Secondary brain injury in trauma patients: The effects of remote ischemic conditioning

Secondary brain injury in trauma patients: The effects ofremote ischemic conditioning

Bellal Joseph, MD, Viraj Pandit, MD, Bardiya Zangbar, MD, Narong Kulvatunyou, MD, Mazhar Khalil, MD,Andrew Tang, MD, Terence O’Keeffe, MD, Lynn Gries, MD, Gary Vercruysse, MD, Randall S. Friese, MD,

and Peter Rhee, MD, Tucson, Arizona

BACKGROUND: Management of traumatic brain injury (TBI) is focused on preventing secondary brain injury. Remote ischemic conditioning(RIC) is an established treatment modality that has been shown to improve patient outcomes secondary to inflammatory insults.The aim of our study was to assess whether RIC in trauma patients with severe TBI could reduce secondary brain injury.

METHODS: This prospective consented interventional trial included all TBI patients admitted to our Level 1 trauma center with an in-tracranial hemorrhage and a Glasgow Coma Scale (GCS) score of 8 or lower on admission. In each patient, four cycles of RICwere performed within 1 hour of admission. Each cycle consisted of 5 minutes of controlled upper limb (arm) ischemiafollowed by 5 minutes of reperfusion using a blood pressure cuff. Serum biomarkers of acute brain injury, S-100B, and neuron-specific enolase (NSE) were measured at 0, 6, and 24 hours. Outcome measure was reduction in the level of serum biomarkersafter RIC.

RESULTS: A total of 40 patients (RIC, 20; control, 20) were enrolled. The mean (SD) age was 46.15 (18.64) years, the median GCS scorewas 8 (interquartile range, 3Y8), and the median head Abbreviated Injury Scale (AIS) score was 3 (interquartile range, 3Y5),and there was no difference between the RIC and control groups in any of the baseline demographics or injury characteristicsincluding the type and size of intracranial bleed or skull fracture patterns. There was no difference in the 0-hour S-100B (p =0.9) and NSE (p = 0.72) level between the RIC and the control group. There was a significant reduction in the mean levels ofS-100B (p = 0.01) and NSE (p = 0.04) at 6 hours and 24 hours in comparison with the 0-hour level in the RIC group.

CONCLUSION: This study showed that RIC significantly decreased the standard biomarkers of acute brain injury in patients with severe TBI.Our study highlights the novel therapeutic role of RIC for preventing secondary brain insults in TBI patients. (J Trauma AcuteCare Surg. 2015;00: 00Y00. Copyright * 2015 Wolters Kluwer Health, Inc. All rights reserved.)

LEVEL OF EVIDENCE: Prospective interventional study, level II.KEY WORDS: Remote ischemic conditioning; traumatic brain injury; S-100B; neuron-specific enolase.

Traumatic brain injury (TBI) remains one of the leadingcauses of death and disability in the United States.1 Sec-

ondary brain injury caused by the complex interplay of in-flammatory mediators induced by a primary insult is the majorcontributing factor for morbidity and mortality after TBI.2

While the primary injury is irreversible, the inflammatorycascade leading to the development of the secondary injurymay be preventable.3 As a result, all the current research in TBIis focused on preventing initiation of this secondary insult.

Remote ischemic conditioning (RIC) is a process wherenormal tissues are subjected to short cycles of ischemia andreperfusion, which have been shown to reduce the sequelae ofan ischemic injury at a remotely injured site.4 RIC has beenshown to improve the outcomes after myocardial infarction,

sepsis, transplantation, reimplantation, and elective neurologicsurgery.5Y11 It is thought to work by releasing endogenoussystemic anti-inflammatory mediators and humoral factors andby using neural pathways, rendering global protection to thebody against subsequent ischemic insults in a remote area.4,12

This protection provided by RIC has two phases, an early(short) phase and a late (prolonged) phase, both of which haveproven to be effective in reducing ischemic size and improvingsurvival.13Y15 However, its role in patients with TBI remainsunclear. The aim of our study was to assess the effects of RIC intrauma patients with severe TBI. We hypothesized that RICreduces the serum level of biomarkers of brain injury in traumapatients with severe brain injury.

PATIENTS AND METHODS

After the approval from the institutional review board atthe College of Medicine at the University of Arizona wasobtained, we performed a prospective interventional study toevaluate the effects of RIC in patients with TBI. All patientswere evaluated and treated at our Level 1 trauma center fromSeptember 2013 to June 2014. Awritten informed consent wasobtained from the patient or patient’s relativewithin 24 hours ofadmission for enrollment in the study.

ORIGINAL ARTICLE

J Trauma Acute Care SurgVolume 00, Number 00 1

Submitted: September 9, 2014, Revised: December 12, 2014, Accepted: January12, 2015.

From the Division of Trauma, Critical Care, Burns, and Emergency Surgery, De-partment of Surgery, University of Arizona Medical Center, Tucson, Arizona.

This study was part of the podium presentation at the 73rd Annual Meeting of theAmerican Association for the Surgery of Trauma, September 11Y13, 2014, inPhiladelphia, Pennsylvania.

Address for reprints: Bellal Joseph, MD, Division of Trauma, Critical Care, Burns,and Emergency Surgery, Department of Surgery, University of Arizona, 1501 NCampbell Ave, Room 5411, PO Box 245063, Tucson, AZ 85724; email:[email protected].

DOI: 10.1097/TA.0000000000000584

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

mailto:[email protected]

Inclusion and ExclusionIn this study, we included all patients with blunt TBI with

Glasgow Coma Scale (GCS) score of 8 or lower and an in-tracranial hemorrhage on the initial brain computed tomogra-phy (CT) scan. Patients who are 18 years or younger, patientson antiplatelet or anticoagulation medications, patients trans-ferred from other institutions, patients dead on presentation,patients undergoing emergent neurosurgical intervention (pa-tients taken to the operating room immediately after initial CTscan in whom RIC could not be performed), and patients whodid not consent for enrollment were excluded from the study.We also excluded patients with concomitant spinal cord inju-ries, patients who died before completion of protocol, andpatients with an intracranial pressure monitor placement beforeperformance of RIC because of the their potential effect onneuronal markers of injury. These patients were excluded toselect a homogenous patient population independent ofconfounding external factors. The presence of the initial in-tracranial hemorrhage was based on the evaluation of the at-tending radiologist, and these findings were confirmed by asingle investigator.

Data PointsWe prospectively recorded the following data points in

each patient: patient demographics, which included age, sex,and mechanism of injury; vital parameters on presentation,which included systolic blood pressure (SBP), heart rate,temperature, and GCS score; neurologic examination on pre-sentation; intoxication (drug or alcohol); details regardingantiplatelet and anticoagulation therapy; intubation, loss ofconsciousness; initial and repeat brain CT scan findings;neurosurgical intervention details; hospital and intensive careunit length of stay (LOS); discharge disposition; GCS score ondischarge; and in-hospital mortality. The Injury Severity Score(ISS) and head Abbreviated Injury Scale (h-AIS) score wereobtained from the trauma registry. We defined neurosurgicalintervention as operative intervention (craniotomy, craniectomy)and/or placement of intracranial pressure monitor. In addition,we also recorded the levels of markers of neuronal injury (neu-ron-specific enolase [NSE] and S-100B) in the blood samplesobtained at 0, 6, and 24 hours. Patients were observed for edema,ecchymosis, thromboembolic disease process, skin breakdown,or development of neurologic complications (loss of sensationor motor function) on the site of application of the cuff duringtheir hospital stay.

Power AnalysisA power analysis was performed to determine the

number of patients required to be included in each group(treatment and control group). The sample size was estimatedbased on previous studies in TBI, which have assessed markersof head injury (NSE and S-100B). The power of the study wascalculated to identify differences in the treatment and controlgroups for a decrease in the level of serum biomarkers on in-jury. After two-sided power analysis, statistical power of 80%,and an > of 0.05, we calculated a sample size of 20 per group(RIC and control groups).

Study Protocol1. Patients meeting inclusion criteria were evaluated on ad-

mission in the trauma bay of our Level 1 trauma center forenrollment in the study.

2. Based on our power analysis, 20 consecutive patients withbrain injury meeting inclusion criteria were enrolled in theRIC group followed by another 20 consecutive TBI pa-tients in the control group.

3. In in the RIC group, ischemic conditioning was performedusing a standard manual blood pressure cuff. The pressurein the blood pressure cuff was maintained at 30 mm Hghigher than the patient’s SBP. Four cycles of ischemicconditioning were performed within 1 hour of admission.Each cycle consisted of 5 minutes of controlled upper limbischemia (cuff up) followed by 5 minutes of reperfusion(cuff down). The total treatment cyclewas 40minutes. Thisstudy protocol was based on the published literature thathas used the concept of RIC in nontrauma patients.16

4. Blood samples were collected in both the groups at 0 hour(before the application of the blood pressure cuff and per-formance of RIC in treatment group), 6 hours, and 24 hours.Blood samples were assessed for serum biomarkers of acutebrain injury, S-100B, and NSE.

5. Patients were followed up through their hospital course toassess for patient outcomes and complications at the site ofattachment of cuff.

Outcome MeasuresThe primary outcome measure was level of markers of

neuronal injury (NSE and S-100B) in both the groups (RIC andcontrol). The secondary outcome was local site complicationsat the site of attachment of cuff. Complications at the site ofattachment of cuff were defined as edema, ecchymosis,thromboembolic disease process, skin breakdown, or devel-opment of neurologic complications (loss of sensation or motorfunction).

Neuronal Biomarkers of InjuryIn this study, we used S-100B and NSE as biomakers of

neuronal injury. These are established markers of brain injuryand are not routinely found in blood. The baseline level of thesemarkers in patients without brain injury is zero. Both NSE(sensitivity, 80%; specificity, 73%) and S-100B (sensitivity,90%; specificity, 78%) have high sensitivity and specificity forassessing neuronal injury. These markers are routinely elevatedafter neuronal injury, and the severity of brain injury has beenshown to be proportional to the level of these markers in bloodstream. In addition, sustained elevation of these markers afterbrain injury is known to be associated with worse outcomes.The levels of neuronal markers of injury in blood were ana-lyzed by enzyme-linked immunosorbent assay using commer-cially available kits (NSE, Biotang Inc., Walham, MA; S-100B,Biovender, Candor, NC).3,17

Statistical AnalysisData are reported as mean (SD) for continuous de-

scriptive variables, as median (range) for ordinal descriptivevariables, and as proportion for categorical variables. We usedMann-Whitney U-test and the Student’s t test to explore for

J Trauma Acute Care SurgVolume 00, Number 00Joseph et al.

2 * 2015 Wolters Kluwer Health, Inc. All rights reserved.


differences in the two groups (RIC and control) for continuousvariables. We used W

2 test to identify differences in outcomesbetween the two groups for categorical variables. Analysisof variance was used to assess for difference in the level ofbiomakers at different time points. The one-way analysis ofvariance test was used to perform repeated-measures analysis.For our study, we considered p G 0.05 as statistically significant.All statistical analyses were performed using the SPSS version21 (IBM, Inc., Armonk, NY).

RESULTS

During the study period, a total of 59 patients wereprospectively screened, of whom 40 patients (RIC, 20; control,20) were included in the study. Of the 19 patients who werescreened and excluded, 4 patients underwent emergent craniot-omy, 3 patients underwent external ventrical drain placementbefore performance of RIC, 3 patients died within 24 hours ofadmission, 5 patients did not consent for enrollment in the study,and the remaining 4 patients were transferred from other hospitals.

Of the patients included in the study (n = 40), the mean(SD) agewas 46.15 (18.64) years, 72.5% (n = 29) were male, themedian GCS score was 8 (range, 3Y8), the mean (SD) admissionSBP was 113.1 (20.6) mm Hg, and the median h-AIS score was3 (range, 3Y5). TheRIC group and the control groupwere similarin demographic characteristics, admission vital parameters, andinjury severity parameters. Table 1 demonstrates the demo-graphic details of the study population.

Subarachnoid hemorrhage (50%) followed by subduralhemorrhage (47.5%) was the most common type of intracranialhemorrhages in both the groups. Skull fractures were seen in52.5% (n = 21) of the patients, of whom 38.1% (n = 8) haddisplaced skull fracture. There was no difference in the type(p = 0.52) and size (p = 0.41) of intracranial hemorrhage andthe type (p = 0.73) of skull fracture between the two groups.Table 2 demonstrates the injury details of the study population.

The mean (SD) 0-hour S-100B value was 429.83(107.96) pg/mL, while the mean (SD) NSE value was 316.22

(77.6) pg/mL for the entire study population. There was nodifference in the 0-hour S-100B (p = 0.9) and NSE (p = 0.72)levels between the RIC and the control group. However, therewas a significant reduction in the levels of S-100B (p = 0.04)and NSE (p = 0.01) at 6 hours and 24 hours in the RIC groupwhen compared with the 0-hour level. However, there was asignificant increase in the levels of S-100B (p = 0.001) andNSE (p = 0.004) at 6 hours and 24 hours in comparison with the0-hour level in the control group. On performing pairedanalysis for comparing level of biomarkers, there was a sig-nificant reduction in the level of NSE and S-100B at 6 hoursand 24 hours in the RIC group compared with the controlgroup. Figures 1 and 2 demonstrate the comparison of markersbetween the RIC and the control group at 0, 6, and 24 hours.

On subanalysis of patients without neurosurgical inter-vention (n = 31: RIC, 16; control, 15), all patients in the RICgroup had a significant decline in the level of S-100B (p = 0.01)and NSE (p = 0.001) at 6 hours and 24 hours in comparisonwith the 0-hour level. Contrastingly, there was a significantincrease in the levels of S-100B (p = 0.001) and NSE (p = 0.04)at 6 hours and 24 hours in comparison with the 0-hour level inall the patients in the control group.

TABLE 1. Demographics

Characteristics RIC (n = 20) No RIC (n = 20) p

Age, mean (SD), y 46.84 (21.7) 45.73 (17.6) 0.64

Q65 y, % (n) 20 (4) 15 (3) 0.67

Male, % (n) 70 (14) 75 (15) 0.72

GCS score, median (IQR) 8 (3Y8) 8 (3Y8) 0.96

ED SBP, mean (SD) 115.8 (21.5) 109.6 (19.7) 0.72

ED heart rate, mean (SD) 84.2 (28.4) 87.9 (23.2) 0.63

ED temperature, mean (SD) 37.3 (1.1) 37.1 (1.6) 0.81

Platelet count, �103, mean (SD) 196.7 (86.4) 201.4 (72.9) 0.75

INR, mean (SD) 1.4 (0.3) 1.3 (0.5) 0.68

Mechanism of injury V

MVC, % (n) 60 (12) 70 (14) 0.51

Fall, % (n) 25 (5) 20 (4) 0.70

ISS, median (IQR) 18 (9Y19) 17 (9Y18) 0.68

Head AIS score, median (IQR) 3 (3Y5) 3 (3Y5) 0.84

ED, emergency department; INR, international normalized ratio; MVC, motorvehicle collision.

TABLE 2. Initial Head CT Scan Findings


Skull fracture, % (n)

Nondisplaced 30 (6) 35 (7) 0.73

Displaced 25 (5) 15 (3) 0.42

Intracranial hemorrhage

SDH, % (n) 45 (9) 50 (10) 0.75

Size , mean (SD), mm 6.4 (3.8) 5.9 (4.5) 0.38

EDH, % (n) 20 (4) 30 (6) 0.47

Size, mean (SD), mm 4.3 (3.7) 4.8 (4.1) 0.41

SAH, % (n) 55 (11) 45 (9) 0.52

IPH, % (n) 15 (3) 10 (2) 0.63

IVH, % (n) 25 (5) 30 (6) 0.72

EDH, epidural hemorrhage; ICH, intracranial hemorrhage; IPH, intraparenchymalhemorrhage; IVH, intraventricular hemorrhage; NC, neurosurgical consultation; SAH,subarachnoid hemorrhage; SDH, subdural hemorrhage.

Figure 1. Comparison between the control and the RIC groupfor the marker S-100B.

J Trauma Acute Care SurgVolume 00, Number 00 Joseph et al.

* 2015 Wolters Kluwer Health, Inc. All rights reserved. 3


Nine patients (RIC, 4; control, 5) underwent neurosur-gical intervention. Of the four patients (craniotomy, 3; intra-cranial pressure monitor placement, 1) in the RIC group thathad intervention, the level of NSE and S-100B increased in thethree patients with craniotomy; however, in the patient withintracranial pressure monitor, the markers remained unchangedat 0, 6, and 24 hours. All the patients in the control group hadsignificant increase in the level of both NSE and S-100 markersat 6 hours and 24 hours in comparison with the 0-hour level. Allpatients who underwent neurosurgical intervention had an in-crease in their levels of NSE and S-100B regardless of RIC.

Table 3 demonstrates the outcomes between the RIC andthe control group. No patient developed complications at thesite of attachment of blood pressure cuff in the study popu-lation. Forty-five percent (n = 18) had progression on repeatbrain CT scan, and neurosurgical intervention was performedin 22.5%9 of the patients after performance of the treatment.There was no difference in the rate of progression (p = 0.52)and neurosurgical intervention (p = 0.69) between the RIC andthe control group. The median discharge GCS score was 13(range, 11Y15), and 40% (n = 16) of the patients weredischarged home. There was no difference in the dischargeGCS score (p = 0.76) and discharge disposition (p = 0.52)among the two groups. The overall mortality rate was 22.5%(n = 9). There was no difference in the mortality rate (p = 0.70)between the RIC and the control group.

DISCUSSION

The mainstay in the management of blunt TBI patientscontinues to be focused on the prevention of secondary braininjury. In this study, we have demonstrated a potential thera-peutic role of a novel treatment modality, RIC in patients withsevere brain injury. In a cohort of severe blunt TBI patients, wefound a significant reduction in the markers of neuronal injuryin patients who received ischemic conditioning. This was incomparison with the control group that had significant in-creases in the biomarkers as expected. This study is a proof ofconcept study, defining the potential role of RIC in TBI. The

use of the concept of RIC in patients with brain injury has thepotential to redefine the management of patients with TBI.

Several studies have shown an acute increase in the levelof both NSE and S-100B after brain injury and a sustainedelevation of these makers until 24 hours after neuronalinsult.18Y20 The initial elevation in the level of biomarkers wasconsistent with the published trends of these markers.3 In ourcontrol group, there was a significant increase in the level ofthese markers during the 24-hour period. This was consistentwith the normal pattern of these markers in patients withneuronal injury. However, we found a significant reduction inthe level of both NSE and S-100B after RIC at 6 hours and24 hours in the RIC group. In addition, there was a significantdifference in the level of both the markers (NSE and S-100B) at6-hour and 24-hours time points between the RIC and thecontrol group. We believe that this difference in the level ofmarkers of neuronal injury may have been contributed by theprotective or therapeutic effects of RIC. The activation ofmultiple inflammatory pathways in the RIC group may haveblunted the secondary inflammatory insults and provided aglobal protection. Although in patients with intervention therewas a rise in the levels of serum markers of neuronal injury, webelieve that the performance of an intervention caused furtherdamage to the neurons and caused elevation of neuronalmarkers. However, the level of biomarkers remained unchangedin the patient with intracranial pressure monitor in the RICgroup and may demonstrate the protective effect of RIC. Webelieve that assessing the true clinical outcomes in patientswith neurosurgical intervention in the RIC group may actuallybetter help to define the role of RIC in this patient population.

Although there are multiple markers of brain injury, inour study, we used the two (NSE and S-100B) well-establishedmarkers of brain injury to assess the effect of RIC in patientswith severe TBI. S-100B is a calcium binding protein, which isfound in the astrocytes, Schwann cells, and the oligodendro-cytes.21 In contrast, NSE is a cytoplasmic neuronal enzyme.22

Several studies have shown the association between thesemarkers and severity of brain injury. Stein et al.17 demonstratedan association between the level of these markers and cerebral

Figure 2. Comparison between the control and the RIC groupfor the marker NSE. *Paired comparison between the RIC andthe control group at each time point.

TABLE 3. Outcomes


RHCT, % (n) 60 (12) 55 (11) 0.75

Progression on RHCT, % (n) 40 (8) 50 (10) 0.52

Neurosurgical intervention, % (n) 20 (4) 25 (5) 0.69

Cuff site complications, % (n) 0 (0) 0 (0) 0.98

LOS

Hospital LOS, median (IQR) 5 (4Y6) 5 (3Y6) 0.37

ICU LOS, median (IQR) 3 (2Y3) 3 (2Y3) 0.54

Discharge GCS score 13 (12Y15) 13 (11Y15) 0.76

Discharge disposition

Home, % (n) 45 (9) 35 (7) 0.52

Rehabilitation center, % (n) 30 (6) 30 (6) 0.91

Skilled nursing facility, % (n) 5 (1) 10 (2) 0.54

Mortality, % (n) 20 (4) 25 (5) 0.70

ED, emergency department; NC, neurosurgical consultation; RHCT, repeat head CT.




hypoperfusion. In addition, in another study, Stein et al.3 alsodemonstrated the association of elevated level of these markersand worse outcomes. Although in our study we did not assesslong-term brain function, the levels of both NSE and S-100Bon admission as well as 6 hours and 24 hours after admis-sion provided evidence to assess the utility of RIC in patientswith severe injury.

The development of secondary brain injury is an im-portant determinant of clinical outcomes in patients with TBI.Secondary brain injury is thought to occur hours to days afterprimary insult. It develops due to the activation of anaerobicglycolysis in the brain leading to initiation of pathophysiologiccascade, which ultimately leads to recruitment of inflammatorycells and release of inflammatory mediators.2,23 Cerebral is-chemia, cerebral hypoxia, systemic hypotension, cerebraledema, and elevated intracranial pressure are considered to beresponsible for the development of secondary injury.23Y26 Al-though guidelines have established the management of hypo-tension, cerebral edema, and intracranial hypertension inpatients with TBI, the optimal therapeutic management of is-chemic injury in TBI patients remains unclear.27Y30 Our studydemonstrates the novel therapeutic role of RIC that may help inthe management of secondary brain injury in patients with TBI.It also proves that RIC can be applied after injury effectivelyand easily without obvious consequences.

The concept of ischemic preconditioning was initiallyestablished by Murry et al.31 who demonstrated the protectiveeffect of ischemic preconditioning on ischemic myocardium.They demonstrated a 75% reduction in the size of the myo-cardial infarct in the ischemic conditioning group in compar-ison with the control group. Since this original study, theprotective benefits of ischemic preconditioning have beenwidely studied in multiple human and animal studies, whichhave established the safety and protective effects of ischemicconditioning in a number of organs and tissues including thekidneys, lungs, liver, skin flaps, ovaries, intestine, stomach,pancreas, and neurons.32Y36 However, to our knowledge, therehave been no studies assessing the role of RIC in TBI patients.Proposed theoretical mechanism of RIC is through the acti-vation of a basic cellular stress response leading to multiplegene expressions and initiation of humoral, neuronal, andsystemic pathways.7,37Y39 Activation of these pathways pro-vides protection against subsequent ischemic injury by im-proving endothelial function and also cerebral blood flow.However, given the similarities in the mechanism of action onRIC and the development of secondary brain injury, we soughtout to evaluate its role in patients with TBI. We found that RICreduced the levels of standard accepted neuronal injurymarkers compared with the control group (no RIC). RIC mayhave a protective or therapeutic role in patients with brain in-jury and may help to prevent secondary brain insults.

There remains a fear of development of local compli-cations when using a blood pressure cuff for providing RIC.However, in our study, we did not identify any complications atthe site of application of the cuff. The results of our study aresimilar to the other studies that demonstrated no local com-plications while performing RIC.12,13,37

The results of this study should be interpreted in contextof its limitations. Although a prospective study, we did not

randomize patients and consecutively enrolled patients in boththe groups, and the sample size is relatively small to assess theimpact of RIC on patient outcomes. We assessed the level ofmarkers in patients’ blood samples for only 24 hours, and thelevel of markers was not assessed for a longer period. Otherknown markers of brain injury such as tau protein and glialfibrillary acidic protein were not assessed.We did not assess thetime of RIC initiation from the time of injury. We did not assessthe impact of RIC on long-term brain function and clinicaloutcomes of the patients. The lack of clinical differences can becaused by multiple reasons and needs further research. We didnot assess the volume of intracranial hemorrhage to comparethe severity of injury among the control and RIC groups. Thestudy population may not be ideal as the TBImay have been toosevere or not severe enough. RIC could not have been extensiveenough or not applied frequently enough or even applied toomuch. These variables need to be tested in both animal andhuman subjects. However, despite these limitations, this studyis the first study to our knowledge to highlight the beneficialeffect of RIC in patients with severe TBI.

CONCLUSION

RIC with a standard blood pressure cuff applied to thearm within 1 hour of patient’s arrival to the trauma centersignificantly decreased the standard biomarkers of acute braininjury. This study demonstrates safety and feasibility of RIC inTBI patients. RIC may have a novel therapeutic role for po-tentially preventing secondary brain insults in TBI patients.Further research assessing the impact of RIC in TBI patients todetermine the ideal patient population and method forperforming RIC could potentially revolutionize the manage-ment of TBI.

AUTHORSHIP

B.J., V.P, R.S.F., and P.R. contributed in the study conception and design.B.J., V.P., M.K., N.K., B.Z., A.T., T.O., G.V., L.G., R.S.F., and P.R.performed the acquisition of data. B.J., V.P., M.K., N.K., B.Z., A.T., L.G.,and P.R. performed the analysis and interpretation of data. B.J., V.P.,M.K., N.K., B.Z., T.O., L.G., R.S.F., and P.R. drafted the manuscript. Allauthors participated in the critical revision.

DISCLOSURE

The authors declare no conflicts of interest.

REFERENCES1. CDCVTBI in the US ReportVTraumatic Brain InjuryVInjury Center

2014. Available at: http://www.cdc.gov/traumaticbraininjury/tbi_ed.html#3.Accessed July 22, 2014

2. Werner C, Engelhard K. Pathophysiology of traumatic brain injury. Br JAnaesth. 2007;99(1):4Y9.

3. Stein DM, Kufera JA, Lindell A,Murdock KR,Menaker J, Bochicchio GV,Aarabi B, Scalea TM. Association of CSF biomarkers and secondary in-sults following severe traumatic brain injury. Neurocrit Care. 2011;14(2):200Y207.

4. Saxena P, Newman MA, Shehatha JS, Redington AN, Konstantinov IE.Remote ischemic conditioning: evolution of the concept, mechanisms, andclinical application. J Card Surg. 2010;25(1):127Y134.

5. Munk K, Andersen NH, Schmidt MR, Nielsen SS, Terkelsen CJ, Sloth E,Bøtker HE, Nielsen TT, Poulsen SH. Remote ischemic conditioning inpatients with myocardial infarction treated with primary angioplasty




http://www.cdc.gov/traumaticbraininjury/tbi_ed.html#3

impact on left ventricular function assessed by comprehensive echocar-diography and gated single-photon emission CT. Circ Cardiovasc Imag-ing. 2010;3(6):656Y662.

6. Lim SY, Hausenloy DJ. Remote ischemic conditioning: from bench tobedside. Front Physiol. 2012;3:27.

7. Konstantinov IE, Arab S, Kharbanda RK, Li J, Cheung MM, CherepanovV, Downey GP, Liu PP, Cukerman E, Coles JG. The remote ischemicpreconditioning stimulus modifies inflammatory gene expression inhumans. Physiol Genomics. 2004;19(1):143Y150.

8. Steiger HJ, Hanggi D. Ischaemic preconditioning of the brain, mechanismsand applications. Acta Neurochir. 2007;149(1):1Y10.

9. Konstantinov IE, Li J, CheungMM, ShimizuM, Stokoe J, Kharbanda RK,Redington AN. Remote ischemic preconditioning of the recipient reducesmyocardial ischemia-reperfusion injury of the denervated donor heart via a KatpchannelYdependent mechanism. Transplantation. 2005;79(12):1691Y1695.

10. Hu S, Dong H-l, Li YZ, Luo ZJ, Sun L, Yang QZ, Yang LF, Xiong L.Effects of remote ischemic preconditioning on biochemical markers andneurologic outcomes in patients undergoing elective cervical decom-pression surgery: a prospective randomized controlled trial. J NeurosurgAnesthesiol. 2010;22(1):46Y52.

11. Sahebally SM, Healy D, Coffey JC, Walsh SR: Should patients takingaspirin for secondary prevention continue or discontinue the medicationprior to elective, abdominal surgery? Best evidence topic (BET). Int JSurg. 2014;12:16Y21.

12. Loukogeorgakis SP, Williams R, Panagiotidou AT, Kolvekar SK, DonaldA, Cole TJ, Yellon DM, Deanfield JE, MacAllister RJ. Transient limbischemia induces remote preconditioning and remote postconditioningin humans by a KATP channelYdependent mechanism. Circulation.2007;116(12):1386Y1395.

13. Wei M, Xin P, Li S, Tao J, Li Y, Li J, Liu M, Li J, Zhu W, Redington AN.Repeated remote ischemic postconditioning protects against adverse leftventricular remodeling and improves survival in a rat model of myocardialinfarction. Circ Res. 2011;108(10):1220Y1225.

14. Andreka G, Vertesaljai M, Szantho G, Font G, Piroth Z, Fontos G, JuhaszED, Szekely L, Szelid Z, Turner MS. Remote ischaemic postconditioningprotects the heart during acute myocardial infarction in pigs. Heart.2007;93(6):749Y752.

15. Loukogeorgakis SP, Panagiotidou AT, Broadbrain MW, Donald A,Deanfield JE, MacAllister RJ. Remote ischemic preconditioning providesearly and late protection against endothelial ischemia-reperfusion injury inhumans: role of the autonomic nervous system. J Am Coll Cardiol.2005;46(3):450Y456.

16. Bøtker HE, Kharbanda R, Schmidt MR, Bøttcher M, Kaltoft AK, TerkelsenCJ, Munk K, Andersen NH, Hansen TM, Trautner S. Remote ischaemicconditioning before hospital admission, as a complement to angioplasty, andeffect on myocardial salvage in patients with acute myocardial infarction: arandomised trial. Lancet. 2010;375(9716):727Y734.

17. Stein DM, Lindell AL, Murdock KR, Kufera JA, Menaker J, BochicchioGV, Aarabi B, Scalea TM. Use of serum biomarkers to predict cerebralhypoxia after severe traumatic brain injury. J Neurotrauma. 2012;29(6):1140Y1149.

18. Berger RP, Pierce MC, Wisniewski SR, Adelson PD, Clark RS, RuppelRA, Kochanek PM. Neuron-specific enolase and S100B in cerebrospinalfluid after severe traumatic brain injury in infants and children. Pediatrics.2002;109(2):e31Ye.

19. De Kruijk J, Leffers P, Menheere P, Meerhoff S, Twijnstra A. S-100B andneuron-specific enolase in serum of mild traumatic brain injury patients. Acomparison with healthy controls. Acta Neurol Scand. 2001;103(3):175Y179.

20. McKeating EG, Andrews PJ, Mascia L: Relationship of neuron specificenolase and protein S-100 concentrations in systemic and jugular venousserum to injury severity and outcome after traumatic brain injury. ActaNeurochir Suppl. 1998;71:117Y119.

21. Rothermundt M, Peters M, Prehn JH, Arolt V. S100B in brain damage andneurodegeneration. Microsc Res Tech. 2003;60(6):614Y632.

22. Marangos P, Schmechel D. Neuron specific enolase, a clinically usefulmarker for neurons and neuroendocrine cells. Annu Rev Neurosci.1987;10(1):269Y295.

23. Maas AI, Stocchetti N, Bullock R. Moderate and severe traumatic braininjury in adults. Lancet Neurol. 2008;7(8):728Y741.

24. Stein SC, Chen XH, Sinson GP, Smith DH. Intravascular coagulation: amajor secondary insult in nonfatal traumatic brain injury. J Neurosurg.2002;97(6):1373Y1377.

25. Chesnut RM, Marshall SB, Piek J, Blunt BA, Klauber MR, Marshall LF:Early and late systemic hypotension as a frequent and fundamental sourceof cerebral ischemia following severe brain injury in the Traumatic ComaData Bank. Acta Neurochir Suppl (Wien). 1993;59:121Y125.

26. Kochanek PM, Clark RS, Ruppel RA, Adelson PD, Bell MJ, Whalen MJ,Robertson CL, Satchell MA, Seidberg NA, Marion DW. Biochemical,cellular, and molecular mechanisms in the evolution of secondary damageafter severe traumatic brain injury in infants and children: lessons learnedfrom the bedside. Pediatr Crit Care Med. 2000;1(1):4Y19.

27. Cooper DJ,Myburgh J, Heritier S, Finfer S, BellomoR, Billot L,Murray L,Vallance S; and the Australian and New Zealand Intensive Care SocietyClinical Trials Group. Albumin resuscitation for traumatic brain injury: isintracranial hypertension the cause of increased mortality? J Neurotrauma.2013;30(7):512Y518.

28. Cooper DJ, Rosenfeld JV, Murray L, Arabi YM, Davies AR, D’Urso P,Kossmann T, Ponsford J, Seppelt I, Reilly P. Decompressive craniectomy indiffuse traumatic brain injury. N Engl J Med. 2011;364(16):1493Y1502.

29. Farahvar A, Gerber LM, Chiu Y-L, Hartl R, Froelich M, Carney N, GhajarJ. Response to intracranial hypertension treatment as a predictor of death inpatients with severe traumatic brain injury: clinical article. J Neurosurg.2011;114(5):1471Y1478.

30. Chesnut RM, Temkin N, Carney N, Dikmen S, Rondina C, Videtta W,Petroni G, Lujan S, Pridgeon J, Barber J. A trial of intracranial-pressuremonitoring in traumatic brain injury. N Engl J Med. 2012;367(26):2471Y2481.

31. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: adelay of lethal cell injury in ischemic myocardium. Circulation. 1986;74(5):1124Y1136.

32. Ali ZA, Callaghan CJ, Lim E, Ali AA, Nouraei SR, Akthar AM, Boyle JR,Varty K, Kharbanda RK, Dutka DP. Remote ischemic preconditioningreducesmyocardial and renal injury after elective abdominal aortic aneurysmrepair: a randomized controlled trial. Circulation. 2007;116(11 Suppl):I98YI105.

33. Kharbanda RK, Li J, Konstantinov I, Cheung M, White P, Frndova H,Stokoe J, Cox P, Vogel M, Van Arsdell G. Remote ischaemicpreconditioning protects against cardiopulmonary bypass-induced tissueinjury: a preclinical study. Heart. 2006;92(10):1506Y1511.

34. Kuntscher MV, Kastell T, Engel H, Gebhard MM, Heitmann C, GermannG. Late remote ischemic preconditioning in rat muscle and adipocutaneousflap models. Ann Plast Surg. 2003;51(1):84Y90.

35. Candilio L, Malik A, Hausenloy DJ. Protection of organs other than theheart by remote ischemic conditioning. J Cardiovasc Med. 2013;14(3):193Y205.

36. Damous LL, Silva SM, Simoes RS, Morello RJ, Carbonel AP, Simoes MJ,Montero EF: Remote ischemic preconditioning on neovascularization andfollicle viability on ovary autotransplantation in rats. Transplant Proc.2008;40:861Y864.

37. Hausenloy DJ, Yellon DM. The therapeutic potential of ischemic condi-tioning: an update. Nat Rev Cardiol. 2011;8(11):619Y629.

38. Konstantinov IE, Arab S, Li J, Coles JG, Boscarino C, Mori A, CukermanE, Dawood F, Cheung MM, Shimizu M. The remote ischemicpreconditioning stimulus modifies gene expression in mouse myocardium.J Thorac Cardiovasc Surg. 2005;130(5):1326Y1332.

39. Kanoria S, Jalan R, Seifalian AM,Williams R, Davidson BR. Protocols andmechanisms for remote ischemic preconditioning: a novel method for re-ducing ischemia reperfusion injury. Transplantation. 2007;84(4):445Y458.

DISCUSSIONDr.DeborahM.Stein (Baltimore,Maryland): Dr. Joseph

and his colleagues present us a small but well-done study onthe use of remote ischemic conditioning in a subset of patientswith severe traumatic brain injury.

The primary outcome measure of this study was reduc-tion in serum biomarkers which are known to be associated




with severity of traumatic brain injury. Dr. Joseph has dem-onstrated to us in an elegant fashion that when remote ischemicconditioning was applied, levels of these biomarkers were lowerat 6 and 24 hours.

Remote ischemic conditioning is not a new concept butthese authors are, in fact, applying it to a novel disease state.Why and how remote ischemic conditioning actually worksnobody really knows.

But whatever the precise mechanism or combination ofmechanisms that provide benefit, there are numerous animalstudies and several human studies that suggest benefit in thisrelatively simple and non-invasive technique most commonlyapplied in cardiac surgery and after myocardial infarction.

However, despite the fact that remote ischemic condi-tioning has been studied since 1986, and a few human studieshave demonstrated benefit on surrogate proximal endpoints, itseems that the concluding paragraphs of all human studiesseem to end with some variation of "our findings merit a largerclinical trial to establish the effects of remote ischemic con-ditioning on clinical outcomes."

If we can’t demonstrate a real benefit in a disease ascommon as myocardial infarction over the past 30 years, whattype of study would we need to design in order to demonstratebenefit in a disease whose outcome is so largely determined bythe primary insult and has as much heterogeneity as traumaticbrain injury?

I do have a few comments and questions for the authors. Iam curious to know why you chose to look at consecutivepatients in each group rather than randomized patients. Howmight this confound your results with respect to the fact that theintervention and control patients actually were not treatedduring the same time period?

Second, we know many interventions to improve trau-matic brain injury have been tried and have failed. Despiteencouraging results in animal studies, a variety of neuro-protective agents have been unsuccessful in demonstrating anyclinical benefit in human trials.

Many of these failures may have to do with the appli-cation of therapies simply too late after injury to have clinicaleffects, given that we know that these secondary insults occurin seconds and minutes following the primary insult.

Much of the success of remote ischemic conditioningand other disease states such as cardiac surgery has been withthe application of the intervention prior to a clinical insult orvery shortly thereafter. I applaud you for applying this therapyso early in your hospital course.

Do you have any data on the time of injury versus theapplication of remote ischemic conditioning to see if, as wesuspect for many other neuroprotective agents, earlier is infact better?

Third, although this study was powered and designedto look at the effect of remote ischemic conditioning on veryproximal outcome-namely serum biomarkers-do you have anydata on the clinical proximal endpoints such as intracranialpressure?

And, lastly and most importantly, I am concerned thatin this small study, albeit it was not remotely powered to detectclinical outcomes, there was no suggestion of any kind ofclinical benefit.

What is your explanation for this and how would youdesign a study going forward? Just a larger study with morepatients? What would the number needed to treat look like? Orare we not looking, perhaps, at the right way of doing remoteischemic conditioning or the right types of patients?

Again, I would like to congratulate the authors on theirwork. I thank the association for the privilege of discussing thispaper. And, Dr. Joseph, I certainly look forward to seeing whatyou do with this in the future. Thank you.

Dr. Walter Biffl (Denver, Colorado): Bellal, that was aninteresting paper. How many of these patients had ICP moni-tors? And did you see an increase in the ICP with this, which Iassume was a painful intervention?

Dr. Aurelio Rodriguez (Wexford, Pennsylvania): Veryprovocative paper. A few years ago somebody presented a paper,"The effect of praying and outcome of patients in the ICU." Itwasdiscussed by Dr. Mattox. You can imagine what he said.

I have a question. You know the way the ACDs, thecompressional devices, work is release of fibrinolysis. Basi-cally, what you are doing is squeezing with the blood pres-sure cuff. Have you measured fibrinolysis in these patients, byany chance?

And my second question is, based on your study wouldyou recommend that the nurses in the ICU stop using intra-arterial line and start using the blood pressure cuffs?Would thatbe your recommendation? Thank you.

Dr. Raul Coimbra (San Diego, California): If you havenot done repeat imaging, do you think that if you had usedadvanced imaging modalities such as complex advanced MRIprotocols or magnetic encephalography you would see a cor-responding decrease in the secondary injury in the brain?

Dr. Bellal Joseph (Tucson, Arizona): Thank you, Dr.Stein, for your comments and questions. So why consecutivepatients instead of randomizing them? I will just say that in ourICU we have TBI protocols to manage patients.

Initially, we didn’t know if this would work, to be honest.Andwe did our first 20RICpatients and did the control later. But,asyoucan see,wedidn’t changeour practice ofTBI over the time.

As far as the timing of traumatic, or remote ischemicconditioning, you know there is a lot of studies that initiallylooked at preconditioning and it is referred to.But if you read a lotof the papers you know pre-, post- and peri-conditioning exists.

At our institution we did a septic mice model and lookedat those factors to see when it was best, you know. And wesaw that two hours post-septic insult was the best time to applythe remote ischemic conditioning. So the real question is, youknow, when or howmany times. I think those are still the milliondollar questions.

We looked at a few patients of how long from injury timeto the actual RIC occurred. And we didn’t have the data onall the patients but it was about 89 minutes from time of injuryto time of ischemia in this study.

As far as secondary outcomes such as ICP, you know,there were only four patients that had ICP monitors in thisstudy. We did not actually look at any other outcomes.

As far as the clinical outcomes in this study, again, this isa pilot study and you know when you look at outcomes in TBIeveryone looks at the GOS as the long-term. It’s kind of like the30-day mortality. What are the right outcomes?




I think as we move forward we’re going to have to look atboth short-term and long-term outcomes, because I think thatsecondary hit does have some sequelae three and six monthsdown the line. So I think that’s important to look at.

You know, again, to be able to alter two well-known orwell-studied biomarkers from a simple blood pressure cuff Ithink may have some promise.

Dr. Biffl, as far as the ICP and the pain, we did not lookspecifically. I can go back and look through whatever ICPpressures we have. But you know it’s interesting you brought upthe point about pain.

A lot of the patients were intubated, sedated and reallycouldn’t tell, but in a fewof the patients I put it on myself. If you

hold a cuff for five minutes at 30 millimeters above whatsystolic is it doesn’t feel good at five minutes, that’s for sure. Sothe pain is a question.

As far as the last comments, we did not measure fibri-nolysis in the group. I didn’t hear the second part of yoursecond question, something about the blood pressure cuff inthe ICU but, you know, I’m not sure.

And then Dr. Coimbra’s points, you know, actually I wishwe would have thought of what you just said prior to doing ourstudy. I think as we move forward that will be something thatwe will do. We will look at advanced MRI protocols or CTAperfusion scans of some sort to kind of better delineate theactual effect in the brain tissue. Thank you.




Secondary brain injury in trauma patients: The effects of remote ischemic conditioning

Documents

Transcript of Secondary brain injury in trauma patients: The effects of remote ischemic conditioning