Advertisement

Evaluation of collateral network near-infrared spectroscopy during and after segmental artery occlusion in a chronic large animal model

Open ArchivePublished:December 11, 2018DOI:https://doi.org/10.1016/j.jtcvs.2018.11.105

      Abstract

      Objective

      Ischemic spinal cord injury remains the most devastating complication after open and endovascular aortic repair. Collateral network near-infrared spectroscopy has been introduced to noninvasively monitor real-time spinal cord oxygenation. In view of recent advancements in endovascular treatment and minimally invasive staged preconditioning before aortic repair, this study sought to evaluate collateral network near-infrared spectroscopy during and after segmental artery occlusion in a chronic porcine model.

      Methods

      Surgery for segmental artery occlusion was performed in 12 juvenile pigs, and bilateral lumbar collateral network near-infrared spectroscopy was recorded. Two intervention groups were designed: Group 1 received subtotal segmental artery occlusion (mimicking reimplantation of crucial segmental arteries with patent T12/T13, N = 5), and group 2 received total occlusion (T4-L5, N = 7). Pigs were monitored over 3 days.

      Results

      All animals were paraplegic during the first 24 hours. The subtotal occlusion group completely recovered, whereas 57% of the total occlusion group remained paraplegic (N = 4/7). After segmental artery occlusion, collateral network near-infrared spectroscopy decreased from 92.3% ± 8% of baseline to 69.3% ± 18% after 10 minutes in the subtotal group (P = .003-.017) and from 90.1% ± 4% to 58.2% ± 9% in the total group (P < .001-.008). Throughout the postoperative period, collateral network near-infrared spectroscopy in the total occlusion group remained lower compared with the subtotal group (<30% baseline threshold, P < .05). Lumbar collateral network near-infrared spectroscopy and neurologic outcome were significantly correlated (R = 0.7, P < .001).

      Conclusions

      Lumbar collateral network near-infrared spectroscopy reacts to occlusion of segmental arteries and correlates with neurologic outcome. The preliminary data suggest that collateral network near-infrared spectroscopy may be a valuable noninvasive tool for detecting imminent spinal cord ischemia during and after aortic procedures involving segmental artery occlusion.

      Key Words

      Abbreviations and Acronyms:

      CN (collateral network), cnNIRS (collateral network near-infrared spectroscopy), MAP (mean arterial pressure), MIS2ACE (minimally invasive staged segmental artery coil and plug embolization), SA (segmental artery), SCI (spinal cord injury), TAAA (thoracoabdominal aortic aneurysm)
      Figure thumbnail fx1
      Scatter graph for cnNIRS and neurologic outcome showing significant positive correlation.
      cnNIRS reacts to SA occlusion. It correlates with neurologic outcome and is lower in total compared with subtotal occlusion.
      Lumbar cnNIRS may be a valuable noninvasive tool for detecting pending spinal cord ischemia during and after aortic procedures involving SA occlusion. Further in-depth analyses are warranted to determine the exact threshold for imminent SCI with regard to absolute cnNIRS values.
      See Commentary on page 165.
      Ischemic spinal cord injury (SCI) remains the most devastating complication after repair of thoracoabdominal aortic aneurysm (TAAA).
      • Koeppel T.A.
      • GA
      • Jacobs M.J.
      DGG Leitlinie-Thorakale und Thorakoabdominelle Aortenaneurysmen.
      • Etz C.D.
      • Luehr M.
      • Kari F.A.
      • Bodian C.A.
      • Smego D.
      • Plestis K.A.
      • et al.
      Paraplegia after extensive thoracic and thoracoabdominal aortic aneurysm repair: does critical spinal cord ischemia occur postoperatively?.
      Despite contemporary perioperative and postoperative adjuncts, the incidence of paraplegia and paraparesis due to SCI after extensive open and endovascular aortic repair (Crawford type II) remains up to 18% in large contemporary series.
      • Koeppel T.A.
      • GA
      • Jacobs M.J.
      DGG Leitlinie-Thorakale und Thorakoabdominelle Aortenaneurysmen.
      • Panthee N.
      • Ono M.
      Spinal cord injury following thoracic and thoracoabdominal aortic repairs.
      In addition to the individual tragedy for the patient, postoperative paraplegia also represents an immense socioeconomic burden.
      National Spinal Cord Injury Statistical Center
      Spinal cord injury facts and figures at a glance.

      Materials and Methods

      Monitoring the Spinal Cord

      Few modalities exist to monitor the spinal cord. Common invasive methods of intraoperative spinal cord viability monitoring include motor-evoked potential and somatosensory-evoked potential.
      • Pillai J.B.
      • Pellet Y.
      • Panagopoulos G.
      • Sadek M.A.
      • Abjigitova D.
      • Weiss D.
      • et al.
      Somatosensory-evoked potential-guided intercostal artery reimplantation in thoracoabdominal aortic aneurysm surgery.
      • Dias-Neto M.
      • Reis P.V.
      • Rolim D.
      • Ramos J.F.
      • Teixeira J.F.
      • Sampaio S.
      • et al.
      Strategies to prevent TEVAR-related spinal cord ischemia.
      Although somatosensory-evoked potential monitoring can be implemented during the postoperative period, motor-evoked potential monitoring on the awake, nonsedated patient is limited.
      • LeMaire S.A.
      • Ochoa L.N.
      • Conklin L.D.
      • Widman R.A.
      • Clubb Jr., F.J.
      • Undar A.
      • et al.
      Transcutaneous near-infrared spectroscopy for detection of regional spinal ischemia during intercostal artery ligation: preliminary experimental results.
      • Demir A.
      • Erdemli O.
      • Unal U.
      • Tasoglu I.
      Near-infrared spectroscopy monitoring of the spinal cord during type B aortic dissection surgery.
      Nevertheless, these modalities usually require significant technical and human resources. Ideally, spinal cord monitoring should reflect perfusion and ultimately tissue oxygenation in real-time to allow for rapid response in hemodynamic (mean arterial pressure [MAP]), cardiopulmonary (atrial fibrillation, ventilation), and cerebrospinal fluid management.

      Collateral Network Concept

      The collateral network (CN) corresponds to an extensive arterial network—largely localized in the paraspinal musculature—fed by branches of the subclavian arteries, hypogastric artery, and directly from aortic segmental arteries interconnected via collaterals both longitudinally and horizontally. These collaterals enable sufficient blood flow to the spinal cord tissue during chronic or acute perfusion loss after extensive segmental artery (SA) sacrifice.
      • Etz C.D.
      • Kari F.A.
      • Mueller C.S.
      • Silovitz D.
      • Brenner R.M.
      • Lin H.M.
      • et al.
      The collateral network concept: a reassessment of the anatomy of spinal cord perfusion.
      • Etz C.D.
      • Kari F.A.
      • Mueller C.S.
      • Brenner R.M.
      • Lin H.M.
      • Griepp R.B.
      • et al.
      The collateral network concept: remodeling of the arterial collateral network after experimental segmental artery sacrifice.

      Collateral Network Near-Infrared Spectroscopy

      Collateral network near-infrared spectroscopy (cnNIRS) has been experimentally and clinically introduced and evaluated for noninvasive (indirect) real-time monitoring of spinal cord perfusion and oxygenation in open TAAA repair.
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
      • Boezeman R.P.
      • van Dongen E.P.
      • Morshuis W.J.
      • Sonker U.
      • Boezeman E.H.
      • Waanders F.G.
      • et al.
      Spinal near-infrared spectroscopy measurements during and after thoracoabdominal aortic aneurysm repair: a pilot study.
      The rationale for cnNIRS is based on the idea that according to the CN concept, blood supply to and oxygenation in the paraspinal vasculature correlate with spinal cord perfusion and oxygenation.
      Our previous clinical and experimental studies have shown that cnNIRS is technically feasible and that lumbar readings correlate with spinal cord perfusion and oxygenation during and after aortic crossclamping and reperfusion.
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
      Contemporary data also suggest that high thoracic cnNIRS (T5-T6/7) does not depict changes in regional oxygenation during distal aortic interventions.
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.

      Study Aim

      Considering the recent introduction of minimally invasive staged SA coil and plug embolization (MIS2ACE) for CN preconditioning before TAAA repair for paraplegia prevention,
      • Dias-Neto M.
      • Reis P.V.
      • Rolim D.
      • Ramos J.F.
      • Teixeira J.F.
      • Sampaio S.
      • et al.
      Strategies to prevent TEVAR-related spinal cord ischemia.
      • Etz C.D.
      • Debus E.S.
      • Mohr F.W.
      • Kolbel T.
      First-in-man endovascular preconditioning of the paraspinal collateral network by segmental artery coil embolization to prevent ischemic spinal cord injury.
      • Von Aspern K.
      • Luehr M.
      • Mohr F.W.
      • Etz C.D.
      Spinal cord protection in open- and endovascular thoracoabdominal aortic aneurysm repair: critical review of current concepts and future perspectives.
      real-time spinal cord oxygenation monitoring becomes increasingly important, particularly because clinical observation of the awake patient represents the only monitoring method during these procedures.
      The aim of this study was to evaluate cnNIRS during and after SA occlusion, focusing on its capability to differentiate between complete and incomplete SA sacrifice and correlation with postoperative neurologic outcome. For this purpose, a chronic large animal model has been designed.
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.

      Materials and Methods

      Ethics Statement

      The chronic experimental animal studies were approved by the Institutional Animal Care and Use Committee in accordance with the local Veterinary Office and the Principles of Laboratory Animal Care.
      An experienced veterinarian was present throughout the experiments.

      Preoperative Preparation and Monitoring Set-up

      Twelve juvenile female pigs (German Landrace, 36-45 kg) were lodged in groups for 5 days preoperatively to acclimate. Animals were examined thoroughly with regard to their physical and neurologic condition. Neurologic assessment was performed according to a modified Tarlov score
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      (Table E1).
      On the day of the operation, the pigs were sedated before being transferred to the operating room. After preoxygenation, anesthesia was supplemented by fentanyl (0.03-0.05 mg/kg fentanyl; Pfizer, New York, NY), and after endotracheal intubation, inhalative isoflurane (1.5%) was administered. Continuous arterial blood pressure monitoring was established via the left femoral artery. Mechanical ventilation was set to 50% oxygen at 20 breaths per minute with a tidal volume adjusted at 8 mL/kg. Body temperature was maintained at 37°C ± 1.5°C. Central venous catheterization of the left jugular vein was performed using a 7F, 20-cm multilumen catheter for additional infusions and blood sampling. The pig was then placed on its right side.

      Near-Infrared Spectroscopy of the Collateral Network

      After shaving of the animal, lumbar cnNIRS optodes were cutaneously placed bilaterally at level L2 to L3. Preliminary experiments have shown that in nonpigmented, shaved pigs weighing less than 55 kg, cnNIRS measurements during ischemia–reperfusion and baseline are not significantly different regarding cutaneous versus subcutaneous optode placement (P = 1.000). Near-infrared spectroscopy signals were continuously recorded using an interface device for noninvasive monitoring of regional tissue (muscle) oxygen saturation of hemoglobin as the ratio between deoxygenated hemoglobin and the sum of deoxygenated and oxygenated hemoglobin (INVOS Oximeter 5100C; Medtronic, Dublin, Ireland). Baseline lumbar cnNIRS values were acquired after preoxygenation and intubation at standard ventilator settings with fraction of inspired oxygen of 0.5.

      Surgical Approach

      Through a lateral thoracotomy via the fifth and ninth intercostal space, the descending thoracic aorta was visualized, and the segmental arteries (SAs) T4 to T13 were exposed. A catheter was introduced into the proximal descending aorta for invasive arterial blood pressure monitoring. An additional catheter was placed into the left atrium. For pressure monitoring, blood sampling and medication administration during the postoperative period catheters were channeled through the fifth intercostal space and subcutaneous tissue and fixated between the shoulders. Abdominal access and exposition of the abdominal aorta and lumbar SAs were achieved via a longitudinal incision.

      Intervention Groups

      Previous studies have demonstrated that the most prominent ischemic spinal cord tissue damage after serial SA occlusion occurs between the lower thoracic (T9/10) and upper lumbar (L2/3) segmental levels.
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
      During open TAAA repair, many centers perform reimplantation of crucial segmental arteries located in this area. To explore the influence of patent perfusion in this region on lumbar cnNIRS—via segmental arteries located centrally at level T12/T13—2 intervention groups were designed. Group 1 received consecutive SA closure with complete lumbar occlusion but patent SAs at levels T12/13 (subtotal, N = 5). Group 2 received consecutive occlusion of all SAs (N = 7) (Figure 1). One animal was originally intended for subtotal occlusion; however, because of an aortic tear and bloody oozing, 2 pairs of SAs at the lower thoracic level needed to be ligated in short succession to avoid blood loss and prolonged ventilation. Because this animal ultimately had all SAs occluded, it crossed over to the total occlusion group (verified during autopsy).
      Figure thumbnail gr1
      Figure 1Subtotal (group 1) = consecutive SA occlusion with patent perfusion of T12 + T13, total (group 2) = complete SA occlusion T4-L5.

      Experimental Sequence and Monitoring Procedure

      Consecutive serial occlusion of the exposed SA (pairs) was performed using vessel clips (Titanium Clips, Teleflex Medical, RTP, Morrisville, NC). Starting from SA T4, occlusion was performed in the caudal direction until L5. MAP (aorta + left femoral artery) and lumbar cnNIRS oxygenation were continuously recorded. Regular arterial blood samples were drawn at 10-minute intervals while continuous oxygenation monitoring via pulse oximetry was recorded. The cnNIRS monitoring system stored oxygenation measurements at 5-second intervals. After SA occlusion, the animals remained sedated in a right-sided position for 2 additional hours at identical ventilator settings to ensure standardized cnNIRS measurements. During this period, drainage tubes (thoracic and abdominal) were placed and wound closure was performed. During the entire procedure, no vasoactive drugs were administered.

      Postoperative Monitoring and Neurologic Assessment

      After extubation and transportation from the operating room to the animal housing facility, blood samples and measurements of invasive arterial blood pressure, core temperature, peripheral oxygenation, and lumbar cnNIRS were recorded hourly by a physician until 8 hours postoperatively. Afterward, arterial blood pressure and blood samples were recorded at 6-hour intervals. Clinical outcome and neurologic evaluation were assessed every 12 hours. Animals were observed for 72 hours. Neurologic recovery was defined according to previous studies
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
      • Bischoff M.S.
      • Scheumann J.
      • Brenner R.M.
      • Ladage D.
      • Bodian C.A.
      • Kleinman G.
      • et al.
      Staged approach prevents spinal cord injury in hybrid surgical-endovascular thoracoabdominal aortic aneurysm repair: an experimental model.
      as a modified Tarlov score greater than 5 and permanent paraplegia/paraparesis 5 or less, respectively. After the experiments were completed, the animals were anaesthetized and euthanized in deep sedation. The spinal cord was removed en bloc immediately after euthanasia before correct catheter placement and SA occlusion was confirmed, and the autopsy resumed.

      Histopathologic Evaluation

      All spinal cords were harvested, and samples from each cord level (identified by nerve root) collected. Evaluation of ischemic cell injury was performed independently and blinded by 2 experienced researchers as previously described and illustrated (Table E2).
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.

      Statistical Analysis

      Data were imported to SPSS (version 17.0; SPSS, Chicago, Ill) for description and analysis. Continuous variables are expressed as mean ± standard deviation. Normal distribution of measurements was determined using the Shapiro–Wilk test, and Q–Q plots were created for verification. Group comparison of lumbar cnNIRS and spinal cord tissue was performed using analysis of variance or Student t test for repeated/correlated measures as appropriate. For correlation analysis, scatter plots were calculated and evaluated using the Spearman's rank correlation coefficient.

      Results

      Clinical Outcome and Paraplegia Rate

      Immediately after the operation all animals—subtotal and total occlusion—presented with severe hind limb motor function loss (modified Tarlov score ≤2) for the first 12 hours and 5 or less until 24 hours after surgery. All animals in the subtotal group recovered within 48 hours (median score at 48 hours = 9; range, 8-10), whereas animals in the total occlusion group needed more time for neurologic recovery (median score at 48 hours = 6; range, 1-8; P = .023). The rate of permanent paraplegia/paraparesis in group 2 (total) was 57% (N = 4). Other adverse events throughout the experiment are listed in Table E3.

      Lumbar Near-Infrared Spectroscopy Measurements

      Analysis of variance between left and right cnNIRS measurements did not show systematic differences (P = 1.000); therefore, the 2 results were averaged to provide a mean estimate of changes.

      Collateral network near-infrared spectroscopy during subtotal segmental artery occlusion (group 1)

      During serial SA occlusion of T4 to L5—sparing SA T12/13—cnNIRS remained stable throughout thoracic SA occlusion, but rapidly decreased from 92.3% ± 8% of baseline to 77.8% ± 9%, 72.4% ± 10%, and 69.3% ± 14% at 1, 5, and 10 minutes after SA occlusion, respectively (P = .003-.017). Within 1 hour after surgery, cnNIRS values increased to 76.0% ± 10%, fluctuating during the first 24 hours—reaching minimum values of 71.2% ± 11% at 6 hours and maximum values of 79.5% ± 10% at 12 hours postoperatively—before steadily increasing to stable values of 86.0% ± 6% and 89.5% ± 9% on days 2 and 3, respectively (Figure 2, group 1/black, and Table E4).
      Figure thumbnail gr2
      Figure 2Comparison of lumbar cnNIRS measurements between subtotal (blue dots) and total (red dots) SA occlusion. Values gradually digress throughout the postoperative period (represented by the regression lines). The regression line of animals from the subtotal group (blue line) remains above the 30% cnNIRS threshold. cnNIRS, Collateral network near-infrared spectroscopy; SA, segmental artery.

      Collateral network near-infrared spectroscopy during total segmental artery occlusion (group 2)

      Analogous to group 1 (subtotal), complete SA occlusion led to a significant, yet more pronounced reduction in cnNIRS with mean differences between groups ranging from 4.4% ± 3.1% to 6.1% ± 3.9% during the first 10 minutes. Starting from 89.4% ± 4% of baseline prior occlusion, cnNIRS in group 2 gradually decreased to 70.5% ± 10%, 65.9% ± 9%, and 59.2% ± 8% at 1, 5, and 10 minutes, respectively (P < .02). After surgery, cnNIRS values increased to 64.3% ± 10% and 61.8% ± 14% from 1 to 24 hours, respectively. At 36 hours postoperatively, lumbar cnNIRS demonstrated a decrease to 57.8% ± 15% baseline before gradually increasing over the remaining surveillance period reaching stable maximum values between 60.5% ± 14% and 63.2% ± 15% (Figure 2, group 2/red, and Table E4).

      Comparison between subtotal and total segmental artery occlusion

      Continuous cnNIRS measurements before SA occlusion did not differ significantly between groups. Postocclusion cnNIRS measurements for the total occlusion group were lower throughout the entire surveillance period compared with the subtotal group, but they did not reach statistical significance until 10 minutes after SA occlusion (P < .05). After 2 days, cnNIRS remained stable in both groups at different oxygenation levels between 86.0% and 89.0% ± 8% of baseline for the subtotal and 60.0% and 63.0% ± 16% of baseline for the total group, respectively (P < .05, Figure 2).

      Mean Arterial Blood Pressure Measurements

      Changes in MAP of the entire cohort ranged from 5% to 7% of baseline. After surgery, MAP did not demonstrate excessive variations (range, 92%-102% of baseline). Comparison of MAP during and after consecutive SA occlusion reveals no significant difference between subtotal and total occlusion (Figure E1; P = .314-.899). Correlation analysis among unaltered MAP, cnNIRS, and neurologic outcome was not significantly correlated (R = 0.1-0.5, P = .227-.950). During the experiment, MAP was not intentionally altered to minimize its influence on cnNIRS measurements.

      Lumbar Collateral Network Near-Infrared Spectroscopy and Neurologic Outcome

      Irrespective of the intervention group, animals that recovered (N = 8) returned to cnNIRS values above 30% of baseline within the first postoperative hour, whereas animals with permanent paraplegia (N = 4, all total occlusion animals) remained below 30% of baseline throughout the entire surveillance period (Figure 3). Immediately after SA clipping until 10 minutes postocclusion, both recovering and paraplegic animals demonstrated comparable lumbar cnNIRS dynamics with significantly reduced levels compared with baseline (P = .001, Table E5). Within 3 days postoperatively, cnNIRS measurements of recovering animals returned to values not significantly different from their individual baseline (91.4% ± 6% vs 88.9% ± 9%, P = .429). However, paraplegic animals' postoperative cnNIRS measurements further decreased relative to their baseline values (88.5% ± 4% vs 37.7% ± 14%, P = .023; Table E5).
      Figure thumbnail gr3
      Figure 3cnNIRS is shown throughout the experiment (top) and the postoperative neurologic status (bottom, Tarlov score). Animals that recovered (left, cnNIRS in blue) and animals that remained paraplegic (right, cnNIRS in green). Recovering animals remain above the 30% cnNIRS threshold throughout the postoperative period (black median line), whereas paraplegic animals remain below the 30% threshold. cnNIRS, Collateral network near-infrared spectroscopy; SA, segmental artery.
      The return of lumbar cnNIRS measurements above 30% baseline preceded neurologic recovery by approximately 24 to 30 hours. Analysis demonstrates a significant positive correlation between cnNIRS measurements and neurologic outcome as illustrated in Figure 4 (R = 0.7, P < .001). For clarification, a scatter graph showing individual cnNIRS point data with regard to subtotal versus total occlusion and recovering versus paraplegic animals has been generated (Figure E2).
      Figure thumbnail gr4
      Figure 4Correlation graph for cnNIRS and neurologic outcome. Values are depicted for each animal at each postoperative time point (N = 12, starting 1 hour postoperatively) showing a significant positive correlation. Different time points are color coded demonstrating the association between recovering animals and higher cnNIRS ratios after 24 hours. High cnNIRS values associated with low Tarlov scores are exclusively measurements within the first 24 postoperative hours, when “temporary” paraplegia was present in all animals. It is demonstrated that high Tarlov scores (>5) are not associated with low cnNIRS measurements (<30% of baseline). cnNIRS, Collateral network near-infrared spectroscopy.

      Histologic Results

      The cervical spinal cord tissue showed no relevant ischemic damage, independent of the intervention group and postoperative neurologic status (Figure 5).
      Figure thumbnail gr5
      Figure 5Degree of histopathologic spinal cord damage; (left) comparing recovering (N = 8, blue) and paraplegic (N = 4, red) animals; (right) comparison between subtotal (N = 5, blue) and total (N = 7, green) SA occlusion. Each colored dot represents a spinal cord tissue sample (cervical = C; thoracic = T, and lumbar = L) with a horizontal line showing the median value according to a tissue damage scoring system from 0 (no damage) to 8 (severe damage). It is noticeable that cord damage of the lower thoracic and lumbar region is more severe in paraplegic animals (left, red) and after total SA occlusion (right, green) compared with recovering animals and those receiving subtotal occlusion, respectively.
      Animals with subtotal occlusion (N = 5) had less severe tissue damage compared with the total occlusion group (N = 7; Figure 5, top). Although differences are evident starting from level T3 downward to level L5, statistical significance is reached only for T5 and T8 (P = .030 and .045).
      Accordingly, because all paraplegic animals were from the total occlusion group (N = 4), mean spinal cord tissue damage was significantly more severe for paraplegic animals compared with animals that recovered (0-4.3, range, 0-8 vs 0-1.9, range, 0-5; P < .001; Figure 5, bottom).

      Discussion

      The presented experiment showed for the first time that lumbar cnNIRS reacts to selective SA occlusion and is capable of differentiating between a subtotal and a total occlusion pattern. Furthermore, evaluation of cnNIRS and neurologic outcome revealed significant positive correlation. Although an exact cutoff value for cnNIRS regarding SCI could not be determined, animals that remained paraplegic had consistent mean cnNIRS values lower than 30% of baseline, whereas animals that recovered returned to levels above 30% of baseline within the first postoperative hour. A short synoptic perspective statement is provided in Video 1.
      Figure thumbnail fx2
      Video 1Short background, conclusion, and perspective statement by the senior author. Video available at: https://www.jtcvs.org/article/S0022-5223(18)33255-0/fulltext.

      Demand for Noninvasive Monitoring

      With the recently introduced endovascular procedure for paraplegia prevention by means of staged CN preconditioning through MIS2ACE,
      • Etz C.D.
      • Debus E.S.
      • Mohr F.W.
      • Kolbel T.
      First-in-man endovascular preconditioning of the paraspinal collateral network by segmental artery coil embolization to prevent ischemic spinal cord injury.
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
      and a general trend toward more extensive endovascular aortic procedures, noninvasive monitoring methods depicting real-time spinal cord perfusion/oxygenation become all the more essential.

      Previous Experience With Collateral Network Near-Infrared Spectroscopy

      Previous experimental and clinical studies on cnNIRS have shown that although in heterogeneous setups and study protocols, it reflects oxygenation changes after aortic crossclamping and distal aortic perfusion in real-time.
      • LeMaire S.A.
      • Ochoa L.N.
      • Conklin L.D.
      • Widman R.A.
      • Clubb Jr., F.J.
      • Undar A.
      • et al.
      Transcutaneous near-infrared spectroscopy for detection of regional spinal ischemia during intercostal artery ligation: preliminary experimental results.
      • Demir A.
      • Erdemli O.
      • Unal U.
      • Tasoglu I.
      Near-infrared spectroscopy monitoring of the spinal cord during type B aortic dissection surgery.
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
      • Boezeman R.P.
      • van Dongen E.P.
      • Morshuis W.J.
      • Sonker U.
      • Boezeman E.H.
      • Waanders F.G.
      • et al.
      Spinal near-infrared spectroscopy measurements during and after thoracoabdominal aortic aneurysm repair: a pilot study.
      • Suehiro K.
      • Funao T.
      • Fujimoto Y.
      • Mukai A.
      • Nakamura M.
      • Nishikawa K.
      • et al.
      Transcutaneous near-infrared spectroscopy for monitoring spinal cord ischemia: an experimental study in swine.
      On the basis of these and other studies, cnNIRS emerged as a promising new and economic method, potentially guiding extensive aortic procedures by giving real-time feedback on spinal cord perfusion.
      • Badner N.H.
      • Nicolaou G.
      • Clarke C.F.
      • Forbes T.L.
      Use of spinal near-infrared spectroscopy for monitoring spinal cord perfusion during endovascular thoracic aortic repairs.

      Sensitivity of Collateral Network Near-Infrared Spectroscopy to Segmental Artery Occlusion

      Lumbar cnNIRS reacted to acute interruption of proximal SA perfusion; however, lumbar cnNIRS first started to decrease once the lower thoracic/lumbar SA inflow was occluded, demonstrating a less severe decrease in measurements with patent lower thoracic SAs. Whether SA occlusion by means of stent-graft implantation or MIS2ACE of only a limited number of arteries can be depicted by lumbar cnNIRS remains to be evaluated. After serial SA occlusion, cnNIRS signals of recovering animals consistently increased throughout the surveillance period, which is in accordance with previous clinical reports and experiments.
      • Etz C.D.
      • Luehr M.
      • Kari F.A.
      • Bodian C.A.
      • Smego D.
      • Plestis K.A.
      • et al.
      Paraplegia after extensive thoracic and thoracoabdominal aortic aneurysm repair: does critical spinal cord ischemia occur postoperatively?.
      • Etz C.D.
      • Kari F.A.
      • Mueller C.S.
      • Brenner R.M.
      • Lin H.M.
      • Griepp R.B.
      • et al.
      The collateral network concept: remodeling of the arterial collateral network after experimental segmental artery sacrifice.
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      • Moerman A.
      • Van Herzeele I.
      • Vanpeteghem C.
      • Vermassen F.
      • Francois K.
      • Wouters P.
      • et al.
      Near-infrared spectroscopy for monitoring spinal cord ischemia during hybrid thoracoabdominal aortic aneurysm repair.
      Moerman and colleagues
      • Moerman A.
      • Van Herzeele I.
      • Vanpeteghem C.
      • Vermassen F.
      • Francois K.
      • Wouters P.
      • et al.
      Near-infrared spectroscopy for monitoring spinal cord ischemia during hybrid thoracoabdominal aortic aneurysm repair.
      demonstrated an immediate linear correlation of cnNIRS and arterial blood pressure after regional SA inflow interruption. However, this linear cnNIRS dependency on arterial blood pressure was not demonstrated in the present study, because no animal experienced relevant blood pressure alterations.
      In an early experiment on spinal cord NIRS monitoring by LeMaire and colleagues,
      • LeMaire S.A.
      • Ochoa L.N.
      • Conklin L.D.
      • Widman R.A.
      • Clubb Jr., F.J.
      • Undar A.
      • et al.
      Transcutaneous near-infrared spectroscopy for detection of regional spinal ischemia during intercostal artery ligation: preliminary experimental results.
      sequential ligation of segmental arteries from T6-L1 led to a prominent decrease at the lower thoracic level (T9-T11). These were significantly lower than measurements from the upper cord (T6-T7), showing a decrease of only 6.3% of baseline (upper cord) compared with 39% (lower cord, P = .026). These findings are consistent with recent studies
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
      attributing the measurement stability of the upper thoracic region to the collateralization via the subclavian/internal thoracic arteries. Whether in addition to lumbar cnNIRS optodes, mid and especially lower thoracic cnNIRS measurements also are necessary for a complete and consistent oxygenation mapping of the CN is yet unclear and is the focus of ongoing research. The limitation of optode placement directly above the vertebrae has been acknowledged by their respective authors, who suggested that measuring surrounding tissue may enable indirect cord monitoring via the paraspinal CN,
      • LeMaire S.A.
      • Ochoa L.N.
      • Conklin L.D.
      • Widman R.A.
      • Clubb Jr., F.J.
      • Undar A.
      • et al.
      Transcutaneous near-infrared spectroscopy for detection of regional spinal ischemia during intercostal artery ligation: preliminary experimental results.
      • Suehiro K.
      • Funao T.
      • Fujimoto Y.
      • Mukai A.
      • Nakamura M.
      • Nishikawa K.
      • et al.
      Transcutaneous near-infrared spectroscopy for monitoring spinal cord ischemia: an experimental study in swine.
      • Luehr M.
      • von Aspern K.
      • Etz C.D.
      Limitations of direct regional spinal cord monitoring using near-infrared spectroscopy: indirect paraspinal collateral network surveillance is the answer!.
      a concept that represents the rationale for current cnNIRS monitoring practice.
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Haunschild J.
      • Hoyer A.
      • Luehr M.
      • Bakhtiary F.
      • Misfeld M.
      • et al.
      Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
      • Luehr M.
      • von Aspern K.
      • Etz C.D.
      Limitations of direct regional spinal cord monitoring using near-infrared spectroscopy: indirect paraspinal collateral network surveillance is the answer!.
      Although sensitivity of cnNIRS after stent deployment during endovascular aortic repair was recently published in 2 reports (total of 3 patients),
      • Badner N.H.
      • Nicolaou G.
      • Clarke C.F.
      • Forbes T.L.
      Use of spinal near-infrared spectroscopy for monitoring spinal cord perfusion during endovascular thoracic aortic repairs.
      • Moerman A.
      • Van Herzeele I.
      • Vanpeteghem C.
      • Vermassen F.
      • Francois K.
      • Wouters P.
      • et al.
      Near-infrared spectroscopy for monitoring spinal cord ischemia during hybrid thoracoabdominal aortic aneurysm repair.
      available data from previous case series and ongoing clinical investigations suggest that lumbar cnNIRS does not reliably pick up relevant changes in oxygenation during endovascular procedures in humans (not experiencing procedure-related SCI).
      • Etz C.D.
      • von Aspern K.
      • Gudehus S.
      • Luehr M.
      • Girrbach F.F.
      • Ender J.
      • et al.
      Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
      • von Aspern K.
      • Bakhtiary F.
      • Misfeld M.
      • Mohr F.W.
      • Etz C.D.
      Indirect monitoring of spinal cord oxygenation by paraspinal near-infrared spectroscopy. Experimental and clinical data (Article in German).
      It is important to note that optode positioning in the aforementioned case reports was at the lower thoracic level (T9-T11), whereas optodes in current clinical investigations are placed bilaterally at the lumbar level (L1-L3).
      The presented study demonstrates that cnNIRS is capable of reflecting CN oxygenation in real-time after selective SA occlusion, also differentiating between complete SA interruption and patency of 2 SA pairs. Nevertheless, the threshold of critical SA sacrifice necessary to be detected by cnNIRS in humans remains unclear. Whether patients exhibiting stable cnNIRS signals also have sufficient perfusion to maintain adequate spinal cord tissue oxygenation, directly via intraspinal collaterals or indirectly via the paraspinal CN,
      • Etz C.D.
      • Kari F.A.
      • Mueller C.S.
      • Brenner R.M.
      • Lin H.M.
      • Griepp R.B.
      • et al.
      The collateral network concept: remodeling of the arterial collateral network after experimental segmental artery sacrifice.
      • Kari F.A.
      • Beyersdorf F.
      Aortic surgery and spinal collateral flow: a call for structured approaches to functional characterization of the intraspinal collateral system.
      needs to be further investigated.

      The Conundrum of Segmental Artery Patency

      Patency of 2 pairs of segmental arteries led to higher cnNIRS signals and expectably less spinal cord tissue damage with accelerated neurologic recovery. The rationale of maintaining perfusion via the lowest thoracic segmental arteries was mainly chosen on the basis of previous studies, demonstrating that this region exhibits the most distinct spinal cord tissue damage after consecutive SA sacrifice.
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
      These findings are consistent with the results of the presented study (most pronounced SCI between thoracic level 8 and lumbar level 2).
      • Etz C.D.
      • Homann T.M.
      • Luehr M.
      • Kari F.A.
      • Weisz D.J.
      • Kleinman G.
      • et al.
      Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
      • Geisbusch S.
      • Stefanovic A.
      • Koruth J.S.
      • Lin H.M.
      • Morgello S.
      • Weisz D.J.
      • et al.
      Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
      This boundary zone—ranging from T9/10 to L2/3—seems to represent the most vulnerable region for ischemic tissue damage. The favorable clinical outcome and less severe spinal cord tissue damage of animals with patent CN inflow to this region underscore this theory, also emphasizing the clinical potential of real-time cnNIRS.
      During recent years, clinical studies have divided the aortic community.
      • David N.
      • Roux N.
      • Douvrin F.
      • Clavier E.
      • Bessou J.P.
      • Plissonnier D.
      • et al.
      Aortic aneurysm surgery: long-term patency of the reimplanted intercostal arteries.
      • Etz C.D.
      • Halstead J.C.
      • Spielvogel D.
      • Shahani R.
      • Lazala R.
      • Homann T.M.
      • et al.
      Thoracic and thoracoabdominal aneurysm repair: is reimplantation of spinal cord arteries a waste of time?.
      • Wynn M.
      • Acher C.
      • Marks E.
      • Acher C.W.
      The effect of intercostal artery reimplantation on spinal cord injury in thoracoabdominal aortic aneurysm surgery.
      Although it becomes more and more apparent that SA reimplantation during open aortic repair does not provide absolute safety, it may offer some protection against postoperative hemodynamic instabilities. With regard to SCI protection, however, this topic remains highly debated among experts.
      At first glance, it may seem that the presented results are in favor of surgical SA reimplantation. However, it is important to emphasize the fundamental differences in clinical SA reimplantation. In this experiment, no aortic crossclamping, SA ostia opening, or time-consuming surgical reimplantation was performed. Therefore, disadvantageous effects of reimplantation that may potentially mitigate adequate preconditioning, such as back-bleeding with steal from the CN or subsequent thrombotic SA occlusion after reimplantation, did not occur. Rather than SA reimplantation, the presented experimental set-up corresponds to a lopsided, open-staged procedure for CN preconditioning. On the basis of these observations, 2 different future experimental and clinical approaches may be considered. Either (1) SAs of the lower thoracic/upper lumbar region should be spared during a first MIS2ACE session to ensure potential superior neurologic integrity or (2) exclusively this region should be occluded first, thereby potentially facilitating a more pronounced stimulus for CN preconditioning and adequate spinal cord ischemia prophylaxis.

      Clinical Implications and Future Perspectives

      The presented findings underscore the potential of lumbar cnNIRS to be a clinically applicable and easy-to-use real-time monitoring modality particularly during the early postoperative period. Of note, all animals that recovered neurologically managed to reestablish cnNIRS oxygenation values above 30% of baseline within 1 hour postoperatively. Whether the predictive value of cnNIRS for neurologic recovery depends on such an oxygenation threshold (eg, 30% of baseline) or is time-dependent (eg, oxygenation recovery within 1 hour postocclusion) cannot be answered conclusively at this point. For clinical practice, however, the demonstrated correlation with neurologic recovery is encouraging, because it is especially important for sedated patients with prolonged ventilation, for whom the risk of delayed paraplegia during the postoperative period seems the greatest.
      • Estrera A.L.
      • Sheinbaum R.
      • Miller C.C.
      • Azizzadeh A.
      • Walkes J.C.
      • Lee T.Y.
      • et al.
      Cerebrospinal fluid drainage during thoracic aortic repair: safety and current management.
      The fact that lumbar cnNIRS depicted oxygenation changes in the CN over a course of 3 days, potentially reflecting the CN collateralization process, also warrants further clinical and experimental investigation.

      Study Limitations

      Because of anatomic and physiologic differences between species (pigs and humans), conclusions drawn from any translational experimental research should be interpreted with caution. When assessing regional tissue oxygenation via cnNIRS, the emitted light penetrates through the paraspinal vasculature/muscles, averaging measurements over the entire area in its optical path. In pigs, the deep paraspinal muscles are arranged in a mainly sagittal, rather than transversal manner; thus, a longitudinally placed optode on a 45-kg pig always measures a larger portion of the CN compared with an average human adult. Yet, this difference may be solved by using larger animals or measuring/averaging values of an equally large portion of the human CN (eg, by placement of additional optodes).

      Conclusions

      Lumbar cnNIRS reacts to occlusion of segmental arteries in real time and correlates with neurologic outcome. Patency of 2—nonreimplanted—pairs of segmental arteries of the lower thoracic region results in a less pronounced lumbar cnNIRS decrease, potentially facilitating maintenance of sufficient spinal cord perfusion. This preliminary data suggests that cnNIRS may be a valuable noninvasive tool for detecting imminent spinal cord ischemia during and after aortic procedures involving segmental artery occlusion. Further in-depth analyses are warranted to determine the exact threshold for imminent spinal cord injury with regard to absolute cnNIRS values.

      Conflict of Interest Statement

      Authors have nothing to disclose with regard to commercial support.

      Appendix

      Figure thumbnail fx3
      Figure E1The invasive mean arterial blood pressures throughout the experiment are shown (subtotal, N = 5 in blue, versus total, N = 7 in red). No significant differences throughout the experiment and in-between groups can be shown. SA, Segmental artery.
      Figure thumbnail fx4
      Figure E2Individual data points for lumbar cnNIRS. Left: Scatter plot showing cnNIRS throughout the experiment comparing subtotal occlusion (blue dots and blue regression line) and total occlusion (red dots and red regression line). Right: Comparison between cnNIRS measurements for recovering animals (blue dots and blue regression line) and paraplegic animals (green dots and green regression line). Recovering and paraplegic animals seem to have substantial overlap intraoperatively. Values gradually digress throughout the postoperative period (represented by the regression lines). cnNIRS, Collateral network near-infrared spectroscopy; SA, segmental artery.
      Table E1Neurologic assessment (modified Tarlov score)
      Modified Tarlov score
      ScoreDescription
      0No voluntary movements
      1Perceptible movements at joints
      2Good movements at joints but inability to stand
      3Ability to get up with assistance but no ability to stand
      4Ability to get up with assistance and stand with assistance
      5Ability to get up with assistance and stand unassisted
      6Ability to get up and stand unassisted <1 min
      7Ability to get up and stand unassisted >1 min
      8Ability to walk <1 min
      9Ability to walk >1 min
      10Complete recovery
      Table E2Ischemic spinal cord damage score with example images
      ScoreDefinitionImage
      0No apparent necrosis
      1Necrosis of single (motor) neurons only
      2Necrosis of 1 posterior horn only
      3Necrosis of both posterior horns only
      4Necrosis of both posterior horns/surrounding white matter
      5Necrosis of both anterior horns only
      6Central necrosis (posterior/anterior horns/parts of white matter)
      7Complete necrosis of grey matter only
      8Complete necrosis of the whole section
      Table E3Adverse events other than paraplegia
      Adverse events during the experiment
      Postoperative dayDescription
      0Intraoperative aortic laceration located in between 2 SAs with need for subsequent ligation (N = 1)
      0-1Temporary difficulties in micturition (N = 3)
      0-3Neurogenic bladder syndrome with overflow incontinence (N = 2, paraplegic animals only)
      3Mild to moderate pleuritis (incidental autopsy finding), normal oxygenation and behavior (N = 2)
      3Pleural effusion not affecting breathing or general health condition (incidental autopsy finding, N = 4)
      SA, Segmental artery.
      Table E4Lumbar collateral network near-infrared spectroscopy measurements for intervention groups (subtotal and total occlusion)
      Procedural stepGroup 1 (subtotal, N = 5)Group 2 (total, N = 7)
      cnNIRScnNIRS (%)MAPcnNIRScnNIRS (%)MAP
      Preoperative baseline62.4 ± 510070.8 ± 762.5 ± 410067.2 ± 4
      Prior SA occlusion57.3 ± 492.3 ± 867.0 ± 455.5 ± 289.4 ± 467.6 ± 4
      Thoracic SA occlusion54.7 ± 488.1 ± 763.2 ± 453.9 ± 386.3 ± 564.0 ± 6
      Lumbar SA occlusion48.6 ± 978.3 ± 1163.6 ± 445.3 ± 572.7 ± 1065.0 ± 3
      1 min postocclusion48.3 ± 977.8 ± 1162.0 ± 443.9 ± 670.5 ± 1065.4 ± 2
      5 min postocclusion44.9 ± 972.4 ± 1160.8 ± 441.1 ± 665.9 ± 964.4 ± 3
      10 min postocclusion43.0 ± 1069.3 ± 1461.2 ± 336.9 ± 559.2 ± 864.6 ± 3
      60 min postocclusion47.2 ± 876.0 ± 1063.0 ± 539.8 ± 564.3 ± 1067.8 ± 4
      12 h postocclusion49.4 ± 579.5 ± 1066.4 ± 339.3 ± 663.6 ± 1368.4 ± 3
      24 h postocclusion47.7 ± 576.8 ± 967.8 ± 238.1 ± 661.8 ± 1369.4 ± 2
      48 h postocclusion53.4 ± 486.0 ± 665.2 ± 239.5 ± 963.2 ± 1568.7 ± 1
      72 h postocclusion55.6 ± 789.5 ± 965.0 ± 337.7 ± 960.5 ± 1672.4 ± 3
      cnNIRS, Collateral network near-infrared spectroscopy; MAP, mean arterial pressure; SA, segmental artery.
      Table E5Lumbar collateral network near-infrared spectroscopy measurements for recovering and paraplegic and group comparisons
      Procedural stepRecovery group (N = 8)Paraplegic group (N = 4)
      cnNIRScnNIRS (%)MAPcnNIRScnNIRS (%)MAP
      Preoperative baseline62.1 ± 510070.5 ± 563.1 ± 310063 ± 2
      Prior SA occlusion56.5 ± 391.4 ± 666.6 ± 455.7 ± 188.5 ± 470 ± 4
      Thoracic SA occlusion53.9 ± 387.1 ± 662.2 ± 454.7 ± 186.9 ± 469.5 ± 4
      Lumbar SA occlusion46.2 ± 675.0 ± 1263.5 ± 347.2 ± 575.1 ± 1167.2 ± 3
      1 min postocclusion45.0 ± 773.1 ± 1362.8 ± 346.3 ± 673.6 ± 1167.3 ± 2
      5 min postocclusion41.3 ± 663.2 ± 1561.8 ± 444.2 ± 570.2 ± 1066.2 ± 3
      10 min postocclusion38.8 ± 971.3 ± 1562.3 ± 439.5 ± 462.8 ± 965.6 ± 2
      60 min postocclusion43.8 ± 776.5 ± 1464.8 ± 640.6 ± 664.7 ± 1167.5 ± 3
      12 h postocclusion46.9 ± 673.7 ± 1467.5 ± 437.0 ± 659.0 ± 1165.5 ± 2
      24 h postocclusion45.2 ± 773.7 ± 1467.6 ± 336.4 ± 458.0 ± 966.5 ± 4
      48 h postocclusion52.8 ± 585.1 ± 666.6 ± 328.0 ± 543.0 ± 970 ± 3
      72 h postocclusion55.2 ± 888.9 ± 969.6 ± 524.5 ± 937.7 ± 1468 ± 3
      Procedural stepcnNIRS (%)Mean difference95% CIP value
      Comparison: Recovery group
       Prior SA occlusion91.4 ± 6
       10 min postocclusion71.3 ± 1528.2 ± 1116.3-38.1.001
       72 h postocclusion88.9 ± 94.2 ± 9−7.2-14.2.429
      Comparison: Paraplegic group
       Prior SA occlusion88.5 ± 4
       10 min postocclusion62.8 ± 925.5 ± 518.8-29.9.001
       72 h postocclusion37.7 ± 1447.9 ± 99.2-79.5.023
      cnNIRS, Collateral network near-infrared spectroscopy; CI, confidence interval; SA, segmental artery.

      Supplementary Data

      References

        • Koeppel T.A.
        • GA
        • Jacobs M.J.
        DGG Leitlinie-Thorakale und Thorakoabdominelle Aortenaneurysmen.
        Europäisches Gefäßzentrum Aachen-Maastricht, 2010 (Available at: http://www.gefaesschirurgie.de/fileadmin/websites/dgg/download/LL_DTAA_und_TAAA_2011.pdf)
        • Etz C.D.
        • Luehr M.
        • Kari F.A.
        • Bodian C.A.
        • Smego D.
        • Plestis K.A.
        • et al.
        Paraplegia after extensive thoracic and thoracoabdominal aortic aneurysm repair: does critical spinal cord ischemia occur postoperatively?.
        J Thorac Cardiovasc Surg. 2008; 135: 324-330
        • Panthee N.
        • Ono M.
        Spinal cord injury following thoracic and thoracoabdominal aortic repairs.
        Asian Cardiovasc Thorac Ann. 2015; 23: 235-246
        • National Spinal Cord Injury Statistical Center
        Spinal cord injury facts and figures at a glance.
        J Spinal Cord Med. 2013; 36: 1-2
        • Pillai J.B.
        • Pellet Y.
        • Panagopoulos G.
        • Sadek M.A.
        • Abjigitova D.
        • Weiss D.
        • et al.
        Somatosensory-evoked potential-guided intercostal artery reimplantation in thoracoabdominal aortic aneurysm surgery.
        Innovations (Phila). 2013; 8: 302-306
        • Dias-Neto M.
        • Reis P.V.
        • Rolim D.
        • Ramos J.F.
        • Teixeira J.F.
        • Sampaio S.
        • et al.
        Strategies to prevent TEVAR-related spinal cord ischemia.
        Vascular. 2017; 25: 307-315
        • LeMaire S.A.
        • Ochoa L.N.
        • Conklin L.D.
        • Widman R.A.
        • Clubb Jr., F.J.
        • Undar A.
        • et al.
        Transcutaneous near-infrared spectroscopy for detection of regional spinal ischemia during intercostal artery ligation: preliminary experimental results.
        J Thorac Cardiovasc Surg. 2006; 132: 1150-1155
        • Demir A.
        • Erdemli O.
        • Unal U.
        • Tasoglu I.
        Near-infrared spectroscopy monitoring of the spinal cord during type B aortic dissection surgery.
        J Card Surg. 2013; 28: 291-294
        • Etz C.D.
        • Kari F.A.
        • Mueller C.S.
        • Silovitz D.
        • Brenner R.M.
        • Lin H.M.
        • et al.
        The collateral network concept: a reassessment of the anatomy of spinal cord perfusion.
        J Thorac Cardiovasc Surg. 2011; 141: 1020-1028
        • Etz C.D.
        • Kari F.A.
        • Mueller C.S.
        • Brenner R.M.
        • Lin H.M.
        • Griepp R.B.
        • et al.
        The collateral network concept: remodeling of the arterial collateral network after experimental segmental artery sacrifice.
        J Thorac Cardiovasc Surg. 2011; 141: 1029-1036
        • Etz C.D.
        • von Aspern K.
        • Gudehus S.
        • Luehr M.
        • Girrbach F.F.
        • Ender J.
        • et al.
        Near-infrared spectroscopy monitoring of the collateral network prior to, during, and after thoracoabdominal aortic repair: a pilot study.
        Eur J Vasc Endovasc Surg. 2013; 46: 651-656
        • von Aspern K.
        • Haunschild J.
        • Hoyer A.
        • Luehr M.
        • Bakhtiary F.
        • Misfeld M.
        • et al.
        Non-invasive spinal cord oxygenation monitoring: validating collateral network near-infrared spectroscopy for thoracoabdominal aortic aneurysm repair.
        Eur J Cardiothorac Surg. 2016; 50: 675-683
        • Boezeman R.P.
        • van Dongen E.P.
        • Morshuis W.J.
        • Sonker U.
        • Boezeman E.H.
        • Waanders F.G.
        • et al.
        Spinal near-infrared spectroscopy measurements during and after thoracoabdominal aortic aneurysm repair: a pilot study.
        Ann Thorac Surg. 2015; 99: 1267-1274
        • Etz C.D.
        • Debus E.S.
        • Mohr F.W.
        • Kolbel T.
        First-in-man endovascular preconditioning of the paraspinal collateral network by segmental artery coil embolization to prevent ischemic spinal cord injury.
        J Thorac Cardiovasc Surg. 2015; 149: 1074-1079
        • Von Aspern K.
        • Luehr M.
        • Mohr F.W.
        • Etz C.D.
        Spinal cord protection in open- and endovascular thoracoabdominal aortic aneurysm repair: critical review of current concepts and future perspectives.
        J Cardiovasc Surg (Torino). 2015; 56: 745-749
        • Etz C.D.
        • Homann T.M.
        • Luehr M.
        • Kari F.A.
        • Weisz D.J.
        • Kleinman G.
        • et al.
        Spinal cord blood flow and ischemic injury after experimental sacrifice of thoracic and abdominal segmental arteries.
        Eur J Cardiothorac Surg. 2008; 33: 1030-1038
      1. Guide for the Care and Use of Laboratory Animals. 8th ed. The National Academies Press, Washington, DC2011
        • Geisbusch S.
        • Stefanovic A.
        • Koruth J.S.
        • Lin H.M.
        • Morgello S.
        • Weisz D.J.
        • et al.
        Endovascular coil embolization of segmental arteries prevents paraplegia after subsequent thoracoabdominal aneurysm repair: an experimental model.
        J Thorac Cardiovasc Surg. 2014; 147: 220-226
        • Bischoff M.S.
        • Scheumann J.
        • Brenner R.M.
        • Ladage D.
        • Bodian C.A.
        • Kleinman G.
        • et al.
        Staged approach prevents spinal cord injury in hybrid surgical-endovascular thoracoabdominal aortic aneurysm repair: an experimental model.
        Ann Thorac Surg. 2011; 92: 138-146
        • Suehiro K.
        • Funao T.
        • Fujimoto Y.
        • Mukai A.
        • Nakamura M.
        • Nishikawa K.
        • et al.
        Transcutaneous near-infrared spectroscopy for monitoring spinal cord ischemia: an experimental study in swine.
        J Clin Monit Comput. 2017; 31: 975-979
        • Badner N.H.
        • Nicolaou G.
        • Clarke C.F.
        • Forbes T.L.
        Use of spinal near-infrared spectroscopy for monitoring spinal cord perfusion during endovascular thoracic aortic repairs.
        J Cardiothorac Vasc Anesth. 2011; 25: 316-319
        • Moerman A.
        • Van Herzeele I.
        • Vanpeteghem C.
        • Vermassen F.
        • Francois K.
        • Wouters P.
        • et al.
        Near-infrared spectroscopy for monitoring spinal cord ischemia during hybrid thoracoabdominal aortic aneurysm repair.
        J Endovasc Ther. 2011; 18: 91-95
        • Luehr M.
        • von Aspern K.
        • Etz C.D.
        Limitations of direct regional spinal cord monitoring using near-infrared spectroscopy: indirect paraspinal collateral network surveillance is the answer!.
        Ann Thorac Surg. 2016; 101: 1238-1239
        • von Aspern K.
        • Bakhtiary F.
        • Misfeld M.
        • Mohr F.W.
        • Etz C.D.
        Indirect monitoring of spinal cord oxygenation by paraspinal near-infrared spectroscopy. Experimental and clinical data (Article in German).
        Gefässchirurgie. 2017; 22: 102-109
        • Kari F.A.
        • Beyersdorf F.
        Aortic surgery and spinal collateral flow: a call for structured approaches to functional characterization of the intraspinal collateral system.
        J Thorac Cardiovasc Surg. 2015; 149: 1675-1680
        • David N.
        • Roux N.
        • Douvrin F.
        • Clavier E.
        • Bessou J.P.
        • Plissonnier D.
        • et al.
        Aortic aneurysm surgery: long-term patency of the reimplanted intercostal arteries.
        Ann Vasc Surg. 2012; 2012: 839-844
        • Etz C.D.
        • Halstead J.C.
        • Spielvogel D.
        • Shahani R.
        • Lazala R.
        • Homann T.M.
        • et al.
        Thoracic and thoracoabdominal aneurysm repair: is reimplantation of spinal cord arteries a waste of time?.
        Ann Thorac Surg. 2006; 82: 1670-1677
        • Wynn M.
        • Acher C.
        • Marks E.
        • Acher C.W.
        The effect of intercostal artery reimplantation on spinal cord injury in thoracoabdominal aortic aneurysm surgery.
        J Vasc Surg. 2016; 64: 289-296
        • Estrera A.L.
        • Sheinbaum R.
        • Miller C.C.
        • Azizzadeh A.
        • Walkes J.C.
        • Lee T.Y.
        • et al.
        Cerebrospinal fluid drainage during thoracic aortic repair: safety and current management.
        Ann Thorac Surg. 2009; 88: 9-15

      Linked Article

      • Commentary: Toward safer aortic surgery: Monitoring spinal cord perfusion with near-infrared spectroscopy
        The Journal of Thoracic and Cardiovascular SurgeryVol. 158Issue 1
        • Preview
          Spinal cord perfusion is provided by segmental arteries that originate from the vertebral, intercostal, lumbar, and hypogastric arteries. This segmental origin of spinal perfusion is supplemented by an extensive intraspinal and paraspinal anastomotic network that provides longitudinal collateral circulation among the horizontally distributed segmental arteries. However, this longitudinal anastomotic network is usually insufficient to maintain spinal perfusion when acute or extensive interruption of the segmental system occurs.
        • Full-Text
        • PDF
        Open Archive