International Spine Intervention Society - 2014 22nd Annual Scientific Meeting Research Abstracts

INTERNATIONAL SPINE INTERVENTION SOCIETY – 201422ND ANNUAL SCIENTIFIC MEETING RESEARCH ABSTRACTS

Adverse Event Rates in Interventional SpineProcedures: A Multi-Institutional StudyCarrie M Carr MD, [email protected],Christopher Plastaras MD2, Matthew J Pingree MD3,Matthew Smuck MD4, Timothy P Maus MD1,Jennifer R Geske MS5, Christine A El-Yahchouchi MD6,Zack McCormick, MD7, David J Kennedy MD4,(1) Mayo Clinic Department of Radiology, Rochester, MN,USA, (2) Department of Physical Medicine andRehabilitation, Penn Spine Center, University ofPennsylvania, Philadelphia, PA, USA, (3) Mayo ClinicDepartment of Physical Medicine and Rehabilitation,Rochester, MN, USA, (4) Department of OrthopedicSurgery, Stanford University, Stanford, CA, USA,(5) Mayo Clinic Department of Health Sciences Research,Division of Biomedical Statistics and Informatics, MayoClinic, Rochester, MN, USA, (6) Department ofAnesthesiology, American University of Beirut MedicalCenter, Beirut, Lebanon, (7) Rehabilitation Institute ofChicago, Northwestern Memorial Hospital, Chicago,IL, USABackground: Interventional procedures directed towardsources of pain in the axial and appendicular musculo-skeletal system are performed with increasing frequency.Despite the presence of evidence-based guidelines forsuch procedures, there are wide variations in practice.Case reports of serious complications such as spinal cordinfarction or infection from spine injections lack appro-priate context and create a misleading view of the risks ofinterventional pain procedures.Objectives: To evaluate the immediate adverse event ratefor interventional spine procedures performed at threeacademic interventional spine practices.Methods: Quality assurance databases at three academicinterventional pain management practices that utilizeevidence-based guidelines (1) were interrogated forimmediate complications from interventional pain pro-cedures. Review of the electronic medical record verifiedor refuted the occurrence of a complication. Same dayemergency department transfers or visits were also iden-tified by a records search.Results: Immediate adverse event data was available for26,062 consecutive procedures. A radiology practice per-formed 19,170 epidural steroid, facet, and sacroiliac injec-tions (2006–2013). A physiatry practice performed 6190epidural steroid, facet, sacroiliac, medial branch blockinjections and medial branch RFA (2004–2009). A second

physiatry practice performed 702 epidural steroid, facet,sacroiliac, medial branch block injections and medialbranch RFA (2009–2010). There were no major compli-cations (permanent neurologic deficit, clinically signifi-cant bleeding [e.g. epidural hematoma] or infection), withany procedure. Overall complication rate was 2.0% (528/26,062). Vasovagal reactions were the most frequent event(1.1%). Twenty patients (>0.1%) were transferred toemergency departments (allergic reactions, chest pain,symptomatic hypertension, vasovagal reaction).Discussion: This study demonstrates that interventionalpain procedures are safely performed with extremely lowimmediate complication rates when evidence-basedguidelines are observed.

References:1. Bogduk N ed. Practice guidelines for spinal diagnosticand treatment procedures, 2nd edition. San Francisco:International Spine Intervention Society; 2013.

Durability of Clinical Effectiveness withDexamethasone in Single Lumbar TransforaminalEpidural Steroid InjectionsFelix E Diehn, MD1 ([email protected]),Jennifer R Geske, MS2, Vance T Lehman, MD1,Naveen S Murthy, MD1, Carrie M Carr, MD1,Timothy J Kaufmann, MD1, Timothy P Maus, MD1,(1) Department of Radiology, Mayo Clinic, Rochester,Minnesota, USA, (2) Department of Health SciencesResearch, Division of Biomedical Statistics and Informatics,Mayo Clinic, Rochester, Minnesota, USABackground Data: The non-particulate steroid dexa-methasone is considered safer than particulate steroidsfor lumbar transforaminal epidural steroid injections(TFESI). Dexamethasone has demonstrated clinicaleffectiveness (1,2), but durability of effectiveness beyond6 months is unproven.Objectives: To retrospectively assess the long-term dura-bility of clinical effectiveness of single lumbar TFESIusing dexamethasone for radicular pain with/withoutradiculopathy.Methods: Retrospective observational study of 750 con-secutive single unilateral lumbar TFESI with dexametha-sone (10 mg). Assessments were: pain Numerical RatingScale (0–10) and Roland-Morris disability questionnaireprior to TFESI and at 3, 6, 9 and 12-months follow-up.

bs_bs_banner

PAIN MEDICINEVolume 15 • Number 8 • 2014

Wiley Periodicals, Inc 1526-2375/0/$15.00/15 1436–1446 doi:10.1111/pme.12540

mailto:[email protected]


Successful pain relief was defined as ≥50% reduction inNRS or pain 0/10; functional success as ≥40% reductionin Roland-Morris score.Results: The mean age of the entire cohort was 67 years;44.9% were male and 46% reported pain duration of >1year. The percentages of patients experiencing successfulpain relief/functional success, respectively, were: at3-months (750 patients) – 35.6%/43.6%; at 6-months(416 patients) – 40.6%/46.2%; at 9-months (246 patients)– 46.8%/52.4%; and at 12-months (170 patients) –41.8%/49.4%.Discussion/Conclusion: The clinical effectiveness oflumbar TFESI performed with dexamethasone wasdurable. Approximately 50% of patients had functionalsuccess at 12 months, despite demographic characteristicsthat would bias against clinical effectiveness (older meanage and high prevalence of chronic pain that suggest asignificant proportion of fixed compressive lesions).

References:1. El-Yahchouchi C. et al. The noninferiority of thenonparticulate steroid dexamethasone vs the particulatesteroids betamethasone and triamcinolone in lumbartransforaminal epidural steroid injections. Pain Med.2013; 14(11):1650–7.2. Kennedy DJ et al. Comparative Effectiveness ofLumbar Transforaminal Epidural Steroid Injections withParticulate Versus Nonparticulate Corticosteroids forLumbar Radicular Pain due to Intervertebral Disc Her-niation: A Prospective, Randomized, Double-Blind Trial.Pain Med. 2014; Epub.

Complication rates of transforaminal andinterlaminar epidural steroid injections:a multi-institutional studyChristine A El-Yahchouchi, MD, [email protected],Christopher T Plastaras, MD2, Timothy P Maus, MD3,Carrie M Carr, MD3, Zack McCormick, MD4,Jennifer R Geske, MS5, Matthew Smuck, MD6,Matthew J Pingree, MD7, David J Kennedy, MD1,(1) Department of Anesthesiology, American University ofBeirut Medical Center, Beirut, Lebanon, (2) Department ofPhysical Medicine and Rehabilitation, Penn Spine Center,University of Pennsylvania, Philadelphia, PA, USA,(3) Department of Radiology, Mayo Clinic, Rochester, MN,USA, (4)Rehabilitation Institute of Chicago, NorthwesternMemorial Hospital, Chicago, IL, USA, (5) Department ofHealth Sciences Research, Division of Biomedical Statisticsand Informatics, Mayo Clinic, Rochester, MN, USA,(6) Department of Orthopedic Surgery, Stanford University,Stanford, CA, USA, (7) Department of Physical Medicineand Rehabilitation, Mayo Clinic, Rochester, MN, USABackground: Transforaminal epidural steroid injections(TFESI) have demonstrated efficacy and effectiveness in

treatment of radicular pain. Despite little evidence ofefficacy or effectiveness, interlaminar epidural steroidinjections (IL) are advocated as a primary therapy forradicular pain due to purported greater safety.Objective: To assess the immediate and delayed compli-cation rates of TFESI and IL injections at three academicmedical centers utilizing International Spine Interven-tion Society practice guidelines.Methods: Quality assurance databases from a Radiologyand two Physical Medicine and Rehabilitation (PMR)practices were interrogated. Medical records werereviewed to verify immediate and delayed complications.Results: There were no neurologic (new fixed deficit) orhemorrhagic (intraspinal/foraminal hematoma) compli-cations in 16,638 consecutive procedures performed in allspine segments (14,956 TFESI and 1,682 IL). Vasovagalreactions occurred in 1.2% of all patients (1.3% TFESI,0.5% IL procedures). Dural punctures occurred in 0.06%of procedures (0.04% TFESI, 0.2% IL). Emergencydepartment transfers occurred in 6 TFESI (0.04%, car-diovascular symptoms or allergic reactions). Allergicreactions occurred in 8 TFESI patients (0.05%).Delayed follow up on PM&R patients (92.5% and78.5%, next business day) and Radiology patients(63.1%, 2 weeks) identified no neurologic, hemorrhagic,or infectious complications. Central steroid response(sleeplessness, flushing, non-positional headache) wasseen in 2.6% of both TFESI and IL patients. Increasedpain was subjectively reported in 2.1% of TFESI patientsand 1.8% of IL patients, but medical record review iden-tified no procedure related complications.Conclusions: Both transforaminal and interlaminar epi-dural steroid injections are safely performed with lowimmediate and delayed complication rates wheninformed by evidence-based procedural guidelines. Bydemonstrating comparable safety, this study suggests thatthe choice between IL and TFESIs can be based ondocumented efficacy and effectiveness and not driven bysafety concerns.

ISIS 22nd Annual Scientific Meeting Abstracts 1437


Synovial cysts confounding access to the dorsalS1 neural foramen in transforaminal epiduralsteroid injectionsChristine A El-Yahchouchi, MD, [email protected],Frank H Willard, PhD2, Jennifer R Geske, MS3,Timothy J Kaufmann, MD4, Felix E Diehn, MD4,John T Wald, MD4, Timothy P Maus, MD4,(1) Department of Anesthesiology, American University ofBeirut Medical Center, Beirut, Lebanon, (2) Departmentof Anatomy, University of New England, Biddeford, Maine,USA, (3) Department of Health Sciences Research, Divisionof Biomedical Statistics and Informatics, Mayo Clinic,(4) Department of Radiology, Mayo Clinic, Rochester,Minnesota, USABackground: Transforaminal epidural steroid injections(TFESI) have demonstrated efficacy and effectiveness intreatment of lumbar radicular pain. The S1 foramenaccounts for approximately 30% of TFESIs; it is impor-tant to identify structures impeding foraminal access.Meningeal (Tarlov) cysts are known to limit access tothe S1 foramen; synovial cysts arising from the L5-S1facet joint have not been previously reported at thissite.Objectives: To characterize the imaging findings, preva-lence, and anatomic origin of synovial cysts within the S1dorsal foramen.Methods: A case series (n = 14) established imaging char-acteristics of S1 synovial cysts (S1SC). Imaging studies of400 patients undergoing TFESI were reviewed for alllesions compromising S1 foraminal access. Cadaveric dis-sections were examined to define the relationship of theinferior recess of the L5-S1 facet to the S1 dorsalforamen.Results: Elderly patients (mean age = 76; 60% male)exhibited S1SC. Eighteen cysts were identified(4 bilateral), most commonly 1–2 cm in diameter. S1SCwere hyperintense on sagittal T2 weighted MRI images,fluid density on CT, dorsal to the S1 nerve. 60% ofS1SC exhibited MRI signs of hemorrhage (thick wall,internal structure); meningeal cysts were pure fluidintensity. Lesions impeded access to S1 dorsal foraminain 5% of 400 imaging studies (16 meningeal cysts, 3synovial cysts, 1 conjoint S1-S2 nerve root). Themultifidus muscle was interposed between the L5-S1facet inferior recess and the S1 dorsal foramen ondissection specimens. There was severe atrophy ofthe ipsilateral multifidus on imaging in 17/18 synovialcysts.Conclusion: The S1 neural foramina should be inspectedon sagittal MRI for potential confounding lesions priorto performing S1 TFESI. Meningeal cysts are morecommon lesions than S1SC. S1SC are seen in elderlypatients with severe multifidus atrophy.

Contralateral Oblique View is Superior to LateralView for Inter-laminar Cervical andCervico-thoracic Epidural AccessJatinder S Gill, MD1, Moris Aner, MD1, John Keel, MD2,Jyotsna Nagda, MD1, Thomas Simopoulos, MD1,(1) Department of Anesthesiology, Beth Israel DeaconessMedical Center, Boston, MA, (2) Physical Medicine andRehabilitation, New England Baptist Hospital, Boston, MAObjective: The purpose of this study was to compare thereliability of the lateral fluoroscopic view and several con-tralateral oblique (CLO) views at different angles in visu-alizing the needle tip and in accurately predicting thelocation of the posterior epidural space in the cervical andcervico-thoracic spine.Design: After the epidural space was accessed but beforeconfirmation with contrast fluoroscopy, we prospectivelyobtained fluoroscopic images at eight different angles.Subsequent contrast injection confirmed epidural spread.Needle visualization and location of needle tip relative tobony landmarks was analyzed.Results: Twenty-four of twenty-seven enrolled patientsunderwent cervical epidural steroid injection (CESI) withthe full complement of investigational imaging. Theneedle tip was clearly visualized in all CLO projections inall 24 subjects. At 50 degrees CLO and MRI measuredCLO, the epidural space was accessed at or just beyondthe ventral laminar margin at the ventral interlaminarline (VILL) or within the proximal half of the predefinedCLO area in all patients. Needle tip was poorly visualizedin the lateral view and the location of the needle tip wasless well defined and independent of the needle locationin AP view.Conclusions: This study provides evidence that contra-lateral oblique view at 50 degrees and at MRI measuredangle is superior to the lateral view during cervical andcervico-thoracic epidural access for the purpose ofneedle tip visualization and in providing a consistentlandmark for the posterior boundary of the epiduralspace. This article introduces the concept of zones todescribe a needle in the epidural space in relation to thebony landmarks.

Does Fibronectin Aggrecan Complex (FAC) PredictResponse to Epidural Steroid Injection?Lisa Huynh, MD, [email protected], Matthew Smuck,MD, Ma Agnes Ith, David J Kennedy, MD, Department ofOrthopaedics, Division of Physical Medicine andRehabilitation, Stanford University, Stanford, CABackground Data: Epidural steroid injection (ESI) is aneffective treatment for pain and disability in patientswith lumbar disc herniation, yet not all patients respondto this treatment. Distinguishing potential respondersfrom non-responders remains a major challenge.

1438 ISIS 22nd Annual Scientific Meeting Abstracts



Clinical and radiographic variables are only weakpredictors.Objective: To investigate the fibronectin-aggrecan G3domain complex (FAC) as a biomarker to predictresponse to ESIs in patients with lumbar disc herniations.Methods: This prospective, consecutive cohort studyenrolled 63 patients electively scheduled for ESIwith lumbar radicular pain and corresponding MRI-confirmed disc herniation. Under fluoroscopic guidance,physiologic saline epidural lavage was performed and col-lected immediately prior to injecting steroid. OswestryDisability Index (ODI) and Numeric Pain Rating Scale(NPRS) were assessed at baseline and 6 weeks post-ESI.ODI improvement of 15 points and NPRS reduction of30% were thresholds for “success” in disability and pain,respectively. FAC was positive (+) if optical density was>0.06 at 450 nm. One consenting participant’s injectionwas cancelled due to hypotension, and two had insuffi-cient lavage samples.Results: FAC status is significantly associated with dis-ability outcomes, with success in 8/8 FAC (+) and 13/52FAC (−) participants (p < 0.0001), yielding 38.1% sensi-tivity (95%CI: 18.2%-61.6%) and 100% specificity(95%CI: 90.9%-100%). If success is defined by bothpain and disability thresholds, sensitivity falls to 26.7%(95%CI: 8%-55.1%) with a slight decrease in specificityto 91.1% (95%CI: 78.8%-97.5%). Mean improvementsdisability were greater in the FAC (+) group than the FAC(−) group (ODI improvement 21 vs. 8 points; p = .007)but not significantly different for pain (NPRS reductionof 41% vs. 15%; p = .19).Conclusion: FAC status predicts disability, but not painoutcomes 6-weeks after ESI in patients with lumbar discherniations.

A Compartive Study of Robotic Assisted CTGuided Interventional Pain Management WithConventional Fluoroscopy ApproachMadhan Kumar, [email protected],Dr. Seena2, Victor Raj1, Anjali Kulkarni3,(1) Department of Pain Management, Global Institute ofPain Management, Global Health City, Chennai TamilNadu, India, (2) Department of Radiology and ImagingSciences, Global Health City, Chennai, India, (3) PerfintHealth Care, Chennai, IndiaAim: To prospectively evaluate the accuracy,speed of needle placement and patient comfort inperforming interventional pain medicine proceduresfor acute and chronic pain during robotic assistedCT-guided procedures and C-arm fluoroscopy guidedprocedures.Materials and Methods: After obtaining InstitutionalEthical Committee approval and written informed

concent from participating patients, 200 patients wereenrolled in the study between January 2012 and January2014. 120 male and 80 female patients with a mean age of49.8 years (range 18–80 years) were enrolled. Patientsunderwent various procedures such as Celiac plexusblock, Lumbar sympathetic block, Dorsal root ganglionblock, Lumbar epidural injection, Thoracic sympa-thetic block, Facet joint block, Stellate ganglion block,Vertebroplasty, Ozone neucleolysis, Superior hypogastricplexus block, Sacro-iliac joint injection, Ganglion Imparblock, Sacral foraminal injection and Caudal injection.Patients were randomly assigned to two groups. Group 1(G1) received CT Guided Robotic Assisted needle place-ment and Group 2 (G2) received C-arm fluoroscopyguided needle placements. The injected medicationswere standardized in both groups. Accuracy of needleplacement, speed of needle insertion and patient comfortwere documented.Results: Statistical analysis of the collected data showedno significant difference in the accuracy of needle place-ment (G1 = 99.8%, G2 = 98%). Significant difference inthe speed of needle placement was observed (G1 = <420seconds, G2 = >600 seconds). Patients’ positionalcomfort also showed significant difference (G1 = 8–10/10 Avg.: 9/10; G2 = 1–8/10 Avg.: 5/10). The procedureswere performed by the same physician trained in inter-ventional pain procedures.Conclusion: Robotic assisted CT-Guided needle place-ments for interventional pain procedures are consider-ably superior to C-arm fluoroscopy guided needleplacements in terms of speed for needle placement andpatient comfort.

Comparison of Facet Joint Activity on 99mTc-MDPSPECT/CT with Facet Joint Signal Change on MRIwith Fat SuppressionVance T Lehman, MD, [email protected],Robert C Murphy, MD1, Louis A Schenck, MS2,Rickey E Carter, PhD2, Geoff B Johnson, MD1,Amy L Kotsenas, MD1 Jonathon M Morris, MD1,Mark A Nathan, MD1, John T Wald, MD1,Timothy P Maus, MD1, (1) Mayo Clinic, Department ofRadiology, Rochester MN, (2) Mayo Clinic Department ofHealth Sciences Research, Division of Biomedical Statisticsand InformaticsSummary of Background Data: Prior studies purportthat both bone scan activity and signal change on MRIwith fat suppression predict facet joint pain and are usedin some practices to help guide facet joint treatment, butthe degree of correlation between these two imagingfindings is unknown.




Objectives: Determine the concordance of facet jointimaging abnormalities on 99mTc-MDP SPECT/CT andfat-suppressed MRI.Methods: All patients who had both a 99mTc-MDPSPECT/CT and an MRI with at least one fat-suppressedT2 or T1 with gadolinium sequence of the lumbar spinewith an indication of pain evaluation within a 180 dayinterval at our institution between 1/1/2008 and 2/19/2013 were identified. Facet joint activity on 99mTc-MDPSPECT/CT and peri-facet signal change on MRI wererated as normal or increased by 3 nuclear medicine phy-sicians and 3 neuroradiologists, respectively. Inter-observer agreement and agreement between the twoexamination types were determined with the kappa andPABAK statistics.Results: 60 patients were identified, 28 (47%) males,with a mean age 49 years (range 12–93 years). Overall,the Cohen value indicated there was no agreementbetween 99mTc-MDP SPECT/CT and MRI (κ = −0.026),although the PABAK values were fair to high at eachspinal level, suggesting that relatively low disease preva-lence lowered the Cohen κ values.Conclusion: Overall, facet joint signal change on fat-suppressed MRI does not always correlate to increased99mTc-MDP activity. MRI and 99mTc-MDP SPECT/CTfor facet joint evaluation should not be considered inter-changeable examinations in clinical practice for identifi-cation of potentially painful facet joints. Neither imagingstudy has been validated as a predictor of a painful joint.

Intradiscal methylene blue treatment fordiscogenic low back painDavid S Levi, MD, [email protected], Scott Horn, DO,Edward Walko, DO, APM Spine and Sports Physicians,Virginia Beach, VABackground: Prior treatments for discogenic low backpain directed at intervertebral discs have been disappoint-ing. One group of researchers, however, reported veryhigh levels of improvement on pain and function after asingle intradiscal injection of methylene blue1.Objective: To assess changes in pain and function inpatients with discogenic low back pain, diagnosed bydiscography, after an intradiscal injection of methyleneblue.Study Design: Prospective trial.Methods: Patients diagnosed with discogenic pain bydiscography underwent a single treatment of intradiscalinjection of methylene blue, determined by prior provo-cation discography.Main Outcome Measures: Pain and function measure-ments were completed at baseline, 1, 2, and 6 monthspost-treatment. Patients were considered a categoricalsuccess based upon a 30% improvement in pain on Visual

Analogue Score (VAS) and function on Oswestry Disabil-ity Index (ODI). Patients were considered a categoricalfailure if they achieved less than 30% improvement inpain and function or pursued other invasive treatmentoptions during the trial period.Results: Sixteen patients underwent the intradiscalmethylene blue injection. Eleven patients underwent asingle level injection. Four patients underwent a two levelinjection and one patient was injected at three levels. Theoverall categorical success rates at 1, 2 and 6 monthspost-injection, were 19%, 21%, and 25% respectively.Conclusion: This small trial failed to demonstrate anoverall clinical success of intradiscal methylene blueinjection for those patients diagnosed with discogenicpain by discography. These findings are in stark contrastto the outstanding results from a recently published, levelone evidence, trial1. Further research is needed to evalu-ate the efficacy of this treatment.1. Peng et al. A randomized placebo-controlled trial ofintradiscal methylene blue injection for the treatment ofchronic discogenic low back pain. Pain 2010;149:124–29.

Radiation dose incurred in the exclusion ofvascular filling in transforaminal epidural steroidinjections: Fluoroscopy, Digital SubtractionAngiography, and CT/FluoroscopyTimothy P Maus, MD, [email protected],Beth Schueler, PhD2, Dayne Magnuson, RT 2,Shuai Leng, PhD2, Felix E Diehn, MD1,(1) Department of Radiology, Mayo Clinic, Rochester,Minnesota, USA, (2) Department of Radiology, Division ofMedical Physics, Mayo Clinic, Rochester, Minnesota, USABackground: Digital subtraction angiography (DSA) hasgreater sensitivity than conventional fluoroscopy (CF) inexclusion of vascular filling during transformational epi-dural steroid injections (TFESIs). Multi-slice, pulsed CTfluoroscopy (CT/F) is also used for this purpose. Theadditional radiation dose incurred using DSA or CT/Frelative to CF is unknown, compromising any risk versusbenefit assessment for the image guidance techniques.Objective: To measure the radiation dose incurred in theevaluation of vascular filling during TFESI using CF,DSA, and CT/F.Methods: Three portable C-arms and a fixed multi-purpose C-arm were evaluated. The radiation dose ratewas measured using an anthropomorphic phantomduring CF and DSA in anterior-posterior positions forcervical and lumbar TFESIs. Effective doses were calcu-lated for 5-second exposures. The effective dosesincurred in the cervical and lumbar spine during 2 CT/Fexposures were calculated based on the reportedvolume CT dose index (CTDIvol) and dose lengthproduct.




Results: DSA imaging increased the effective doseincurred over CF with portable C-arms (medium doserate) by 2.5–4.3 fold for cervical TFESI and 2.3–4.2 foldfor lumbar TFESI. The incremental dose incurred withDSA ranged from 4.0–7.7 μSv in the cervical region andfrom 22–38 μSv in the lumbar spine. CT/F increased theincurred dose 19 fold in the cervical region and 8.0 foldin the lumbar region (incremental doses 49 μSv and140 μSv respectively) relative to CF.Conclusion: The use of DSA imaging to exclude vascularuptake during TFESI increases radiation dose over CF.CT/F incurs additional dose beyond most DSA. Mini-mizing radiation dose by limiting DSA and CT/F use tospine segments or clinical situations involving higher riskmay be desirable. However, the incremental radiationdoses incurred by DSA or CT/F are of such low magni-tude that health risks cannot currently be estimated.

Factors Associated with Pain Reduction after aFirst-time Transforaminal Epidural Steroid Injectionfor Lumbrosacral Radicular PainZachary McCormick, MD, [email protected],Daniel Cushman MD1, Cynthia Garvan PhD2,Ellen Casey MD,1 David J. Kennedy MD3,Christopher Plastaras MD4, (1) Department of PM&R,Northwestern McGaw Medical Center/The RehabilitationInstitute of Chicago, Chicago, IL, (2) College of Nursing,University of Florida, Gainesville, FL, (3) Department ofOrthopaedic Surgery, Stanford University, Palo Alto, CA,(4) Department of PM&R, Hospital of the University ofPennsylvania, Philadelphia, PABackground Data: Clinical factors associated with a suc-cessful outcome after transforaminal epidural steroidinjection (TFESI) are not well defined.Objective: To identify clinical factors associated withpain improvement after a TFESI for the treatment oflumbosacral radicular pain.Methods: This was a retrospective cohort study of 188adults who underwent a fluoroscopically-guided TFESIfor lumbosacral radicular pain. Pain reduction from pre-injection to follow-up was measured by visual analoguescale (VAS). Patients were grouped by: worsened painor no relief (≤0%), pain relief but less than 50% (>0%to <50%), or significant pain relief (≥50%). Differencesin clinical factors were compared among these groups.Results: The mean (SD) duration of pain prior to injec-tion and time to follow-up after injection were 45.8 (81.0)weeks and 20.0 (14.2) days, respectively. Significantlymore patients who experienced ≥50% pain relief atfollow-up reported higher pre-injection pain on the VASand McGill Pain Inventory (MPI) (p < 0.0001 and p =0.0358, respectively) as well as no worsening of their painwith walking (p = 0.0161). A positive femoral stretch was

associated with ≥50% pain relief at follow-up (p =0.0477). No significant differences were found betweenpain reduction and clinical factors including a radiologicdiagnosis of disc herniation versus stenosis or otherneural tension signs on physical examination includingthe straight leg raise and the slump sit tests.Conclusions: Greater baseline pain on the VAS andMPI, a history of a lack of worsening pain with walking,and a physical exam finding of a positive femoral stretchtest, predict more frequent pain reduction after TFESIfor lumbosacral radicular pain at short-term follow up.Greater baseline pain on the MPI and a history of a lackof worsening pain with walking predict a magnitude of>50% pain reduction.

Fluoroscopically Guided LumbosacralTransforaminal Epidural Steroid Injections:Adverse Events and Predictive FactorsZachary McCormick MD, [email protected],Christopher Plastaras MD,2 Cynthia Garvan PhD3,Donald Macron MD4, Anand Joshi MD5,Gary Chimes MD6, Wesley Smeal MD7,Joshua Rittenberg MD8, David J. Kennedy MD9,(1) Department of PM&R, Northwestern McGaw MedicalCenter/The Rehabilitation Institute of Chicago, Chicago, IL,United States, (2) Department of PM&R, University ofPennsylvania, Philadelphia, PA, United States, (3) Collegeof Nursing, University of Florida, Gainesville, FL, UnitedStates, (4) Department of Neurological Surgery, StonyBrook Neurosciences Institute, Stony Brook, NY, UnitedStates, (5) Department of Orthopaedics, Duke UniversityMedical Center, Durham, NC, United States, (6) LakeWashington Sports and Spine. Bellevue, WA, United States,(7) Alegent Health Physical Medicine and RehabilitationClinic, Omaha, NE, United States, (8) Kaiser PermanenteDepartment of Physical Medicine and Rehabilitation,Oakland, CA, (9) Stanford University Department ofOrthopeadics, Palo Alto, CABackground Data: While the types and incidence ofadverse events associated with TFESI has been described,prospective investigation is reported in only one study.No study has used a standardized questionnaire to solicitadverse events (AEs) from patients in order to capture anaccurate range and incidence of complications.Objective: To prospectively and systematically identifythe types and incidence of AEs associated with TFESI.Additionally, to evaluate demographic and clinical factorsthat may predict a higher risk of an AE.Methods: This was a retrospective cohort study from amulti-physician academic PM&R clinic. Patients, ages19–89, who underwent a fluoroscopically-guided TFESIfor lumbosacral radicular pain between 2004–2008 wereincluded. AE data were prospectively collected using a




survey both immediately and at 24–72 hours followingTFESI. The relationship of adverse events with gender,age, trainee presence, steroid type, pre-procedure visualanalogue scale (VAS), systolic blood pressure, fluoros-copy time and corticosteroid injectate volume wereanalyzed with chi-square, Fisher’s exact or Wilcoxonrank sum two-sided tests. Logistic regression analysis wasalso performed.Results: In 1,295 consecutive patients undergoing 2,025TFESIs procedures, Immediate AEs and delayed AEsoccurred after 182 (9.2%) and 305 (20.0%) injections,respectively. The most common immediate AEs were:vasovagal reaction (4.2%) and interrupted procedurefrom intravascular flow (1.7%). Common delayed AEsinclude: pain exacerbations (5.0%), injection site soreness(3.9%), facial flushing/sweating (1.8%), headache (3.9%),and insomnia (1.6%). Significant associations were iden-tified between adverse events and gender, age, pre-procedure VAS, steroid type, volume of corticosteroid,and fluoroscopy time.Conclusions: Lumbosacral TFESI is a safe procedure.Permanent adverse events were not found in our sample.The most AEs associated with TFESI include vasovagalepisodes, interrupted procedure from intravascular flow,pain exacerbation, injection site soreness, headache, andinsomnia.

Ultrasound Guided Lumbar Medial BranchBlock-The Second LookJoseph Park, MD, FRCP, FIPP, [email protected],Ali Zahran, MD2, (1) Michael De Groote School of Medi-cine, McMaster University, Hamilton, Ontario, Canada,(2) Hamilton Health Sciences in Hamilton, Ontario, CanadaBackground and Objectives: Over the last decade, it hasbeen suggested that diagnostic lumbar medial branchblock (LMBB) can be done with ultrasound guidedtechnique as effectively as conventional fluoroscopictechnique with success rate 89–95%1,2 and 62 % onobese patients (BMI = 30)3. It has many advantages overconventional technique. This clinical prospective obser-vational study reexamines the feasibility and validity ofultrasound guided diagnostic LMBB.Methods: 22 patients with chronic low back pain under-went ultrasound guided LMBB twice. The low frequencycurved ultrasound probe was used (2–6 MHz). Theneedle tip was verified with fluoroscopy. The time taken,satisfaction, pain score, complications, accuracy, andsuccess rate were measured.Results: Average BMI of patients were 27.9. Average sat-isfaction and pain score were 7.7 and 5.6 respectively. 29% of patients complained of significant pain. Averagetime taken was 17.4 min. There were no significantcomplications. Success rate was 42% with distance

from the target 7.2 mm. (1st trial = 38 %, 2nd trial:46%).Conclusions: Despite many advantages of using ultra-sound guided LMBB, this study showed that it was notfeasible due to high intensity of pain. It failed the validityand reliability test as success rate was 42 % which lowerthan previous studies1,2,3. Success rate of blocking atL5-S1 was even lower. It took much longer than conven-tional fluoroscopic LMBB. Further reassessment is war-ranted in the future study.

References1. Manfred Greher et al: Ultrasound-guided lumbarFacet Nerve Block. Anesthesiology May 2004;100:5:1242–1248.2. Jae-Kwang Shim et al: Ultrasound-guided LumbarMedial-Branch Block: A Clinical Study With Fluoros-copy Control. Regional Anesthesia and Pain Medicine.2006 Oct.:31:5:451–454.3. Rauch S et al: Ultrasound-guided lumbar medialbranch block in obese patients: a fluoroscopicallyconfirmed clinical feasibility study. Reg Anesth PainMedicine 2009;34:340–342.

Autogenous Point of Care Bone MarrowConcentrate (BMC) for the Treatment of LumbarDegenerative Disc Disease (DDD): IRB ControlledProspective StudyKenneth A. Pettine, MD, [email protected],The Orthopedic Stem Cell Institute, Johnstown, COBackground: Current treatment options for moderateto severe symptomatic lumbar DDD include fusion orconservative therapy. Autogenous BMC may provide anon-surgical option for this condition.Objectives: Evaluate the safety and feasibility of autog-enous BMC for non-surgical treatment of lumbar DDD.Methods: Twenty-six patients were injected with autog-enous BMC at one or two levels. Symptomatic discs werediagnosed by MRI scanning or discography. Pfirrmangrades and patient demographics will be discussed.Every patient had a pre procedure ODI, VAS, and physi-cal exam. The patients were followed prospectively at 6weeks, 3 months, 6 months and 1-year, with repeat MRIat one year.Sixty ml of bone marrow aspirate was obtained throughpercutaneous aspiration of the posterior iliac wing,and concentrated with the ART21 system. Typically,2–3 ml of BMC was injected into each symptomaticlumbar nucleus with a standard two-needle discographytechnique.Results: No complications were associated with the iliacwing aspirate or disc injection. Pre and post injectionODI and VAS scores will be discussed. Four patients hadless than 25% improvement in ODI and VAS 6 months




post injection. Two underwent re-injection and are sig-nificantly improved clinically at 1 year; 2 patients electedto proceed with lumbar fusion.Patients receiving >2,000 CFU-F/cc experienced areduction of 70.3% for ODI and 68.9% for VAS at 12months post-treatment. Twenty patients received a 12month follow-up MRI, 8 patients improved at least oneor two pfirrman grade and no patients regressed. Seven ofthe 8 patients who improved radiographically received>2,000 CFU-F/cc.Conclusion: Preliminary results indicate autogenousBMC may provide a clinical benefit as a non-surgicaltreatment of DDD in the lumbar spine.

Interventional gene vector delivery to theperipheral nervous system: Preclinical testing ofCT-guided procedures in pigs and imaging basedmodeling in humansJosef Pleticha, MD, [email protected],Andreas S Beutler, MD1, W Michael Hooten, MD1,Timothy P Maus, MD2, (1) Department of Anesthesiology,Mayo Clinic, Rochester, MN, USA, (2) Department ofRadiology, Mayo Clinic, Rochester, MN, USABackground: Gene therapy of the peripheral nervoussystem (PNS) is approaching clinical translation as anovel treatment strategy for chronic pain. Gene transfercan modulate both opioid and non-opioid nociceptivepathways, expanding the therapeutic targets of currentlyavailable analgesic agents. Although well established inrodents, attaining robust and anatomically selective genetransfer in a large animal model is a remaining challenge.Objective: To deliver gene vector to the PNS ofa translational large animal model (domestic swine) withdocumentation of transduction in the target neuraltissue.Methods: CT-fluoroscopy (CTF) guided vector deliveryvia intrathecal (IT) and intraganglionic (IG) routes. TheIT space was accessed at its lateral recess (the swine spinalcord extends to S2). For IG delivery, a myelogram visu-alized the dorsal root ganglia (DRG), which were tar-geted by a custom-designed injection apparatus (steppedneedle) allowing convection-enhanced delivery of thevector directly into the DRG parenchyma1.Results: The feasibility of both routes for gene transferto the PNS was validated by administering adeno-associated virus gene vectors, which resulted in efficienttransduction of the targeted sites. Imaging based model-ing of the IG route suggested that it may be translatableto humans for clinical use.Conclusion: CTF guided delivery of gene vector to thePNS of a large animal model resulted in successful trans-duction of target tissue. There are no technical barriers totargeting human neural tissue. This holds promise for

treatment of diffuse (IT delivery) or segment specific (IGdelivery) chronic pain.1. Pleticha J, Maus TP, Christner JA, et al. Minimallyinvasive convection enhanced delivery of biologics intodorsal root ganglia: Validation in the pig model and pro-spective modeling in humans: Technical note. J Neuro-surg 2014.

When NOT to refer patients for diagnostic facetjoint blocks: Derivation of a clinical decision rulein the diagnosis of cervical facet joint painGeoff M Schneider, PT, PhD, [email protected],Gwendolen Jull, PT, PhD2, Kenneth Thomas, MD, MHSc3,Ashley D Smith, PT 4, Carolyn Emery, PT, PhD5,Peter Faris, PhD6, Chad Cook, PT, MBA, PhD7,Bevan Frizzell, MD8, Paul Salo, MD3, (1) GraduateSciences Education, Faculty of Medicine, University ofCalgary, Calgary, Alberta, Canada, (2) School of Healthand Rehabilitation Sciences, The University of Queensland,Queensland, Australia, (3) Department of Surgery, Facultyof Medicine, University of Calgary, Calgary, Alberta,Canada, (4) School of Health and Rehabilitation Sciences,The University of Queensland, Queensland Australia,(5) Sport Injury Prevention Research Centre, University ofCalgary, Calgary, Alberta, Canada, (6) AHS Research -Research Excellence Support Team, Foothills Medical Centre,Calgary, Alberta, Canada,(7) Division of Physical Therapy, Walsh University, NorthCanton, Ohio, (8) Department of Radiology, Faculty ofMedicine, University of Calgary, Calgary, Alberta, CanadaSummary of Background Data: In the United States, theuse of facet joint procedures has increased dramaticallyover the past 15 years. Optimal use of healthcareresources may begin with identifying patients that aresuitable for diagnostic blocks.Objective: To derive a clinical decision rule (CDR) toidentify patients best suited for cervical diagnostic facetjoint blocks.Methods: Consecutive patients (n = 125) with neck painparticipated in this prospective study. All patients under-went a standardized and reliable clinical examinationprior to receiving diagnostic facet joint blocks.1 Allpatients underwent the reference standard controlled,comparative medial branch blocks (MBB). The physi-cians performing the MBB were blinded to the findingsof the clinical examination. Multivariate regression analy-ses were performed and diagnostic accuracy statisticswere calculated for the CDR’s.Results: A CDR including the findings of the manualspinal examination (MSE), palpation for segmentaltenderness (PST), and extension-rotation test (ERT)demonstrated a specificity of 84% (95% CI: 77–90) andpositive likelihood ratio of 4.94 (95% CI: 2.80–8.20). The




PST and MSE possessed a sensitivity of 94% (95% CI:90–98) and 92% (95% CI: 88–97) respectively. Negativefindings on the PST were associated with a negativelikelihood ratio of 0.08 (95% CI: 0.03–0.24).Discussion/Conclusion: The MSE and PST may beuseful tests in identifying patients that are not suitable fordiagnostic facet joint blocks. Further research is neededto validate the CDR’s prior to their routine use in clinicalpractice.1. Schneider GM, Jull G, Thomas K, et al. Intrarater andInterrater Reliability of Select Clinical Tests in PatientsReferred for Diagnostic Facet Joint Blocks in the Cervi-cal Spine. Arch Phys Med Rehabil. Aug 2013;94(8):1628–1634.

Low Levels of Pain Catastrophization andDisability Predicts Success of CervicalRadiofrequency Neurotomy in Individuals WithChronic Whiplash SymptomsAshley D Smith, AD, [email protected],Gwen Jull1, Geoff M. Schneider, PT, PhD2,Bevan Frizzell2, Al Hooper, RA2, Michele Sterling,PT, PhD3, (1) Division of Physiotherapy, NHMRC Centreof Clinical Excellence Spinal Pain, Injury and Health,University of Queensland, Brisbane, Australia, (2) Facultyof Medicine, University of Calgary, Calgary, Alberta,Canada, (3) Centre of National Research on Disability andRehabilitation Medicine Griffith Health Institute, GriffithUniversity, Gold Coast, AustraliaBackground: Sensory hypersensitivity and psychologicalfactors of chronic whiplash associated disorders (WAD)have been shown to be modulated by cervical radiofre-quency neurotomy (cRFN) [1,2]. However, it is unknownwhich of these features predict successful response tocRFN.Objectives: To determine if clinical features (physicaland psychological) of WAD predict successful responseto cRFN.Methods: 53 individuals with chronic WAD who under-went cRFN participated in this longitudinal study pre/post cRFN. Individuals were categorized into risk stratafor widespread sensory hypersensitivity - WSH (presenceor absence), and together with posttraumatic stresssymptom severity (PDS scale), pain catastrophizing(PCS) and disability (NDI), entered as predictor variablesinto a multivariate logistic regression model (forwardsstepwise fashion). Accuracy statistics were performed.The dichotomized outcome measure was Global Ratingof Change ≥ 4 (‘SUCCESS’) or <4 (less-SUCCESS) withcRFN.Results: Low pain catastrophizing and low disabilityscores both predicted ‘SUCCESS’ of cRFN. ROC analy-ses correctly predicted 95–98% of those with a successful

outcome (sensitivity), but only 23% of those without a‘successful’ outcome (specificity).Discussion: The results of this study indicate that thepresence of WSH or posttraumatic stress symptoms donot predict response to cRFN. Low levels of pain cata-strophizing and disability remained significant predictorsfor a successful response to cRFN.

References:1. Smith AD, Jull G, Schneider G, Frizzell B, HooperRA, Sterling M. Cervical Radiofrequency NeurotomyReduces Central Hyperexcitability and Improves NeckMovement in Individuals with Chronic Whiplash. PainMed 2014;15(1):128–141.2. Smith A, Jull G, Schneider G, Frizzell B, Hooper A,Dunne-Proctor R, Sterling M. Cervical RadiofrequencyNeurotomy Reduces Psychological Features in Individu-als with Chronic Whiplash Symptoms. Pain Physician2014;In Press.

Challenging perceptions: Does access to workers’compensation influence neuromodulation therapyoutcomes?Paul Verrills MD, Bruce Mitchell MD, David Vivian MDand Adele Barnard PhD, Metro Pain Clinic, Melbourne,AustraliaBackground: The perception in the medical community,based almost entirely on anecdote, is that compensablepatients do not respond well to interventional pain treat-ment and in particular, to spinal cord stimulation (SCS).Previous publications demonstrating poor outcomes inthe compensation setting have fundamental flaws whichmay dangerously impact on these patients access toappropriate medical management.Objective: To investigate, in a clinical setting, whetheroutcomes markedly differ for privately insuredversus compensable patients receiving neuromodulationtherapy.Methods: Over five years, we treated 411 patients witheither conventional spinal SCS, high frequency SCS(HF-10 SCS) or peripheral nerve field stimulation(PNFS). Of these, 288 were privately insured and 123were recipients of workers’ compensation. Outcomesincluding pain (NPRS), analgesics, employment status,disability (Oswestry Disability index (ODI)), and patientsatisfaction. Mean follow-up was 10.6 ± 11.3 (3–60)months. Study was IRB approved.Results: A significant reduction in post-treatment NPRSwas observed in both privately insured (P) and compen-sable (C) patients (P: 7.3 ± 1.8 vs. 3.1 ± 2.3, p ≤ 0.05; C:7.2 ± 1.7 vs. 3.3 ± 2.2, p ≤ 0.05). Moreover, the reductionin NPRS for all three therapies did not vary (P: PNFS 4.3± 2.5; SCS 4.5 ± 2.6; HF-10 SCS 3.7 ± 2.1 vs. C: PNFS4.5 ± 2.1; SCS: 4.0 ± 1.7; HF-10 SCS 3.1 ± 2.0). Analgesic



use reduced in 75% of private and 66% of compensablepatients and this correlated with pain relief (P: r = 0.46,p = 0.001; C: r = 0.44, p = 0.001). Reductions in ODIdisability and high treatment satisfaction was reported.Among applicable subjects, approximately 48% of pri-vately insured and 33% of compensable patients reportedan improved capacity for paid employment.Conclusion: In a large cohort, we have clearly demon-strated comparable outcomes in both privately insuredand compensable patients. This study supports unbiasedaccess for all patients to appropriate neuromodulationtherapies.

Stem Cell Transplantation for Regeneration of theAnimal Intervertebral Disc: Systematic Review andMeta-analysisZhen Wang, [email protected], Shanmiao Gou2,3,Marisa J. Terry2, Jason P. Dauffenbach2,Mathew J. Pingree2,3, William D. Mauck3,Randy A. Shelerud2,4, Timothy P. Maus5,Carman M. Terzic2, Wenchun Qu2,3,4, (1) Center for theScience of Healthcare Delivery, (2) Department of PhysicalMedicine and Rehabilitation, (3) Department ofAnesthesiology Pain Division, (4) Spine Center,(5) Department of RadiologyIntroduction: Management of intervertebral disc (IVD)degenerative disease is challenging, as it is accompaniedby irreversible loss of IVD cells. Stem cell transplantationto the disc has shown promise in decelerating, arresting,or reversing the degenerative process. Multiple pre-clinical animal trials have been conducted, but with het-erogeneous designs and conflicting outcomes. To assessthe effect of stem cell transplantation, a systematic reviewof global literature and a meta-analysis was performed.Materials and Methods: A comprehensive literaturesearch was conducted through November 19th, 2013.Inclusion criteria consisted of controlled animal trials.Two reviewers screened abstracts and full texts. Disagree-ments were resolved by a third reviewer. Random effectsmodels were constructed to pool standardized meandifferences (SMD). Heterogeneity between studies wasestimated using the I2 statistic.Results: Thirty four studies were included. In therandom effects model, statistically significant differenceswere found with the stem cell group exhibiting increaseddisc height index (SMD = 3.83, 95% CI, 2.30–5.35; p <0.001), increased MRI T2 signal intensity (SMD = 1.92,95% CI, 1.24–2.60; p < 0.001), increased Type II collagenmRNA expression (SMD = 3.75, 95% CI, 1.54–5.96;p < 0.001), and decreased histologic disc degenerationgrade (SMD = −3.03, 95% CI, −4.09–1.97; p < 0.001).There was statistical heterogeneity between studies( p < 0.0001).

Discussion: This systematic review and meta-analysissupports the utility of stem cell transplantationin positively impacting imaging, physiologic, and bio-chemical markers of disc degeneration. An unavoidablelimitation is that all studies were performed onquadruped animals with largely non-weight bearingspines.Conclusion: Stem cells transplanted to the IVD in quad-ruped animals decelerate, arrest, or reverse the IVDdegenerative process. Further studies in human clinicaltrials will be needed to understand if such benefit can betranslated to bipedal humans.

Preliminary Results of Patients Treated withPercutaneous Hydrodiscectomy for RadiculopathySecondary to Herniated Nucleus PulposusThomas C Kowalkowski, DO, [email protected],Interventional Pain & Physical Medicine Clinic, Sartell,MN, USABackground Data: Lumbar discherniation is one of theleading causes of lower back pain and sciatica. If conser-vative treatments such as physical therapy and epiduralsteroid injections fail, microdiscectomyis often con-sidered. Since microdiscectomy was first described,advancements in minimally invasive techniques, includ-ing percutaneous mechanical lumbar discectomy, toreduce intradiscal pressure on the nerve root haveevolved. Percutaneous hydrodiscectomy is performedusing an approach similar to standard discography tomechanically remove disc material with a high velocity,non-thermal saline fluid jet through a cannulated system.Objectives: To evaluate clinical outcomes of patientstreated with percutaneous hydrodiscectomy.Methods: Single center retrospective study on consecu-tive patients that failed conservative managementfor radiculopathy secondary to subligamentous lumbarherniated nucleouspulposus treated with percutaneoushydrodiscectomy at a single lumbar level.Results: Fifteen patients (73% male), mean age of 48.7years (range 26–83) that failed conservative treatmentincluding transforaminal epidural steroid injectionsunderwent percutaneous hydrodiscectomy without com-plications. Fourteen patients (93%) had improvement inback pain and radiculopathy and met the short-termclinical success criteria of improvement within 6 monthspost procedure. Mean VAS decreased from 70.7 to 30(p = 0.0002) and mean ODI improved from 41.1% to22.8% (p = 0.0007) at last follow-up. Four patientsthat had successful short-term outcomes subsequentlyhad additional intervention; 2 spinal cord stimulators,1hydrodiscectomy at a second level and index level, and 1artificial disc replacement.




Conclusions: These preliminary results demonstratepercutaneous hydrodiscectomy is a viable non-surgicaltreatment option in a select group of patients withradiculopathy secondary to subligamentous lumbar her-

niated nucleouspulposus. Larger prospective controlledstudies are needed to further refine patient selection andvalidate the long-term benefits of this promising newtechnology.


Documents

International Spine Intervention Society - 2014 22nd Annual Scientific Meeting Research Abstracts