Novel approaches to_spinal_cord_protection_during.15

C

REVIEW

CURRENTOPINION Novel approaches to spinal cord protection during

thoracoabdominal aortic interventions

opyright © Lippincott Will

www.co-anesthesiology.com

a b c
John G.T. Augoustides , Marc E. Stone , and Benjamin Drenger
Purpose of review

Spinal cord ischemia after thoracoabdominal aortic interventions is a devastating complication because itsignificantly worsens the perioperative morbidity and mortality. Long-term outcome is also affected becauseof medical complications which are directly related to the neural deficits. Paraplegia has significantmedical, social, and financial aspects. Limited mobility, the need for assistance in activities of daily living,makes paraplegia an important target for prevention. An understanding of spinal cord blood supply, riskfactors for spinal ischemia, and strategies for spinal cord rescue in this setting can help minimize thenegative outcome effects of this important complication.

Recent findings

The vascular supply of the spinal cord is via an extensive collateral arterial network with multiple auxiliaryarterial supplies. Risk factors for spinal cord ischemia include extensive aortic repair, prior aortic repair,spinal cord malperfusion on clinical presentation, systemic hypotension, acute anemia, prolonged aorticclamping, and vascular steal. Spinal rescue strategies include systemic hypothermia, endovascular aorticrepair, permissive systemic hypertension, cerebrospinal fluid drainage, pharmacologic neuroprotection,and intensive neuromonitoring.

Summary

The progression of spinal cord ischemia after thoracoabdominal aortic interventions can frequently bearrested before irreversible infarction results. This spinal cord rescue depends on the early detection andimmediate multimodal intervention to maximize spinal cord oxygen supply. The devastating outcomesassociated with spinal infarction in this setting offset the risks and knowledge gaps currently associated withcontemporary interventions.

Keywords

cerebrospinal fluid drainage, collateral arterial network, evoked potentials, hypothermia, spinal cord ischemia

aDepartment of Anesthesiology and Critical Care, Perelman School ofMedicine, University of Pennsylvania, Philadelphia, Pennsylvania,bDepartment of Anesthesiology, Icahn School of Medicine, Mount SinaiMedical Center, New York, New York, USA and cDepartment of Anes-thesiology and Critical Care Medicine, Hadassah-Hebrew UniversityMedical Centers, Jerusalem, Israel

Correspondence to John G.T. Augoustides, MD, FASE, FAHA, AssociateProfessor, Cardiothoracic Section, Anesthesiology and Critical Care,Dulles 680, HUP, 3400 Spruce Street, Philadelphia, PA 19104-4283,USA. Tel: +1 215 662 7631; fax: +1 215 349 8133; e-mail: [email protected]

Curr Opin Anesthesiol 2014, 27:98–105

DOI:10.1097/ACO.0000000000000033

INTRODUCTION

Spinal cord protection remains clinically importantafter thoracoabdominal aortic interventions (TAAIs)because spinal cord ischemia is still common andsignificantly worsens the perioperative morbidityand mortality [1,2,3

&

]. Spinal cord ischemia is man-ifested in the conscious patient by the developmentof lower extremity weakness or during surgeryby the attenuation of spinal cord signals withmonitoring of somatosensory-evoked potentialsand motor-evoked potentials [4,5,6

&&

]. The aorticrepair techniques in the contemporary era can beopen, endovascular, or hybrid. Total endovascularrepair may include the use of fenestrated orbranched endovascular stent components to notonly exclude the aneurysm, but also to preserveaortic branch perfusion. Hybrid TAAIs typicallyinvolve nonfenestrated endovascular repair ofdiseased aorta combined with open surgical

iams & Wilkins. Unautho

transposition of aortic branches to preserve criticalorgan perfusion [7,8]. This review is focused on therecent approaches to protect the spinal cord duringTAAI from an anesthesiology perspective. It willhighlight the risk factors for spinal cord ischemiaand the principal spinal cord preservation strategies.

rized reproduction of this article is prohibited.

Volume 27 � Number 1 � February 2014

mailto:[email protected]

mailto:[email protected]

KEY POINTS

� Spinal cord ischemia still occurs afterthoracoabdominal aortic interventions, regardless of thetechnique or aortic disease. It is considered adevastating perioperative complication.

� Spinal cord ischemia occurs when oxygen supply isinsufficient to meet the neural oxygen demand.

� The detection of spinal cord ischemia relies onneuromonitoring with evoked potentials in theanesthetized patient and on neurologic examination inthe awake patient.

� The vascular supply to the spinal cord is via the anteriorand posterior spinal arteries, that are supported by acollateral arterial network, which has cephalic, central,and caudal inputs.

� The risk factors and therapeutic interventions for spinalcord ischemia after thoracoabdominal aorticinterventions can largely be understood in terms of theireffects on the spinal collateral arterial network andspinal cord perfusion pressure.

� Rescue from spinal cord ischemia in thoracoabdominalaortic interventions is frequently possible with amultimodal protocol.

Novel approaches to spinal cord protection Augoustides et al.

RISK FACTORS FOR SPINAL CORDISCHEMIA: THE IMPORTANCE OF THESPINAL COLLATERAL ARTERIALNETWORK

The cause of spinal cord ischemia after TAAI is a netoxygen debt in the spinal cord, where neural oxygensupply is hampered by the surgical intervention.Spinal cord protection after TAAI depends on anunderstanding of the risk factors for spinal cordischemia, summarized in Table 1. An importantmechanism underlying these risk factors is the com-promise of the spinal collateral arterial network(SCAN), as outlined in the comments in Table 1.The SCAN concept has resulted in large part fromthe bench and bedside research by Griepp and col-leagues [9,10].

In 1882, Adamkiewicz described the spinal cordvascular plexus of the anterior and posterior spinalarteries with augmentation from cephalic, central,and caudal input [5,9,10]. Griepp et al. showed thatperfusion in these spinal arteries is supported bycomplex and intertwined collateral network, withcephalic contribution to the SCAN from thebrachiocephalic arteries and with significantcontribution from the vertebral arteries. The centralarterial input to the SCAN is from multiple seg-mental aortic branches such as the intercostaland lumbar arteries. This central arterial supply to

Copyright © Lippincott Williams & Wilkins. Unau

0952-7907 � 2014 Wolters Kluwer Health | Lippincott Williams & Wilk

the SCAN is scarcely dispersed in the lower thoracicarea, but frequently includes a large arterial branchat the level of the lower thoracic called the arteryof Adamkiewicz or the arteria radicularis magna.This augmented supply from this large segmentalartery is vital, given the small diameter of theanterior spinal artery in the mid-to-lower thoracicarea (T8 to L1). The caudal arterial supply to theSCAN is from the internal iliac arteries andtheir branches. The SCAN concept explains thatspinal cord perfusion stems from multiple intercon-nected sources that require adequate perfusionpressure and vascular tone throughout the arterialnetwork.

STRATEGIES FOR SPINAL CORDPROTECTION

The contemporary strategies for perioperative spinalcord protection can frequently be understood astherapeutic consequences of the SCAN concept.Although they are summarized in Table 2, they willnow be more fully explored in this section.

Deep hypothermia

Deep hypothermia for spinal cord protection is anestablished technique in TAAI as it minimizes theneural oxygen demand during aortic reconstructionwhen oxygen delivery is compromised [11

&&

,12&

].The definition of deep hypothermia by the recentexpert consensus is a systemic temperature rangeof 14.1–20.0 degrees Celsius, ideally measured atthe nasopharynx [13

&

]. In addition to spinal cordprotection, deep hypothermic circulatory arrest(DHCA) offers multiple advantages in open TAAIsuch as minimal dissection of periaortic tissues,elimination of the requirement for sequential aorticclamping, easy access to the aortic arch, and excel-lent visceral organ protection [11

&&

,12&

]. In a recentsingle-center evaluation of open thoracoabdominalaneurysm repair (n¼243: Crawford extent I 26%;Crawford extent II 40%; Crawford extent III 34%)utilizing DHCA, the incidence of spinal cord ische-mia was 5.3% (13 patients: nine with paraplegia andfour with paraparesis) [14]. Emergency surgery wassignificantly associated with spinal cord ischemiacompared with elective surgery (16.7 vs. 3.9%;P¼0.04) [14]. Fehrenbacher and colleagues havealso recently demonstrated in a similar large con-temporary series (n¼343) with DHCA that theincidence of spinal cord ischemia was 1.1–3.2%,depending on the aortic disease [15,16]. These datademonstrate that experienced perioperative teamscan achieve excellent spinal cord protection inextensive open TAAI with DHCA.

thorized reproduction of this article is prohibited.

ins www.co-anesthesiology.com 99

C

Table 1. Risk factors for spinal cord ischemia during thoracoabdominal aortic interventions

Risk factors Comment

Complicated aortic dissection with visceral malperfusion,e.g., acute type B dissection

In acute complicated aortic dissection, there may be spinal cordischemia, renal and intestinal ischemia, because of regionalmalperfusion due to aortic branch occlusion or exclusion from thedissection process.

Total extent of thoracoabdominal repair, e.g., CrawfordExtent II aneurysm repair has a very high risk of spinal cordcompromise

The risk of spinal cord ischemia increases with the extent of aorticrepair because of greater loss of segmental arterial spinal cordsupply.

Prior thoracoabdominal aortic segment repair, e.g., openrepair of abdominal aortic aneurysm; endovascular repairof the descending thoracic aorta

Prior aortic repair aggravates the risk of spinal cord ischemiabecause of the loss of spinal cord arterial segmental supplyfrom the prior procedure. In this setting, the spinal cord hascompromised reserve of its collateral arterial network.

Extent of preservation of spinal segmental arterial supply,e.g., intercostals arteries; lumbar arteries

The risk of spinal cord ischemia may be reduced if fenestratedendograft is used, or in open aortic repair if intercostal arteriesare reimplanted into the aortic graft

Duration of aortic clamping In the absence of distal aortic perfusion, the duration of aorticclamping to allow open aortic replacement often correlated withthe duration of spinal cord hypoperfusion and hence ischemia.Distal aortic perfusion techniques will not completely mitigate therisk of spinal cord ischemia.

Acute anemia Significant acute anemia is typically associated with severe bleed-ing. Acute anemia aggravates spinal cord ischemia by impairingoxygen delivery as a consequence of low hemoglobin mass.

Systemic hypotension Hypotension or circulatory collapse significantly compromises spinalcord perfusion and thus may precipitate spinal cord ischemia fromdecreased oxygen delivery.

Systemic vasodilation with vascular steal Vasodilators such as sodium nitroprusside have been utilized forcontrol of hypertension associated with aortic clamping. Thesystemic vasodilation resulted in vascular steal, compromisingspinal perfusion pressure to precipitate spinal cord ischemia. Backbleeding from open branches into the surgical field in open repairwill also cause the steal phenomenon.

Adapted with permission from [5] (no copyright required).

Cardiovascular anesthesia

Thoracic endovascular aortic repairExtensive meta-analysis has suggested that acrossdiverse aortic diseases, thoracic endovascular aorticrepair (TEVAR), per se, compared with open repairsignificantly decreases the risk of spinal cord ische-mia in TAAI [5,7,8]. A recent analysis of TEVAR vs.open repair in DeBakey Type III chronic dissections(n¼89) from two experienced medical centers dem-onstrated that TEVAR significantly reduced the risksof perioperative mortality (0 vs. 10.3%) and spinalcord ischemia (0 vs. 12.1%) [17]. Contemporarymeta-analysis of hybrid TAAI has demonstrated invery high-risk patients (n¼660: 19 studies) thathybrid techniques may reduce mortality and spinalcord ischemia, although the investigators con-cluded that further high-quality trials are indicatedto clarify the extent of these outcome advantages[18].

A large single-center analysis of TAAI (n¼406)over 11 years demonstrated an unexpectedly lowincidence of spinal cord ischemia, namely 2.7%

opyright © Lippincott Williams & Wilkins. Unautho

100 www.co-anesthesiology.com

with a permanent neurological deficit of only1.5% [19

&&

]. Independent risk factors for spinal cordischemia in this series were consistent with Table 1:previous abdominal aortic aneurysm repair [oddsratio 4.8 (OR); 95% confidence interval (CI) 1.4–17.2; P¼0.026], extensive coverage of the descend-ing thoracic aorta (OR 3.6; 95% CI 1.1–12.3;P¼0.038), and implantation of thoracoabdominalbranched or fenestrated grafts (OR 9.5; 95% CI 1.2–50.6; P¼0.032) [19

&&

]. The investigators postulatedthat the extensive arterial network, the SCAN, asexplained by Griepp and colleagues, is the likelyexplanation for the low incidence of spinal cordincidence in their reported TEVAR experience.Although TEVAR results in coverage of central spi-nal segmental arterial supply, back bleeding is pre-vented, and in this fashion prevents vascular steal inthat part of the vascular network. Furthermore, theinvestigators pointed out that collateral perfusionfrom the paravertebral muscle compartment willalso augment spinal segmental perfusion in the



Table 2. Spinal cord protection strategies during thoracoabdominal aortic interventions

Risk factors Comment

Systemic hypothermia, e.g., deep hypothermic circulatoryarrest; moderate hypothermia with left heart bypass; mildhypothermia with native circulation

Hypothermia reduces spinal oxygen demand and thus extends itstolerance of limited oxygen supply. The aortic repair can beconducted under deep hypothermia with circulatory arrest,moderate hypothermia with circulatory support or mildhypothermia with native circulation. Permissive mild hypothermiashould be induced with cautious to avoid the development ofarrhythmias. Deep hypothermia maximizes spinal cord protectionfrom cooling because of maximal suppression of neuralmetabolism and hence minimizes oxygen demand.

Thoracoabdominal endovascular repair, including hybridtechniques, e.g., endovascular stenting with fenestratedor branched grafts; endovascular stenting with opendebranching procedures

Endovascular aortic repair lowers the risk of spinal cord ischemiabecause of beneficial effects on the spinal collateral arterialnetwork. These benefits may be further enhanced if the extensiveaortic procedure can be staged.

Systemic hypertension, e.g., maintaining the mean arterialpressure between 80 and 100 mmHg with titratedintravenous norepinephrine infusion

Permissive systemic hypertension increases spinal cord perfusionpressure. The technique should be applied both, during surgeryand on arrival to the intensive care unit. It is important during thistherapy to monitor for adequate cardiac output and bleeding fromdisrupted suture lines.

Cerebrospinal fluid drainage (CSF), e.g., lumbarsubarachnoid catheter

Drainage of CSF increases spinal cord perfusion pressure. The safeconduct of this intervention is essential to minimize seriouscomplications.

Pharmacologic neuroprotection These drugs increase neural tolerance of ischemia. The supportingevidence is weak at best. Options include steroids, papaverine,lidocaine, magnesium and minocycline.

Intensive neuromonitoring e.g.,. neurologic examination;somatosensory-evoked potentials; motor-evoked potentials

Early detection of spinal cord ischemia is a clinical emergency. Aspinal rescue protocol should be urgently implemented in order torecover spinal cord function.

Adapted with permission from [5] (no copyright required).


TEVAR segment. This muscle compartment is typi-cally perfused from the second dorsal branch of thespinal segmental artery, and thus provides bloodsupply to the segmental arterial branches at theiraortic origins where the TEVAR procedure was per-formed [9,19

&&

,20]. Given the very low incidence ofspinal cord ischemia over time in their series, theinvestigators gradually abandoned routine utiliz-ation of cerebrospinal fluid (CSF) drainage in theirTEVAR practice, a principle that will be discussedfurther in the subsequent section of this article[19

&&

]. Given the beneficial effects of TEVAR onmorbidity and mortality in TAAI, it is likely thatthis therapeutic approach will significantly expandin the near future.

Maintenance of spinal cord perfusionpressure: the importance of permissivehypertensionSpinal cord perfusion pressure is defined as thedifference between systemic mean arterial pressureand CSF pressure or right atrial pressure whichever isgreater [2,3

&

,4,5]. Consequently, any increase inright atrial pressure to pressures higher than CSFpressure will decrease the spinal cord perfusionpressure. Raised right atrial pressure may result from



increased preload and increases in intrathoracicpressure because of positive end-expiratory pressure.Furthermore, if mean arterial pressure is elevated,there will be a higher spinal cord perfusion pressureand enhanced perfusion of the spinal cord throughthe SCAN. This principle is central in the perioper-ative management of patients undergoing TAAI topreserve spinal cord perfusion and protect againstspinal cord ischemia. An important tenet of modernspinal cord protection during TAAI (whether openor endovascular) includes intentionally maintain-ing the perioperative mean arterial blood pressure inthe high physiologic range, namely 80–100 mmHg[21,22

&&

].Maintenance of high pressure in the collateral

spinal network for at least 24–48 h postoperatively(or longer if there are apparent neurological deficits)is important to the success of a spinal cord protec-tion strategy, as progressive postoperative remodel-ing of the SCAN has been demonstrated within 48 h[9,20]. These acute adaptations in the SCAN includeincreases in the diameter of the anterior spinalartery and dilation of the epidural arterial networkwithin 24 h, and a realignment of the arteriolesparallel to the spinal cord by 5 days postoperatively[9,20]. These adaptations in the SCAN allow acute



C


expansion of its capacity for enhanced spinal cordperfusion. Permissive perioperative systemic hyper-tension augments spinal cord perfusion afterTAAI, giving the SCAN the critical time to expandcapacity to compensate for the loss of segmentalarterial input resulting from the aortic repair[19

&&

,20,21,22&&

,23]. This strategy acts as part of abridge to recovery of the SCAN in the perioperativeperiod.

Drainage of cerebrospinal fluidPerfusion pressurization of the SCAN is often insuf-ficient to fully protect the spinal cord and preventparaplegia. Perioperative drainage of CSF to maxi-mize spinal cord perfusion pressure (by minimizingthe resistance to afferent spinal cord blood supply) istypical in modern TAAI. The 2010 American HeartAssociation guidelines strongly recommend CSFdrainage (Class I, Level B evidence) in TAAI withhigh risk of spinal cord ischemic injury [23].

In practice, although CSF drainage is primarilyutilized in high-risk patients, the final decisiondepends on the team discussion and institutionalpractice. Proactive CSF drainage as a routine part of aperioperative protocol in TAAI with TEVAR hasrecently been demonstrated in a single-center study(n¼94, 2005–2012) to be very effective with a 1.1%incidence of spinal cord ischemia [24]. A secondapproach, namely selective CSF drainage for post-operative symptomatic spinal cord ischemia afterTEVAR, was associated with a 1.4% rate of perma-nent spinal cord ischemia [25]. A third approach,namely selective preoperative CSF drainage in high-risk TEVAR patients, was evaluated in a recent largesingle-center series (n¼381, 2002–2012) [26]. Theincidences of permanent and temporary spinal cordischemia in this series were 1.8 and 4.7%, respect-ively [26]. A single-center analysis of spinal cordischemia after TEVAR (n¼424, 2002–2010) againconfirmed the low incidence of 2.8% with a 75%rate of complete recovery, with early diagnosis andprompt institution of permissive hypertensionalone or with CSF drainage [26]. In multivariateanalysis, chronic renal insufficiency was the onlyindependent predictor of spinal cord ischemia (OR4.39; 95% CI 1.2–16; P¼0.029) [26]. Further trialsare required to confirm renal dysfunction as a pre-dictor for spinal cord ischemia after endovascularTAAI.

A systematic review (n¼4936, 46 studies)recently evaluated whether preoperative CSF drain-age in TEVAR reduces the risk of spinal cord ische-mia [27]. The incidence of spinal cord ischemia inthis meta-analysis was 3.89% (95% CI 2.95–4.95%),with no clear reductions evident with routine pre-operative drainage, selective preoperative drainage,



or no preoperative drainage [27]. An updatedCochrane meta-analysis evaluated CSF drainage inopen TAAI for aortic aneurysm (n¼287, threerandomized controlled trials before 2005) [28

&

].The overall benefit of CSF drainage in this limiteddataset was an 80% reduction in the relative risk ofspinal cord ischemia. A meta-analysis demonstratedan OR for CSF drainage of 0.48 (95% CI 0.25–0.92)[28

&

]. In summary, although CSF drainage is indi-cated in most extensive open TAAI, its role in endo-vascular TAAI is evolving, given the significantlylower risk of spinal cord ischemia. Current guide-lines recommend selective preoperative CSF drain-age in patients at high risk for spinal cordcompromise after TEVAR, but this will depend onthe institutional protocol [22

&&

,23]. CSF drainage isthe mainstay of rescue therapy for spinal cord ische-mia after TAAI, especially if there is incompleterecovery after several hours of permissive hyper-tension [22

&&

,23].

MANAGEMENT OF CEREBROSPINALFLUID DRAINAGE

Regardless of the specific protocol, the widelyaccepted goal is to maintain CSF pressure less than10–15 mmHg [2,5,29

&&

,30&&

,31]. Although this CSFpressure goal is widely accepted, it should also betitrated to clinical effect and the overall trend inspinal cord perfusion pressure. For example, if theawake patient has normal lower extremity musclepower, then CSF pressure management becomes lessimportant because at this timepoint the spinal cordis intact.

Once a CSF drainage catheter has been placed,the CSF pressure can be maintained at a defined goalby monitoring the CSF pressure (either continu-ously or intermittently) and intentionally drainingCSF until either the desired pressure is achieved or aspecific volume has been drained. This procedure isof particular importance in open TAAI, as appli-cation of aortic cross clamp (AXC) may induce asharp rise in CSF pressure. It occurs in spite ofeffective distal perfusion via left-sided bypass andeffective cardiac preload reduction. A possible expla-nation for the increase in CSF pressure is that AXCinduces a sympathetically mediated vasoconstric-tion in the systemic and spinal vasculature. Thisincreased tone in the thin-walled spinal veins maydecrease the critical closing pressure needed toprovoke collapse of the radiculospinal veins as theypass through the dura, with consequent venousengorgement and increase in CSF pressure [32].Possible explanations for the increase in CSF pres-sure after stent graft deployment might be attrib-uted to the same cause of sympathetic spinal




stimulation or to the local acidosis-induced veno-constriction, which occurs in the presence ofinadequate spinal cord perfusion.

(1)

Co

0952

If the CSF pressure remains above goal followingdrainage, a decision needs to be made if morewill be drained or other maneuvers performed tooptimize spinal cord perfusion pressure.Though the amount that can ‘safely’ be drainedat a given time has not conclusively been deter-mined, many experienced centers are limitingdraining to a specified volume of CSF each hour,but limit hourly maximum drainage to less than25 ml to avoid complications such as intracra-nial hematoma from torn subdural bridgingveins [29

&&

,30&&

,33,34].
(2) Setting the drainage unit at a specific height on a
bedside pole and allowing the CSF to drain freelyas needed. For example, if one levels the drainageunit to 13 cm above the phlebostatic axis, CSFwill continue to drain as long as CSF pressureexceeds 10 mmHg (1.3 cm H2O¼1 mmHg). Theadvantage of this common method is the auto-matic maintenance of the desired CSF pressure,but profound vigilance is required where thisprotocol is in use because a serious disadvantageis the potential for excessive drainage (e.g., if thedrainage unit falls to the floor), which may pre-dispose to serious complications [30

&&

,34].

Prior to discontinuing and pulling out the CSFdrainage, capping off the drain for 24 h prior to theremoval of the catheter may allow the CSF pressureto normalize and ascertain whether the patient isfree from spinal cord ischemia as spinal cordperfusion pressure subsequently decreases [27].Although this approach is consistent with ourunderstanding of the SCAN, there are sparse datato support this practice.

COMPLICATIONS OF CEREBROSPINALFLUID DRAINAGE

The complications of CSF catheter insertion andsubsequent drainage include catheter fracture,infection, CSF leak, abducens nerve palsy, neuraxialhematoma, and intracranial hematoma [30

&&

,34].One of the feared complications of CSF drainageis bleeding, with subsequent potential epidural orsubdural hematoma. Clearly, a frankly ‘bloody tap’during attempted CSF drain placement will necessi-tate postponement of the aortic repair, but subduralbleeding can also occur as a result of CSF drainageitself because of the tearing of subdural bridgingveins. Excessive CSF drainage was associated withintracerebral hematomas and with significant

pyright © Lippincott Williams & Wilkins. Unau

-7907 � 2014 Wolters Kluwer Health | Lippincott Williams & Wilk

morbidity and mortality [33]. A recent large analysisfrom a high-volume center (n¼504, 2005–2009)demonstrated a 2.8% incidence of intracranial hem-atoma after CSF drainage for open TAAI: 72% ofthese cases were subdural hematomas [30

&&

].Furthermore, the incidence of postdural punctureheadache was 9.7%, of whom 34.6% required epi-dural blood patch for clinical resolution [30

&&

]. Inmultivariate analysis, connective tissue disorder wasan independent predictor for postdural punctureheadache (OR 3.08; 95% CI 1.33–7.13). The inves-tigators concluded that the risk of intracranial hem-atoma was modest, and that epidural blood patchshould be considered early in the management ofpostdural puncture headache [30

&&

].In a smaller, recent, Swedish experience (n¼84,

2009–2012), CSF drainage in endovascular TAAI wasassociated with a 3.6% incidence of serious bleedingcomplications (one epidural and two subdural hem-atomas, two of which required neurosurgical inter-vention) [35]. In summary, CSF drainage has seriousbut uncommon complications [33–36]. The clinicalindication for CSF drainage in TAAI must balancethe surgical risks of spinal cord ischemia with therisks of CSF drainage. In endovascular TAAI, the roleof this neuroprotective intervention will likely be ona selective basis, given the significantly lower risk ofspinal cord ischemia in this setting.

Pharmacologic neuroprotectionThe search for the ideal neuroprotective agent in thesetting of spinal cord ischemia after TAAI has beenvigorous and sustained [2,5]. As the onset of spinalischemia in TAAI is typically known, the neuropro-tective agent could be administered intravenouslyor intrathecally either prophylactically or duringischemia and reperfusion of the spinal cord. Encour-aging results from the laboratory experiments havedemonstrated promising roles for an array of agentssuch as allopurinol, activated protein C, adenosine,barbiturates, carbamazepine, lidocaine, magnes-ium, mannitol, naloxone, papaverine, prostaglan-dins, steroids, and volatile anesthetics [5]. Despiteall these candidate agents, an ideal agent for clinicaladministration has yet to be found.

A recent clinical trial from a high-volume center(n¼330; 250 exposed to papaverine, 2002–2010)evaluated the neuroprotective effect of intrathecalpapverine as part of a multimodal protocol for spinalcord protection during open TAAI [37

&

]. The ration-ale for this intervention is that papaverine, as avasodilator, will increase spinal cord perfusion,particularly in the narrowed lower thoracic area,when given in the intrathecal space. Exposure topapaverine as part of a multimodal spinal protectionprotocol in this trial significantly reduced the risk of



C


permanent paraplegia (3.6 vs. 7.5%; P¼0.01) andparaparesis (1.6 vs. 6.3%; P¼0.01) [37

&

]. The inves-tigators concluded that papaverine likely providesadditional neuroprotection in this setting.

The multiplicity of neuroprotective agentspoints to an inadequate understanding of ischemiaand reperfusion in the spinal cord after TAAI from apharmacologic perspective. Recently, investigatorshave focused attention on the resident macrophagesin the spinal cord (the microglia) that are thought toplay a central role in the development of neuraldeath during spinal cord ischemia [38,39,40

&

]. Alaboratory study in rats demonstrated that exposureto the macrolide antibiotic, minocycline, duringspinal cord ischemia from thoracic aortic occlusionnot only inhibits spinal microglia, but also signifi-cantly protects against the clinical development ofparaplegia [40

&

]. This promising study has identifieda novel cellular target for further study and possiblya clinical trial in the future.

In summary, although a multiplicity of agentshave shown promise in the laboratory, there are todate no robust randomized clinical trials supportingthe efficacy of a single agent for spinal cord neuro-protection in TAAI. It appears that steroids are themost frequently administered agents for neuropro-tection during open TAAI, but this practice is basedon limited evidence [5]. Given the low risk of spinalcord ischemia and expanding clinical applicationsof endovascular TAAI, it is likely that this practicewill gradually disappear unless future clinical trialsstrongly support an agent.

Intensive neuromonitoring of the spinal cord

In the awake patient after TAAI, serial neurologicassessment willdetect the developmentof spinal cordischemia. This diagnosis should urgently trigger aprotocol for spinal cord rescue, including permissivesystemic hypertension with or without CSF drainage[1,2,3

&

]. In theanesthetizedandhence uncooperativepatient during TAAI, intraoperative detection of spi-nal cord ischemia typically relies on neuromonitor-ing with somatosensory-evoked potentials andmotor-evoked potentials [4,5,6

&&

,41,42].The intraoperative detection of spinal cord

ischemia during TAAI is critical because it signifi-cantly correlates with the development of paraple-gia and poor perioperative outcome [41,42]. Theintraoperative diagnosis by transcranial motor-evoked potentials (tcMEPs) gives the team theopportunity to intervene to enhance spinal cordperfusion and save the spinal cord. It is essential totailor the anesthetic plan so as to maximize thepreservation of neural signals both for somatosen-sory-evoked potentials and tcMEPs [4,5,6

&&

,41,42]. A



critical disadvantage of tcMEPs is the high sensitivityto volatile anesthetics and muscle relaxants. Thus,when tcMEP monitoring is planned, intravenousanesthesia should be used with minimal volatileanesthetics or muscle relaxants. This will minimizethe interference with the signal profiles from medi-cations and maximize the detection of spinal cordischemia. Anesthetic interventions to maximizespinal cord oxygen delivery in this setting includesystemic hypertension, augmented CSF drainage,and red blood cell transfusion for correction of ane-mia [22

&&

,23]. Surgical interventions to maximize thespinal protection will depend on the stage and type ofTAAI but might include augmentation of distal aorticperfusion, control of bleeding, reimplantation ofintercostal arteries, and staging of the TAAI to allowrecovery of the SCAN [5,43–45]. In summary, com-prehensive evaluation of spinal cord functionremains essential during the entire perioperativeperiod in patients undergoing TAAI. In high-riskpatients for spinal cord injury, spinal cord monitor-ingwith evokedpotentials is indicated.Thedetectionof spinal cord ischemia should urgently trigger inter-ventions for spinal cord rescue, based on institutionalprotocol, patient factors, and team discussion. Theimportance of clinical experience, clear communi-cation, and seamless teamwork all contribute signifi-cantly to prompt and successful reversal of spinalcord ischemia in this challenging setting.

CONCLUSION

The recent progress in TAAIs has resulted in betterfreedom from spinal cord ischemia for our patients,although this devastating complication still occurs.The promising areas of investigation in the nearfuture will likely involve refinements in endovascu-lar aortic techniques and a multimodality approach,both for enhanced spinal cord protection in thesechallenging aortic procedures.

Acknowledgements

Financial support: Institutional.

Conflicts of interest

There are no conflicts of interest.

REFERENCES AND RECOMMENDEDREADINGPapers of particular interest, published within the annual period of review, havebeen highlighted as:
& of special interest&& of outstanding interest
1. Desart K, Scali ST, Feezer RJ, et al. Fate of patients with spinal cord ischemiacomplicating thoracic endovascular aortic repair. J Vasc Surg 2013; 58:635–642.




2. Vaughn SB, Lemaire SA, Collard CD. Case scenario: anesthetic considera-tions for thoracoabdominal aortic aneurysm repair. Anesthesiology 2011;115:1093–1102.

3.&

Becker DA, McGarvey ML, Rojvirat C, et al. Predictors of outcome in patientswith spinal cord ischemia after open aortic repair. Neurocrit Care 2013;18:70–74.

This interesting study highlights in a contemporary series from an experiencedcenter the negative outcome effects of spinal cord ischemia after extensive openaortic repairs.4. Drenger B, Parker S, McPherson RW, et al. Spinal stimulation evoked

potentials during thoracoabdominal aortic aneurysm surgery. Anesthesiology1992; 76:689–695.

5. Augoustides JGT. Spinal cord protection strategies. In: Subramaniam K, ParkKW, Subramaniam B, editors. Anesthesia and perioperative care for aorticsurgery. New York, USA: Springer; 2011. pp. 239–256; Chapter 11..

6.&&

Sloan T, Edmonds HL, Koht A. Intraoperative electrophysiologic monitoring inaortic surgery. J Cardiothorac Vasc Anesth 2013; 27:1364–1373.

This review article discusses in detail the details of somatosensory and motorevoked potentials for spinal cord monitoring in thoracoabdominal aortic proce-dures.7. Andritsos M, Desai ND, Grewal A, et al. Innovations in aortic disease

management: the descending aorta. J Cardiothorac Vasc Anesth 2010;24:523–529.

8. Coady MA, Ikonomidis JS, Cheung AT, et al. Surgical management ofdescending thoracic aortic disease: open and endovascular approaches: ascientific statement from the American Heart Association. Circulation 2010;121:2780–2804.

9. Etz CD, Kari FA, Mueller CS, et al. The collateral network concept: areassessment of the anatomy of spinal cord perfusion. J Thorac CardiovascSurg 2011; 141:1020–1028.

10. Bischoff MS, Di Luozzo G, Griep EB, et al. Spinal cord preservation inthoracoabdominal aneurysm repair. Perspect Vasc Surg Endovasc Ther2011; 23:214–222.

11.&&

Griepp RB, Di Luozzo G. Hypothermia for aortic surgery. J Thorac CardiovascSurg 2013; 145:S56–S58.

This article discusses in detail the evidence supporting hypothermia for spinal cordprotection in thoracoabdominal aortic procedures.12.&

Kouchoukos NT. Thoracoabdominal aortic aneurysm repair using hypother-mic cardiopulmonary bypass and circulatory arrest. Ann Cardiothorac Surg2012; 1:409–411.

This interesting study highlights in a contemporary series from an experiencedcenter the excellent outcomes with deep hypothermic circulatory arrest afterextensive open aortic repairs.13.&

Yan TD, Bannon PG, Bavaria J, et al. Consensus on hypothermia in aortic archsurgery. Ann Cardiothoracic Surg 2013; 2:163–168.

This expert consensus study presents a rationale for the classification of hypother-mia in aortic surgery.14. Kouchoukos NT, Kulik A, Castner CF. Outcomes after thoracoabdominal

aortic aneurysm repair using hypothermic circulatory arrest. J Thorac Cardio-vasc Surg 2013; 145:S139–S141.

15. Corvera JS, Fehrenbacher JW. Open repair of chronic aortic dissectionsusing deep hypothermia and circulatory arrest. Ann Thorac Surg 2012;94:78–81.

16. Fehrenbacher JW, Siderys H, Terry C, et al. Early and late results ofdescending thoracic and thoracoabdominal aortic aneurysm open repair withdeep hypothermia and circulatory arrest. J Thorac Cardiovasc Surg 2010;140:S154–S160.

17. Leshnower BG, Szeto WY, Pochettino A, et al. Thoracic endografting reducesmorbidity and remodels the thoracic aorta in DeBakey III aneurysms. AnnThoracic Surg 2013; 96:914–921.

18. Canaud L, Karthikesalingam A, Jackson D, et al. Clinical outcomes of singleversus staged hybrid repair for thoracoabdominal aneurysm. J Vasc Surg2013; 58:1192–1200.

19.&&

Zipfel B, Buz S, Redlin M, et al. Spinal cord ischemia after thoracic stent-grafting: causes apart from intercostals artery coverage. Ann Thorac Surg2013; 96:31–38.

This interesting article discusses how thoracoabdominal endovascular aorticrepair evolved at a high-volume institution over time, based on an understandingand clinical application of the spinal collateral arterial network concept.20. Etz CD, Kari FA, Mueller CS, et al. The collateral network concept: remodeling

of the arterial collateral network after experimental segmental arterial sacrifice.J Thorac Cardiovasc Surg 2011; 141:1029–1036.

21. Ullery BW, Wang GJ, Low D, et al. Neurological complications of thoracicendovascular aortic repair. Semin Cardiothorac Vasc Anesth 2011; 15:123–140.

22.&&

Grabenwoger M, Alfonso F, Bachet J, et al. Thoracic endovascular repair(TEVAR) for the treatment of aortic diseases: a position statement from theEuropean Association for Cardio-Thoracic Surgery (EACTS) and the Eur-opean Society of Cardiology (ESC), in collaboration with the EuropeanAssociation of Percutaneous Cardiovascular Interventions (EAPCI). Eur HeartJ 2012; 33:1558–1563.

This comprehensive guideline presents a guideline-based endovascular approachto thoracoabdominal aortic diseases with recommendations based on clear levelsof evidence and clinical benefit.



23. Hiratzka LF, Bakris GL, Beckman JA, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM guidelines for the diagnosis and manage-ment of patients with thoracic aortic disease. A report of the American Collegeof Cardiology/American Heart Association Task Force on Practice Guide-lines, American Association for Thoracic Surgery, American College ofRadiology, American Stroke Association, Society of Cardiovascular Anesthe-siologists, Society of Cardiovascular Angiography and Interventions, Societyof Interventional Radiology, Society of Thoracic Surgeons and Society ofVascular Medicine. J Am Coll Cardiol 2010; 55:e27–e129.

24. Bobadilla JL, Wynn M, Tefera G, et al. Low incidence of paraplegia afterthoracic endovascular aneurysm repair with proactive spinal cord protectiveprotocols. J Vasc Surg 2013; 57:1537–1542.

25. Keith CJ, Passman MA, Carignan MJ, et al. Protocol implementation ofselective postoperative lumbar spinal drainage after thoracic aortic endograft.J Vasc Surg 2012; 55:1–9.

26. Hanna JM, Andersen ND, Hamza A, et al. Results with selective preoperativelumbar drain placement for thoracic endovascular aortic repair. Ann ThoracSurg 2013; 95:1968–1975.

27. Ullery BW, Cheung AT, Fairman RF, et al. Risk factors, outcomes, and clinicalmanifestations of spinal cord ischemia following thoracic endovascular aorticrepair. J Vasc Surg 2011; 54:677–684.

28.&

Wong CS, Healy D, Canning C, et al. A systematic review of spinal cord injuryand cerebrospinal fluid drainage after thoracic aortic endografting. J VascSurg 2012; 56:1438–1447.

This comprehensive meta-analysis demonstrates the low incidence of spinal cordischemia in endovascular aortic techniques, despite the low quality of evidence.29.&&

Khan SN, Stansby G. Cerebrospinal fluid drainage for thoracic and thora-coabdominal aortic aneurysm surgery. Cochrane Database Syst Rev 2012;10:CD003635.

This comprehensive meta-analysis demonstrates the benefits of cerebrospinal fluiddrainage in open thoracoabdominal interventions.30.&&

Youngblood SC, Tolpin DA, LeMaire SA, et al. Complications of cerebrospinalfluid drainage after thoracic aortic surgery: a review of 504 patients over5 years. J Thorac Cardiovasc Surg 2013; 146:166–171.

This is the best recent review of lumbar drain complications after thoracic aorticsurgery from an experienced and high-volume center.31. Bilal H, O’Neill B, Mahmood S, et al. Is cerebrospinal fluid drainage of benefit

to neuroprotection in patients undergoing surgery on the descending thoracicaorta or thoracoabdominal aorta. Interact Cardiovasc Thorac Surg 2012;15:702–708.

32. Drenger B, Parker SD, Frank SM, Beattie C. Changes in cerebrospinal fluidpressure and lactate concentrations during thoracoabdominal aortic aneur-ysm surgery. Anesthesiology 1997; 86:41–47.

33. Dardik A, Perler BA, Roseborough GS, et al. Subdural hematoma afterthoracoabdominal aortic aneurysm repair: An underreported complicationof spinal fluid drainage? J Vasc Surg 2002; 36:47–50.

34. Fedorow CA, Moon MC, Mutch AC, et al. Lumbar cerebrospinal fluid drainagefor thoracoabdominal aortic surgery: rationale and practical considerations formanagement. Anesth Analg 2010; 111:46–58.

35. Mehmedagic I,ReschT,AcostaS.Complications tocerebrospinal fluiddrainageand predictors of spinal cord ischemia inpatientswithaortic disease undergoingadvanced endovascular therapy. Vasc Endovasc Surg 2013; 47:415–422.

36. Leyvi G, Ramachandran S, Wasnick JD, et al. Risks and benefits of cere-brospinal fluid drainage during thoracoabdominal aortic aneurysm surgery.J Cardiothor Vasc Anesth 2005; 19:392–399.

37.&

Lima B, Nowicki ER, Blackstone EH, et al. Spinal cord protection strategiesduring descending and thoracoabdominal aortic aneurysm repair in themodern era: the role of intrathecal papaverine. J Thorac Cardiovasc Surg2012; 143:945–952.

This clinical trial presents evidence supporting papaverine as part of a multimodalprotocol for spinal cord protection in thoracoabdominal aortic procedures.38. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of

inflammatory cells. J Leukoc Biol 2010; 87:779–789.39. Drenger B, Fellig J, Barzilay Y. Minocycline-induced spinal cord protection

from aortic occlusion-related ischemia in the rabbit model. Anesth Analg2010; 110:S-39.

40.&

Smith PD, Bell MT, Puskas F, et al. Preservation of motor function after spinalcord ischemia and reperfusion injury through microglial inhibition. Ann ThoracSurg 2013; 95:1647–1653.

This interesting clinical trial presents the laboratory evidence supporting minocy-cline as part of multimodal protocol for spinal cord protection in thoracoabdominalaortic procedures.41. McGarvey ML. Effective tool or necessary evil: intraoperative monitoring during

thoracic aneurysm repairs. J Clin Neurophysiol 2012; 29:154–156.42. BeckerDA,McGarveyML,RojviratC,etal.Predictorsofoutcome inpatientswith

spinal cord ischemia after open aortic repair. Neurocrit Care 2013; 18:70–74.43. Hsu CC, Kwan GN, van Driel ML, et al. Distal aortic perfusion during

thoracoabdominal aneurysm repair for prevention of paraplegia. CochraneDatabase Syst Rev 2012; 3:CD008197.

44. Ullery BW, Wang GJ, Woo EY. No increased risk of spinal cord ischemia indelayed AAA repair following thoracic aortic surgery. Vasc Endovasc Surg2013; 47:85–91.

45. Hughes GC, Andersen ND, Hanna JM, et al. Thoracoabdominal aortic aneur-ysm: hybrid repair outcomes. Ann Cardiothorac Surg 2012; 1:311–319.



Novel approaches to_spinal_cord_protection_during.15

Health & Medicine

Transcript of Novel approaches to_spinal_cord_protection_during.15