In this prospective cohort study, 101 patients were enrolled from December 2016 to October 2019 at the Department of Vascular Surgery in Shanghai Changzheng Hospital. The inclusion criteria were (1) age above 18 years, (2) at least 1 supra-aortic branch encroached by thoracic lesions, (3) landing zone not long enough (≤15 mm) for fixation of the aortic endograft so coverage of the LSA or LCCA had to be performed, (4) aortic disease confirmed by at least 1 radiologic examination (e.g., computed tomography angiography [CTA], magnetic resonance angiography [MRA]), (5) patent supra-aortic branch, and (6) type I/II aortic arch. The exclusion criteria were (1) cardiopulmonary and renal insufficiency contraindicating general anesthesia (according to the anesthesiologist), (2) severe infection causing high fever or organ dysfunction, (3) allergy to contrast medium, (4) adverse cardiovascular or cerebrovascular event within 3 months before intervention, (5) occlusion or stenosis of the supra-aortic branch or severe twisting of the arteries to be fenestrated, (6) no appropriate peripheral access, (7) Stanford A aortic lesions, (8) type III aortic arch and (9) Marfan syndrome. The protocol for this study was approved by the participating hospital’s Ethics Committee, and informed consent was obtained from each participant.

The enrolled patients underwent in situ laser, needle, or QF fenestration according to their order of admission. All interventions were performed by advanced interventional radiologists. The procedures were performed under general anesthesia. Two ProGlide vascular sutures (Perclose ProGlide, Abbott Inc, USA) were preset for right femoral artery puncture, and a 12-F sheath was inserted. A 7-F sheath (pre-left carotid artery fenestration) was punctured and inserted via the left common carotid artery, and a 6-F 55 cm renal artery sheath (Flexor® Check-Flo® Introducer, Cook, Inc, USA) was punctured and inserted into the left brachial artery. A 5-F labeled pig-tail catheter was introduced into the sheath of the right femoral artery to the ascending aorta, and the aortic size was measured. A Lunderquist super-stiff guide wire was inserted, and the pigtail catheter was withdrawn to the ascending aorta through the left brachial artery or carotid artery sheath for real-time intraoperative angiography. After angiographic evaluation of the aortic lesions, aortic stents of appropriate size (Capitivia, Medtronic, USA/Ankura, Lifetech, China) were introduced along the super stiff guide wire, accurately positioned, and released to repair thoracic aortic lesions and cover the supra arch branch arteries. Advanced through the 0.035 inch support catheter, the needle (reversed end of a 0.018 inch guidewire) or intravenous laser catheter (LFK-SLT30, Rafcon, China) or QF (Suzhou Innomed Medical Device, China) was introduced along the supra arch branch artery and guided to the branch artery for stent graft fenestration. After fenestration, the needle, intravenous laser catheter, or QF was advanced into the aortic lumen. The fenestration instruments were withdrawn, and another angiogram was made to confirm that the needle, intravenous laser catheter, or QF was positioned in the aortic cavity, and a 0.035 inch guide wire was introduced along the catheter to the ascending or descending aorta. A balloon (Mustang, Boston Scientific Corporation, USA) of appropriate size to expand the fenestration was positioned. The stent deployed in the super arch artery had a 5 mm protrusion in the aortic arch, and the distal end in left subclavian artery did not cover the orifice of the vertebral artery. Post dilatation was carried out if necessary. Finally, aortography was performed to check that the intervention was carried out successfully.

After the intervention, follow-up was conducted at 1 month, 6 months, and annually thereafter. The following perioperative outcomes were assessed: success rate of fenestration; average durations of intervention, fluoroscopy, rupture, and guide wire passage after fenestration (from guide wire touching the membrane to passing through the hole); and amount of contrast agent used. Follow-up outcomes included 30-day mortality, type I and III endoleaks, retrograde dissection, mortality, stent patency rate, stent morphology, secondary re-interventions, and severe cardiovascular and cerebrovascular events. Computed tomographic angiography of the aorta was performed to evaluate the rates of branch stenting, internal leakage rate, and secondary intervention.

The data were analyzed with SPSS 22.0 software. The data were expressed as mean values ± SD. Continuous variables were presented as median and range in the case of nonparametric distributions, and comparisons were made using the Mann-Whitney test. Continuous variables are presented as means ± standard deviations in cases of parametric distributions, and comparisons were made using the independent t-test. Categorical variables were compared using the Chi-square test or Fisher’s exact test and reported as frequencies (percentages). A p value of <0.05 was considered statistically significant.

In all, 101 patients underwent in situ fenestration during TEVAR between May 2016 and October 2019 (85 males and 16 females; mean age, 67 ± 10 years; 71 acute onset). In all, 34 patients underwent 40 in situ laser fenestrations during TEVAR (aortic dissection: n = 21, aortic aneurysm: n = 5, aortic ulcer: n = 8); 36 patients underwent 43 in situ needle fenestrations during TEVAR (aortic dissection: n = 22, aortic aneurysm: n = 6, aortic ulcer: n = 8); and 31 patients underwent 37 in situ QF fenestrations during TEVAR (aortic dissection: n = 19, aortic aneurysm: n = 5, aortic ulcer: n = 7). The baseline characteristics, including risk factors and comorbidities, were not significantly different among the three groups (Table 1).

TEVAR was technically successful in 94 patients (Figure 2). The technical success rates of the laser, needle, and QF fenestration groups were 94.1%, 94.4%, and 100%, respectively, showing an insignificant difference (p > 0.05). In the in situ laser fenestration group, one procedure failed due to an awkward fenestration angle of the laser guide wire. In the in situ needle fenestration group, one procedure failed because the needle had failed to puncture the membrane. After correction of mixed factors such as age and gender, variance analysis and Kruskal-Wallis test results showed the average operation, fluoroscopy, fenestration, and guide wire passage durations after fenestration were shorter in the QF fenestration group (p < 0.001). Moreover, the average amount of contrast agent in the QF fenestration group was less than that in the needle fenestration and laser fenestration group (p < 0.001) (Table 2).

All the in situ fenestrated arteries were found to be patent on postoperative follow-up CTA imaging and clinical symptoms.

The overall Perioperative complication rates of the laser, needle, and QF fenestration groups were 5.9%, 5,6%, and 0% (p > 0.05). In the perioperative period, no cerebral infarction, myocardial infarction, transient ischemic attacks, or other neurologic complications occurred. In the in situ laser fenestration group, one patient had a sheath injury due to laser heat (Figure 3A), and one patient presented with intraoperative vertebral artery ischemia. In the in situ needle fenestration group, one patient presented with an access-site hematoma, and one patient presented with a femoral arterial thrombus. No endoleak (include type I and type III), false lumen thrombosis, or subsequent remodeling of the aorta was evident.

A total of 89 (88.1%, 89/101) patients were followed for a median of 12.6 ± 1.6 months. At the 1-year follow-up, the overall postoperative complication rates of the laser, needle, and QF fenestration groups were 3.3%, 6.5%, and 0% (p > 0.05). In the laser fenestration group, there was one death due to postoperative ST-segment elevation myocardial infarction, with no direct association with the intervention. In the needle fenestration group, one patient developed occlusion of the bridge stent at 3-month follow-up. The occlusion was recanalized after intervention (Figure 3B–D). There were no complications in the QF group. No reverse tear or endoleak (type I and type III) was observed in any of the three groups (Table 3, Figure 4).