Skeletal Radiol (2011) 40:149–157

DOI 10.1007/s00256-010-0984-3

REVIEW ARTICLE

The lumbar facet joint: a review of current knowledge:

Part II: diagnosis and management
Gerard P. Varlotta & Todd R. Lefkowitz &
Mark Schweitzer & Thomas J. Errico & Jeffrey Spivak &
John A. Bendo & Leon Rybak

Received: 18 April 2010 / Revised: 13 May 2010 / Accepted: 14 May 2010 / Published online: 26 June 2010
# ISS 2010

Abstract This article is the second article in a two-part Keywords Lumbar facet joint . Zygapophyseal joint pain .
review on lumbar facet joint pathology. In this review, we Facet-mediated pain . Lumbar intra-articular steroid
discuss the current concepts and controversies regarding the injection . Lumbar medial branch block .
proper diagnosis and management of patients presenting Lumbar medial branch neurotomy
with presumed facet-mediated lower back pain. All efforts
were made to include the most relevant literature from the
fields of radiology, orthopaedics, physiatry, and pain Introduction
management. Our focus in this article is on presenting the
evidence supporting or refuting the most commonly Renewed interest in the lumbar facet joint as a source of
employed injection-based therapies for facet-mediated spinal pain has resurfaced with the refinement of interven-
lower back pain. tional pain procedures and the development of motion
preservation surgical technology. Non-surgical treatment
options available for patients have expanded from intra-
articular facet joint injections to include procedures such
as medial branch blocks and selective medial branch
G. P. Varlotta : T. R. Lefkowitz (*) neurotomies. The reduction of mechanical lower back pain
NYU School of Medicine, and paravertebral spasm after these procedures may
Rusk Institute of Rehabilitation Medicine,
enhance a patient’s spinal strengthening and stabilization
317 East 34th Street, 5th Floor,
New York, NY 10016, USA program.
e-mail: [email protected]

M. Schweitzer
Diagnosis
The Ottawa Hospital, The University of Ottawa,
501 Smyth,
Module S-1 Ottawa, ON, Canada Much has been written about the diagnosis and treatment of
lumbar zygapophyseal joint pain. A review of the relevant
T. J. Errico
literature found conflicting evidence in support of a
NYU School of Medicine,
530 First Avenue, Suite 8U, relationship between radiographic facet joint abnormalities
New York, NY 10016, USA and facet-mediated pain. This may partly be due to the poor
reliability of the lumbar “facet joint syndrome” diagnosis
J. Spivak : J. A. Bendo : L. Rybak
given to patients presenting with primary lower back pain
NYU School of Medicine,
301 East 17th Street, complaints [1]. The “pseudoradicular” referral patterns of
New York, NY 10010, USA the lumbar facet joints may mimic the pain felt from a
150 Skeletal Radiol (2011) 40:149–157

herniated disc and may make differentiating between the had no place in the diagnosis of facet-mediated lower back
two conditions difficult [2]. In fact, zygapophyseal joint pain.
pain may be accompanied by hamstring tightness that limits Pneumaticos et al. used bone scintigraphy with single
straight-leg raising, further confusing the diagnosis with photon emission computed tomography (SPECT) to help
sciatica. After excluding other common etiologies of lower identify patients who might benefit from facet joint
back pain including discogenic pain and nerve root injections [14]. Bone scintigraphy with SPECT has been
impingement, unilateral or bilateral symptoms radiating to shown to increase the sensitivity of depicting actively
one or both buttocks, groins, and thighs and stopping above remodeling bony lesions [15, 16]. The authors prospective-
the knee may be presumed pain of facet origin [2–5]. ly studied 47 patients with lower back pain who were
Exacerbating factors are many and may include psychoso- scheduled for facet joint injections and randomized them
cial stress, increased physical activity or inactivity, lumbar into two groups. Patients in group A underwent bone
extension with or without rotation, and prolonged standing scintigraphy with SPECT prior to their injections. Patients
or sitting [6]. in group B proceeded directly with their injections without
Although CT is the established imaging modality for the bone scintigraphy at the spinal levels indicated by their
radiological diagnosis of lumbar facet joint osteoarthritis, referring physician. Those individuals in group A were
its ability to accurately identify pain originating from the further subdivided into two groups: group A1, where
lumbar facet joints is less certain. Certain investigators, patients received injections only at the levels which were
including Carrera et al., have demonstrated a relationship positive on the bone scan and group A2, where patients had
between the degenerative changes seen on CT and facet- a negative bone scan; these patients were injected similarly
mediated pain by using controlled anesthetic blocks of the to those in group B.
lumbar zygapophyseal joints [7, 8]. The authors inferred The patients in group A1 had a significantly higher
that the facet joints were the primary pain generators if the change in their pain scores at 1-month follow-up compared
patient’s pain relief lasted at least as long as the half-life of to those patients in groups B and A2. Interestingly, in the
the anesthetic agent used. They found CT to be sensitive for patients with positive bone scans, the number of facet joints
facet-mediated pain, but not specific. Other investigators, injected decreased from the original 60 to 27. The authors
including Fairbank et al., Jackson et al., and Revel et al., also performed a cost analysis that showed a $326
did not find such a relationship [9–11]. reduction in Medicare costs per patient with the use of
To help determine the prevalence and clinical features of SPECT imaging. As a result of their work, the authors
the lumbar facet syndrome, Schwarzer et al. evaluated 176 demonstrated that bone scintigraphy with SPECT is helpful
consecutive patients with chronic lower back pain with in identifying lower back pain patients who may benefit
diagnostic intra-articular facet blocks [12]. Of these from facet joint injections at least in the short term.
patients, 47% had a positive response to screening Dolan and colleagues also investigated whether utilizing
injections at one or more levels, but only 15% had at least SPECT imaging in the diagnosis of facet-mediated lower
a 50% response to confirmatory injections when a longer- back pain improves the outcomes of patients treated with
acting anesthetic agent was used. The authors found that intra-articular facet joint injections [15]. The authors
patients’ responses to the anesthetic blocks were not evaluated 58 patients with lower back pain of presumed
associated with any single clinical feature or combination facet origin by SPECT imaging. Twenty-two patients had
of clinical features. They therefore questioned the existence uptake of the isotope in their facets. This group of scan-
of the lumbar facet syndrome, but not the capacity of the positive patients was compared to 36 scan-negative controls
facet joints to produce pain. that had tender facet joints on physical examination. The
Schwarzer et al. in another prospective cross-sectional two groups’ responses to fluoroscopically-guided facet joint
study evaluated the validity of CT in diagnosing pain injections were evaluated over three follow-up sessions.
arising from the lumbar zygapophyseal joints [13]. The The scan-positive patients showed decreased pain scores on
authors evaluated 63 patients with lower back pain of three separate pain inventories at 1 month and on one pain
greater than 3 months duration with CT and facet joint inventory at 3 months. The scan-negative patients’ pain
injections. Three blinded radiologists rated all facet joint scores remained unchanged. A total of 95% of scan-
images from L3-4 through L5-S1 for osteoarthritic positive patients reported improvement at 1 month and
changes and generated a composite score. Injections were 79% at 3 months, which was significantly greater than the
single-blinded, fluoroscopically controlled intra-articular scan-negative controls. Tenderness on physical examination
blocks, given 1 week apart, with the use of placebo did not correlate with increased uptake on the SPECT scan.
injections that the patients consented to. Inter-observer Although facet joint osteoarthritis was a more common
agreement was variable with ICCs ranging from 0.34–0.66 finding in the scan-positive patients, there was no
among the three reviewers. The authors concluded that CT corresponding increased uptake on SPECT imaging. The
Skeletal Radiol (2011) 40:149–157 151

relationship between radiographically abnormal facet joints, level, the interspinous ligament and muscle, and the
increased uptake on bone scan and SPECT imaging, and multifidus muscle [22, 23]. The medial branch of L5,
facet-mediated pain certainly warrants further study. How- however, comes off the L5 dorsal ramus late at the inferior
ever, scintigraphy in its various permutations remains an portion of the L5-S1 facet and then travels in the groove
excellent window into active disease related to bone between the SAP of S1 and the sacral ala [24–26]. With this
remodeling and hyperemia. in mind, the L5 dorsal ramus is the target nerve for a
The difficulty in diagnosing facet-mediated pain from diagnostic block, not its medial branch [17].
the history and physical examination alone, coupled with Bogduk strongly recommends performing controlled,
the failure of present-day imaging tools to accurately comparative blocks at more than one level to prove
predict which patients with degenerative changes will zygapophyseal joint pain because single diagnostic blocks
ultimately become symptomatic, leaves controlled, diag- carry a false-positive rate between 25 and 41% [27–29]. It
nostic blocks as the only means of making a putative has been reported that patients can have a placebo response
diagnosis [17–21]. These blocks can be accomplished in to a single block, even when the injection is administered
one of two ways. First, intra-articular injections of an subcutaneously [20]. Genuine zygapophyseal joint pain is
anesthetic agent with a known half-life can be performed highly likely, however, if the patient’s pain is completely
after obtaining an appropriate arthrogram of the targeted relieved each and every time that the joint is blocked. As
joint (Fig. 1). Despite the initial enthusiasm of this stated previously, the duration of a patient’s pain relief
approach, these injections have never been tested for should be concordant with the duration of action of the
validity as a diagnostic test [17]. The second more reliable anesthetic used.
method is via blocking the small nerve fibers which Ideally, less than 100% pain relief after a MBB is
innervate the facet joints, a medial branch block (MBB). considered a spurious response as it implies the presence of
A single zygapophyseal joint is innervated by two medial multiple pain generators. The studies conducted to date
branches from the dorsal primary ramus of one spinal level show that patients do not commonly have multiple sources
and the level immediately above. The segmental nomen- of spinal pain. Combinations of discogenic pain and facet-
clature is post-fixed (expanded inferiorly by one level) mediated pain or facet-mediated pain and sacroiliac joint
whereby the L3 and L4 medial branches innervate the L4-5 pain have been reported to occur in less than 5% of patients
facet joint and the L4 medial branch and the dorsal ramus [30, 31]. When the origin of a patient’s pain can be
branch of L5 spinal nerve innervate the L5-S1 facet joint identified, it is usually one or the other of these structures
[22, 23]. [17]. Clinically, patients who present with overlapping
A typical lumbar medial branch courses around the neck patterns of referred pain into the lower back, buttocks,
of the superior articular process (SAP) from the vertebrae thighs, or legs often confuse the examiner into thinking that
below and then travels onto the lamina where it divides, there is more than one source of spinal pain present. Painful
giving off branches to the caudal facet joint, the joint at that spasms in the lumbar paravertebral muscles or gluteal
muscles, if present, for example, should direct the examiner
to screen for underlying facet joint or sacroiliac joint
disease.

Management

The prevalence of true zygapophyseal joint pain thus

depends on whether controlled, diagnostic blocks are
utilized in any one study. When they are used, prevalence
rates of 15% among injured workers [12], 40% in one older
population without pre-existing trauma [20], and 45% in a
heterogeneous population of patients seen in a pain clinic
have all been reported [21, 28, 29]. These studies have been
criticized though for using only 50% pain relief as the
criterion for a positive block [18]. If complete pain relief
was used as the criterion standard, then the prevalence rates
Fig. 1 A fluoroscopic oblique image of the lumbar spine during a
would likely be considerably lower.
facet injection demonstrating an arthrogram with contrast outlining the Intra-articular injections for the treatment of lower back
joint pain without a pre-existing diagnosis of zygapophyseal
152 Skeletal Radiol (2011) 40:149–157

joint pain are steeped in controversy. Proponents of this who reported marked or very marked improvement (42% of
intervention can trace their treatment rationale back to the steroid group and 33% of placebo group) was not
original paper by Mooney and Robertson in 1976 describ- statistically different at 1 and 3 months.
ing 100 consecutive patients with lower back discomfort The results at 6 months revealed that more patients
treated with intra-articular facet joint injections consisting treated with methylprednisolone reported greater clinical
of depomedrol and local anesthetic [32]. When an analysis improvement, less pain, and less physical disability.
of pooled studies on the administration of intra-articular Bogduk commented that Carette’s 6-month follow-up data
steroids was performed nearly 30 years later, the results included patients in both groups who reported late (but
showed that slightly less than half of all patients with an positive) responses to the injections [18]. These patients,
initial positive response maintained that response for a time however, had not reported similar improvements at earlier
period of 3–6 months [18]. follow-ups. Bogduk argued that the differences in pain at
A notable exception to this pattern of response was 6 months were spurious and not attributable to a sustained
reported by Lynch and Taylor in their study of 17 patients steroid effect. If the analysis of the raw data were limited to
undergoing intra-articular facet joint injections with meth- only those patients who reported a marked improvement at
ylprednisolone for the treatment of presumed zygapophy- the 1-month follow-up, the actual percentage of patients
seal joint pain [33]. Fifty-three percent of their patients with continued relief at the 6-month mark would not be
reported immediate, complete relief of pain after their statistically different from those patients in the saline group.
injections. The percentage of successful outcomes then As reflected by the literature, most clinicians who
increased to 82% at the 3- and 6-month follow-up mark, as perform intra-articular steroid injections to treat patients
partial-responders gradually became pain free. It was not with lower back pain do so without first utilizing controlled
apparent from a critical appraisal of the published study, diagnostic anesthetic blocks to prove zygapophyseal joint
however, how many of the pain-free patients were from the pain. Lilius and Carette found no evidence to support
original 17 treated with intra-articular injections and how sustained pain relief attributable to the intra-articular
many were from the group treated with extra-articular injection of corticosteroids. Other critics of the Carette
injections [18]. study, however, who support the use of intra-articular
Lilius et al. compared the outcomes of patients treated steroids point out that the authors did not use true
for lower back pain with intra-articular steroid injections, controlled blocks to select their patients as only 50% pain
intramuscular steroid injections, and intra-articular injec- relief was required for study entry [38, 39]. Bogduk
tions of normal saline [34–36]. Twenty-eight patients countered that holding Carette and colleagues to a standard
received intra-articular injections of bupivacaine and of practice that most practitioners do not adhere to
methylprednisolone, 39 patients received intramuscular themselves is disingenuous and weakens their criticism of
injections of bupivacaine and methylprednisolone, and 42 the study.
patients received intra-articular injections of normal saline The dissonance between the immediate response rates
[35]. There was an immediate decrease in pain reported from intra-articular steroid injections and the prevalence of
after the injections in all three groups with VAS scores zygapophyseal joint pain reported in observational studies
quickly returning to baseline over the following 6 weeks. suggests that these results are inflated by false-positive
Although there were a small proportion of patients treated responses [18]. Carrera and Williams, for example, reported
with normal saline that maintained prolonged pain relief a 68% immediate response rate to the injection of
compared to those treated with intra-articular steroids, there methylprednisolone and lignocaine [8]. In an earlier study,
were never any statistically significant differences reported Carrera reported a 65% immediate response rate [7]. The
between the three groups. previously referenced study by Lynch and Taylor reported a
The only referenced study which used controlled 53% immediate response rate [33]. Comparing these
diagnostic blocks prior to the administration of intra- percentages to those of controlled studies of prevalence
articular steroids was performed by Carette et al. [37]. using proper diagnostic blocks confirms these suppositions.
Carette and co-workers followed 95 patients with chronic Therapeutic MBBs have been systematically evaluated
lower back pain for 6 months after being randomized to in only a few high-quality studies. Manchikanti et al.
either intra-articular steroid injections with methylprednis- conducted a prospective, randomized, double-blind con-
olone or isotonic saline. Study eligibility requirements trolled trial of 60 patients with chronic lower back pain
ensured that participants had at least 50% pain relief after [40]. All study participants met diagnostic criteria for
anesthetic infiltration of the targeted joints. At 1-month lumbar zygapophyseal joint pain by means of comparative,
follow-up, none of the outcome measures studied, including controlled diagnostic facet joint nerve blocks with 1%
pain, functional status, and back flexion were statistically lidocaine followed by 0.25% bupivacaine of tender joints
different between the two groups. The subset of patients identified on physical examination. The study cohort was
Skeletal Radiol (2011) 40:149–157 153

divided into four treatment groups each consisting of 15

patients. Groups I and II comprised the non-steroid arm of
the study. Group I patients acted as controls with blocks of
bupivacaine, while group II patients underwent blocks with
both bupivacaine and sarapin. Patients in groups III and IV
comprised the steroid arm of the study. Group III patients
underwent blocks with bupivacaine and betamethasone
while group IV patients had blocks with bupivacaine,
sarapin, and betamethasone. Results showed significant
improvements in pain and functional status in all treatment
groups at 3, 6, and 12 months of follow-up compared to
baseline measurements. There were no significant differ-
ences reported between the steroid and non-steroid groups
in duration of pain relief obtained. In a 1-year follow-up to
this study, Manchikanti and colleagues reported significant
pain relief and functional improvement in 120 patients
following three to four MBBs with an average duration of
pain relief of 15 weeks per injection [41].
The authors acknowledge that there have been no claims
made as to the effect of MBBs on peri-articular inflamma-
Fig. 2 A schematic showing needle placements adjacent to the medial
tion. Manchikanti et al. did not disprove a possible branches at the third and fourth lumbar vertebrae (L3, L4) and the
inflammatory etiology of facet-mediated lower back pain dorsal rami at the fifth lumbar (L5) and first sacral (S1) vertebrae
as intra-articular facet blocks with corticosteroids were not (inset: S1, detail). Reproduced from Gofeld M et al. Pain Physician
part of their study’s protocol. The pain arising from an 2007; 10:291–299. Permission obtained from Holly Long, American
Society of Interventional Pain Physicians, Editorial Services Coordi-
inflamed facet joint with an intact capsule may be nator. [email protected] or 1-270-554-9412 ext. 230
alleviated, albeit for a relatively short period, after an
intra-articular steroid injection as the capsule itself is richly
innervated with nociceptive and autonomic nerve fibers
capable of transmitting pain centrally [42–44]. Medial
branch anesthetic blocks with or without steroids, will
interrupt the flow of nociceptive traffic along the course of
the treated nerves based on the effect of the anesthetic agent
alone (Figs. 2 and 3). Head-to-head comparisons of intra-
articular facet joint steroid injections and medial branch
anesthetic blocks are needed to help resolve this issue.
One may ask, then, is there a procedure used to treat true
lumbar zygapophyseal joint pain that can stand up to
current scientific scrutiny? The answer is equivocally yes
and is found through review of the literature on lumbar
medial branch neurotomies. The history behind this
intervention shows that it has evolved from the “facet
denervation” procedure first espoused by Shealy [45–48].
Shealy developed a method to percutaneously coagulate the
medial branches innervating suspected painful facet joints
using a radiofrequency electrode. Subsequent dissection
studies, however, showed that the anatomical basis for facet
denervation was flawed as there were no nerves where
Shealy and others had placed their electrodes [49, 50]. Fig. 3 An oblique fluoroscopic image of the lumbar spine showing a
Shealy’s technique for facet denervation was corrected needle at the junction of the L5 transverse process and SAP for an L4
after further anatomical study and renamed lumbar medial medial branch block. The target point for an L5 dorsal ramus block is
marked by a white dot. Reproduced from Bogduk N et al. The Spine
branch neurotomy (LMBN) by Bogduk and colleagues in Journal 2008; 8(1):56–64. Permission obtained from Elsevier through
1980 [25] (Fig. 4). The thermal coagulation employed by Copyright Clearance Center. [email protected] or 1-877-
way of a medial branch neurotomy denatures nerve proteins 622-5543
154 Skeletal Radiol (2011) 40:149–157

Fig. 4 AP and lateral radio-

graph showing setup for an L4
medial branch block. The
electrode is seen crossing the
neck of the SAP of L5.
Reproduced from Bogduk N
et al. The Spine Journal 2008;
8(1):56–64. Permission
obtained from Elsevier through
Copyright Clearance Center.
[email protected]
or 1-877-622-5543

offering a far longer lasting clinical effect than a simple Six randomized controlled trials have been published to
medial branch anesthetic block. At temperatures above 80° date [57–62]. Four of these studies have been criticized for
C, motor and sensory fibers are affected in a nonselective their methodological and technical flaws, including failure
manner, interrupting the conduction of nociceptive to employ comparative or placebo-controlled blocks and
impulses, thereby, relieving pain [51, 52]. lack of a validated anatomical technique. The prospective,
The International Spine Intervention Society (ISIS) rec- non-randomized study by Dreyfuss et al. is notable for its
ommends placing the electrodes parallel along the course of strict patient selection criteria, comparative double diag-
the medial branches rather than perpendicular to them [53]. nostic blocks, and anatomically correct lesioning technique
This technique allows for a greater length of the target [61]. The authors performed pre- and post-lesioning EMG
nerves to be coagulated and has been shown to correlate with of the multifidus muscles on 15 patients with chronic lower
the duration of pain relief achieved [54]. The procedure seals back pain to ensure accuracy of their neurotomies.
the nerves in situ without creating a gap across which the Approximately 60% of study patients achieved at least
nerves can regenerate. Healing is delayed by several months 90% pain relief at 12 months, while 87% achieved at least
as the coagulated segments must be repaired by endocellular 60% pain relief. Pain relief was associated with denervation
processes. Patients who test positive with lumbar medial of the multifidus muscles in those segments where the
branch blocks are ideal candidates for LMBN. These medial branches had been coagulated. Dreyfuss’ study
individuals are usually older patients without significant further demonstrated that pre-lesioning electrical stimula-
psychosocial morbidity who have had discogenic pain and tion of the medial branches is unnecessary, if the electrode
sacroiliac joint pain ruled out by other means [53]. is accurately placed radiographically, at the groove of the
Numerous small studies of LMBN have been done with SAP between the intervertebral foramen and the mamillo-
variable but mostly positive results. Vad et al. conducted a accessory ligament [62].
prospective, non-randomized study involving 12 male A relatively large prospective clinical audit performed by
baseball pitchers with sports-related lower back pain Gofeld et al. over a 10-year period from January 1991 to
diagnosed with lumbar MBBs [55]. All participants had December 2000 looked at the response of 174 patients with
failed prior conservative treatments including intra-articular chronic lower back pain treated with LMBN [63]. Some
zygapophyseal joint injections with steroids and local 32% of patients were considered treatment failures because
anesthetic. The mean duration of pain relief achieved after they either experienced no pain relief or had pain relief
radiofrequency neurotomy was 1.3 years and 83% of lasting less than 6 months. Approximately 68% of patients
players were able to return to their pre-procedure level of reported good to excellent results after a 6-month follow-
baseball pitching. Mogalles et al. studied 15 patients with up. Of this subgroup of patients, 96% reported pain relief
chronic facet-mediated lower back pain diagnosed via for 6–12 months, 43% reported relief for 12–24 months,
controlled, comparative MBBs [56]. After undergoing laser and 2% for more than 24 months. Median duration of pain
denervation of the medial branches, eight out of 15 patients relief among the entire clinical cohort of 174 patients was
experienced complete relief and six experienced more than 9 months, and 12 months in those patients who maintained
50% pain relief. good to excellent results for at least 6 months.
Skeletal Radiol (2011) 40:149–157 155

Nath and colleagues performed a randomized con- trolled, comparative anesthetic blocks of the lumbar
trolled study of LMBN in 40 patients with chronic lower medial branches remain the most reliable method of
back pain (20 active and 20 controls). The authors diagnosing facet-mediated pain. LMBN is the most
mandated three separate positive facet blocks as part of studied intervention to date with the most evidence-
their inclusion criteria, in addition to using a multiple based support. Intra-articular corticosteroid injections,
lesioning technique at each facet level to provide although commonly practiced, lack support in the
effective denervation. The active treatment group had medical literature. As mentioned above, intra-articular
significantly greater improvements as compared to the facet blocks to decrease peri-articular inflammation may
placebo group in select quality of life variables, global be effective but need to be compared against anesthetic
perception of improvement, and overall pain. They also MBBs.
had statistically significant improvements in their back Future directions in the study of lumbar facet disease
and leg pain, as well as their lower back and hip range of include refinement of the existing facet joint grading
motion. Pre-lesioning sensory deficits and ankle reflex systems. Efforts are currently underway to improve their
abnormalities also normalized post-procedure. reliability for the purpose of selecting patients for lumbar
Cohen et al. compared the success rates of LMBN by total disc arthroplasty [66]. It is hoped that a more reliable
way of two different pain relief thresholds obtained after grading system will enable practitioners to better choose
diagnostic MBBs [64]. The conventional greater than 50% amongst the non-surgical treatment options available for
pain relief threshold was compared against a more patients with zygapophyseal joint pain.
stringently proposed greater than or equal to 80% pain
relief threshold. In this retrospective, multi-center clinical
analysis, 262 patients with chronic lower back pain who
Conflict of Interest The authors declare that they have no conflict
underwent LMBN and reported at least 50% pain relief of interest.
after a lumbar MBB were included. Subjects were divided
into a partial pain relief group (≥50% but <80%) and a
near-complete pain relief group (≥80%). One hundred and
References
forty-five patients obtained partial pain relief after the MBB
and 117 patients obtained near-complete pain relief. Out-
1. Jackson RP. The facet syndrome. Myth or reality? Clin Orthop
comes between the two groups were compared after Relat Res. 1992;279:110–21.
controlling for 14 demographic and clinical variables and 2. Kalichman L, Hunter DJ. Lumbar facet joint osteoarthritis: a review.
were not significantly different, with 52% of patients in the Semin Arthritis Rheum. 2007;37:69–80.
3. Helbig T, Lee CK. The lumbar facet syndrome. Spine.
partial pain relief group reporting ≥50% relief from the
1988;13:61–4.
radiofrequency denervation procedure as compared to 56% 4. Lippitt AB. The facet joint and its role in spine pain. Management
of patients in the near-complete pain relief group. with facet joint injections. Spine. 1984;9:746–50.
No other forms of conservative treatment including 5. Nelemans PJ, de Bie RA, de Vet HC, Sturmans F. Injection
therapy for subacute and chronic benign low back pain. Spine.
medications, physical therapy, or manual therapy have been
2001;26:501–15.
evaluated for efficacy in lumbar zygapophyseal joint pain 6. Eisenstein SM, Parry CR. The lumbar facet arthrosis syndrome.
[17]. One study by Morrisette et al. examined the thermal Clinical presentation and articular surface changes. J Bone Joint
effects of ultrasound on the peri-articular tissues of the Surg Br. 1987;69:3–7.
7. Carrera GF. Lumbar facet joint injection in low back pain and
lumbar spine [65]. The authors found that continuous
sciatica: preliminary results. Radiology. 1980;137:665–7.
1-MHz ultrasound applied at either 1.5-W/cm2 or 2.0-W/ 8. Carrera GF, Williams AL. Current concepts in evaluation of the
cm2 intensity has the capacity of heating tissue adjacent to lumbar facet joints. CRC Crit Rev Diagn Imag. 1984;21:85–
the L4-5 facet joint. As expected, the higher-intensity 104.
9. Fairbank JCT, Park WM, McCall IW, O’Brien JP. Apophyseal
ultrasound treatments resulted in greater and faster temper-
injection of local anesthetic as a diagnostic aid in primary low-
ature rises. The implication of the deep-heating effects of back pain syndromes. Spine. 1981;6:598–605.
ultrasound for the treatment of zygapophyseal joint pain 10. Jackson RP, Jacobs RR, Montesano PX. Facet joint injection in low-
was not explored. back pain: a prospective statistical study. Spine. 1988;13:966–71.
11. Revel ME, Listrat VM, Chevalier XJ, et al. Facet joint block for
low back pain: identifying predictors of a good response. Arch
Phys Med Rehabil. 1992;73:824–8.
Conclusions 12. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine G, Bogduk N.
Spine. 1994;19:1132–7.
13. Schwarzer AC, Wang S, O’Driscoll D, Harrington T, Bogduk N,
We have reviewed the most relevant literature on the
Laurent T. The ability of computed tomography to identify a
diagnostic and therapeutic treatment options for patients painful zygapophyseal joint in patients with chronic low back
presenting with lumbar zygapophyseal joint pain. Con- pain. Spine. 1995;20:907–12.
156 Skeletal Radiol (2011) 40:149–157

14. Pneumaticos SG, Chatziioannou SN, Hipp JA, Moore WH, Esses 37. Carette S, Marcoux S, Truchon R, Grondin C, Gagnon J, Allard Y,
SI. Low back pain: prediction of short-term outcome of facet joint et al. A controlled trial of corticosteroid injections into facets
injection with bone scintigraphy. Radiology. 2006;238:693–8. joints for chronic low back pain. N Engl J Med. 1991;325
15. Dolan AL, Ryan PJ, Arden NK, Stratton R, Wedley JR, Hamann (14):1002–7.
W, et al. The value of SPECT scans in identifying back pain likely 38. Manchikanti L, Staats PS, Singh V, et al. Evidence-based practice
to benefit from the facet joint injection. Br J Rheumatol. guidelines for interventional techniques in the management of
1996;35:1269–73. chronic spinal pain. Pain Physician. 2003;6:3–81.
16. Holder LE, Machlin JL, Asdourian PL, Links JM, Sexton CC. 39. Slipman CW, Bhat AL, Gilchrist RV, et al. A critical review of the
Planar and high-resolution SPECT bone imaging in the diagnosis evidence for the use of zygapophyseal injections and radio-
of facet syndrome. J Nucl Med. 1995;36:37–44. frequency denervation in the treatment of low back pain. Spine J.
17. Bogduk N. Evidence-informed management of chronic low back 2003;3:310–6.
pain with facet injections and radiofrequency neurotomy. The 40. Manchikanti L, Manchikanti KN, Manchukonda R, Cash KA,
Spine Journal. 2008;8:56–64. Damron KS, Pampati V, et al. Evaluation of lumbar facet joint
18. Bogduk N. A narrative review of intra-articular corticosteroid nerve blocks in the management of chronic low back pain:
injections for low back pain. Pain Med. 2005;6(4):287–96. preliminary report of a randomized, double-blind controlled
19. Schwarzer AC, Derby R, Aprill CN, et al. Pain from the lumbar trial: clinical trial NCT00355914. Pain Physician. 2007;10
zygapophyseal joints: a test of two models. J Spinal Disord. (3):425–40.
1994;7:331–6. 41. Manchikanti L, Singh V, Falco FJ, Cash KA, Pampati V.
20. Schwarzer AC, Wang S, Bogduk N, McNaught PJ, Laurent R. Lumbar facet joint nerve blocks in managing chronic facet joint
Prevalence and clinical features of lumbar zygapophyseal joint pain: one-year follow-up of a randomized, double-blind
pain: a study in an Australian population with chronic low back controlled trial: clinical trial NCT00355914. Pain Physician.
pain. Ann Rheum Dis. 1995;54:100–6. 2008;11(2):121–32.
21. Manchikanti L, Pampati V, Fellows B, Baha AG. The inability of 42. Giles LG, Hasrvey AR. Immunohistochemical demonstration of
the clinical picture to characterize pain from facet joints. Pain nociceptors in the capsule and synovial folds of human zygapo-
Physician. 2000;3:158–66. physeal joints. Br J Rheumatol. 1987;26:362–4.
22. Bogduk N, Wilson AS, Tynan W. The human lumbar dorsal rami. 43. Giles LGF, Taylor JR. Innervation of the lumbar zygapophyseal
J Anat. 1982;134:383–97. joint folds. Acta Orthop Scand. 1987;58:43–6.
23. Bogduk N. Innervation of the lumbar spine. Spine. 1983;8:286– 44. Ashton IK, Ashton BA, Gibson SJ, Polak JM, Jaffray DC,
93. Eisenstein SMJ. Morphological basis for low back pain: the
24. Bogduk N, Twomey LT. Clinical anatomy of the lumbar spine. demonstration of nerve fibers and neuropeptides in the lumbar
2nd ed. London: Churchill Livingstone; 1991. facet joint capsule but not in the ligamentum flavum. J Orthop
25. Bogduk N, Long DM. Percutaneous lumbar medial branch Res. 1992;10:72–8.
neurotomy. A modification of facet denervation. Spine. 45. Shealy CN. Facets in back and sciatic pain. Minn Med.
1980;5:193–200. 1974;57:199–203.
26. Derby R, Bogduk N, Schwarzer A. Precision percutaneous 46. Shealy CN. The role of the spinal facets in back and sciatic pain.
blocking procedures for localizing spinal pain. Part I: the posterior Headache. 1974;14:101–4.
lumbar compartment. Pain Digest. 1993;3:89–100. 47. Shealy CN. Percutaneous radiofrequency denervation of spinal
27. Schwarzer AC, Aprill CN, Derby R, et al. The false-positive rate facets. J Neurosurg. 1975;43:448–51.
of uncontrolled diagnostic blocks of the lumbar zygapophyseal 48. Shealy CN. Facet denervation in the management of back sciatic
joints. Pain. 1994;58:195–200. pain. Clin Orthop. 1976;115:157–64.
28. Manchikanti L, Pampati V, Fellows B, Bakhit CE. Prevalence of 49. Bogduk N, Long DM. The anatomy of the so-called “articular
lumbar facet joint pain in chronic low back pain. Pain Physician. nerves” and their relationship to facet denervation in the treatment
1999;2:59–64. of low back pain. J Neurosurg. 1979;51:172–7.
29. Manchikanti L, Pampati V, Fellows B, Bakhit CE. The 50. Bogduk N, Long DM. Percutaneous lumbar medial branch
diagnostic validity and therapeutic value of lumbar facet joint neurotomy. A modification of facet denervation. Spine.
nerve blocks with or without adjuvant agents. Curr Rev Pain. 1980;5:193–200.
2000;4:337–44. 51. Fs L, Goldring S. The effect of radiofrequency current and heat on
30. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine G, Bogduk N. peripheral nerve action potential in the cat. J Neurosurg.
The relative contributions of the disc and zygapophyseal joint in 1968;29:42–7.
chronic low back pain. Spine. 1994;19:801–6. 52. Frohling MA, Schole W, Wolburg-Bucholz K. Nonselective nerve
31. Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in fiber damage in peripheral nerves after experimental thermocoa-
chronic low back pain. Spine. 1995;20:31–7. gulation. Acta Neurochir. 1998;140:1297–302.
32. Mooney V, Robertson J. The facet syndrome. Clin Orthop. 53. International Spine Interventional Society. An algorithm for the
1976;115:149–56. investigation of low back pain. In: Bogduk N, editor. Practice
33. Lynch MC, Taylor JF. Facet joint injection for low back pain. guidelines for spinal diagnostic and treatment procedures. San
J Bone Joint Surg. 1986;68B:138–41. Francisco, CA: International Spinal Intervention Society; 2004. p.
34. Lilius G, Laasonen EM, Myllynen P, Harilainen A, Salo L. 87–94.
Lumbar facet joint syndrome. Significance of non-organic signs. 54. Lau P, Mercer S, Govind J, Bogduk N. The surgical anatomy of
A randomized placebo-controlled clinical study. Rev Chir Orthop lumbar medial branch neurotomy. Pain Med. 2004;5:289–98.
Reparatrice Appar Mot. 1989;75:493–500. 55. Vad VB, Cano WG, Basrai D, Lutz GE, Bhat AL. Role of
35. Lilius G, Laasonen EM, Myllynen P, Harilainen A, Gronlund G. radiofrequency denervation in lumbar zygapophyseal joint syno-
Lumbar facet joint syndrome: a randomized clinical trial. J Bone vitis in baseball pitchers: a clinical experience. Pain Physician.
Joint Surg. 1989;71(4):681–4. 2003;6(3):307–12.
36. Lilius G, Harilainen A, Laasonen EM, Myllynen P. Chronic 56. Mogalles AA, Dreval ON, Akatov OV, Kuznetsov AV, Rynkov IP,
unilateral back pain: predictors of outcome of facet joint Plotnikov VM, et al. [Percutaneous laser denervation of the
injections. Spine. 1990;15:780–2. zygapophyseal joints in the pain facet syndrome] – Article in
Skeletal Radiol (2011) 40:149–157 157

Russian. Zh Vopr Neirokhir IM N N Burdenko 2004 Jan–Mar; management of chronic low back pain: a randomized double-blind
(1): 20–25. trial. Spine. 2008;33(12):1291–7.
57. van Kleef M, Barendse GA, Kessels A, Voets HM, Weber WE, de 62. Dreyfuss P, Halbrook B, Pauza K, Joshi A, McLarty J, Bogduk N.
Lange S. Randomized trial of radiofrequency lumbar facet Efficacy and validity of radiofrequency neurotomy for chronic
denervation for chronic low back pain. Spine. 1999;24:1937–42. lumbar zygapophyseal joint pain. Spine. 2000;25:1270–7.
58. Gallagher J, Petriccione Di Vadi PL, Wedley JR, Hamman W, 63. Gofeld M, Jitendra J, Faclier G. Radiofrequency denervation of
Ryan P, et al. Radiofrequency facet joint denervation in the the lumbar zygapophyseal joints: 10-year prospective clinical
treatment of low back pain: a prospective controlled double-blind audit. Pain Physician. 2007;10:291–9.
study to assess its efficacy. Pain Clin. 1994;7:193–8. 64. Cohen SP, Stojanovic MP, Crooks M, Kim P, Schmidt RK,
59. Leclaire R, Fortin L, Lambert R, Bergeron YM, Rossignol M. Shields CH, et al. Lumbar zygapophyseal (facet) joint radio-
Radiofrequency facet joint denervation in the treatment of low frequency denervation success as a function of pain relief during
back pain: a placebo-controlled trial to assess efficacy. Spine. diagnostic medial branch blocks: a multicenter analysis. Spine J.
2001;26:1411–6. 2008;8(3):498–504. Epub 2007 Jun 18.
60. van Wijk RM, Geurts JW, Wynne HKJ, Hammink E, Buskens E, 65. Morrisette DC, Brown D, Saladin ME. Temperature changes in
Lousberg R, et al. Radiofrequency denervation of lumbar facet lumbar peri-articular tissue with continuous ultrasound. J Orthop
joints in the treatment of chronic low back pain: a randomized, Sports Phys Ther. 2004;34(12):754–60.
double-blind, sham lesion-controlled trial. Clin J Pain. 66. Stieber J, Quirno M, Cunningham M, Errico TJ, Bendo JA. The
2005;21:335–44. reliability of computed tomography and magnetic resonance
61. Nath S, Nath CA, Pettersson K. Percutaneous lumbar zygapophy- imaging grading of lumbar facet arthropathy in total disc
seal (facet) joint neurotomy using radiofrequency current, in the replacement patients. Spine. 2009;34(23):E833–40.