Total aortic arch replacement using the J-graft open stent graft for distal aortic arch aneurysm: report from two centres in Japan

Abstract

OBJECTIVES

The open-style stent graft technique has been changing the strategy for true distal arch aneurysms extending to the descending aorta. Our mid-term results of surgical repair using a J-graft open stent graft are presented.

METHODS

Between May 2015 and June 2020, 69 patients with a distal arch aneurysm (53 males, median age 74 years) underwent total arch replacement combined with J-graft open stent deployment. All 59 surviving patients were followed for a median follow-up period of 1.8 (0.6–3.6) years.

RESULTS

Antegrade deployment was successfully performed in all patients without any difficulties. The deployed device was securely fixed at the target area, and it initiated thrombus formation. The diameter of the excluded aneurysm was decreased in 54 patients (91.5%) during the follow-up period. There were no type I endoleaks, but there were 3 type II endoleaks; 2 of the 3 type II endoleaks disappeared during the follow-up period. Additional endovascular operations were performed in 3 patients. There were 10 in-hospital deaths (14.5%), and the incidences of stroke, spinal cord injury and distal embolism were 11.6%, 5.8% and 2.9%, respectively. The 1- and 3-year survival rates were 84.8% and 79.4%, respectively, and the 1- and 3-year freedom from reintervention rates were 97.2% and 81.3%, respectively.

CONCLUSIONS

The J-graft open stent graft was easy to deploy, and it could shift the distal anastomosis to a more proximal side. The mid-term performance of this device was good. It has the potential to provide one-stage repair.

Abstract

Aortic arch aneurysms involving the proximal descending aorta remain among the great challenges in cardiovascular surgery.

INTRODUCTION

Aortic arch aneurysms involving the proximal descending aorta remain among the great challenges in cardiovascular surgery. They may require more than one surgical intervention before completion. To reduce the surgical invasiveness and improve the clinical outcomes, open-style stent graft techniques have been gaining worldwide attention [1–8]. The Japanese-made J-graft open stent graft (JGOS; Japan Lifeline Co., Tokyo, Japan) has been commercially available since 2014; it was developed as a device exclusively for use in open-style deployment without increasing technical complexity and was designed to enhance trackability to the aortic wall due to its oval-shaped, nitinol wire-woven structure [3–6]. Although the effect of this device has been widely recognized in cases of acute type A aortic dissection, its utility for true distal aortic arch aneurysms involving the proximal descending aorta is less clear [3–6]. The present study was designed to evaluate the safety, efficacy and mid-term results of total arch replacement combined with the JGOS for true distal aortic arch aneurysms extending to the proximal descending aorta.

MATERIALS AND METHODS

Aneurysmal definition and patient selection

Aneurysms were evaluated by preoperative contrast-enhanced computed tomography (CT). In the present study, the recommendation for avoiding spinal cord injury related to open-style stent graft deployment was followed; the stent end should be above the eighth thoracic vertebral level (Th) [1, 5]. Therefore, the patients were selected using the following criteria. Only patients with atherosclerotic aortic arch aneurysms were enrolled, and dissecting, infectious, inflammatory or traumatic aneurysms were excluded. The aneurysms originated from the ascending aorta or the aortic arch, extending to the proximal descending aorta within the Th 7 level. Patients with descending aortic aneurysms that extended beyond the Th 8 level were excluded. A shaggy aorta was defined as follows: (i) thrombus was measured in non-aneurysmal aortic segments (<40 mm); (ii) atheroma thickness ≥5 mm; and (iii) irregular atheroma surface showing finger-like projections. If the patient had all 3 findings, the diagnosis of shaggy aorta was confirmed [9].

Methods

The 2 Institutional Review Boards approved this study and waived the requirement for individual consent because of its retrospective design. Between September 2015 and June 2020, a total of 69 consecutive patients underwent total arch replacement combined with antegrade JGOS deployment at the National Defense Medical College Hospital and Saiseikai Utsunomiya Hospital. This study was a two-centre, non-blinded study. The patients’ records were reviewed retrospectively. Clinical follow-up ended on 30 June 2020 and was 100% complete. The primary end point was the composite incidence of in-hospital death and neurological complications, such as cerebral infarction and spinal cord injury. In-hospital death was defined as death from any cause during the period between the operation and discharge from the hospital. Distal embolism was defined as systemic multiple embolisms to the visceral organs or extremities, with or without cerebral and spinal cord infarctions. The secondary end points related to the effectiveness of JGOS were defined as follows: the presence or absence of any type of endoleak, kinking, structural fracture or migration of the graft and changes in the diameters of the excluded aneurysms. The excluded aneurysmal diameters and their change rates were measured at the site where the aneurysm showed its greatest dimension. Postoperative thrombus formation was defined as no leak of contrast medium at the excluded aneurysm in the delayed phase on CT. These parameters were evaluated by CT at regular intervals after the operation.

Stent deployment

The selection of the JGOS was based on the preoperative CT (Fig. 1). The length and diameter were selected in the following manner: the distal landing zone needed to be >30 mm in the straight descending aorta. The diameter was selected to be 10% larger than the native aortic diameter where it was planned to deploy the distal stent end. The JGOS was deployed antegradely from the transected aortic stump during the hypothermic circulatory arrest period (Fig. 2). During device deployment, no imaging modalities were used, such as endoscopy, fluoroscopy or transoesophageal echocardiography.

Figure 1:

Preoperative and postoperative computed tomography images. (A) The distal aortic arch aneurysm extending to the proximal descending aorta. (B) The line indicates the estimated length from the point of the expected aortic stump to the distal end of the stent graft. (C) The aneurysm is excluded by the J-graft open stent and it also shows good trackability to the curved descending aorta.

Figure 2:

The structure of the J-graft open stent graft. (A) The principle of J-graft open stent deployment: (a) the delivery system is inserted into the aorta; (b) deployment is started and the stent part begins to expand; (c) the graft part is deploying; and (d) the stent graft is just released. (B) Before insertion, the insertion length of the device is confirmed at the operative field, and a mark is put on it to avoid too much insertion. (C) Delivery system is inserted up to the mark.

Surgical technique

Preoperative cerebrospinal fluid drainage was not done routinely. Cardiopulmonary bypass was started by arterial cannulation of the ascending aorta or right subclavian artery and bicaval venous cannulation. Myocardial protection was achieved with intermittent antegrade or retrograde cold-blood cardioplegic solution, depending on the condition of the patient’s ascending aorta. Near-infrared spectroscopy was used for cerebral monitoring. When the rectal temperature reached 25°C, circulatory arrest was started. The aortic arch was usually transected circumferentially just distal to the left common carotid artery origin. Then, cerebral protection was started by antegrade selective perfusion. Cooling was maintained by rectal temperatures of 20–25°C. After deployment of the JGOS to the descending aorta, the graft portion of the JGOS was trimmed and incorporated with the native aortic stump wall. It was then anastomosed together with the proximal four-branched prosthesis. The left subclavian artery was ligated at its origin. Then, the left subclavian artery was reconstructed by the side branch of the proximal prosthesis in an end-to-end or end-to side fashion, depending on the condition of the patient’s left subclavian artery. Finally, the proximal anastomosis was performed.

Statistical analysis

Data were collected and analysed retrospectively. Categorical variables are presented as counts and percentages. Continuous variables are presented as medians with interquartile range. GraphPad Prism (version 7.04; GraphPad Software, San Diego, CA, USA) was used to estimate survival and freedom from reoperation with the Kaplan–Meier method. Other statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria) [10]. Specifically, EZR is a modified version of R commander designed to add statistical functions frequently used in biostatistics. Univariable analysis was performed using the Fisher’s exact test. Candidate variables for univariable analysis were selected on the basis of clinical relevance. A multivariable logistic regression model was used to identify the risk factors for mortality. The factors found to be significant on univariable analysis (log-rank test, P < 0.10) were entered into a subsequent multivariable analysis. Cox proportional hazards regression models were used to identify the predictors of all-cause mortality and neurological complications during the follow-up period. Significance was defined by a P-value of <0.05.

RESULTS

Preoperative data

The patients’ preoperative characteristics are shown in Table 1. The patients’ median age was 74 years (71–78), and 53 patients (76.8%) were male. Aneurysmal type was fusiform in 54 patients (78.3%) and saccular in 15 patients (21.7%). In 9 patients (13.0%), the aneurysm extended to the ascending aorta, and in 17 patients (24.6%), the lesions reached the proximal aortic arch. Preoperative comorbidities included coronary artery disease in 17 patients (24.6%), and renal insufficiency in 15 patients (21.7%). Four patients (5.8%) had a left recurrent laryngeal nerve palsy, and 15 patients (21.7%) had a history of stroke. Four patients (5.8%) had undergone previous cardiac surgery, and 7 patients (10.1%) had previous abdominal aortic surgery. A shaggy aorta was seen in 24 patients (34.8%). The median calculated EuroSCORE II was 14.8% (9.2–20.0).

Table 1:

Preoperative characteristics of the study patients

Variables
Number of patients	69
Age (years), median (IQR)	74 (71–78)
Male gender, n (%)	53 (76.8)
Body mass index (kg/m2), median (IQR)	23.7 (21.7–25.3)
Aneurysmal type, n (%)
Fusiform	54 (78.3)
Saccular	15 (21.7)
Shaggy aorta	24 (34.8)
Aortic lesion, n (%)
Ascending-proximal descending	9 (13.0)
Proximal arch-proximal descending	17 (24.6)
Distal arch-proximal descending	43 (62.3)
History of median sternotomy, n (%)	4 (5.8)
Previous aortic operation, n (%)
Ascending aorta	2 (2.9)
Aortic arch	0
Descending aorta	1 (1.4)
Thoraco-abdominal	0
Infrarenal abdominal aorta	7 (10.1)
Preoperative comorbidity, n (%)
Coronary artery disease	17 (24.6)
Chronic obstructive pulmonary disease	19 (27.5)
Renal insufficiency	15 (21.7)
History of stroke	15 (21.7)
Preoperative left recurrent nerve palsy	4 (5.8)
Preoperative EuroSCORE II, median (IQR)	14.8 (9.2–20.0)

IQR: interquartile range.

Operative data

Operative data are shown in Table 2. The median times of cardiopulmonary bypass, selective cerebral perfusion, lower body circulatory arrest and cardiac ischaemia were 289 min (236–315), 119 min (88–142), 76 min (61–87) and 198 min (153–213), respectively. A total of 15 concomitant surgical procedures were performed, as follows: 8 coronary artery bypass grafting, 5 valve surgery and 2 arrhythmia surgery. The aortic arch was transected between the brachiocephalic artery and the left carotid artery in 13 patients (18.8%), between the left carotid artery and the left subclavian artery in 51 patients (73.9%) and just distal from the left subclavian artery in 5 patients (7.2%). The JGOS was successfully implanted into each of the 69 patients antegradely without any complications. The median diameter and length of the implanted JGOS were 31 mm (29–33) and 90 mm (90–90), respectively. The left subclavian artery was reconstructed in 67 patients (97.1%).

Table 2:

Operative data of the study patients

Variables
Operative time (min), median (IQR)	476 (426–531)
Cardiopulmonary bypass time (min), median (IQR)	289 (236–315)
Selective cerebral perfusion time (min), median (IQR)	119 (88–142)
Lower body circulatory arrest time (min), median (IQR)	76 (61–87)
Cardiac ischaemic time (min), median (IQR)	198 (153–213)
Transected point of the aortic arch, n (%)
Between brachiocephalic artery and left carotid artery	13 (18.8)
Between left carotid artery and left subclavian artery	51 (73.9)
Just distal from left subclavian artery	5 (7.2)
Diameter of the four-branch-prosthesis, median (IQR)	28 (28–30)
Diameter of stent graft
25 mm, n (%)	4 (5.8)
27 mm, n (%)	4 (5.8)
29 mm, n (%)	22 (31.9)
31 mm, n (%)	18 (26.1)
33 mm, n (%)	9 (13.0)
35 mm, n (%)	7 (10.1)
37 mm, n (%)	2 (2.9)
39 mm, n (%)	3 (4.3)
Diameter of the J-graft open stent graft, median (IQR)	31 (29–33)
Length of stent graft
60 mm, n (%)	8 (11.6)
90 mm, n (%)	46 (66.7)
120 mm, n (%)	15 (21.7)
Length of the J-graft open stent graft, median (IQR)	90 (90–90)

IQR: interquartile range.

Postoperative data

Postoperative data are shown in Table 3. All deployed stent grafts were securely fixed at the target area, and thrombosis was formed at the space between the aneurysmal intima and stent graft. There were no patients with a type I endoleak, kinking or migration of the stent graft. The position at the distal stent end was as follows: at the Th 5 level in 13 (18.8%); at the Th 6 level in 23 (33.3%); at the Th 7 level in 24 (34.8%); at the Th 8 level in 8 (11.6%); and at the Th 9 level in 1 patient (1.4%). The distal end was positioned at the Th 8 level or a more proximal site in 68 patients (98.5%). Three patients developed type II endoleaks. One patient who had a persistent type II endoleak had undergone additional left subclavian artery embolization. The type II endoleaks of the other 2 patients disappeared spontaneously during the follow-up period. There were 10 neurological complications. Stroke occurred in 8 patients (11.6%); 6 of the 8 patients who developed stroke died in hospital. Paraplegia occurred in 4 patients (5.8%); 2 of them had multiple distal embolisms including cerebral infarctions. New-onset left recurrent laryngeal nerve palsy occurred postoperatively in 1 patient (1.4%). There were 10 in-hospital deaths; the mortality rate was 14.5%. The causes of the in-hospital deaths were cerebral infarction in 6, mediastinitis in 2 and pneumonia in 2 patients. Analyses of risk factors associated with in-hospital mortality and neurological complications are shown in Table 4. Based on the analyses, chronic obstructive pulmonary disease (P = 0.02) and shaggy aorta (P = 0.02) were identified as independent risk factors.

Table 3:

Postoperative data of the study patients

Variables
Ventilation time (h), median (IQR)	18 (10–240)
ICU stay (days), median (IQR)	4 (1–35)
Postoperative complications, n (%)
Stroke	8 (11.6)
Spinal cord injury	4 (5.8)
Distal embolism	2 (2.9)
Mediastinitis	2 (2.9)
Re-exploration for bleeding	1 (1.4)
New onset of left recurrent nerve paralysis	1 (1.4)
Mean follow-up period (years), median (IQR)	1.8 (0.6–3.6)
Postoperative computed tomography, n (%)
Type I endoleak	0
Type II endoleak	3 (4.3)
Distal landing zone, n (%)
Th5	13 (18.8)
Th6	23 (33.3)
Th7	24 (34.8)
Th8	8 (11.6)
Th9	1 (1.4)
Additional operations, n (%)
Left subclavian artery occlusion	1 (1.7)
TEVAR	2 (3.4)
In-hospital mortality (n = 10), n (%)
Cerebral infarction	6 (8.7)
Pneumonia	2 (2.9)
Mediastinitis	2 (2.9)

ICU: intensive care unit; IQR: interquartile range; TEVAR: thoracic endovascular repair.

Table 4:

Risk factors associated with in-hospital death and neurological complications

Variables	Univariable	Multivariable
	P-value	Odds ratio	95% CI	P-value
Age > 75 years	0.77
Gender (male)	1.00
COPD	0.03	7.34	1.33–40.6	0.02
Diabetes mellitus	0.33
Ischaemic heart disease	0.01	2.27	0.37–13.7	0.37
Stroke	0.47
Renal failure (Cr > 1.5 mg/dl)	0.01	3.36	0.63–17.9	0.16
Shaggy aorta	0.0002	10.6	1.58–71.4	0.02
Reoperation	1.00
CPB time (>301 min)	0.21
Distal landing zone (>Th7)	0.54

CI: confidence interval; COPD: chronic obstructive pulmonary disease; CPB: cardiopulmonary bypass; Cr: creatinine; Th: thoracic vertebral level.

Mid-term follow-up

A total of 59 patients (85.5%) survived the first operation. The median follow-up period was 1.8 years (0.6–3.6). No patient developed new onset of any type of endoleak, structural fracture or migration of the stent graft during the follow-up period. On a Cox proportional hazards regression analysis to identify risk factors for all-cause mortality and neurological complications during the follow-up period, a shaggy aorta was identified as the only independent risk factor (hazard ratio 6.4, 95% confidence interval 1.3–31.6; P = 0.02). Survival rate at 1 and 3 years was 84.8% and 79.4%, respectively (Fig. 3A). Freedom from death and neurological complications at 1 and 3 years was 80.6% and 77.2%, respectively (Fig 3B).

Figure 3:

(A) Kaplan–Meier survival curve for all patients who underwent operative procedures (n = 69). (B) Freedom from death and neurological complications during the follow-up period (n = 69).

Additional operations

Additional endovascular interventions were required in 3 patients (5.1%) (Table 3). One patient underwent left subclavian artery embolization due to a persistent type II endoleak and re-expansion of the excluded aneurysm. The other 2 patients had endovascular stent extension at the distal end of the JGOS due to downstream aneurysmal dilatation. There were no complications or in-hospital deaths related to the additional interventions. Freedom from aortic reintervention at 1 and 3 years was 97.2% and 81.3%, respectively (Fig. 4).

Figure 4:

Freedom from aortic reintervention for all patients who were discharged from the hospital (n = 59).

Change in diameter of the excluded aneurysms

The excluded aneurysm was completely thrombosed in 58 survivors (98.3%) at the end of the follow-up period, and its diameter showed shrinkage in 54 patients (91.5%). Four patients showed no change in the excluded aneurysmal diameter, and only 1 patient developed re-expansion due to a persistent type II endoleak from the patient’s left subclavian artery. The median change rate of shrinkage in the excluded aneurysmal diameter was 90.1% in patients with fusiform aneurysms and 69.2% in patients with saccular aneurysms during the follow-up period.

DISCUSSION

Open-style stent graft techniques have been spreading worldwide recently [1–8, 11–14]. The JGOS is an open-style stent graft that was designed based on the following concept in Japan: (i) easy to deploy and expand, without increasing technical complexity and prolonging the duration of ischaemia; (ii) it avoids the use of fluoroscopy; (iii) firmly fixates to the aortic wall without graft migration; and it has (iv) high flexibility and good trackability to the aortic wall [3–6]. The cut end of the proximal unstented graft is incorporated with the distal native aortic stump, which is sewn together with a proximal four-branched prosthesis. Therefore, problems associated with endovascular therapy, such as type Ia endoleaks, are less likely to occur. The biggest advantage of the JGOS is ‘proximalization’ of the distal anastomosis site [6, 11]. This advantage gives surgeons several benefits: (i) providing easy and reliable distal anastomosis, which can avoid bleeding; (ii) avoidance of left thoracotomy, which can prevent decreased respiratory function; and (iii) minimum dissection requirements around the aortic arch, which can avoid injury to the surrounding tissues, such as the left recurrent laryngeal nerve. These benefits are particularly important in cases in which the distal anastomosis site is originally deep. Therefore, the JGOS has a potential to shorten lower body circulatory arrest time and reduce the incidence of re-exploration [11]. In fact, although the reported re-exploration rate due to bleeding in total arch replacement has ranged from 3% to 30%, it was significantly reduced to only 1.4% (1 case) in the present study [1, 2]. Laryngeal malfunction after total arch replacement can cause respiratory complications, especially in elderly patients [12]. Based on the literature, the rate of left recurrent laryngeal nerve palsy due to total arch replacement has ranged from 3% to 20% [1, 12]. In the present series, the incidence was reduced to only 1.4% (1 patient). Although this patient did not develop aspiration pneumonia after the operation, 1 patient who had left recurrent laryngeal nerve palsy preoperatively developed severe aspiration pneumonia after the surgery, which resulted in in-hospital death. Therefore, we believe that it is important to avoid left recurrent laryngeal nerve damage during total arch replacement as much as possible, and we should pay attention to the patient’s recurrent laryngeal nerve condition continuously. The other advantage of the JGOS is that it has the potential to provide complete repair with a single-stage operation for selected patients with an aortic arch aneurysm that extends to the proximal descending aorta [1, 2, 4–7, 14]. Furthermore, if the condition requires additional surgical intervention, the JGOS can provide an ideal landing zone or anastomosis site for further endovascular or open reintervention [14]. The JGOS contributes to the less invasiveness of this surgery for such complex aortic disease.

Assessment of the primary end point

In-hospital mortality

According to the published reports, in-hospital mortality of total arch replacement with an open-style stent graft has ranged from 0% to 17.2% [1–8, 10–15]. In the present study, the in-hospital mortality rate was 14.5%, which was almost the same as the reported incidence and the predicted mortality calculated by EuroSCORE II (14.8%). The major cause of in-hospital mortality was cerebral infarction; 8 patients (11.6%) developed cerebral infarction and 6 (75%) of them died in-hospital. The rate of stroke was not significantly higher than in other reports, but the development of intraoperative cerebral infarction was strongly related to in-hospital mortality. In the present study, multivariable analysis showed that chronic obstructive pulmonary disease and shaggy aorta were significant risk factors for in-hospital death. To reduce in-hospital mortality, total arch replacement combined with the JGOS for patients with chronic obstructive pulmonary disease or a shaggy aorta should be carefully considered. According to the Cox proportional hazards regression analysis, a shaggy aorta was identified as the only independent risk factor for all-cause mortality and neurological complications during the follow-up period.

Rates of spinal cord injury, distal embolism and cerebral infarction

It is well known that the open-style stent graft technique has a potential risk of spinal cord injury; it is reported that incidence has ranged from 0% to 21.1% [1–5, 11–15]. Although the cause of spinal cord injury is still not fully understood, it might be related to the length of the stent graft, which is recommended to be above the Th 8 level. In the present study, 4 patients (5.8%) developed spinal cord injuries. The rate of spinal cord injury was not significantly higher than in other reports. The distal end of the stent graft of these 4 patients was the Th 5 level in 1, the Th 6 level in 1 and the Th 8 level in 2 patients. None of them exceeded the Th 8 level [1–6]. In these 4 patients, there was no history of previous graft replacement, and there was no significant difference in the lower body circulatory arrest time between patients who developed spinal cord injury (71 min) and those who did not (75 min). Hypotensive events were not recorded in the 4 patients during the operative and postoperative periods. On the other hand, there were 24 patients (34.8%) with a shaggy aorta in the present study population, and 3 of the 24 patients developed spinal cord injury. Moreover, 2 of the 3 patients with a shaggy aorta who developed spinal cord injury had systemic multiple distal embolisms and cerebral infarctions. Although multivariable analysis showed that the presence of a shaggy aorta was a preoperative risk factor for in-hospital mortality, the presence of a shaggy aorta was not a significant risk factor for multiple systemic embolisms including spinal cord injury, distal embolism or cerebral infarction in the present study [1–5, 13]. Koizumi et al. [5] mentioned that a shaggy descending aorta was a risk factor for distal embolism. The mechanism is presumed to involve peeling away of atherosclerotic plaques from the aortic wall when the stent graft is deployed. In the present study, 8 patients (11.6%) developed cerebral infarctions. Of these 8 patients, 4 had a shaggy aorta. The reported incidence of cerebral infarction following this type of operation has ranged from 2.5% to 21% [1–7, 9–13]. The rate of stroke was not significantly higher than in other reports. It was unclear whether the use of the JGOS increased the incidence of stroke. However, considering the distribution of the cerebral infarction in the 8 patients, most of the cerebral infarctions developed in the basilar artery area. Deployment of the JGOS might be related to the development of cerebral infarctions. We should have completed the left subclavian artery closure before deployment of the JGOS. The development of stroke strongly affected in-hospital mortality. We must take measures to prevent the development of cerebral infarction as much as possible, and this could provide a significant reduction of in-hospital mortality. Based on these results, the indication for total arch replacement combined with the JGOS for patients with a shaggy aorta should be considered carefully.

Assessment of the secondary end points

None of the patients developed a type Ia or Ib endoleak or kinking or migration of the JGOS during the follow-up period. Complete thrombus formation at the excluded aneurysm was achieved in 98.2% of surviving patients. The diameters of the excluded aneurysms showed shrinkage in 91.2% of the patients at the end of the follow-up period. Type II endoleaks were seen in 3 patients at discharge, but only 1 patient required additional left subclavian artery embolization. A total of 3 patients required additional endovascular therapy during the follow-up period, including the above-mentioned patient. The other 2 patients had additional endovascular repairs due to downstream aneurysmal dilatation during the follow-up period. Based on these results, the performance of the JGOS in aneurysmal exclusion was high; it was reliable and primarily curative. The JGOS has the potential to provide one-stage complete surgical repair in selected patients with an aortic arch aneurysm that extends to the proximal descending aorta. Furthermore, the JGOS provides an excellent landing zone if additional distal endovascular reintervention is required. However, in-hospital mortality of 14.5% was higher than the mortality calculated from the Japan SCORE database (3.5–9.9%). We have to reduce in-hospital mortality by taking all possible measures, including careful patient selection.

Limitations

There are some limitations in this study. First, it was a retrospective study at 2 centres, it was not a randomized study, the sample size was small and the follow-up period was short. Second, there were small differences in the surgical procedures. However, the main surgical principles were uniform among the participating surgeons, including how to deploy the JGOS or anastomose the prosthesis, and in the use of selective antegrade cerebral perfusion for brain protection.

CONCLUSIONS

Based on the present study results, one cannot make any firm conclusion about the performance of the JGOS due to the small sample size. However, based on the findings of the present study, the JGOS was safely and easily deployed into the target area in all enrolled patients in an antegrade fashion through the opened aortic arch without any trouble. This device was able to shift the distal anastomosis line to the more proximal side. The performance of the JGOS in achieving aneurysmal exclusion was good, and its mid-term results were good. In selected patients with combined aortic arch and descending aortic aneurysms, the JGOS approach has a potential to provide one-stage repair. On the other hand, measures to prevent the development of neurological complications during the operation are needed, and this could provide a significant reduction in the in-hospital mortality. Long-term follow-up is needed to further verify the safety and efficacy of the JGOS.

Conflict of interest: none declared.

Author contributions

Koji Tsutsumi: Data curation; Formal analysis; Supervision; Writing—original draft; Writing—review & editing. Osamu Ishida: Investigation. Nozomu Yamanaka: Formal analysis. Visualization. Kanako Hayashi: Data curation. Kenichi Hashizume: Data curation; Resources; Supervision.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Michael Grimm, Samuel Heuts and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

ABBREVIATIONS

CT

Computed tomography

JGOS

J-graft open stent graft

Th

Thoracic vertebral level