Spinal Cord Protection During Thoracic and Thoracoabdominal Endovascular Aortic Repair: 5-Year Results of a Preventive Protocol Including Prophylactic Cerebrospinal Fluid Drainage in High-Risk Patients

Abstract

Purpose:

Spinal cord injury (SCI) is a devastating complication of thoracoabdominal aortic (TAA) repair. The use of prophylactic cerebrospinal fluid drainage (CSFD) as part of a protective protocol during endovascular repair is controversial. This article reports the results of the prophylactic use of CSFD as part of the of a prevention protocol implemented in 2016.

Methods:

Retrospective review of spinal cord outcomes (SCI rate and CSFD-related complications) in patients treated endovascularly for TAA disease at a single institution from 2016 (implementation of an institutional SCI risk reduction protocol) to 2021. Patients were classified as high risk (≥2 factors), intermediate risk (1 factor), or low risk (0 factor). Only high-risk patients without contraindications underwent a prophylactic CSFD placement.

Results:

One hundred eighty-one patients were analyzed (124 males; 69.6 years): 130 (69%) aneurysms (n=24 thoracic, n=28 Crawford 1–2–3, and n=78 Crawford 4/pararenal), 35 (19.9%) chronic aneurysmal dissections, and 16 (8.8%) acute complicated type B dissections. Interventions were staged in 31 (17.2%) cases, and consisted of 74 (41%) Thoracic EndoVascular Aneurysm Repair (TEVAR) and 107 (59%) Fenestrated Branched EndoVascular Aneurysm Repair (F-BEVAR). Sixty-nine (38.1%) patients were identified as being at high risk of SCI and CSFD was used prophylactically in 64 of them (4 failures and 1 contraindication). Spinal cord injury occurred in 8 cases (4 paraparesis, 4 paraplegias including 2 permanent), of which 3 had a prophylactic CSFD and 5 underwent rescue drainage. In addition, 4 patients developed SCI related to prophylactic CSFD (intradural hematoma), resulting in 1 paraparesis and 3 paraplegias. Other CSFD-related complications were mild (6) or moderate (2), for a total of 12 complications (17%). Factors associated with major drain complications were: curative anticoagulation 36 hours after drain removal (n=1), multiple punctures (n=1), platelet count <100 000 at drain removal (n=1), and bipolar disorder (n=2). Overall, 4 patients had permanent paraplegia and 1 had sphincter dysfunction at the last follow-up. Mean follow-up was 17 months. Mortality was 4.4% at 30 days and 13.3% at 18 months, including 3 (1.6%) aortic-related deaths.

Conclusions:

With the protocol we used to protect the spinal cord, we report results comparable with the SCI literature and highlight the risks associated with prophylactic CSFD use, which requires a better understanding of contraindications.

Keywords

aortic aneurysm endovascular aortic repair aorta TEVAR FEVAR cerebrospinal fluid drainage spinal cord injury spinal cord ischemia paraplegia

Introduction

Spinal cord ischemia (SCI) is a frightening and devastating complication of thoracic and thoracoabdominal aortic aneurysm (TAAA) repair and dissection.^1–5 Permanent paraplegia significantly impairs patients’ quality of life and is associated with poorer long-term survival.⁶ The incidence of SCI following open surgical repair has decreased in historical series. The incidence of spinal cord injury (SCI) with open surgical repair has decreased from 15-20% in historical control series to 6-10%, with evidence from prospective randomized trials that prophylactic cerebrospinal fluid drainage (CSFD) reduces the rate of SCI. Although intuitively less invasive than open repair, endovascular thoracic and TAAA repair is still associated with high rates of SCI (1.5%–2.5% after Thoracic EndoVascular Aneurysm Repair (TEVAR),⁷ and up to 10% after endovascular type II TAAA repair).^1,4–6

Several risk factors for SCI following endovascular aneurysm repair have been described, including coverage of the left subclavian or hypogastric arteries, renal insufficiency, perioperative hypotension, previous abdominal aortic aneurysm repair, and coverage of a long segment of the thoracic aorta, particularly if the distal descending portion (T7–T12) is involved.^3,4,6 As a result, a proactive spinal cord protection protocol associated with high-volume practice is a priority to reduce the incidence of SCI.⁸

Previous studies have suggested that a bundled intervention including a combination of parameters to improve spinal cord oxygenation may improve neurological outcome by reducing the risk of SCI.^7–9 Selective prophylactic CSFD has been widely used in these bundled interventions and recent guidelines recommend its prophylactic use when extensive thoracic aortic coverage (>200 mm) is planned (grade IIa).¹⁰ However, several studies have recently reported 10% of any type of complications associated with its use, mostly minor but sometimes severe, such as spinal hematoma. Therefore, the widespread use of CSFD remains controversial.^10–12

Since 2016, our multidisciplinary aortic team has been implementing a protocol for spinal cord protection interventions to reduce the risk of SCI (including blood pressure, glycemic and hemoglobin control, neuromonitoring, staged procedures, subclavian and hypogastric artery revascularization, and prophylactic CSFD for patients deemed at high risk of SCI according to our protocol). In this setting, we conducted a retrospective, single-center study with the aim of investigating the real-life outcomes achieved during the first 5 years (January 2016–January 2021) of practice following the implementation of a proactive spinal cord protection protocol.

Methods

Population

All consecutive patients who underwent TEVAR or Fenestrated Branched EndoVascular Aneurysm Repair (F-BEVAR) for dissection or aneurysm repair at our institution were retrospectively identified in a prospectively maintained database. Clinical data were collected in an anonymized Excel database to de-identify private information. This study was in accordance with the tenets of the Declaration of Helsinki. Data collection was carried out in accordance with French data protection laws, after obtaining the patients’ consent to the anonymous collection of their data on the standard informed consent form provided by our institution. All patients were informed of the possibility of retrospective data collection for clinical research purposes. Institutional Review Board’s approval was not required due to the anonymized use of patient data.

Data Collection

Medical history was reviewed for the presence of cardiovascular risk factors (hypertension, hyperlipidemia, tobacco/alcohol use, and body mass index), comorbidities (diabetes mellitus, coronary artery disease, chronic obstructive pulmonary disease (COPD), and chronic renal failure), previous aortic repair, and Marfan or Ehlers-Danlos disease. Indication for surgery, additional procedures (subclavian and hypogastric artery embolization or coverage with or without revascularization), surgical setting (elective or urgent/emergent), American Society of Anesthesiology (ASA) class, and Revised Cardiac Risk Index (RCRI) for preoperative risk ¹³ were also reviewed. For CSFD, we collected data on indication for drainage, timing of placement (prophylactic vs curative), failure of catheter insertion, duration of CSFD, all dysfunction, and complication types with description. Data on any symptoms of postoperative SCI, its management, occurrence of any hemodynamic problem, and need for hemoglobin transfusion, immediate, and long-term recovery were also recorded.

Protocol for the Management of Thoracoabdominal Disease and Indication for CSFD

The spinal cord protection protocol was based on several strategies, including surgical staging and preservation of the collateral network (left subclavian and internal iliac arteries, except when technically not feasible), blood pressure management (intraoperative mean arterial pressure [MAP]>80 mmHg and up to 100–110 mmHg in case of postoperative neurological event), target hemoglobin ≥10 mg/dL preoperatively and postoperatively. To limit the number of prophylactic CSFD insertions and the associated complications, a preoperative scoring system has been introduced for the use of CSFD as either a prophylactic or curative tool. According to our protocol and the literature,^3,4,6–9 7 major risk factors for SCI have been identified: the need to cover aortic segments from Th7 to Th12; extended thoracic aortic coverage >200 mm; thoracoabdominal coverage >40 mm above the celiac trunk; previous open or endovascular thoracic or abdominal aortic surgery; occlusion of an internal iliac or left subclavian artery without revascularization; previous episodes of SCI symptoms during previous conventional or endovascular aortic surgery; emergency treatment (within 15 days of the onset of aortic symptoms). Patients were categorized as high risk (≥2 factors), intermediate risk (1 factor), or low risk (0 factor). Prophylactic CSFD was used only in patients classified as high risk for SCI with no medical contraindications. For patients without prophylactic CSFD, a strategy of early clinical assessment and salvage therapeutic drainage was adopted.

Contraindications to prophylactic CSFD were: urgent procedures; anatomical difficulties related to vertebrae or osteosynthesis devices or other metal implants at the site of CSFD insertion; coagulation abnormalities (prothrombin time <60%, international normalized ratio >1.5, activated partial thromboplastin time ratio >1.20, antifactor Xa level >0. 10 UI/ml, direct oral anticoagulant [DOAC]-specific antifactor Xa>30 ng/ml) or platelet count <100 000; non-suspended anticoagulant (at least 5 days before drain placement) or antiplatelet medication (at least 3 days for aspirin and 5 days for clopidogrel); history of hemiplegia due to cerebral hemorrhage; fever or headache prior to surgery; condition with high risk of brain involvement (Arnold-Chiari malformation, acute intracranial hypertension) or local/systemic infection within 7 days prior to CSFD placement. Since 2020, our protocol has changed. Patients on long-term anticoagulation are now considered a contraindication for prophylactic CSFD. The perioperative management strategy for anticoagulant therapy (DOAC or vitamin K antagonist) was managed according to French guidelines.¹⁴ Subcutaneous heparin injection was stopped at least 12 hours and intravenous heparin infusion at least 4 hours before CSFD placement or removal.

The drain was placed the day before surgery by an experienced anesthetist to ensure an adequate delay between the lumbar puncture and intraoperative heparin injections. From 2016 to 2020, drains were occluded and patients were returned to the surgical ward without monitoring of the drain patency. Since 2020, our center has been equipped with an automated medical device (LiquoGuard®7 pump, Möller Medical GmbH, Fulda, Germany) to ensure minimal drainage and to confirm the patency of the drain. From this time, patients were systematically managed in intensive care unit (ICU) from the day before surgery. Cerebrospinal fluid pressure was continuously monitored during surgery. An individualized target approach was used according to the basal cerebrospinal fluid (CSF) pressure, with drainage only if there was an increase of 5 mmHg from baseline. Before the introduction of LiquoGuard®, we used passive drainage with manual manipulation of the drain (Codman EDS 3™ device). Since 2020, we have used an active drainage approach including a pressure-controlled CSFD with a flow rate between 10 and 20 ml/h in case of overpressure.

During the procedure, the range for MAP was 80 to 100 mmHg after stent graft deployment, and the target was higher at 100 to 120 mmHg in the event of SCI symptoms with norepinephrine infusion. The hemoglobin transfusion threshold was 10 g/dL. Due to lack of good evidence, we excluded rescue medications, such as mannitol, methylprednisolone, and naloxone from our protocol.

Postoperatively, patients were monitored in the ICU for at least 48 hours, with neurological assessment every 2 hours. If signs of SCI were diagnosed, a rapid subtraction of 20 ml of CSF was performed. If a second extraction maneuver was required, a curative CSFD was inserted. In the event of persistent neurological deficit, urgent angiotomography or magnetic resonance imaging (MRI) was planned to exclude spinal hematoma. After catheter removal, patients rested in the supine position for 6 hours and continued close neurological monitoring in the surgical unit for a further 48 hours.

Outcomes of Interest and Definition

Our primary outcome was to analyze the rate and severity of SCI (defined as any new motor or sensory deficit of the lower limbs in the absence of documented intracerebral hemispheric events) and to describe complications associated with CSFD, including hemorrhagic (puncture-site bleeding, subarachnoid hemorrhage, and subdural/epidural hematoma), mechanical (catheter occlusion, disconnection, or fracture), neurological (CSF leak and headache), or infectious (meningitis and abscess) complications.

We have adopted the classification of CSFD complications used by Kärkkäinen et al:¹³

Mild, indicating that the complication occurred but resolved spontaneously/with nominal intervention, without prolonging the hospital stay and without causing permanent disability;

Moderate, indicating that the complication required intervention, prolonged hospitalization by more than 24 hours or caused minor permanent disability;

Severe, indicating a complication requiring major surgical or medical intervention, associated with prolonged convalescence and prolonged/permanent disability, or resulting in death.

Our secondary outcomes were

primary technical success (defined as successful device delivery and deployment without surgical conversion, mortality, type I or III endoleak or graft limb occlusion);

primary clinical success (defined as clinical success without the need for additional or secondary surgical/endovascular procedures) and assisted primary clinical success (defined as clinical success with the use of an additional or secondary endovascular procedure);

the rate of overall postoperative complications (requiring surgery or not), endoleak and death (aortic-related or not).

Statistical Analysis

Data were collected in a computer database and analyzed using statistical software (IBM SPSS Statistics 21, 2012, IBM Corp.). Results are expressed as mean (standard deviation of mean) for continuous variables or counts (percentages) for categorical variables.

Results

Characteristics and Anatomical Features of the Patients

A total of 181 consecutive patients were included in the analysis (124 [68.5%] male; mean age=69.6 years, range=49–93), including 165 (84.2%) aneurysm repairs (130 atherosclerotic aortic aneurysms and 35 chronic aneurysmal dissections), and 16 (8.8%) complicated acute or subacute type B aortic dissections. Atherosclerotic aortic aneurysm extension was thoracic in 24 cases (18.5%), Crawford 1 in 1 (0.8%), Crawford 2 in 14 (19.8%), Crawford 3 in 13 (10%), and Crawford 4/pararenal in 78 (60%). The median preoperative aneurysm diameter was 62.3 mm (range=54–110 mm). Patients with dissection presented with 35 (19.9%) chronic aneurysmal dissections and 16 (8.8%) acute or subacute complicated type B aortic dissections. Of these 51 patients with complicated aortic dissection, 16 (31.3%) required coverage of the entire aorta and 35 (68.7%) required coverage limited to the descending thoracic aorta. Preoperatively, 10 (14.5%) patients had an occluded renal artery and 2 (1.1%) had an occluded hypogastric artery.

Previous aortic surgery was reported in 76 (42%) patients: 25 open repair or EndoVascular Aneurysm Repair (EVAR), 12 ascending aortic replacement, 16 aortic arch surgery, 9 frozen elephant trunks, 33 TEVAR or surgical treatment of the descending thoracic aorta. Approximately, one third of patients had coronary artery disease (n=5, 30.4%) and about half had hyperlipidemia (n=85, 47%); mean preoperative estimated glomerular filtration rate (eGFR) was 71.9 (range=22–175). Other comorbidities, ASA class, and RCRI score are shown in Table 1.

Table 1.

Demographics.

Baseline data	Total n = 181 (100%)	Patients at high risk* for SCIn = 69 (38%)	Patients at medium or low risk* for SCI n = 112 (62%)
Sex
Male	124 (68.5)	56 (81.2)	68 (60.7)
Female	57 (31.5)	13 (18.8)	44 (39.3)
Age	69.6 years (range=49–93)	67.3 (range=28–87)	71.1 (range=45–93)
Indication for surgery
Aneurysm	130 (69)	43 (62.3)	87 (77.7)
Crawford 1	1	1 (1.4)	—
Crawford 2	14	10 (14.5)	4 (3.6)
Crawford 3	13	11 (15.9)	2 (1.8)
Crawford 4/pararenal	78	9 (13)	69 (61.6)
Thoracic	24	13 (18.8)	11 (9.8)
Acute or subacute complicated type B aortic dissection	16 (8.8)	3 (4.3)	13 (11.6)
Chronic aneurysmal dissection	35 (19.9)	23 (33.3)	12 (10.7)
Thoracoabdominal repair	10 (28.5)	9 (13)	1 (0.9)
Thoracic	25 (71.4)	14 (20.2)	9 (8)
Shaggy aorta	3 (1.6)	2 (2.9)	1 (0.9)
ASA class
2	44 (24.3)	20 (29)	24 (21.4)
3	114 (63)	41 (59.4)	73 (65.2)
4	23 (12.7)	8 (11.6)	15 (13.4)
RCRI score
2	98 (54.1)	41 (59.4)	58 (51.8)
3	59 (35.6)	18 (26.1)	40 (35.7)
4	24 (13.3)	10 (14.5)	14 (12.5)
Comorbidities
Dyslipidemia	85 (47)	30 (43.5)	55 (49.1)
Hypertension	127 (70.2)	51 (73.9)	76 (67.9)
Alcohol use^a	4 (2.2)	2 (2.9)	2 (1.8)
Tobacco use^a	88 (48.6)	34 (49.3)	54 (48.2)
Diabetes	30 (16.6)	13 (18.8)	17 (15.2)
BMI>30	48 (26.5)	15 (21.7)	30 (26.8)
Atrial fibrillation	21 (11.6)	7 (10.1)	14 (12.5)
CAD	55 (30.4)	17 (24.6)	38 (33.9)
Stroke/TIA	16 (8.8)	5 (7.2)	11 (9.8)
COPD	35 (19.3)	15 (21.7)	20 (17.9)
CKD	18 (9.9)	7 (10.1)	11 (9.8)
Marfan/Ehlers-Danlos	7 (3.9)	4 (5.8)	3 (2.7)
PAD	12 (6.6)	3 (4.3)	9 (8)
Prior cancer	2 (1.1)	5 (7.2)	12 (10.7)
Anticoagulant therapy	22 (12.2)	5 (7.2)	17 (15.2)
Antiplatelet medication	78 (43.1)	29 (42)	49 (43.8)

Data are reported as n (%) or mean±standard deviation.

Abbreviations: ASA, American Society Anesthesiologist; AVK, antivitamin K; BMI, body mass index; CAD, coronary heart disease; CKD, chronic kidney disease (glomerular filtration rate <30 ml/min); COPD, chronic obstructive pulmonary disease; PAD, peripheral arterial disease; RCRI, Revised Cardiac Risk Index; TIA, transient ischemic attack.

Active user or weaned for less than 3 years

Risk of SCI is reported according to our Institutional protocol.

Surgical Procedures

Procedures included 74 (41%) TEVAR, 94 (52%) FEVAR, and 13 (7%) BEVAR. Staged interventions were performed in 31 (17.1%) cases, of which 23 patients were identified as high risk for SCI. Additional procedures were performed in 58 cases: 43 (23.7%) left subclavian artery targeting and revascularization, 7 (3.9%) hypogastric artery embolization, and 8 (4.4%) iliac branch devices. No additional renal or lumbar artery embolization was performed at the time of surgery. Twelve (11%) patients had perioperative hemodynamic problems and 18 (9.9%) required blood transfusion. Table 2 highlights all procedural data.

Table 2.

Intraoperative and Postoperative Data.

Intraoperative and postoperative data	Total 181 (100%)
Setting of the procedure
Urgency	4 (2.2)
Emergency	25 (13.8)
Elective	152 (84)
Type of procedure
TEVAR	74 (41)
FEVAR	94 (52)
BEVAR	13 (7)
Staged procedure	31 (17.1)
Procedure with operation time >3 hours	92 (50.8)
Additional procedures
Subclavian revascularization	43 (23.7)
Hypogastric artery embolization	7 (3.9)
Iliac branch device	8 (4.4)
Internal iliac artery surgical revascularization	2 (1.1)
Mean operative time	222 minutes (range=50–580)
Primary technical success	93.9%
Primary clinical success	94.5%
Assisted primary clinical success	100%
Need for surgical conversion	1 (0.6)
In hospital complications treated with medical management*	48 (26.5)
CV events	7 (3.9)
Pulmonary	15 (8.3)
Stroke/TIA	3 (1.7)
Kidney failure^a	10 (5.5)
Digestive events^b	3 (1.6)
Pulmonary embolism/deep venous thrombosis	2 (1.1)
Infection	7 (3.9)
Retrograde aortic dissection	1 (0.6)
In-hospital complication requiring surgery*
Hemorrhage	4 (2.2)
Access site (hematomas, pseudoaneurysm)	2 (1.1)
Lower limb compartment syndrome	4 (2.2)
Lower limb ischemia	2 (1.1)
Visceral ischemia	3 (1.7)
Endoleak type 1A/B	1 (0.6)
Endoleak type 3	1 (0.6)
Stent fracture	1 (0.6)
Death	29 (16)
In-hospital	6
Within 30 days	8
During follow-up period	21
Aortic-related	3
Length of stay
ICU	3.3 days (range=1–43)
Conventional unit	9.7 days (range=3–43)
Follow-up	491 days (range=1-3375)
Re-intervention rate within 30 days	4.4%
Re-intervention rate within 90 days	7.2%

Data are reported as n (%) or median.

Abbreviations: BEVAR, Branched EndoVascular Aneurysm Repair; FEVAR, Fenestrated EndoVascular Aneurysm repair; ICU, Intensive Unite Care; TEVAR, Thoracic EndoVascular Aneurysm Repair.

One more complication; CV, cardiovascular; TIA, transient ischemic attack.

Organic or non-organic failure. ^bEither cholecystitis, acute mesenteric ischemia or ischemic colitis.

Lumbar Drainage Parameters

Sixty-nine (38.1%) patients were identified as being at high risk for SCI according to our protocol (Table 3). In this group, 26 (37.7%) patients presented with dissection (3 acute and 23 chronic), while 43 (62.3%) underwent treatment for atherosclerotic aneurysm (13 thoracic, 1 Crawford 1, 10 Crawford 2; 2 Crawford 3, and 9 Crawford 4/pararenal; Table I). Twenty-five (36.2%) underwent intentional left subclavian artery (LSA) coverage and revascularization, 6 (8.7%) required hypogastric artery embolization (iliac branch devices were not technically feasible in these patients), 2 (2.9%) had severely shaggy aorta, and 44 (63.8%) had a history of previous aortic surgery. The CSFD was inserted prophylactically in 64 patients, was not technically feasible in 4 patients and had a contraindication in 1 patient (Table 4). Perioperative hemodynamic problems occurred in 8 (11.9%) of these patients, requiring transfusion in half of the cases (n=4). One drain was not working at the time of surgery (TEVAR) without consequence.

Table 3.

Risk Factors for SCI Identified in Our Protocol.

Risk factors for SCI	Patients
Risk factors for SCI	n	%
Need to cover the aortic segments from Th7 to Th12	37	53.6
Extended thoracic aorta coverage >200 mm	27	39.1
Occlusion of one internal iliac artery without revascularization	6	8.7
Thoracoabdominal coverage >40 mm above the coeliac trunk	29	24
Past experience of SCI during prior aortic repair	3	4.3
Emergent treatment (within 15 days from the onset of the aortic symptoms)	10	14.5
Prior open or endovascular thoracic or abdominal aorta surgery	33	47.8

Table 4.

CSF Drainage Data.

CSFD data	Total (n=181)
CSF drainage insertion	69 (38.1)
Timing of CSF drainage insertion
Preoperative	64 (92.7)
Postoperative	5 (7.3)
Failure in catheter positioning	4 (5.8)
CSF drainage use duration, days (range)	2.2 (1–3)
SCI	12 (6.6)
Cause of SCI
Related to the procedure / perioperative course	8 (66.6)
Related to the spinal drain	4 (33.4)
Severity of SCI symptoms
Mild	0
Moderate	5 (41.7)
Severe	7 (58.3)
Neurological deficit at the last follow-up	5 (41.7)
CSF drainage complications	12 (17.4)
Severity of CSF drainage complications*
Mild	5 (42)
Moderate	4 (33)
Severe	3 (25)

Data are reported as n (%) or median.

Abbreviations: CSF, cerebrospinal fluid; SCI, spinal cord injury.

CSF drain-related complications were categorized on the risk of morbidity to the patient: Mild = therapeutic unnecessary, Moderate = therapeutic (medical or surgical) necessary, Severe = threatened vital status.

Spinal Cord Injury

A total of 12 (6.6%) patients developed postoperative paraplegia or paraparesis. All were treated for atherosclerotic aneurysm (2 Crawford type 2, 3 Crawford type 3, 4 Crawford type 4, and 1 thoracic aneurysm) except 1 patient treated for acute type B aortic dissection. None of these events occurred during a first-stage procedure. Procedures included 5 FEVAR, 3 TEVAR, 3 BEVAR, and 1 FEVAR+iliac branch. Anatomical and procedural details are given in Table 5. These patients had several risk factors for spinal cord ischemia: 9 extensive aortic coverage or previous aortic surgery; 4 hemodynamic problems or transfusion in the perioperative period; 5 long procedure time; 2 shaggy aortas, and 1 hypogastric embolization. Of the 12 patients who experienced an SCI, only 7 of them were considered at high risk of SCI before surgery and received a prophylactic CSFD.

Table 5.

Clinical data of patients who have develop SCI.

Sex and age at the procedure	Indication for surgery and type of procedure	ASA class	Risk factor for SCI	Timing for CSFD insertion	Cause of SCI	Type of deficit and evolution*	Post-operative complication and cause of death
M, 71 yo	Type IV TAAA 4 FEVAR	2	Hypertension, COPD, need for hypogastric coverage, prolonged procedure, prior TEVAR	Preoperative	L2-L3 subdural hematoma subsequent CSFD insertion	SCI grade 2; total recovery	None
M, 84 yo	Type IV TAAA 4 FEVAR+IBG	3	Hypertension, COPD, CKD, hypogastric artery embolization, severe shaggy aorta, perioperative anemia and need for transfusion, prolonged procedure	Postoperative	Postoperative SCI	SCI grade 3C; no recovery	None
M, 86 yo	Type IV TAAA 4 FEVAR	3	Hypertension, emergent procedure, perioperative anemia and need for transfusion, perioperative hemodynamic issues	Postoperative	Hemodynamic issues	SCI grade 1; total recovery	Acute renal failure; death from mesenteric ischemia 1 year after surgery
M, 51 yo	Acute type B aortic dissection, TEVAR	4	COPD, emergent procedure, extended thoracic aorta coverage	Preoperative	Acute complication of dissection	SCI grade 3C; no recovery
M, 70 yo	Type IV TAAA 4 FEVAR	3	Hypertension, shaggy aorta	Postoperative	Postoperative SCI	SCI grade 1; total recovery	Urinary tract infection, need for lower limb fasciotomy
F, 79 yo	Type III TAAA 5 BEVAR	2	Hypertension, extensive aorta coverage, prolonged procedure, thoracoabdominal coverage > 40 mm above the coeliac trunk;	Preoperative	L2-L5 subdural hematoma subsequent to CSFD insertion	SCI grade 1; total motor recovery persistence of sphincter disorders	Pneumoniae
F, 65 yo	Type III TAAA 3 BEVAR	3	Hypertension, prior ascending aorta replacement, perioperative hemodynamic issues, prolonged procedure, TEVAR	Postoperative	Postoperative SCI	SCI grade 1; total recovery	None
M, 65 yo	Type III TAAA 4 FEVAR	3	Hypertension, prior aortic dissection, prior TEVAR, perioperative anemia, and need for transfusion, prolonged procedure	Postoperative	Postoperative SCI	SCI grade 1; total recovery	None
F, 68 yo	TYPE II TAAA, 2F/2 BEVAR	2	Extensive aorta coverage, prolonged procedure, TEVAR	Preoperative	Postoperative SCI	SCI grade 1; total recovery	None
F, 65 yo	Type II TAAA, 4 FEVAR	3	Need for subclavian artery coverage and transposition, prior TEVAR	Preoperative	Postoperative SCI	SCI grade 1; total recovery	None
F, 54 yo	Thoracic aneurysm, TEVAR	3	COPD, prior aortic dissection, surgical arch replacement, extensive aorta coverage	Preoperative	L2 L3 subdural hematoma subsequent to CSFD insertion	SCI grade 3C; no recovery	Respiratory failure
M, 70 yo	Thoracic aneurysm, TEVAR	3	COPD, prior surgical treatment of AAA, need to cover the aortic segments from Th8 to Th12	Preoperative	Subdural hematoma subsequent to CSFD insertion	SCI grade 3C; no recovery	Death from hemorrhagic stroke 7 months after surgery

Abbreviations: ASA, American Society Anesthesiologist; BEVAR, Branched EndoVascular Aneurysm Repair; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CSFD, cerebrospinal fluid drainage; SCI, spinal cord injury; FEVAR, Fenestrated EndoVascular Aneurysm Repair; IBG, iliac branch graft; TEVAR, Thoracic EndoVascular Aneurysm Repair.

Neurological outcomes are reported according to Oderich et al.¹⁵

In 4 of these 7 patients with prophylactic CSFD, spinal cord hemorrhage related to CSFD insertion or removal was the etiology of the injury. One hematoma was surgically evacuated with no improvement in symptoms. The factors identified with major CSFD complications were: curative anticoagulation 36 hours after drain removal (n=1), multiple punctures (n=1), platelet count <100 000 undiagnosed before drain removal (n=1), and bipolar disorder (n=2).

The remaining 5 patients were considered at intermediate risk and underwent CSFD as a salvage procedure in conjunction with transfusion and hemodynamic optimization. This resulted in complete regression of paraparesis in 4 of 5 patients.

Of the 8 patients who experienced SCI related to the procedure, 4 had transient paraparesis and 4 had paraplegia, 2 of which were permanent. Of the 4 patients with SCI related to CSFD, 1 had paraparesis and 3 had paraplegia, including 2 with permanent paraplegia. Overall, 5 of the 12 patients had permanent disability at the last follow-up (2 permanent paraplegia, 1 paraparesis and 1 sphincter impairment). Details are shown in Table 5.

CSFD-Related Complications

Twelve patients (17%) had CSFD-related complications, including 6 minors (2 headaches, 3 leaks, and 1 occlusion); 2 moderates (headaches requiring blood patch); and 4 majors (4 intradural hematomas as described above). No cases of cerebral hemorrhage, infection, or catheter fracture/disconnection were reported (Table 4).

Secondary Outcomes

Primary technical and clinical success, primary assisted clinical success and endoleak rates are shown in Table 2. Forty-eight (26.5%) patients experienced a postoperative complication that was managed medically; 18 (9.9%) developed a complication requiring surgery. The mean hospital stay was 9.7 days (range=3–43) and the mean ICU stay was 3.3 days (range=1–43). Mean follow-up was 17 months. The mortality rate was 4.4% at 30 days and 13.3% at 18 months. A total of 3 patients died of an aortic-related complication (1.6%) during the study period (Table 5).

Discussion

Spinal cord injury is a dreaded complication that can occur after endovascular or open repair of thoracic and TAA disease, mainly depending on the extent of aortic pathology.^1,4,5,12 Despite some well-recognized tools for spinal cord protection (staging, preservation of the left subclavian and hypogastric arteries, hemodynamic control), none of them completely prevent the risk of SCI.^14,16–20

In this article, we report the results after 5 years of implementing a proactive spinal protection protocol with selective use of prophylactic CSFD in a single center with a growing experience in spinal cord protection and extensive aortic repair. At our institution, CSFD is used in conjunction with a bundled spinal cord protection procedure according to a protocol that was introduced in 2016 and will be updated again in 2018 and 2020. It is well known that bundled perioperative clinical care protocols can improve neurological outcomes in endovascular aortic repairs.^8,21,22 Cerebrospinal fluid drainage is one of the components, but it remains controversial because it may be associated with a number of serious complications that are rarely reported and usually described only summarily after endovascular repair.¹³ Most studies have analyzed bundles only as fixed packages, rather than breaking down the contributions of individual components. This probably explains the difficulty in attributing the benefit of prophylactic CSFD. To improve selection, we proposed a pragmatic approach to the indication of CSFD by introducing a specific preoperative score that could clearly identify patients at higher risk of SCI. According to our experience and the literature, SCI is more common in patients with COPD, over 70 years of age or those with renal insufficiency, hypertension, and degenerative aneurysms.^21–23 However, all these “patient” factors appear to be of little relevance in defining a “high risk” category, as they are too frequently found and ultimately related to atherosclerosis, the main disease involved in degenerative vascular pathologies. Furthermore, perioperative hypotension (MAP<70 mmHg with or without blood loss) and perioperative transfusion are often associated with prolonged or urgent procedures and remain important unpredictable risk factors for SCI prior to surgery. On the other hand, the “classic” major predictable risk factors are the sacrifice of the left subclavian or hypogastric artery, extensive aortic coverage and shaggy aorta.^21,24–27

A 2-stage procedure for ischemic preconditioning of the spinal cord and improvement of the collateral network to reduce the rate of SCI is recommended in the case of long coverage.^12,20,28 Therefore, in our practice, we revascularize the subclavian artery and the hypogastric artery whenever coverage is required and revascularization is technically feasible. In a randomized controlled trial of open TAAAs, Coselli et al showed that the use of CSFD significantly reduced the incidence of SCI.²⁶ Similarly, in a recent meta-analysis of 12 245 patients including open TAAAs and descending thoracic aortic aneurysms, the pooled rates of postoperative outcomes were 5.7% (95% CI=4.3–7.5) and 3.0% (95% CI=2.1–4.2) for permanent and temporary SCI, respectively. Prophylactic CSFD was associated with a lower operative mortality rate (p<0.001).²⁹

In this setting, our main observations were:

A large number of patients are not eligible for prophylactic CSFD due to anatomical reasons; medication use or for procedures in an emergency situation;

When prophylactic insertion is performed, CSFD is used postoperatively in only a few cases.

The overall rate of CSFD-related complications is approximately 6.6%, with a severe presentation occurring in 1.6% of cases. In our article, one third of postoperative neurological disorders were drain-related SCI associated with intradural hematoma (Table 4). These results are similar to other large reports.²⁰

Previous studies have suggested that coagulation status influences the overall complication rate after intrathecal catheter insertion. To prevent puncture site bleeding and further subdural hematoma formation, the platelet threshold for drain insertion should be ≥ 100 000/mm³ with an international normalized ratio <1.3 and a normal activated partial thromboplastin time. It is also essential to stop antiplatelet drugs at least 7 days before the procedure, intravenous heparin at least 4 hours before the procedure and subcutaneous heparin 8 to 10 hours before the procedure, in accordance with the American Society of Regional Anesthesia and Pain Medicine recommendations for insertion of intrathecal catheters.³⁰

In a large multicenter analysis of daily practice in CSFD, the vast majority of respondents discontinued clopidogrel before drainage insertion, but only 7% of them also discontinued aspirin, as shown in our protocol.³¹ We recommend further safety of drain placement by excluding all patients treated with prior anticoagulant therapy. Extra care should be taken when it is claimed that patients will benefit from a strategy of prophylactic drainage.

The question of the best timing for CSFD (prophylactic or curative) and its effectiveness remains unanswered. Currently, there is a wide range of practice for CSFD in different institutions. Although prophylactic use is associated with a longer stay in the intensive care unit and restricted movement, the main advantage is that it ensures prompt treatment of any complications.^4,17,32 However, the patient may be exposed to drainage-related complications (up to 11% of cases), including: spinal hematoma in 0.8%, headache in 3.9%, intracranial hematoma in 1.5%, CSF leak in 0.9%, drain fracture in 0.1%, meningitis in 0.1%, puncture site bleeding in 2.1%, unspecified neurological deficit in 0.6%, and death in 0.3% of cases, without even knowing if CSFD is actually used postoperatively.^13,33 In addition, treatment of these complications (ie, sphenopalatine ganglion block or blood patch) can be more or less rapid and may fail or cause additional iatrogenic complications (infection, meningitis, arachnoiditis, seizures, hearing or vision loss, radicular pain, and neural deficits).^34,35

Aucoin et al showed that therapeutic CSFD was associated with worse long-term neurological outcomes (79% of patients with permanent paraplegia at discharge vs 54% of patients with prophylactic use) and overall survival (50±10% vs 71±9%) compared with prophylactic drains.³⁶ According to the literature, curative use of CSFD could be considered a valuable option only if insertion can be performed within 1 to 2 hours of the onset of SCI symptoms. Therefore, anesthetic procedures should be standardized, unnecessary sedation should be avoided, and regular neurologic examinations should be performed after the procedure.³⁷

We acknowledge several limitations of this study: it was retrospective, non-randomized and performed at a single institution. Given our limited patient population, a conclusive analysis of CSFD strategies will require multi-center collaboration to pool data for more reliable conclusions.

Conclusion

Paraparesis/paraplegia is a terrible complication after TEVAR and B/FEVAR. Standardized surgical and anesthetic protocols are essential to minimize the risk of SCI. The analysis of our practices has highlighted the risks associated with the prophylactic use of CSFD, especially in the presence of altered coagulation status, which would require further strengthening of the exclusion criteria for prophylactic drainage. Its use requires a better definition of patients at high risk of developing a catheter-related complication (long-term anticoagulation, hostile spine) and the need for strict control of the patient’s coagulation parameters, especially during withdrawal.

Footnotes

Author Contributions

Conception and design: OB, LB, BM

Analysis and interpretation: OB, LB, GG, BM

Data collection: OB, TCL, AH, BM

Writing the article: OB, LB, BM

Critical revision of the article: OB, LB, TLC, AH, GG, BM

Final approval of the article: OB, LB, TLC, AH, GG, BM

Statistical analysis: OB

Obtained funding: Not applicable.

Overall responsibility: BM

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: BM is a consultant for COOK Medical, GORE, Philips. LB recieved grant from Baxter ™.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Sponsoring Societies

None.

ORCID iD

Blandine Maurel

References

Coselli

Bozinovski

Lemaire

SA.

Open surgical repair of 2286 thoracoabdominal aortic aneurysms. JVS. 2013;57(2):521–524.

Zipfel

Zaefferer

Riambau

, et al. RESTORE II Study Group. worldwide results from the RESTORE II on elective endografting of thoracic aneurysms and dissections. JVS. 2016;63(6):1466–1475.

Kato

Motoki

Isaji

, et al. Spinal cord injury after endovascular treatment for thoracoabdominal aneurysm or dissection. Eur J Cardiothorac Surg. 2015;48(4):571–577.

Buth

Harris

Hobo

, et al. Neurologic complications associated with endovascular repair of thoracic aortic pathology: incidence and risk factors: a study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg. 2007;46(6):1103–1110.

Peppelenbosch

Vermeulen Windsant

Jacobs

Tordoir

Schurink

GW.

Open repair for ruptured abdominal aortic aneurysm and the risk of spinal cord ischemia: review of the literature and risk-factor analysis. EJVES. 2010;40:589–595.

Heidemann

Kölbel

Kuchenbecker

, et al. Incidence, predictors, and outcomes of spinal cord ischemia in elective complex endovascular aortic repair: an analysis of health insurance claims. J Vasc Surg. 2020;72(3):837–848.

Pasqualucci

Al-Sibaie

Vaidyan

KPT

, et al. Epidural corticosteroids, lumbar spinal drainage, and selective hemodynamic control for the prevention of spinal cord ischemia in thoracoabdominal endovascular aortic repair: a new clinical protocol. Adv Ther. 2020;37(1):272–287.

Scali

Kim

Kubilis

, et al. Implementation of a bundled protocol significantly reduces risk of spinal cord ischemia after branched or fenestrated endovascular aortic repair. J Vasc Surg. 2018;67(2):409.e4–423.e4.

Augoustides

Stone

Drenger

Novel approaches to spinal cord protection during thoracoabdominal aortic interventions. Curr Opin Anaesthesiol. 2014;27(1):98–105.

10.

Riambau

Böckler

Brunkwall

, et al. Editor’s choice management of descending thoracic aorta diseases. Eur J Vasc Endovasc Surg. 2017;53(1):4–52.

11.

Maurel

Delclaux

Sobocinski

, et al. The impact of early pelvic and lower limb reperfusion and attentive peri-operative management on the incidence of spinal cord ischemia during thoracoabdominal aortic aneurysm endovascular repair. Eur J Vasc Endovasc Surg. 2015;49(3):248–254.

12.

Juszczak

Murray

Koutsoumpelis

, et al. Elective fenestrated and branched endovascular thoraco-abdominal aortic repair with supracoeliac sealing zones and without prophylactic cerebrospinal fluid drainage: early and mediumterm outcomes. Eur J Vasc Endovasc Surg. 2019;57(5):639–648.

13.

Kärkkäinen

Cirillo-Penn

Sen

, et al. Cerebrospinal fluid drainage complications during first stage and completion fenestrated-branched endovascular aortic repair. J Vasc Surg. 2020;71(4):1109–1118.

14.

Lee

Marcantonio

Mangione

, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major non-cardiac surgery. Circulation. 1999;100:1043–1049.

15.

Oderich

Forbes

Chaer

, et al. Reporting standards for endovascular aortic repair of aneurysms involving the renal-mesenteric arteries. J Vasc Surg. 2021;73(1s):4S–52S.

16.

Albaladejo

Bonhomme

Blais

, et al. Management of direct oral anticoagulants in patients undergoing elective surgeries and invasive procedures: updated guidelines from the French Working Group on Perioperative Hemostasis (GIHP)—September 2015. Anaesth Crit Care Pain Med. 2017;36(1):73–76.

17.

Bisdas

Panuccio

Sugimoto

, et al. Risk factors for spinal cord ischemia after endovascular repair of thoracoabdominal aortic aneurysms. J Vasc Surg. 2015;61(6):1408–1416.

18.

Katsargyris

Is it really time to eliminate prophylactic cerebrospinal fluid drainage in TAAA endovascular repair. Eur J Vasc Endovasc Surg. 2019;57(5):649.

19.

Fillinger

Greenberg

McKinsey

, et al. Reporting standards for thoracic endovascular aortic repair (TEVAR). J Vasc Surg. 2010;52(4):1022–331033.

20.

Khan

Smalley

Nesvick

, et al. The use of lumbar drains in preventing spinal cord injury following thoracoabdominal aortic aneurysm repair: an updated systematic review and meta-analysis. J Neurosurg Spine. 2016;25(3):383–393.

21.

Keith

Jr Passman

Carignan

, et al. Protocol implementation of selective postoperative lumbar spinal drainage after thoracic aortic endograft. J Vasc Surg. 2012;55(1):1–8.

22.

Hnath

Mehta

Taggert

, et al. Strategies to improve spinal cord ischemia in endovascular thoracic aortic repair: outcomes of a prospective cerebrospinal fluid drainage protocol. J Vasc Surg. 2008;48(4):836–840.

23.

Zhang

Sun

, et al. Intercostal artery reconstruction: the simple and effective technique on spinal cord protection during thoracoabdominal aortic replacement. Ann Vasc Surg. 2016;34:62–67.

24.

Cambria

Davison

JK.

Regional hypothermia with epidural cooling for spinal protection during thoracoabdominal aneurysm repair. Semin Vasc Surg. 2000;13:315–324.

25.

Svensson

Patel

Robinson

, et al. Influence of preservation or perfusion of intraoperatively identified spinal cord blood supply on spinal motor evoked potentials and paraplegia after aortic surgery. JVS. 1991;13:355–365.

26.

Coselli

LeMaire

Köksoy

, et al. Cerebrospinal fluid drainage reduces paraplegia after thoracoabdominal aortic aneurysm repair: results of a randomized clinical trial. J Vasc Surg. 2002;35(4):631–639.

27.

Scali

Wang

Feezor

, et al. Preoperative prediction of spinal cord ischemia after thoracic endovascular aortic repair. J Vasc Surg. 2014;60(6):1481–1490.

28.

Etz

Weigang

Hartert

, et al. Contemporary spinal cord protection during thoracic and thoracoabdominal aortic surgery and endovascular aortic repair: a position paper of the vascular domain of the European Association for Cardio- Thoracic Surgery. Eur J Cardiothorac Surg. 2015;47(6):943–957.

29.

Khan

Naik

Hameed

, et al. Open repair of descending thoracic and thoracoabdominal aortic aneurysms: a meta-analysis. Ann Thorac Surg. 2020;110(6):1941–1949.

30.

Khemlani

Schurink

Buhre

, et al. Cerebrospinal fluid drainage in thoracic and thoracoabdominal endovascular aortic repair: a survey of current clinical practice in European Medical Centers. J Cardiothorac Vasc Anesth. 2022;36(5):1318–1325.

31.

Narouze

Benzon

Provenzano

, et al. Interventional spine and pain procedures in patients on antiplatelet and anticoagulant medications (second edition): guidelines From the American Society of Regional Anesthesia and Pain Medicine, the European Society of Regional Anaesthesia and Pain Therapy, the American Academy of Pain Medicine, the International Neuromodulation Society, the North American Neuromodulation Society, and the World Institute of Pain. Reg Anesth Pain Med. 2018;43(3):225–262.

32.

Khoynezhad

Donayre

Bui

, et al. Risk factors of neurologic deficit after thoracic aortic endografting. Ann Thorac Surg. 2007;83(2):S882–S889; discussion S890.

33.

Amabile

Grisoli

Giorgi

, et al. Incidence and determinants of spinal cord ischaemia in stent-graft repair of the thoracic aorta. Eur J Vasc Endovasc Surg. 2008;35(4):455–461.

34.

Chan

Desai

Hwang

NC.

Cerebrospinal fluid drains: risks in contemporary clinical practice. J Cardiothorac Vasc Anesth. 2022;36(8 pt):2685–2699.

35.

Giaccari

Aurilio

Coppolino

, et al. Peripheral nerve blocks for postdural puncture headache: a new solution for an old problem. In Vivo. 2021;35(6):3019–3029.

36.

Aucoin

Bolaji

Novak

, et al. Trends in the use of cerebrospinal drains and outcomes related to spinal cord ischemia after thoracic endovascular aortic repair and complex endovascular aortic repair in the Vascular Quality Initiative database. J Vasc Surg. 2021 Oct;74(4):1067–1078

37.

Czerny

Eggebrecht

Sodeck

, et al. Mechanisms of symptomatic spinal cord ischemia after TEVAR: insights from the European Registry of Endovascular Aortic Repair Complications (EuREC). J Endovasc Ther. 2012;19(1):37–43.