Minimally-interventional therapeutic procedures in the spine:an evidence-based review
Ioannis Karnezis FRCS(Orth)Orthopaedic and Spinal Surgeon,Director, Back Care network (Athens)
invasive interventional techniques for the
clinical trials al articles were categorised
practicing clinician from the principles of
clinical practice) as all clinical decisions
The aim of the present article is to study
the peer-reviewed medical literature andevaluate the current evidence oncommon minimally-interventionaltherapeutic spinal procedures. Table I: The minimally-invasive therapeutic interventional spinal procedures considered in thepresent study
2. Selective nerve root injections3. Epidural adhesiolysis4. Zygapophyseal (facet) joint injections5. Zygapophyseal (facet) joint denervation6. Sacroiliac joint injections7. Intradiscal Electrothermal Therapy (IDET)8. Percutaneous lumbar endoscopic discectomy9. Vertebroplasty and Kyphoplasty
Table II: ‘Levels of Evidence’ of published studies
High-quality randomized controlled trial with statistical y significantdifference or no statistical y significant difference but narrow confidenceintervals,orSystematic review of Level-I randomized controlled trials (and studyresults were homogeneous).
Lesser-quality randomized controlled trial (e.g., <80% follow-up, no blinding,or improper randomization),orProspective comparative study,orSystematic review of Level-II studies or Level-I studies with inconsistentresults.
Case-control study,orRetrospective comparative study,orSystematic review of Level-III studies. (Centre for Evidence-Based Medicine, Headington, Oxford OX3 7LF, United Kingdom)Table III: ‘Grades of Recommendation’ of treatment methods
consistent level 2 or 3 studies or extrapolations from level 1 studies
level 4 studies or extrapolations from level 2 or 3 studies
level 5 evidence or troublingly inconsistent or inconclusive studies of anylevel
(Centre for Evidence-Based Medicine, Headington, Oxford, OX3 7LF, United Kingdom)
corticosteroid close to the site ofsymptomatic pathology which is
followed in clinical practice, namely (a)
(either the caudal epidural injection, when
interlaminar epidural injection, when the
(one year) pain relief in cases of cervical
nerve root block) which involves delivery
of corticosteroid selectively to the level
syndrome (grade of recommendation: A).
for their use include acute radicular pain,
further evaluation by level-I studies.
post-lumbar laminectomy (‘failed back’)syndrome (1). However, there is ongoing
corticosteroid close to the site of origin
pain of short-term duration only (2,3) or
no significant reduction in pain relief and
injection was also found to be inferior to
(therapeutic injection) is performed.
Although the role of diagnostic selective
nerve root injections in the differential
there is significant controversy about the
in the latter pathology a short-lived (one
caudal epidural injections without steroid
to acute disc prolapse rather in cases of
symptomatic lumbar spinal stenosis (8). Table IV: Level of Evidence of studies on therapeutic nerve root injections
no statistical differences in primary outcome variables (pain, function);
reduction in the need for spinal surgery (‘surrogate end point’)
discectomy) syndrome (grade ofrecommendation: A).
However, two level-I studies (randomizedcontrolled trials) addressing the
in studies of case series without control
steroid injection) and control (localanesthetic only injection) groups for
chronic lumbar radicular pain (17) and the
Postoperative epidural fibrosis is said to
‘failed back surgery’ syndrome (18).
nerve root injections also failed to show
cervical nerve root steroid injections (19).
selective steroid nerve root injections for
statistically significant effect of steroid
tissue believed to be responsible for the
selective) interspinous epidural injections
endoscopy (‘myeloscopy’) to allow three-
dimensional visualization of the contents
selective caudal epidural steroid injection
root injections of corticosteroids have no
effect on the long-term natural history of
literature (Table V) reveals four level-I
radiculopathy or the 'failed back’ (post-
symptomatic postoperative epiduralfibrosis epidural adhesiolysis is
significantly more effective than placebo
be the result of repeated procedures, the
effects of a single procedure lasting for
an average of around 3 to 4 months (24).
from 80-97% at 3 months to 47-72% at 12months (21-23). However, this appears to
Table V: Level of Evidence of studies on epidural adhesiolysis
successful results with repeated procedures ranges from 80-97% (3 months)
to 47-72% (12 months);effects of a single procedure last for average 3 to 4 monthssuccessful results with both non-endoscopic and endoscopic adhesiolysis
also shows similar successful results forsymptomatic postoperative epidural
endoscopic adhesiolysis (25-27). Final y,
there is one level-IV study (28) reporting
contrast material before injection usually
months is an effective minimally-invasive
bilaterally usually at multiple levels of
Comment: as the majority (five out of the
results of facet joint injections for lower
back pain of facet origin. The first trial
between three patient groups havingintra-articular and peri-articular injection
neck pain (‘facet joint syndrome’).
Research into the role of facet joints in
spinal pain has shown that cervical facet
groups having intra-articular injection of
recent, trial (n=200 randomized patients)
having intra-articular injections with or
of 12 months from the procedure (34).
joints of an affected motion segment.
literature (Table VI) shows that there are
relief following facet injections that has
controlled trials) on the results of medial
response after facet block injections. The two largest of these randomized
Facet joints are innervated by the medial
neurotomy’ or simply ‘facet neurotomy’, is
the nerve supply to a painful facet joint
improvement in functional disability. The
base of the transverse processes alongthe course of the medial branch before
Table VI: Level of Evidence of studies on medial branch (facet) neurotomy
no difference in level of pain, physical activities and analgesic intake;short-term improvement in function but no further difference in level of
pain and functional disability;significant al eviation of pain and functional disability for 12 months aftertreatmentLumbar: pain relief for average 4-12 months in 60-87% of patients;
Cervical: pain relief for average 7-9 months in 80% of patients;Thoracic: pain relief form 76% at 1 year to 69% at 3 years (one study)
the published data on the results offacet neurotomy for back pain of facet
what has been described as ‘facet joint
synovitis’ in younger athletic patients (41).
lasting effectiveness (up to three years)
secondary to facet joint involvement (44).
and related disability for two years after
Conclusion: there is limited evidence that
sacroiliac joint injection of steroids can
has been attributed to the ‘sacroiliac joint
an effective treatment for back pain offacet joint origin.
Intradiscal Electrothermal Therapy (IDET)
can be at least temporarily (average 9 to
cervical or thoracic pain of facet origin
by nociceptive nerve fibers that havebeen shown to grow into the annulus of
the affected disc. During the procedure a
energy are transferred to a broad section
spondylarthropathic origin (‘sacroiliac
joint syndrome’). Sacroiliac joint injections
of nociceptive fibers. This is believed to
annular fissures, reduction of disc bulge
and desensitization of the pain receptors
material to confirm intra-articular position
of the needle before injection usual y of
steroid (therapeutic procedure). If helpful,
The value of sacroiliac joint injection of
anaesthesia and light sedation. The latter
literature shows that there are no level-I
studies (randomized controlled trials) on
structures. With the patient in the prone
the results of sacroiliac joint injection
patients with low back pain attributed to
the ‘sacroiliac joint syndrome’. There is
contact with the posterior annular wal .
strictly defined patients. Interestingly, the
single available level-II (prospective non-
literature (Table VII) shows that there are
only two level-I studies (randomizedcontrolled trials) on the results of IDETfor discogenic lower back pain. The firststudy (52) showed no significant changein outcome measures in the IDET or thecontrol group at 6 months and ittherefore no significant benefit from IDET
Table VII: Level of Evidence of studies on Intradiscal Electrothermal Therapy
No significant benefit from IDET over placebo;
IDET appears to provide worthwhile relief in a smal proportion ofstrictly defined patientsPain reduced by half in 54% of patients;
Pain relieved completely in 20% of patients;Long-term results enduringSuperior to conventional conservative care in pain reduction in
substantial proportion of carefully selected cases
published data on the results of IDET for
discogenic lower back pain representlevel-IV evidence (case series studies
practical y limited to level-IV (case series
regarding the effect of IDET treatment of
ranged from half of the patients beingdissatisfied with the outcome (58), early
surgical method for the treatment ofsymptomatic disc prolapse (herniation). It
literature (Table VIII) shows that the vast
resolution of neurological deficit) in 77%
Table VIII: Level of Evidence of studies on percutaneous endoscopic lumbar discectomy
higher (not statistical y significant) rate of symptom resolution compared
to open microsurgical discectomyno difference in clinical outcome between percutaneous endoscopic
lumbar discectomy and open microdiscectomyexcel ent / good clinical outcome (pain relief and resolution of
neurological deficit) in 77% to 94% of cases
studies that fall into the level-III evidencecategory (retrospective comparative
open microdiscectomy (68-69). Thesingle published randomized controlled
higher rate of resolution of sciatica, low-
height and often prolonged disability.
from compression osteoporotic fractures.
of studied cases. Of note is that no other
since the time of publication of the latter
position by insertion of a special cannula
through the pedicle of the vertebra or via
persistent segmental instability and pain,
operative (surgical morbidity) methods of
trauma, short hospitalization, accelerated
literature (Table IX) shows that there is
Vertebroplasty and bal oon Kyphoplasty.
appreciation of the limitations of thetraditional non-operative (poor restoration
Table IX: Level of Evidence of studies on Vertebroplasty and balloon Kyphoplasty
useful for early pain relief assisting rehabilitation;
no difference with nonoperative management beyond 12 months(or less)rapid, significant, sustained improvement in back pain, function and
published data on the results ofVertebroplasty and bal oon Kyphoplasty
persist beyond 12 months (77-80). Final y,
a level-II (prospective comparative) study
(71-74). This also holds true for cases of
Review of the level-II evidence (deriving
epidural corticosteroid injection in themanagement of sciatica. Br JRheumatol 1988;27(4):295-9.
repeatedly-performed epiduraladhesiolysis for the ‘post lumbar
recommendation: B). More prospective,randomized controlled studies are
stenosis do not improve afterepidural steroid injection. Clin J Pain1998;14(2):148-51.
B, et al. Forceful sacrococcygealinjections in the treatment ofpostdiscectomy sciatica. Acontrolled study versusglucocorticoid injections. Joint BoneSpine 2001;68(1):43-9.
Jamison RN. Effects of epiduralsteroid injection on pain due tolumbar spinal stenosis or herniateddisks: a prospective study. ArthritisCare Res 1998;11(4):291-7.
9. Stav A, Ovadia L, Sternberg A, et al.
Cervical epidural steroid injection forcervicobrachialgia. Acta Anaesthesiol
Backache (3rd edition). Baltimore:Williams and Wilkins. Chapter 13:
10. Thomas E, Cyteval C, Abiad L, et al.
interspinous corticosteroid injectionin discal radiculalgia - a prospective,
for nerve root compression. Arandomised, controlled trial. J Bone
Farrow D. Comparison of caudalsteroid epidural with targeted steroid
effect of nerve-root injections on theneed for operative treatment of
epidural steroids: an outcome study.
23. Veihelmann A, Devens C, Trouillier H,
low back pain: a randomized,double-blind trial. Pain Physician
efficacy of corticosteroids inperiradicular infiltration for chronic
lumbar laminectomy syndrome: aone-year outcome study and cost
patients with failed back surgerysyndrome: a comparison of three
solutions. Clin J Pain 1999;15(2):132-5.
et al. Effectiveness of percutaneousadhesiolysis with hypertonic saline
19. Anderberg L, Annertz M, Persson L, et
radiculopathy: a prospective andrandomised study. Eur Spine J
Transforaminal Ventral EpiduralAdhesiolysis. Pain Physician2001;4(3):273-279.
injection of steroid/local anaestheticduring epiduroscopy potentially
30. McCulloch J, Transfeldt E. Macnab’s
31. Lilius G, Laasonen EM, Myllynen P, et
41. Vad VB, Cano WG, Basrai D, et al.
clinical experience. Pain Physician2003;6(3):307-12.
series of consecutive patients. PainMed 2005;6(4):282-6.
back pain: a randomized clinical trial.
neurotomy for chronic neck pain. Neurosurgery 1999;45(1):61-7.
study with minimum 1-year follow up. Pain Physician 2006;9(2):97-105.
36. Leclaire R, Fortin L, Lambert R, et al.
45. Barlocher CB, Krauss JK, Seiler RW.
J Neurosurg 2003;98(Suppl 1):14S-20S.
discogenic low back pain. Spine J2004;4(1):27-35.
from internal disc disruption. Spine J2002;2(5):343-50.
55. Lee MS, Cooper G, Lutz GE, et al.
joints: a prospective clinical trial. Int
Lumbar Discogenic Pain: A Minimum2-Year Clinical Outcome Study. Pain
48. Luukkainen R, Nissila M, Asikainen E,
et al. Periarticular corticosteroidtreatment of the sacroiliac joint in
prospective outcome study with aminimum 2-year follow-up. Spine
49. Maugars Y, Mathis C, Berthelot JM, et
al. Assessment of the efficacy ofsacroiliac corticosteroid injections in
57. Endres SM, Fiedler GA, Larson KL.
function and reducing chronic lowback pain in selected patients.
Kautiainen HH, et al. Efficacy ofperiarticular corticosteroid treatment
of the sacroiliac joint. Clin ExpRheumatol 2002;20(1):52-4.
for sacroiliac joint syndrome. Am JPhys Med Rehabil 2001;80(6):425-32.
placebo for the treatment of chronicdiscogenic low back pain. Spine
61. Choi G, Lee SH, Raiturker PP, et al.
Percutaneous endoscopicinterlaminar discectomy for
53. Pauza KJ, Howell S, Dreyfuss P, et al.
contained disc herniations. Neurosurgery 2005;57(4 Suppl):357S-
endoscopic disc and spine surgery:foraminal approach. Surg Technol Int
endoscopic discectomy: surgicaltechnique and preliminary results
treatment of foraminal andextraforaminal lumbar disc
71. Garfin SR, Buckley RA, Ledlie J.
symptomatic vertebral bodycompression fractures results in
64. Kafadar A, Kahraman S, Akboru M.
and quality of life for elderly patients.
fractures. Spine 2006;31(10):1113-8.
73. Han KR, Kim C, Eun JS, Chung YS.
herniations. J Spinal Disord Tech2006;19(5):338-43.
outcome, and prognostic factors of43 consecutive cases. Spine
Schroeder RJ, et al. Percutaneousbal oon kyphoplasty in the treatment
microdiscectomy: a matched cohortanalysis. Mt Sinai J Med
76. Masala S, Fiori R, Massari F, Simonetti
G. Vertebroplasty and kyphoplasty:new equipment for malignant
69. Schizas C, Tsiridis E, Saksena J.
Clark WA. Clinical outcomes afteracute osteoporotic vertebralfractures: a 2-year non-randomisedtrial comparing percutaneousvertebroplasty with conservativetherapy. Med J Aust 2006;184(3):113-7.
et al. Reduction of pain and fractureincidence after kyphoplasty: 1-yearoutcomes of a prospectivecontrolled trial of patients withprimary osteoporosis. Osteoporos Int2005;16(12):2005-12.
al. Treatment of painful vertebralfractures by kyphoplasty in patientswith primary osteoporosis: aprospective nonrandomizedcontrolled study. J Bone Miner Res2005;20(4):604-12.
80. Diamond TH, Champion B, Clark WA.
Management of acute osteoporoticvertebral fractures: a nonrandomizedtrial comparing percutaneousvertebroplasty with conservativetherapy. Am J Med 2003;114(4):257-65.
81. Grohs JG, Matzner M, Trieb K, Krepler
P. Minimal invasive stabilization ofosteoporotic vertebral fractures: aprospective nonrandomizedcomparison of vertebroplasty andballoon kyphoplasty. J Spinal DisordTech 2005;18(3):238-42.
The Binomial Distribution: Logistic RegressionA common type of data that is encountered is when there is a binary outcome. Examples includewhether a subject survives or dies from some treatment. Statisticians usually talk about theoutcomes ( Yi ) from the i th trial ( i =1,., n ) as being "success" ( Yi =1) and "failure" ( Yi =0). These areassumed to be stochastic, so there is a
Köllensperger M, Geser F, Seppi K, Stampfer-Kountchev M, et al. Red flags for multiple system atrophy. Mov Disord. 2008 Jun 15;23(8):1093-9. Abele M, Minnerop M, Urbach H, Specht K, Klockgether T. Sporadic adult onset ataxia of unknown etiology : a clinical, electrophysiological and imaging study. J Neurol. 2007 Oct;254(10):1384-9. Weyer A, Abele M, Schmitz-Hübsch T, Schoch B, Frings M, Ti