomy with a left pleurotomy or a left anterolateral thoracotomy [6, 7); a clam-shell incision [8]; and a left posterolateral thoracotomy. [9] Recently, anterolateral thoracotomy with partial sternotomy (ALPS), another one-stage approach, has been employed for redo cases or complex aortic pathologies, such as aortic infection or shaggy aorta, in our institution [10, 11].

Our goal was to review our surgical experiences with the ALPS approach in patients with complex characteristics.

MATERIALS AND METHODS

Ethical statement

This study was approved by the institutional review board of the Kobe University School of Medicine on 4 January 2023 (IRB number: # B220178).

Patient selection

From October 2019 to November 2023, a total of 23 patients underwent one-stage repair of a complex pathology of the aortic arch through ALPS. The patients’ profiles were retrospectively reviewed and are shown in Table 1. Aortic events were defined as hospital death and major adverse cardiovascular events after the surgeries.

Surgical procedure

Patients were intubated using a double-lumen endotracheal tube and placed in a right semi-recumbent position. A skin incision was made from the bottom of the xiphoid to the anterior axillary line at the third intercostal space with a convex curved line (Fig. 1A-1C). The thorax was entered through the third intercostal space, and a partial lower sternotomy was performed at the same level (Fig. 1A). The left internal thoracic artery was ligated, and cardiopulmonary bypass (CPB) was established with a bicaval drain (Fig. 2). An arterial cannula was placed in the ascending aorta, and when necessary, an additional cannula was placed in the femoral artery. The left ventricle was vented through the right upper pulmonary vein. After tympanic and rectal temperatures dropped to 23°C and 30°C, respectively, cardiac arrest was achieved by antegrade cardioplegia and maintained by antegrade and retrograde cardioplegia.

Before the aortic arch was opened, the descending aorta was clamped to perfuse the lower body from the femoral arterial cannula with a flow rate of 1.0–2.0 l min⁻¹. After the aortic arch was opened, selective antegrade cerebral perfusion (SCP) from inside the aortic arch was established using a 15 Fr balloon cannula (Sumitomo Bakelite Co, Ltd, Tokyo, Japan) in the brachiocephalic artery and 12 Fr balloon cannulas in the left common carotid (LCA) and left subclavian artery (LSA) with the total SCP flow from 10 to 12 ml kg⁻¹min⁻¹ adjusted to maintain a perfusion pressure between 30 and 40 mmHg.

In cases of aorta-related infection, radical resection of all infected tissue, including the aortic wall, infected prosthetic graft or stent grafts and the oesophagus [in the case of an aorto-oesophageal fistula (AEF)], was performed. After aggressive debridement, a distal aortic anastomosis was performed on the descending aorta, and then antegrade perfusion of the lower body and rewarming were started through a branch graft. A proximal anastomosis was then performed to the ascending aorta with a sealed four-branched graft (J graft, Japan Lifeline, Tokyo or Triplex, Terumo Corporation, Tokyo, Japan), and coronary perfusion was resumed.

The arch vessels were reconstructed in the order brachiocephalic artery, LCA and LSA, completing the graft-to-graft anastomosis. Continuous regional cerebral oxygen saturation was monitored using transcranial near-infrared spectroscopy. In cases of aorta-related infection, a rifampicin-soaked, gelatin-impregnated Dacron graft (Gelweave, Vascutek, Renfrewshire, Scotland, UK) was used for aortic reconstruction because homograft availability is limited in Japan. Furthermore, the xeno-pericardial tube grafts were not utilized due to lack of evidence in our institution. Finally, omentopexy was performed on the same day as the aortic reconstruction, and/or the installation of the latissimus dorsi muscle flap was performed several days later. In 1 patient who underwent a redo Bentall procedure (case 16), the ascending-to-descending aorta was replaced as described previously after aortic root reconstruction with coronary button reimplantation using the Svensson technique [12]. In case of AEF, the infected aorta was first resected under CPB. Then, oesophagectomy with the radical resection of all infected tissue was performed just before the aortic reconstruction was done under CPB. The oesophagus was partially resected at the level of the diaphragm and closed with a stapler by the oesophageal surgeon. These patients were fed through the gastrostomy tube after oesophagectomy until oesophageal reconstruction was performed. After returning to a normal life, about 6 months after the oesophagectomy, the reconstruction of the oesophagus was planned [13].

Statistical analyses

The JMP software program for Macintosh, version 15.2.1 (SAS Institute, Cary, NC, USA) was used for the statistical analyses. All continuous data are expressed as the mean ± standard error. The Kaplan–Meier method was used to analyse the overall survival and freedom from aortic events. A P- value < 0.05 was considered significant for all statistical comparisons.

RESULTS

In this study, 100% of the patients were followed up. The mean follow-up time was 2.2 ± 2.4 years. The mean age was 61.9 ± 16.7  (range: 21–79) years. Nineteen patients were male (82.6%). The mean BMI was 23.5 ± 4.2 kg/m². Fifteen patients had previously undergone a surgical or endovascular intervention (65.2%). In the 23 patients, the aortic pathologies were as follows: 11 patients had the following aorta-related infections: AEF: 4: graft infection; 6: native aortic infection; 1: aortic dissection. Nine patients (acute type A with shaggy aorta: 1, chronic type A: 1, acute type B on chronic type A [rupture]: 1, chronic type B: 5, chronic type B with shaggy aorta: 1), non-dissecting aneurysm with shaggy aorta in 1 patient, and coarctation of the aorta (CoA) in 2 patients. The extent of aortic replacement was as follows: 18 patients underwent replacement from either the sinotubular junction or the ascending aorta to the descending aorta; 1, from the aortic root to the descending aorta and 3, from the aortic arch between the LCA and LSA to the descending aorta, and 1 patient had a descending aortic replacement (Table 1; Fig. 3). Six patients underwent omentopexy, and 3 underwent latissimus dorsi flap installation. One patient had both an omentopexy and a latissimus dorsi flap installation because of the poor omentum.

Three patients died in the hospital (13.0%). The causes of the hospital deaths were pneumonia, pancreatitis and mediastinitis, respectively. The mean operating, CPB and myocardial ischaemic times were 778 ± 176, 324 ± 89 and 166 ± 74 min, respectively. The mean number of transfusions of red cell concentrate, of fresh frozen plasma and of platelet concentrate was 20.1 ± 8.7 units, 16.9 ± 9.0 units and 20.4 ± 8.2 units, respectively. The length of stay in the postoperative intensive care unit was 7.6 ± 1.6 days. The length of stay in the hospital was 43.2 ± 44.7 days. Two patients died during follow-up of brain infarctions (2.1 months after the operation) and rectal cancer (9.7 months after the operation). The overall survival and freedom from aortic events were 73.3%± 10.2% and 74.1%±10.2%, respectively, at the 2-year follow-up (Figs. 4 and 5). Tracheostomy, re-exploration for bleeding and recurrent laryngeal nerve palsy occurred in 8 (34.7%), 2 (8.6%) and 3 (13.0%) patients, respectively. The aortic events were as follows: brain infarction in 2 patients, mediastinitis, respiratory failure due to pneumonia, detachment of the coronary button and pancreatitis in 1 patient each. Aorta-related late deaths were not observed, and there was an absence of recurrent infection.

DISCUSSION

TEVAR has dramatically changed the strategy for aortic operations because it is minimally invasive [14]. Similarly, surgical procedures using the frozen elephant trunk (FET) have also been frequently applied to open arch repair for aortic dissection and aneurysms [15]. However, with the increasing number of patients who undergo aortic repair with TEVAR or FET, the number of complications associated with stent grafts has also increased despite simultaneous technological advancements. As a result, the number of patients who require open conversion after aortic repair with TEVAR or FET has risen [16–18]. Patients who require open conversion exhibit manifestations similar to those who have complex pathology of the aortic arch.

The strategy for open conversion should be individually tailored. In our institution, a staged approach has not been applied for the primary treatment of a complex pathology of the aortic arch due to concerns about patient death and the risk of rupture in between operations. We previously reported that a one-stage approach with the appropriate protection of multiple organs achieved acceptable outcomes [7, 9].

The ALPS approach has recently been used in our institution as the optimal strategy for managing patients with complex pathology of the aortic arch or shaggy aorta. Uchida et al. reported cases that were successfully treated using the ALPS approach [19]. Of note, the ALPS approach offers a few advantages. First, it provides excellent exposure from the aortic root to the descending aorta. In this study, we were able to perform redo Bentall safely (case 16). In addition, in patients with an aorta-related infection, radical resection of all infected tissue, including the aortic wall, prosthetic graft or stent grafts and oesophagus, is important [13, 20, 21], and a staged operation is not allowed, given the associated characteristics. The ALPS approach enables surgeons to perform not only aggressive debridement but also oesophagectomy due to the excellent wide exposure. In fact, it took just 16 min for an experienced thoracic surgeon to perform an oesophagectomy in case 18 (Fig. 6A and 6B). Notably, 11 of the 23 patients in this study had aorta-related infections, including 4 patients with an AEF and 6 with a graft infection. Case 17 (AEF) died of mediastinitis. However, this instance of mediastinitis had nothing to do with the AEF. No recurrent infection was observed. We believe that the wide exposure provided by the ALPS approach contributed to the excellent outcomes.

Second, the ALPS approach can ensure the secure protection of multiple organs. The CPB can be established in an ordinary manner via a median sternotomy. Eighteen patients in our series had an arterial cannula in both the ascending aorta and femoral artery. One patient had a cannula in the axillary artery and the femoral artery. Three patients had a cannula in the femoral artery alone and 1 had it in the ascending aorta alone. Particularly in patients with a shaggy aorta, antegrade perfusion should be performed to prevent embolic events caused by retrograde perfusion from the femoral artery [21]. Fukuda et al. reported that, to protect against embolic complications in patients with a shaggy aorta, the following 3 points are important: selecting a safe antegrade arterial perfusion site, never touching the diseased aorta and debriding the anastomotic site [22]. Two major cannulation sites have been identified for safe antegrade arterial perfusion: central perfusion from the ascending aorta or aortic arch and peripheral perfusion from the axillary artery [13, 22]. From this perspective, the ALPS approach has the advantage of being able to establish safe antegrade perfusion. However, embolic complications were not avoided in 1 of the 4 patients with shaggy aorta in our study (case 21). This patient developed pancreatitis and embolic cerebral infarction and died 2.5 months after the operation. The reason case 14 developed cerebral infarction 3 days after the operation is unknown. Furthermore, myocardial protection is reliably obtained by both antegrade and retrograde cardioplegia. Left ventricular venting can be performed through the right upper pulmonary vein. In our study, both antegrade and retrograde cardioplegia were used in 15 patients. In terms of brain protection, SCP is securely established. If the ostium of an arch vessel is extremely atherosclerotic, the arteriotomy can be extended from the diseased ostium to a relatively clear distal portion [13]. In addition, we can maintain the perfusion of the lower body by clamping the descending aorta and performing perfusion from the femoral artery, thereby decreasing the side effects of hypothermia. Lower body perfusion was performed in 21 patients in this study.

Left posterolateral thoracotomy is one of the most beneficial approaches for an extensive thoracic arch aneurysm, as we previously reported [9]. However, this approach is associated with difficulties in establishing SCP and myocardial protection. Furthermore, as noted previously, it is difficult to completely protect against the embolization by retrograde perfusion from the femoral artery in patients with a shaggy aorta. Joo et al. reported that cases involving the aortic arch resulted in significantly worse outcomes with respect to open conversion following TEVAR [17] because of total circulatory arrest and the requirement of manipulation of the diseased arch. It is difficult to address these issues with a median sternotomy and a left posterolateral thoracotomy. However, the ALPS approach is a useful procedure to resolve these problems. Kouchoukos et al. described excellent outcomes with the clamshell approach, which also provides superior exposure of the entire thoracic aorta [23]. However, the clamshell approach may cause respiratory failure due to the bilateral thoracotomy. In addition, destruction of both internal thoracic arteries is needed [4]. Therefore, Furukawa et al. recommended the semi-clamshell approach, because it might be less invasive than the clamshell approach, thus reducing the risks associated with the original approach [24]. Respiratory complications must be considered with the ALPS approach as well. In our study, 1 patient received extracorporeal membrane oxygenation because of severe pneumonia 14 days after the operation; whether or not the pneumonia was related to the ALPS approach is unclear. Notably, this patient died 3 months after the operation. In contrast, Ishikawa et al. and Uchino et al. reported no significant difference regarding respiratory complications between the ALPS approach and a standard median sternotomy [25, 26]. In the ALPS approach, the ribs are not routinely disarticulated and transected, and the sternum is completely repaired (Fig. 1B and 1C). Almost all patients become free of surgery-related pain within 2 months. This result might explain the lower incidence of respiratory complications. Furthermore, with ALPS, the mid-descending aorta is easily exposed without pressure manipulation. The decreased degree of invasiveness for the left lung and the partial lower sternotomy might help stabilize the left thorax and preserve postoperative respiratory function [25]. Czerny et al. reported another approach for extensive thoracic aortic aneurysms, namely, a median sternotomy combined with a left fourth intercostal space incision [27]. They described a technique of double arterial perfusion with double-clamping of the descending thoracic aorta to avoid lower body hypothermic circulatory arrest. This concept is similar to our strategy. However, our approach avoids full sternotomy and rib transection. We believe that this approach diminishes the incidence of postoperative respiratory complications.

Although mean blood product transfusion volumes were large and the mean operating time was long, 15 cases were redo operations and 11 patients had an aortic infection, including 6 with a prosthetic graft infection and 4 with an AEF. Redo operations and management of infection are more complex, difficult and time-consuming, particularly when an AEF is present. Furthermore, 7 patients also underwent omentopexy and 8 underwent tracheostomy. A pre-existing or early tracheostomy may confer a considerable risk of deep sternal wound infection when a median sternotomy approach is used. Therefore, ALPS may be preferable in patients who already have a tracheostomy or who may need one. We do not hesitate to perform a tracheostomy in a patient who requires one when using the ALPS approach.

LIMITATIONS

Several limitations associated with the present study warrant mention. It was a single-centre, retrospective study with a small sample size. Further prospective, multicentre studies with a large sample population are therefore required to validate our findings and investigate long-term outcomes.

CONCLUSION

The ALPS approach not only enables the protection of multiple organs but also provides satisfactory exposure from the aortic root to the descending aorta. In the current series, there were many critical patients with complex pathologies. Although the hospital mortality rate remained high, there were no aorta-related late deaths and no recurrent infections. We believe that the ALPS approach is one of the beneficial options for patients with complex pathology of the aortic arch and shaggy aorta, although further improvements are required in patients with severe shaggy aorta.

FUNDING

The authors have nothing to disclose regarding commercial support.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.

CONFLICTS OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

Katsuhiro Yamanaka: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Validation; Visualization; Writing—Original draft; Writing—Review & Editing. Shota Hasegawa: Data curation; Investigation; Writing—Review & Editing. Ryo Kawabata: Investigation; Writing—Review & Editing. Hironaga Shiraki: Investigation; Writing—Review & Editing. Shunya Chomei: Investigation; Writing—Review & Editing; Taishi Inoue: Data curation: Investigation; Writing—Review & Editing. Takanori Tsujimoto: Investigation; Writing—Review & Editing. Shunsuke Miyahara: Formal analysis; Methodology; Software; Writing—Review & Editing. Hiroaki Takahashi: Supervision; Writing—Review & Editing. Kenji Okada: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Supervision; Validation; Visualization; Writing—Review & Editing.

Presented at the 35th Annual Meeting of the European Association for Cardio-Thoracic Surgery, Barcelona, Spain, 13–16 October 2021.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• AEF

  aorto-oesophageal fistula

 
• ALPS

  anterolateral thoracotomy with partial sternotomy

 
• CPB

  cardiopulmonary bypass

 
• FET

  frozen elephant trunk

 
• LCA

  left common carotid artery

 
• LSA

  left subclavian artery

 
• SCP

  selective antegrade cerebral perfusion

 
• TEVAR

  thoracic endovascular aortic repair