Prevalence of 'Pure' Lumbar Facet Joint Pain in Patients with Chronic Low Back Pain +MP3
Research Review By Dr. Michael Haneline©
Audio:
Date Posted:
June 2022
Study Title:
The Prevalence of “Pure” Lumbar Zygapophysial Joint Pain in Patients with Chronic Low Back Pain
Authors:
MacVicar J, MacVicar A & Bogduk N
Author's Affiliations:
Southern Rehabilitation Institute, Christchurch, New Zealand.
Publication Information:
Pain Medicine 2021; 22(1): 41–48. doi: 10.1093/pm/pnaa383
Background Information:
The prevalence of lumbar zygapophysial joint (Z joint, or facet) pain in patients with chronic low back pain is quite variable, with estimates ranging from 5% or less (1, 2) to 45% (3, 4), with several other estimates in between.
Z-joint pain is often diagnosed by injecting an anesthetic into the painful joint(s) or the medial branches that innervate the suspected joint(s) to see if the pain is relieved (termed diagnostic block).
What may have led to the diverse prevalence rates in the various studies is the use different criteria for case definition, including whether the patients experience 50% relief, 80% relief, or complete relief of pain after a diagnostic block to signify that the condition is present and whether controlled blocks (i.e. placebo [normal saline] vs. anesthetic) were used. When complete relief of pain is required, studies typically report very low prevalence rates, whereas they are higher when partial relief is used as the diagnostic criterion.
There is also inconsistency about whether age is factor in the prevalence of lumbar Z-joint pain. Some studies have reported its prevalence to be very low in young patients (as low as 2%) yet very high in older patients, 30% to 40% in those over the age of 65 (5, 6). On the other hand, other studies have reported that age is not a factor in lumbar Z-joint pain prevalence (7).
No previous study on the prevalence of lumbar Z-joint pain has based its estimate on complete relief of pain following placebo controlled diagnostic blocks. Confidence in the diagnosis is much stronger when a study’s criteria require complete relief of pain and the use of placebo controls which virtually eliminates false positives, occurring in 25% to 45% of patients when uncontrolled blocks are used.
Therefore, the objective of this study was to estimate the prevalence of lumbar Z-joint pain using stringent operational criteria, testing the following 2 hypotheses:
Z-joint pain is often diagnosed by injecting an anesthetic into the painful joint(s) or the medial branches that innervate the suspected joint(s) to see if the pain is relieved (termed diagnostic block).
What may have led to the diverse prevalence rates in the various studies is the use different criteria for case definition, including whether the patients experience 50% relief, 80% relief, or complete relief of pain after a diagnostic block to signify that the condition is present and whether controlled blocks (i.e. placebo [normal saline] vs. anesthetic) were used. When complete relief of pain is required, studies typically report very low prevalence rates, whereas they are higher when partial relief is used as the diagnostic criterion.
There is also inconsistency about whether age is factor in the prevalence of lumbar Z-joint pain. Some studies have reported its prevalence to be very low in young patients (as low as 2%) yet very high in older patients, 30% to 40% in those over the age of 65 (5, 6). On the other hand, other studies have reported that age is not a factor in lumbar Z-joint pain prevalence (7).
No previous study on the prevalence of lumbar Z-joint pain has based its estimate on complete relief of pain following placebo controlled diagnostic blocks. Confidence in the diagnosis is much stronger when a study’s criteria require complete relief of pain and the use of placebo controls which virtually eliminates false positives, occurring in 25% to 45% of patients when uncontrolled blocks are used.
Therefore, the objective of this study was to estimate the prevalence of lumbar Z-joint pain using stringent operational criteria, testing the following 2 hypotheses:
- that “pure” lumbar Z-joint pain has a nonzero prevalence; and
- that the prevalence of lumbar Z-joint pain is significantly less than current estimates.
Pertinent Results:
There were 2 cohorts of patients, corresponding to 2 slightly different diagnostic protocols that were used, with 67 patients in the first cohort and 139 in the second. No patients who underwent blocks were excluded from the sample and none were lost to follow-up.
Only 20 out of the 67 patients in cohort 1 reported complete relief of pain and underwent control blocks. Nine of these 20 patients (45%) reported no relief both after placebo blocks and after a second local anesthetic block. Four (20%) experienced relief from the placebo but not from a second local anesthetic block. No patient had relief from both the local anesthetic block and placebo. Only 7 patients (35%) were completely relieved of pain after the second local anesthetic block but not after placebo, representing only 10% of the initial sample.
51 out of the 139 patients in cohort 2 reported complete relief of pain from their initial screening blocks and were eligible to proceed to control blocks. However, 12 patients did not proceed for reasons such as the insurer declined to pay, the procedure was too painful, the patient preferred to pursue surgery, etc. and another 9 patients completed only one of the control blocks for various reasons.
5 patients experienced complete relief of pain after each of a total of three local anesthetic blocks and 13 patients were relieved by their second local anesthetic block, but not when normal saline was used. Thus, 18 patients satisfied the criteria for a positive response, representing 13% of cohort 2.
The 7 cases from cohort 1 were combined with the 18 cases from cohort 2 and averaged, resulting in a mean-weighted prevalence of 12%. The lower 2 lumbar segments were most commonly affected, either one segment or both. All age groups except older adults were affected.
Screening blocks were positive in 30% of cohort 1, 37% of cohort 2, and 34% of the combined sample. However, these proportions fell to 10%, 13%, and 12% respectively after control blocks, representing a false-positive rate of 65% for the initial screening blocks.
Only 20 out of the 67 patients in cohort 1 reported complete relief of pain and underwent control blocks. Nine of these 20 patients (45%) reported no relief both after placebo blocks and after a second local anesthetic block. Four (20%) experienced relief from the placebo but not from a second local anesthetic block. No patient had relief from both the local anesthetic block and placebo. Only 7 patients (35%) were completely relieved of pain after the second local anesthetic block but not after placebo, representing only 10% of the initial sample.
51 out of the 139 patients in cohort 2 reported complete relief of pain from their initial screening blocks and were eligible to proceed to control blocks. However, 12 patients did not proceed for reasons such as the insurer declined to pay, the procedure was too painful, the patient preferred to pursue surgery, etc. and another 9 patients completed only one of the control blocks for various reasons.
5 patients experienced complete relief of pain after each of a total of three local anesthetic blocks and 13 patients were relieved by their second local anesthetic block, but not when normal saline was used. Thus, 18 patients satisfied the criteria for a positive response, representing 13% of cohort 2.
The 7 cases from cohort 1 were combined with the 18 cases from cohort 2 and averaged, resulting in a mean-weighted prevalence of 12%. The lower 2 lumbar segments were most commonly affected, either one segment or both. All age groups except older adults were affected.
Screening blocks were positive in 30% of cohort 1, 37% of cohort 2, and 34% of the combined sample. However, these proportions fell to 10%, 13%, and 12% respectively after control blocks, representing a false-positive rate of 65% for the initial screening blocks.
Clinical Application & Conclusions:
This study showed that only 12% of patients with chronic lower back pain who were referred to a physician specializing in pain management and the use of joint injections met the criteria necessary to be diagnosed with Z-joint (facet) pain. This prevalence is markedly less than what has been previously reported in the literature. The authors attributed this discrepancy to the stringent criteria for case definition that were used, which required complete relief of lumbar Z-joint pain following randomized, placebo-controlled medial branch blocks.
This information will enable clinicians to formulate more accurate diagnoses in patients with chronic lower back pain and assist in developing appropriate treatment plans.
This information will enable clinicians to formulate more accurate diagnoses in patients with chronic lower back pain and assist in developing appropriate treatment plans.
Editor’s comment: So, what does this study mean for chiropractors? Our main intervention, spinal manipulation, is thought to target the facet joints primarily (even though the mechanism of action for SMT likely goes well beyond the joints!). This study suggests that the prevalence of “pure” facet pain in CLBP is lower than we thought. This shouldn’t surprise us, as it is well known that LBP is a multi-factorial condition, which often requires a multi-factorial treatment approach, of which SMT is normally only one component. Also – there are many other potential pain generators and mechanisms that can affect humans! Remember, the identification of an exact pain generator is a separate (and perhaps less important?) issue from whether facilitating proper movement in a spine, whether segmentally or globally, can have a positive impact on a patient’s condition. At this point, SMT remains a reasonable component of a multi-pronged approach to the management of chronic LBP.
Study Methods:
This was a practice-based study in which patients from one practice who met the inclusion criteria were sequentially enrolled. The procedures that were utilized were usual in that practice as well as in the national administrative system in which they operated. Since it was a cross sectional prevalence study, there were no control or comparison groups.
To be included in the study, patients had to have had back pain for more than 3 months and clinical features that would likely lead to a diagnosis of lumbar Z-joint pain, including constant, dull, aching pain, unilateral or bilateral, in the lumbar or lumbosacral region, without or without referred pain into the buttock or lower limb (8). Patients were excluded if serious causes of back pain (ex. tumors or infections) were evident on imaging, they had lancinating pain into the lower limb or neurological signs suggestive of radicular pain or radiculopathy, as well as patients who were pregnant.
Patients were informed that a single block could not establish a diagnosis and if positive, would need to be tested with a second and third block using different agents, including a placebo. To establish a diagnosis, blocks had to be target specific, patients had to have complete relief of pain, the response had to be repeated with a second local anesthetic block, and the patient had to distinguish the effects of the local anesthetic from a placebo injection.
Blocks were administered based on the distribution of the patient’s pain and published data on the location of painful joints, most commonly at the L4–L5 and L5–S1 levels. Patients with bilateral pain received blocks on both sides.
Medial branch blocks were initially performed at L3, L4 (for the L4–L5 joint) or at L4, L5 (for the L5–S1 joint) as a screening to determine if the patient was a candidate for intra-articular blocks. If the initial blocks at L3, L4 or L4, L5 did not relieve the patient’s pain, a second screening block was performed. If still no relief of pain, no further investigations were undertaken. Patients whose pain was in the middle or upper lumbar spine received blocks directed at segmental levels that correlated with the center of their pain distribution.
Patients whose initial screening blocks relieved the pain completely were given placebo control blocks as a test of the first response. A local anesthetic agent was used for the first block to provide evidence that the patient’s pain could be completely relieved. Next, the patient was randomly assigned to receive either normal saline placebo or another type of local anesthetic agent that was not used for the first block. The patient then received a third block comprised of the agent they did not receive for the second block. The use of 2 local anesthetics tested for consistency of response and the placebo tested whether the patient could distinguish an active control from an inactive control. Patients had to experience complete relief of pain with the anesthetic injections and no relief with the placebo to be considered as having Z-joint pain.
Another protocol was used later in the study that utilized fully randomized placebo-controlled blocks in which there was a possibility that the patient would not receive a placebo. This protocol more rigorously controlled for guessing. Under this protocol, patients received either lidocaine or bupivacaine for the initial screening block. For the second and third blocks, the patient was randomly assigned to receive either lidocaine or bupivacaine or normal saline. The patient could receive the same or a different local anesthetic on two or three occasions and normal saline on one or no occasion.
To be included in the study, patients had to have had back pain for more than 3 months and clinical features that would likely lead to a diagnosis of lumbar Z-joint pain, including constant, dull, aching pain, unilateral or bilateral, in the lumbar or lumbosacral region, without or without referred pain into the buttock or lower limb (8). Patients were excluded if serious causes of back pain (ex. tumors or infections) were evident on imaging, they had lancinating pain into the lower limb or neurological signs suggestive of radicular pain or radiculopathy, as well as patients who were pregnant.
Patients were informed that a single block could not establish a diagnosis and if positive, would need to be tested with a second and third block using different agents, including a placebo. To establish a diagnosis, blocks had to be target specific, patients had to have complete relief of pain, the response had to be repeated with a second local anesthetic block, and the patient had to distinguish the effects of the local anesthetic from a placebo injection.
Blocks were administered based on the distribution of the patient’s pain and published data on the location of painful joints, most commonly at the L4–L5 and L5–S1 levels. Patients with bilateral pain received blocks on both sides.
Medial branch blocks were initially performed at L3, L4 (for the L4–L5 joint) or at L4, L5 (for the L5–S1 joint) as a screening to determine if the patient was a candidate for intra-articular blocks. If the initial blocks at L3, L4 or L4, L5 did not relieve the patient’s pain, a second screening block was performed. If still no relief of pain, no further investigations were undertaken. Patients whose pain was in the middle or upper lumbar spine received blocks directed at segmental levels that correlated with the center of their pain distribution.
Patients whose initial screening blocks relieved the pain completely were given placebo control blocks as a test of the first response. A local anesthetic agent was used for the first block to provide evidence that the patient’s pain could be completely relieved. Next, the patient was randomly assigned to receive either normal saline placebo or another type of local anesthetic agent that was not used for the first block. The patient then received a third block comprised of the agent they did not receive for the second block. The use of 2 local anesthetics tested for consistency of response and the placebo tested whether the patient could distinguish an active control from an inactive control. Patients had to experience complete relief of pain with the anesthetic injections and no relief with the placebo to be considered as having Z-joint pain.
Another protocol was used later in the study that utilized fully randomized placebo-controlled blocks in which there was a possibility that the patient would not receive a placebo. This protocol more rigorously controlled for guessing. Under this protocol, patients received either lidocaine or bupivacaine for the initial screening block. For the second and third blocks, the patient was randomly assigned to receive either lidocaine or bupivacaine or normal saline. The patient could receive the same or a different local anesthetic on two or three occasions and normal saline on one or no occasion.
Study Strengths / Weaknesses
The protocols used in this study were meticulous and likely reflect a more accurate estimate of the prevalence of Z- joint pain in patients with chronic lower back pain. However, since this was an observational study in which there were no comparison or control groups, non-randomized patients were derived from only one practice, and minimal blinding was implemented, applying its findings to other populations has limitations.
Additional References:
- Jackson R, Jacobs R, Montesano P. Facet joint injection in low back pain: A prospective study. Spine 1988; 13(9): 966–71.
- Carette S, Marcoux S, Truchon R, et al. A controlled trial of corticosteroid injections into facet joints for chronic low back pain. New Engl J Med 1991; 325(14): 1002–7.
- Manchikanti L, Pampati V, Fellows B, et al. Prevalence of lumbar facet joint pain in chronic low back pain. Pain Physician 1999; 2(3): 59–64.
- Manchikanti L, Pampati V, Fellows B, et al. The inability of the clinical picture to characterize pain from facet joints. Pain Physician 2000; 3(2): 158–66.
- DePalma M, Ketchum J, Saullo T. What is the source of chronic low back pain and does age play a role? Pain Med 2011; 12(2): 224–33.
- DePalma M, Ketchum J, Saullo T. Multivariable analyses of the relationships between age, gender, and body mass index and the source of chronic low back pain. Pain Med 2012; 13(4): 498–506.
- Schwarzer A, Wang S, Bogduk N, et al. Prevalence and clinical features of lumbar zygapophysial joint pain: A study in an Australian population with chronic low back pain. Ann Rheum Dis 1995; 54(2): 100–6.
- Bogduk N. On the definitions and physiology of back pain, referred pain, and radicular pain. Pain 2009; 147(1–3): 17–9.
