Cool-Shot Technique to Protect Spinal Cord during Thoracoabdominal Aortic Replacement

Abstract

Deep hypothermia helps protect the spinal cord, but is invasive. Here, we present a method to avoid reperfusion injury by selectively circulating cold blood under high pressure to the intercostal artery during reperfusion after intercostal artery reconstruction. Of the 23 patients who underwent thoracoabdominal aortic aneurysm open repair, one died. The motor evoked potential disappeared during aortic clamping in nine patients. Six patients recovered completely from aortic clamping release, two showed recovery >50% and one achieved full recovery 3 months later. Permanent motor impairment did not occur. This method could prevent reperfusion injury and paraplegia following thoracoabdominal aortic aneurysm surgery.

 

MissingLinkText

Keywords

thoracoabdominal aortic aneurysm - cool-shot technique - spinal cord protection - reperfusion injury

Introduction

Spinal cord injury is one of the complications that surgeons must seriously avoid in patients undergoing thoracic and abdominal aortic aneurysm repair, causing permanent paralysis at a rate of 2.0 to 10.8%.[1] [2] [3] [4]

We have used cold blood reperfusion as a method of protection against spinal cord ischemia to prevent spinal cord nerve tissue reperfusion injury. We have found this technique to be particularly useful when intraoperative motor evoked potential (MEP) monitoring showed a weak response and spinal cord perfusion failure was strongly suspected. Herein we present a case series wherein this technique was performed.

 

MissingLinkText

Patients and Methods

We performed open aortic surgery with intercostal artery (ICA) reconstruction in 23 patients at our hospital from January 2016 to December 2022. All patients underwent preoperative multislice computed tomography scan angiography of the Adamkiewicz artery (AKA).[5]

We inserted a cerebrospinal fluid drainage (CSFD) catheter between the lumbar vertebral bodies one day preoperatively. Muscle MEPs (using an MEE-2000; Nihon Kohden, Tokyo, Japan) were recorded intraoperatively. Muscle MEPs were obtained in the abductor pollicis brevis and abductor hallucis muscles of the feet using subcutaneous needle electrodes. A 50% decrease in MEP amplitude from the baseline amplitude obtained immediately before the interventions was considered a significant change; MEP amplitude loss was deemed an alarm signal for spinal ischemic changes.[6] No neuromuscular blocking agents were administered during MEP monitoring.

We conducted partial cardiopulmonary bypass through cannulation of the left femoral artery and femoral vein ([Fig. 1A]). We routinely used mild-to-moderate hypothermia (32–34°C, rectal temperature). Circulating flow during cardiopulmonary bypass was slowly titrated up to 3.5 L/min and maintained after aortic clamping. Subsequently, we used segmental-staged aortic clamping to execute cross-clamping, first clamping the proximal side of the left subclavian artery and middle descending aorta. The abdominal aorta was clamped above the celiac artery branches. We measured MEPs 5 minutes after aortic clamping. Of these, the ICA between Th8 and Th12 was selected for reimplantation based on preoperative diagnosis.

We adopted the following strategies:

• Upper body circulation was maintained at 80 to 100 mmHg at radial artery pressure with autologous cardiac output, while lower body circulation was maintained at hypothermic blood flow (1.0–1.5 L/min) with cardiopulmonary bypass circulation.

• We re-perfused the AKA from a 9-mm branch of an artificial graft with a blood temperature of 20°C and perfusion pressure of 80 to 100 mmHg. We raised blood temperature to 34°C within approximately 15 minutes (cool-shot technique) ([Fig. 1A]).

• Mean blood pressure of 100 mmHg was enforced if MEPs weakened or disappeared during clamping ([Fig. 1B]).

Fig. 1 Mechanical cardiopulmonary support during surgery for thoracoabdominal aortic aneurysm. (A) Modification of the cardiopulmonary bypass during surgery for thoracoabdominal aortic aneurysm. (B) The end-to-side anastomosis is created between the Bentall graft and abdominal Y-shaped graft. Reperfusion methods after Adamkiewicz artery reconstruction are as follows. Blood cooled to 20°C is perfused from the artificial vascular side branch before releasing the proximal aortic clamp following intercostal artery reconstruction. The distal anastomosis or abdominal organ branches are reconstructed during 15 to 20 minutes of rewarming, and the proximal clamping is subsequently released. Red, black, yellow, and blue arrows indicate the proximal aortic clamping, distal aortic clamping, site of intercostal artery reconstruction, and abdominal vessel branching, respectively. White and purple arrows indicate the site of blood delivery and direction of cooled blood delivery to the intercostal artery, respectively.

CSFD was continued for 24 hours postoperatively.

[Table 1] presents patient characteristics. One patient (4.5%) died postoperatively due to pneumonia caused by multidrug-resistant Pseudomonas aeruginosa.

Table 1 Preoperative baseline characteristics

Age (year)	59.7 ± 12.6 (28–74)
Male sex	20 (86.9%)
Heritable thoracic aortic disease: Marfan syndrome	3 (13.6%)
Hypertension	20 (90.9%)
Dyslipidemia	12 (54.5%)
Diabetes mellitus	3 (13.6%)
Estimated glomerular filtration rate (mL/min/1.73 m2)	71.9 ± 20.6
Prior cerebrovascular accident	2 (9.1%)
Left ventricular ejection fraction (%)	69.5 ± 4.7
EuroSCORE II	5.4 ± 4.8
Type of aortic disease
True aortic aneurysm	6 (27.3%)
Aortic dissection type I	8 (45.5%)
Aortic dissection type IIIb	6 (27.3%)
Type of Crawford classification I	4 (18.2%)
II	10 (45.5%)
III	7 (31.8%)
IV	1 (4.5%)
Adamkiewicz artery branches at
Th 7–9	3
Th 10–12	16
L1–2	3
Reattachment of intercostal arteries	22 (100%)
6th to 12th intercostal arteries	3 (13.6%)
8th to 12th intercostal arteries	8 (36.4%)
9th to 12th intercostal arteries	2 (9.1%)
10th to 12th intercostal arteries (plus first lumbar artery)	4 (18.2%)
11th to 12th intercostal arteries	2 (9.1%)
12th intercostal artery to second lumbar arteries	3 (21.1%)
Operative time (minutes)	539 ± 114
Cardiopulmonary bypass time (minutes)	176 ± 58
Aortic clamp time (minutes)	120 ± 44
minimum rectal temperature	35.1 ± 0.6
Operative death	1 (4.5%)
Cerebrovascular accident	4 (18.2%)
Acute renal failure	2 (9.1%)
Cardiac complication	0
Pulmonary complication	3 (13.6%)
Infection	3 (13.6%)
Bleeding requiring reoperation	0
Spinal cord injury	1 (4.5%)
Persistent paraparesis or paraplegia	0
Transient paraparesis	1 (4.5%)
Length of hospital stay (days)	25.8 ± 19.4

Abbreviation: EuroSCORE II, European System for Cardiac Operative Risk Evaluation.

During ICA reconstruction, MEP was attenuated or absent in nine patients. However, MEP improved in all but one case (case 1: [Fig. 2A–D]). Although MEP did not improve in one patient intraoperatively, it improved 1 month later. This patient underwent coronary artery bypass surgery via the left internal thoracic artery, aortic arch replacement, and Y-grafting for an abdominal aortic aneurysm (case 2: [Fig. 3A–D]). Moreover, the patient experienced severe hypotension due to a postoperative blood transfusion allergy and difficulty maintaining blood pressure. Nevertheless, motor palsy improved after 1 month, with a slight depth perception abnormality remaining. We attributed the leading cause of spinal cord failure paralysis to the lack of collateral blood flow during aortic clamping and failure to maintain blood pressure after reconstruction.

Fig. 2 Case 1: a 72-year-old man presenting with a chronic type IIIb dissecting aortic aneurysm (Crawford type II). (A) Preoperative findings (Crawford type II). The thoracic descending aorta has a maximum diameter of 63 mm, while the abdominal aorta shows a diameter of 48 mm. The Adamkiewicz artery branches off the descending aorta at the 11th–12th thoracic vertebral level (black arrow). (B) First postoperative three-dimensional (3D) CT. We opt for descending aortic replacement as our first procedure (light blue line coverage). (C) 3DCT following the final surgery. We perform artificial vascular replacement for the thoracoabdominal aortic aneurysm in the second surgery (yellow-green line). The intercostal artery is reconstructed as an island between the 8th and 12th thoracic vertebrae. The black arrows indicate the reconstructed Adamkiewicz artery. (D) Alteration in motor response evoked potential (MEP): MEP in the right leg is lost during reconstruction of the intercostal artery (red arrow), but recovered after revascularization (blue arrow). CT, computed tomography.

Fig. 3 Case 2: a 73-year-old man with Behçet's disease and multiple aortic saccular aneurysms. (A) Preoperative findings. The coronary arteries show 90% stenosis in the left anterior descending branch and an aneurysm in the left subclavian artery. The Adamkiewicz artery is branched from the 10th to 11th intercostal artery (black arrow). The patient also undergoes artificial vascular replacement (Y-graft) for an abdominal aortic aneurysm. (B) First postoperative three-dimensional CT. We initially perform total arch aortic replacement and coronary artery bypass (left internal thoracic artery–left anterior descending branch: light blue arrow). (C) 3DCT following the final surgery. Artificial vessel replacement is performed for the thoracoabdominal aortic aneurysm in the second surgery (yellow-green line). The intercostal artery is reconstructed as an island between the 9th and 12th thoracic vertebrae. Black arrows indicate the reconstructed Adamkiewicz artery. (D) Alterations in motor response evoked potentials (MEPs): MEPs in the left leg disappear during intercostal artery reconstruction (red arrow) and, unfortunately, they do not recover after revascularization (blue arrows).

Although we used the CSFD during surgery, we were unable to prove its efficacy in this study because we were not able to use it in all patients based on the established criteria; it was only used according to the protocol after surgery.

 

MissingLinkText

Discussion

Excellent results were achieved with this technique; there were no instances of post-discharge paraplegia, despite momentary MEP loss in some patients. Although hypothermia is helpful for spinal cord protection, whole-body hypothermia during thoracoabdominal aortic aneurysm surgery is invasive. Hence, we employed the cool-shot technique for local cooling of the spinal nerves.

The degree of hypothermia employed may be mild (30–34°C) or severe (15–20°C) depending on the institution. Deep hypothermic circulatory arrest (DHCA) is typically reserved for complex thoracic aortic procedures. In large centers where DHCA is routinely performed, patient outcomes are comparable to those of endovascular repair.[7] [8] [9] However, adapting DHCA in older patients is challenging and may increase blood loss. Therefore, we devised a method in which surgery is performed under mild hypothermia, with only the spinal cord under deep hypothermia.

Although epidural cooling to prevent reperfusion injury is effective, it is invasive and complex.[10] Hence, we investigated a strategy commencing with local cooling perfusion after ICA reconstruction and gradually raising spinal cord blood delivery temperature.

This study has certain limitations. First, the sample size was small. Second, only results in cases of reconstructed ICAs were reported; the extent of spinal cord cooling using the cool-shot technique after reconstruction remains unknown.

 

MissingLinkText

Conclusion

We devised a regional spinal-cord cooling technique using the spinal cool-shot technique following ICA reconstruction that is effective in additional spinal-cord protection. This technique may prevent reperfusion injury, and early and delayed postoperative paralysis in patients with intraoperative MEP loss during thoracoabdominal aortic aneurysm surgery.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declared.

Data Availability Statement

The data associated with this manuscript are not publicly available; however, they can be provided by the corresponding author upon reasonable request.

• References

• 1 Coselli JS, LeMaire SA, Preventza O. et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg 2016; 151 (05) 1323-1337
• 2 Gaudino M, Khan FM, Rahouma M. et al. Spinal cord injury after open and endovascular repair of descending thoracic and thoracoabdominal aortic aneurysms: a meta-analysis. J Thorac Cardiovasc Surg 2022; 163 (02) 552-564
• 3 Murana G, Castrovinci S, Kloppenburg G. et al. Open thoracoabdominal aortic aneurysm repair in the modern era: results from a 20-year single-centre experience. Eur J Cardiothorac Surg 2016; 49 (05) 1374-1381
• 4 Norton EL, Orelaru F, Ahmad RA. et al. Hypothermic circulatory arrest versus aortic clamping in thoracic and thoracoabdominal aortic aneurysm repair. J Card Surg 2022; 37 (12) 4351-4358
• 5 Imamura Y, Kin H, Yoshioka K, Tabayashi A, Saitoh D. Thoracoabdominal aortic aneurysm repair based on pre- and postoperative evaluation of the artery of Adamkiewicz. Eur J Cardiothorac Surg 2022; 62 (05) ezac196
• 6 Tanaka Y, Kawaguchi M, Noguchi Y. et al. Systematic review of motor evoked potentials monitoring during thoracic and thoracoabdominal aortic aneurysm open repair surgery: a diagnostic meta-analysis. J Anesth 2016; 30 (06) 1037-1050
• 7 Amabile A, Lewis E, Costa V, Tadros RO, Han DK, Di Luozzo G. Spinal cord protection in open and endovascular approaches to thoracoabdominal aortic aneurysms. Vascular 2023; 31 (05) 874-883
• 8 Weiss AJ, Lin HM, Bischoff MS. et al. A propensity score-matched comparison of deep versus mild hypothermia during thoracoabdominal aortic surgery. J Thorac Cardiovasc Surg 2012; 143 (01) 186-193
• 9 Kulik A, Castner CF, Kouchoukos NT. Outcomes after thoracoabdominal aortic aneurysm repair with hypothermic circulatory arrest. J Thorac Cardiovasc Surg 2011; 141 (04) 953-960
• 10 Tabayashi K, Motoyoshi N, Saiki Y. et al. Efficacy of perfusion cooling of the epidural space and cerebrospinal fluid drainage during repair of extent I and II thoracoabdominal aneurysm. J Cardiovasc Surg (Torino) 2008; 49 (06) 749-755

Publication History

Received: 14 January 2024

Accepted: 22 April 2024

Accepted Manuscript online:
03 May 2024

Article published online:
11 June 2024