Results of a multidisciplinary spinal cord ischemia prevention protocol in elective repair of Crawford's extent I-III thoracoabdominal aneurysms by fenestrated and branched endografts

INTRODUCTION

Fenestrated and branched endografting (F/B-EVAR) is an established technique to treat thoracoabdominal aortic aneurysms (TAAAs) in patients at high surgical risk with specific anatomical characteristics^[1,2]. Satisfactory results in terms of technical and clinical success have been reported in the literature at mid-term follow-up, even in very challenging scenarios, such as urgent situations or in cases with previous aortic repair^[3].

Despite the progress in overall postoperative results, spinal cord ischemia/infarction (SCI) remains a possible catastrophic complication after F/B-EVAR for TAAAs, leading to a significant reduction in quality of life and survival^[4]. In a recent systematic review and meta-analysis, the pooled incidence of SCI after F/B-EVAR was found to be 13%^[5]. Previous studies reported an incidence of SCI up to 35%^[6], with a higher risk related to factors such as urgent/emergent repair, previous aortic surgery, Crawford’s extent I-III TAAAs, and loss of subclavian/hypogastric arteries^[7-9].

In the last decades, a number of preoperative, intraoperative and postoperative strategies including surgical, anesthesiological and medical adjuncts have been proposed in order to reduce the incidence of SCI after F/B-EVAR in TAAAs^[10-14]. CSFD^[15] and intraoperative neuromonitoring with SSEPs/MEPs^[1,16-17] have also been extensively investigated, but their efficacy is still debated in the literature.

The aim of the present study was to report the results of a dedicated multidisciplinary SCI prevention protocol, consisting of surgical, anesthesiological and neurological measures, for elective endovascular repair of Crawford’s extent I-III TAAAs by F/B-EVAR.

METHODS

RESULTS

A total of 104 patients underwent elective endovascular repair in the period of study, with 6 (6%) presenting with Crawford’s extent I, 51 (49%) with extent II, and 47 (45%) with extent III. Table 2 provides a summary of demographic and clinical characteristics of the population (71% male, mean age 73 ± 6 years). Aneurysm characteristics and procedural details are reported in Table 3. Forty-eight (46%) patients have had a previous aortic procedure, mostly a surgical aortic repair (39 out of 48 patients). All the procedures were performed under general anesthesia. A single-stage repair was performed in 20% of cases, with 80% undergoing a staged repair, according to endograft design, anatomical and clinical characteristics. A fenestrated device was used in 18% and a branched device in 68% of cases, with a custom-made device with fenestrations and branches chosen in the remaining 14% of cases. To ensure an adequate proximal and distal sealing zone, a thoracic endograft was deployed in 95 cases, 22 during a previously planned step and 73 simultaneously with the thoracoabdominal module release. Thirteen supra-aortic trunks debranching were also needed, and an iliac branch device was used in 18 patients. Prophylactic spinal drainage was performed overall in 90% of cases; and in 81% of staged repairs, it was also used in the second stage. The mean ICU stay was 5 ± 5 days and the mean hospital stay was 25 ± 22 days.

Postoperative events are reported in Table 4. Eight patients developed SCI: 2 transitory, with complete regression of symptoms, and 6 with different severity degrees of permanent deficits. Among the latest, 3 cases had permanent paraplegia. Details of the 8 patients with SCI are specified in Table 5. Five cases of cerebral hemorrhage were detected, 2 among SCI patients (one having a post-traumatic cerebral hemorrhage that occurred after the aortic repair). Postoperative cardiac and pulmonary morbidity were reported in 6 cases, respectively. A renal function worsening of any degree occurred in 21 (20%) of the patients, 3 requiring hemodialysis (2 permanently).

Three patients died within 30 postoperative days, while other 4 during a prolonged/complicated hospitalization (overall in-hospital mortality 7%). The mean follow-up was 30 ± 18 months. Overall estimated 3-year survival was 62%, with a significant difference in survival at 2 years of follow-up between patients with and without postoperative SCI (SCI: 18% vs. no-SCI: 69%; P < 0.001), as shown in Figure 1.

Figure 1. Kaplan-Meier estimate of survival in patients treated for Crawford’s extent I-III patients, with or without SCI.

DISCUSSION

The dedicated multidisciplinary SCI prevention protocol in elective F/B-EVAR for Crawford’s I-III TAAAs analyzed in this paper led to encouraging rates of SCI (8% overall SCI, 6% permanent impairment with 3% paraplegia).

The beneficial effect of a bundled protocol for SCI prevention was already shown by Scali et al., who compared the results of F/B-EVAR before and after the introduction of a dedicated protocol for SCI prevention, including cerebrospinal fluid drainage, blood pressure control, transfusion strategy, and pharmacological adjuncts (steroids, naloxone)^[23]. They found a significant reduction in SCI rate from 13% to 3% (P = 0.007), with even more significant results in Crawford’s extent I-III TAAA (19% vs. 4%, P = 0.004). Moreover, a subsequent beneficial effect on 1-year survival was obtained, with an increase from 90% to 99% after the introduction of the protocol (P = 0.05), although a possible influence by a combination of factors such as the natural learning curve of the surgeons may have occurred.

As a matter of fact, a study on Vascular Quality Initiative data published in 2021 by Aucoin et al. also showed a decrease in SCI rates over the study period (2014-2019), despite an unchanged use of prophylactic CSFD^[4]. This finding suggests that other measures included in the protocols over the years may contribute to better outcomes.

In our series, the combined 30-day/in-hospital mortality was significantly higher in patients with SCI (P:0.032). Moreover, patients with SCI had a lower survival rate than patients without SCI at follow-up (18% and 69% at 2 years, respectively).

Similar results were reported by Heidemann et al. In their multicenter retrospective cohort study including 877 patients treated with F/B-BEVAR for a juxta-/para-renal aneurysm or a TAAA, SCI occurred in 10.7% of cases^[24]. Among all the SCI cases reported, 37% occurred after 30 days from the endovascular treatment. In their study, SCI was not associated with a higher in-hospital/30-day mortality, but with later mortality (14.7% of 90-day mortality in patients presenting SCI compared to 1.1% of those without SCI, P > 0.05). The authors suggest that these results may be due to the effectiveness of the intensive care units, with a worse outcome occurring in the patients transferred to other clinical settings.

Our results failed to demonstrate a clear correlation between prophylactic CSFD and SCI (P:1.0).

The effectiveness of prophylactic CSFD has been recently questioned even in TAAA at high SCI risk^[4], due to the incidence of CSFD complications, usually divided into major and minor. Major complications include intracranial hemorrhage, spinal hematoma, meningitis, and CSFD fracture requiring neurosurgical intervention. The main minor complications are reflex hypotension during catheter insertion, spinal headache, minimal presence of blood in the CSFD catheter, and non-functional CSFD.

In the 2023 multicenter retrospective study, Marcondes et al. reported the results of 541 patients with TAAA extent I-III endovascularly treated without the use of prophylactic CSFD^[25]. The authors reported an overall incidence of SCI of 8%, with 2% of permanent paraplegia. A rescue CSFD was used only in 4% of all patients, with only 0.3% of major drain-related complications.

In 2018, Dijkstra et al. published a review of preventive strategies for SCI after thoracic and thoraco-abdominal aortic repair^[12]. They included 43 studies for a total of 7,168 patients, all retrospective cohort studies, mainly non-comparative, based on very heterogeneous populations of thoracic or thoracoabdominal aneurysms, dissections, penetrating aortic ulcers, and traumatic injuries. Overall, a specific SCI prevention protocol was used in about 77% of the studies included in the review. Generally, SCI is more often transient than permanent (5.7% vs. 2.2%, respectively), with the transient SCI rate ranging from 0.3% to 31% and the permanent SCI rate from 0.3% to 21%, without a time trend^[12]. Permanent SCI was 3.6% in the thoracoabdominal aneurysm subgroup at pooled analyses. The more counterintuitive result reported by this study was the high rate (8%) of SCI affecting TAAA repairs with CSFD. The Authors suggest that this may be related to a subgroup of patients at very high risk for SCI. Also counterintuitive was the finding of an association between a permissive endoleak and a rather high (5%) transient SCI rate. The final suggestion was to selectively use CSFD only in high-risk patients and to promote international multicenter prospective and high-quality studies to define a universally accepted preventive protocol for SCI.

To assess risks and benefits related to the use of selective CSFD in F/B-EVAR for Crawford’s extent I-IV TAAA, Kitpanit et al. analyzed 106 consecutive patients treated from 2014 to 2019 in a prospective physician-sponsored investigational device exemption study^[8]. Despite an overall low rate of SCI (3.8%), the authors found a high incidence of CSFD-related major complications, including spinal hematoma, subarachnoid and cerebellar hemorrhage, and a spinal drain fracture. There was a significant increase in intensive unit care and hospital length of stay for patients with CSFD. The authors concluded by discouraging a routine use of prophylactic CSFD and instead emphasized the need for prospective randomized trials on the use of CSFD in the endovascular treatment of TAAA.

Recently, the US Aortic Research Consortium (US-ARC)^[26] published the results of a 65-question survey on the methods and protocols to prevent SCI in high-risk patients. These high-risk patients were identified in the presence of Crawford’s extent I-III, a shaggy aorta, a previous EVAR or open infra-renal aortic repair, or an abnormal pelvic or vertebral perfusion. The 8 principal investigators differed in the answers principally on the timing of the resumption of antihypertensive medications, on the duration of the hemoglobin goals, and on the management of CSFD. Particularly, the investigators using prophylactic CSFD in high-risk patients for SCI were 6 of 8 (75%), with one of the 6 changing the practice during the study to only CSFD placement as a rescue maneuver for SCI onset. The US-ARC concluded with a consensus on the beneficial role of the CSFD for the prevention of SCI, although underlying the need for clinical trials to obtain rigorous scientific data on its preoperative prophylactic use.

On this point, Aucoin et al. reported worse neurologic outcomes and lower survival with the therapeutic CSFD for SCI symptoms onset compared with prophylactic CSFD^[26]. These findings again highlight the necessity of randomized controlled trials to compare prophylactic and therapeutic CSFD. Conversely, in the systematic review of Pini et al., the pooled SCI rate was 13% for symptomatic CSFD and 14% for prophylactic CSFD (P:0.87)^[5].

Among major complications, intracranial hemorrhage affected 5 patients in our series (2 also presenting SCI). Previous research conducted by our group^[27] already indicated that ICH after F/B-EVAR does occur mainly in patients with CSFD, and that a platelet count reduction greater than 60%, chronic kidney disease, and a liquor drainage higher than 50 mL are strongly associated with ICH, independently from the urgent or selective setting. Statistical analysis failed to show a difference between manual or auto-draining, which have always been selected according to anesthesiologists’ preference and availability of the devices. Although 50 mL seemed to be an unrealistic threshold already in 2021 when these findings were published, our group focused on strict and careful CSFD management.

Permissive endoleak was first described by Reilly and Chuter in 2010^[28], when they successfully reversed the symptoms of SCI in a patient who received an endovascular repair of a type II Crawford’s extent TAAA. The authors obtained a Ib endoleak by placing a balloon-expandable stent between the distal portion of the infrarenal endograft and the aortic wall. Three months later, a Palmaz stent was used to solve the endoleak and complete the procedure. Thereafter, this idea was developed^[29] to leave an interstep intentional endoleak, and to avoid abrupt cessation of spinal cord perfusion through intercostal and lumbar arteries, while enhancing collateral circuit formation. The main techniques to stage a TAAA endovascular repair are three: a first isolated thoracic endograft placement^[30], a minimally invasive segmental artery coil embolization^[31,32], and a TASP^[20]. Depending on aneurysm anatomical characteristics, in our series, a multi-staged TAAA repair was realized whenever possible. Not-staged cases were usually patients with a higher anesthesiological risk or with larger aneurysmal diameters. Among the preventive measures introduced by the present multidisciplinary protocol, treatment staging reached a statistical significance, in accordance with previous reports in the literature^[33,34]. Given the positive effect of staging in preventing SCI, the absence of TAAA rupture between the two steps in the presented series, and also considering that the second stage can often be performed under local anesthesia, further prospective studies about this preventive strategy should be performed, in order to validate these conclusions.

In their 2021 systematic review and meta-analysis on the occurrence of SCI after TAAA endovascular repair, Pini et al. reported a lower pooled SCI rate after staged compared with non-staged repair (9% vs. 18%, respectively; P = 0.02), independently from the method and timing of staging^[5]. More recently, Dias-Neto et al. published an analysis of the data from 24 centers of the ARC, with 1947 extent I-III TAAA electively treated with a staged approach from 2006 to 2021^[35]. The staging strategies (proximal thoracic endografting, TASP, MISACE, and combinations of these) allow lower rates of mortality and/or permanent paraplegia at 30 days or within hospital stay, and higher 1- and 3-year survival.

Recent studies focused on the results of different anesthesiological choices in the endovascular repair of TAAA give us several insights into the different available possibilities^[35-36]. A detailed discussion of every aspect of a SCI prevention protocol together with the anesthesiology team for each patient is fundamental for clinical success^[36].

In 2022, Monaco et al. compared first the short-term results of F/B-EVAR performed under general anesthesia with sedation with those performed under monitored anesthesia care (MAC) in addition to local anesthesia, finding that the type of anesthesia seemed to have no effect on procedure success, perioperative morbidity, or mortality in patients undergoing F/BEVAR, despite a higher need of inotropes/vasopressors to treat intraoperative hypotension with general anesthesia^[37]. In 2023, Monaco et al. compared the results of F/B-EVAR under MAC with remifentanil-based sedation with those using dexmedetomidine instead, finding a worse patient satisfaction with the latter. Moreover, remifentanil was associated with less hemodynamic effect than dexmedetomidine^[38].

Considering the high complexity of F/BEVARprocedures and their potential long surgical time, general anesthesia was used in all cases reported in the present study.

Concerning intraoperative SSEPs/MEPs monitoring, no large randomized controlled trials are currently available, at the best of our knowledge. In a retrospective review of 1,214 thoracic and TAAA^[16] treated either in open fashion or endovascularly between 2000 and 2013 in 12 Japanese centers, 631 patients received intraoperative MEPs monitoring and the outcome was compared with the outcome of 583 patients treated without neuromonitoring. MEPs failed to improve the outcome. The low number of cases performed under SSEPs/MEPs until 2022 in our department precludes a meaningful consideration of its role in the secondary prevention of SCI.

This study has limitations primarily concerning its retrospective nature, which includes potential biases related to data collection and incomplete medical records, and the small size of SCI group that may not fully represent the broader SCI population, potentially limiting general conclusions. Further research with larger and more diverse SCI groups is necessary to validate and extend these findings.

In conclusion, dedicated multidisciplinary SCI prevention protocol in elective F/B-EVAR for Crawford’s I-III TAAAs is feasible and safe, with encouraging rates of SCI (8% overall SCI, 6% permanent impairment with 3% paraplegia). The 30-day mortality (3%), cardiopulmonary morbidities (6%), and dialysis rate (3%) were satisfactory, as well as the estimated survival at 3 years (62%). Patients with SCI had a significantly lower survival (18% vs. 69%) at 2 years.

Further high-quality scientific data are needed to define the role of prophylactic or therapeutic CSFD. Although defining the efficacy of individual SCI prevention measures is not easy, treatment staging has been widely associated with lower rate of SCI.