Secondary Logo

Journal Logo

Intra-articular Triamcinolone Versus Hyaluronate Injections for Low Back Pain With Symptoms Suggestive of Lumbar Zygapophyseal Joint Arthropathy: A Pragmatic, Double-Blind Randomized Controlled Trial

Annaswamy, Thiru, M., MD, MA; Armstead, Corey, MD; Carlson, Luke, MD; Elkins, Nicholas, J., DO; Kocak, Denizen, BS; Bierner, Samuel, M., MD, MRM

American Journal of Physical Medicine & Rehabilitation: April 2018 - Volume 97 - Issue 4 - p 278–284
doi: 10.1097/PHM.0000000000000879
Original Research Articles

Objective The aim of the study was to compare hyaluronate with triamcinolone injections in treating chronic low back pain suggestive of lumbar zygopophyseal joint arthropathy.

Design This was a prospective, double-blind, randomized controlled trial. Thirty subjects were randomly assigned to receive bilateral L3-S1 lumbar zygopophyseal joint injections with triamcinolone (KA) or Synvisc-One (HA). Pain (visual analog scale) and Pain Disability Questionnaire scores at 1, 3, and 6 mos were evaluated.

Results No significant intergroup differences (P > 0.05) in outcomes were noted in the 30 recruited subjects. For KA/HA (baseline; 1 mo; 3 mos; 6 mos), visual analog scale scores were the following: 70 (15)/74 (10); 58 (29)/45 (25); 58 (29)/56 (25); and 59 (28)/63 (24), respectively. Pain Disability Questionnaire scores were the following: 100 (23)/102 (28); 77 (30)/74 (34); 87 (26)/74 (36); and 96 (25)/79 (25). Overall percent improvement at 6 mos for KA was 51 (35) and for HA was 42 (33) (P = 0.51). Synvisc-One group visual analog scale scores improved significantly (70 [20]–45 [25] at 1 mo, P = 0.008). Pain Disability Questionnaire scores improved at 1 mo (100 [23]–77 [30], P = 0.009) in the KA group and at all time points in the HA group (102 [28]–74 [34] at 1 mo, P = 0.002; 74 [36] at 3 mos, P = 0.037; 79 at 6 mos [median = 52–99.5], P < 0.001). Medians and quartiles were used in statistical analysis when data did not pass normality.

Conclusions Patients with chronic low back pain suggestive of lumbar lumbar zygopophyseal joint arthropathy responded similarly to triamcinolone or hyaluronate injections. Synvisc-One group showed significant short- and long-term functional improvement and short-term pain improvement; KA group showed only significant short-term functional benefit and no significant short- or long-term pain improvement.

From the VA North Texas Health Care System, Physical Medicine & Rehabilitation Service, Department of Physical Medicine & Rehabilitation, UT Southwestern Medical Center at Dallas, Dallas, Texas (TMA); River City Rehabilitation and Spine Specialist, Columbus, Georgia (CA); VA North Texas Health Care System, Interventional Radiology, Dallas, Texas (LC); Department of Physical Medicine & Rehabilitation, UT Southwestern Medical Center at Dallas, Dallas, Texas (NJE); UT Southwestern Medical School, Dallas, Texas (DK); and Department of Physical Medicine & Rehabilitation, University of Nebraska Medical Center, Omaha, Nebraska (SMB).

All correspondence should be addressed to: Thiru M. Annaswamy, MD, MA, 4500, S Lancaster Rd, #117 Dallas, TX 75216.

VA North Texas Health Care System, Physical Medicine & Rehabilitation Service, is the institution where the research was conducted.

This article is the report of a clinical trial. However, this clinical trial was conducted in 2012–2014, at which point, there were no requirements for all clinical trials to be registered. Therefore, this RCT does not have a registration with the clinical trials registry.

The abstract of this research was previously presented at: Annaswamy TM, Armstead C, Avraham R, et al. Scientific Paper Presentations: Intra-articular triamcinolone vs. hyaluronate injections for lumbar zygapophyseal joint arthropathy: a pragmatic, double blind randomized controlled trial. Paper presented at: 2015 Annual Assembly of AAPMR, September 2015; Boston, MA. PM&R. 7;9S:91.

Annaswamy TM, Armstead C, Avraham R, et al. Scientific Paper Presentations: Intra-articular Triamcinolone versus Hyaluronate Injections for Lumbar Zygapophyseal Joint Arthropathy: A Pragmatic, Double Blind Randomized Controlled Trial. Abstracts of Scientific Papers and Posters Presented at the Annual Meeting of the Association of Academic Physiatrists. Paper presented at: American Journal of Physical Medicine & Rehabilitation, March 2015; San Antonio, TX, 94:3 (Online Supplement); a6–a7.

Financial disclosure statements have been obtained, and no conflicts of interest have been reported by the authors or by any individuals in control of the content of this article.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (

Lumbar facet syndrome is commonly defined as low back pain (LBP) associated with arthropathy of the di-arthrodial synovial lumbar facet joint(s). The facet joint, an articulation of the superior and inferior articular processes of adjacent vertebral bodies, is composed of both articular and subchondral cartilage and synovial fluid encased by an inner synovial membrane and outer joint capsule. The facet joint, synonymous with zygapophyseal joint (ZJT), has long been recognized as a source of LBP. Hirsch et al.1 in 1963 confirmed this theory by evaluating distribution of pain elicited by intra-articular injection of hypertonic saline. In healthy individuals, the intervertebral disc bears most of the load in the spine, but as the disc degenerates, increasing loads are shifted onto the facet joints and the joint capsule,2 leading to an inflammatory response in the ZJT.3

Pain signals emanating from the inflamed ZJT are relayed through medial branches of the dorsal rami. Denervation of these branches with radiofrequency ablation is commonly performed with varying degrees of success; however, a more direct approach to alleviate ZJT pain is to inject corticosteroids or other agents into the joint in an attempt to reduce inflammation.3

Corticosteroids decrease swelling and tenderness in addition to increasing viscosity through reductions in inflammatory responses and an increase in hyaluronic acid (HA) concentration in the joint.4 Reports of efficacy of corticosteroid injections vary widely.5,6 Despite the lack of rigorous evidence-based support, corticosteroid injections continue to be widely used in the treatment of lumbar facet syndrome.

Hyaluronic acids used for viscosupplementation are theorized to treat arthropathy by reproducing the viscoelasticity of synovial fluid, preventing and possibly reversing the destruction of cartilage and bone by reducing chondrocyte apoptosis.7 Hyaluronic acid facilitates the synthesis of matrix proteins, keratin sulfate, and proteoglycans and acts on the inflammatory pathway.8 Additional explanations for the mechanism of pain reduction include possible elevation of the depolarization threshold in surrounding nerves and reduction of substance P.8,9 In both inflamed and noninflamed joints, Synvisc was found to reduce the impulse discharge of articular nerves by an average of 50%.9 Although the specific contribution of each of these effects on overall clinical outcome is unclear, most clinical trials of HA demonstrate improvement in pain, function, or both. Although Food and Drug Administration approved only for treating knee osteoarthritis,10 studies have suggested HA viscosupplementation to be effective in various joints.11–15

Low back pain may be a result of facetogenic, myofascial, discogenic, or neurogenic etiologies. Consequently, efficacy of facet joint interventions addressing the pain source can be influenced by the diagnostic certainty of facetogenic pain from the ZJT as the source of pain. Lumbar facet syndrome can be diagnosed clinically or procedurally. The clinical diagnosis is based on a constellation of symptoms and signs including the following: paravertebral lumbar pain, absence of radiation of pain below the knee, LBP exacerbated by lumbar extension and rotational movement, and a normal neurological examination.16 Although no single element from the previous set of symptoms and signs is adequate to diagnose lumbar facet syndrome, a clinical diagnosis can be made when multiple elements of the aforementioned characteristics are present.17–20 Imaging, such as x-ray, computed tomography scan, or magnetic resonance imaging, may further aid in the diagnosis, but normal imaging does not preclude lumbar facet joint syndrome.21 The procedure-based diagnosis of lumbar facet syndrome is made by performing medial branch blocks using local anesthetic of different duration of effect and evaluating for concordant pain response. Pain reduction of duration consistent with the anesthetic used is considered confirmation that lumbar pain is in facetogenic in etiology. Although procedure-based diagnosis of facet pain is recommended to provide objective support for the diagnosis of facetogenic pain, especially in research trials, it is not always considered a standard of care in all practice settings because of practical considerations including time, access, cost, and inconvenience to patients.

Although a few previous studies have investigated the use of HA in lumbar facet joints, results have been variable.7,22,23 The reasons for this variability include different diagnostic criteria used, lack of control group,22 different HA products studied, and different outcome measures evaluated. Therefore, to better evaluate the comparative efficacy of HA in treating facetogenic LBP, we designed a pragmatic study to compare the efficacy of injected HA versus corticosteroids in the treatment of clinically diagnosed lumbar facetogenic pain.

Back to Top | Article Outline


Study Objectives

The study objective was to compare the effectiveness of sodium hyaluronate (HA) injections with triamcinolone injections in improving pain and function in patients with clinically diagnosed lumbar facet joint arthropathy causing LBP.

Back to Top | Article Outline

Study Design

This was a prospective, double-blind, randomized, controlled trial (RCT), comparing the effectiveness of intra-articular ZJT injections of HA with triamcinolone for pain relief, functional improvement, patient satisfaction, and clinical safety for ZJT-mediated chronic low back pain. This clinical trial was conducted in 2012–2014, at which point, there were no requirements for all clinical trials to be registered. Therefore, this RCT does not have a registration with the clinical trials registry.

Back to Top | Article Outline

Study Population

Subjects were recruited from patients with chronic low back pain who were referred to a Veterans Affairs hospital-based, spine clinic.

Back to Top | Article Outline

Study Criteria

Study subjects had to meet the following criteria: for inclusion criteria, patients with axial LBP (no radiation past knee), physical examination demonstrating paramedian lumbar tenderness without midline tenderness, pain reproduced with hyperextension and lateral rotation (facet loading), and radiographic evidence of ZJT degeneration (criteria consistent with the study by Helbig and Lee16 on the probability of pain relief with ZJT injection); and for exclusion criteria, patients with a history or presence of alcohol or drug abuse or unstable psychosis, subjects who are unable to understand informed consent or have a high probability of noncompliance, intra-articular treatment of any joint with corticosteroids or glycosaminoglycans within 3 mos, or with a HA-based product within 6 mos before injection, and patients with previous lumbar surgery or radiculopathy (pain radiating past knee), known allergy or contraindication to administered agents, critical skin conditions at injection site or other confounding medical conditions such as nonosteoarthritic joint disease, immune deficiencies, malignant disease, use of anticoagulants, or joint infection.

Back to Top | Article Outline

Power Analysis

A priori power analysis was performed on pain reduction, the main outcome measure. To obtain a mean intergroup difference on the visual analog scale (VAS) of 30 (20) mm at an α error of 0.05, for a power of more than 80%, the sample size would need to be 12 patients per group for a total of 24 patients. Accounting for six dropouts, we estimated enrolling 30 patients in our study.

Back to Top | Article Outline

Institutional Review Board Approval

This study was reviewed and approved by our facility's institutional review board. All study participants provided written informed consent. This study conforms to all CONSORT guidelines and reports the required information accordingly (see Checklist, Supplemental Digital Content,

Back to Top | Article Outline


After obtaining informed consent, each potential study subject underwent a history and physical examination in the spine clinic to determine whether they met all study criteria. Baseline data were then obtained from all eligible subjects including demographic information, height, weight, and duration of pain. Baseline data on pain (100-mm VAS) and disability (Pain Disability Questionnaire [PDQ]) were then obtained. Randomization was performed by a research pharmacist using computer-generated random numbers who then provided the blinded medication in a syringe wrapped in opaque labeling. A block randomization of alternating assignments was used to determine to which treatment group each subject was assigned. The blocking factor was unknown to the investigators and research coordinator. Patients were randomized to either the Kenalog group (KA) or the Synvisc-One group (HA). On the scheduled date of the procedure, bilateral L3-4, L4-5, and L5-S1 ZJTs were fluoroscopically injected with the study medication using standard injection technique, and contrast confirmation, by an independent, highly experienced interventional radiologist who was not involved in the preprocedure or postprocedure assessments. Interventional radiologist had performed hundreds of facet injections before the study. All injections were performed by the same radiologist. Patients in the KA group received 1 ml of triamcinolone (10 mg/ml of Kenalog) in each joint, and patients in the HA group received 1 ml of Synvisc-One (8 mg of Hylan GF-20 per vial) in each joint, for a total volume of 6 ml injected per patient. Study subjects followed up at 1, 3, and 6 mos after the procedure when outcome data including pain (VAS) and functional status (PDQ) were collected. In addition, an overall patient satisfaction score (global percent improvement) was obtained at the 6-mo visit. Once enrolled, subjects were asked to make no changes to their treatment, including physical therapy and medications they were taking before beginning the study. If they were previously receiving physical therapy, they were allowed to complete it.

Back to Top | Article Outline

Outcome Measures and Primary Data Analysis

Back pain from the ZJT can be quantified in a number of ways including a visual analog scale and a PDQ. The PDQ allows patients to rate their ability to complete specific tasks on a scale ranging from having no problems with the task to being unable to complete it at all. Primary outcome measures were pain assessed on a 100-mm VAS and functional disability assessed on the PDQ. These outcome measurements were assessed at baseline, 1, 3, and 6 mos after the procedure. Secondary outcomes measure was patient satisfaction obtained at 6 mos after the procedure.

Outcome measures in the two groups were compared using the t test (Mann–Whitney U test for nonnormally distributed data). Change in outcome measures over time in each study group was assessed using the paired t test (Wilcoxon signed rank test for nonnormally distributed data).

Back to Top | Article Outline

Secondary Data Analyses

A general linear model multivariate analysis of variance with four multivariate tests (Pillai's Trace, Wilks' Lambda, Hotelling's Trace, and Roy's Largest Root) was performed to determine the significance of baseline variables and group assignment in determining eventual outcome. The cohort as a whole (regardless of group assignment) was also analyzed to determine the total change in outcomes for the 6-mo period of the study.

Back to Top | Article Outline


A total of 30 subjects (3 females, 27 males), with a mean age of 58.33 yrs, mean weight of 207 lb, and a mean height of 69 inches, were enrolled in the study (Table 1). At baseline, enrolled subjects had a mean duration of LBP of 182 mos (>15 yrs), a mean pain score of 67 mm on the VAS (KA = 65 mm, HA = 51 mm), and a mean PDQ score (higher scores = more disability, range 0–160: of 101 [KA = 100, HA = 102]) (Table 1). Mean ages for each group were the following: KA, 58 yrs; HA, 59 yrs. Mean body mass index was 34.67 in female subjects and 30.29 in male subjects. The KA group was 80% male with a mean age of 57.7 yrs of age, duration of LBP of 172.2 mos, and mean weight of 210.5 lb. The HA group was 100% male with a mean age of 58.9 yrs of age, duration of LBP of 186 mos, and a mean weight of 203.6 lb. There were no significant differences in baseline variables with the exception of sex. The high percentage of male participants can be explained by the study being performed at a Veterans Affairs (VA) medical center, which has a predominately male population. All subjects were veterans, many with a history of service-connected injury that resulted in their first onset of LBP. This could explain why many patients reported 40+ yrs of LBP even though they likely developed lumbar facet arthropathy much later.



Five subjects dropped out of the study (three from the HA group, two from the KA group). Four subjects dropped out because of noncompliance with study visits, and one changed his mind regarding study participation after enrolling. One subject was withdrawn from the study after intervention, because he developed symptoms of neurogenic claudication and was no longer eligible for study participation. The final study included 13 subjects in the KA group and 12 in the HA group.

All subjects underwent their randomly assigned intervention (15 Synvisc injections and 15 Kenalog injections) without any complications or adverse events except for the following variances. In four subjects, one facet joint each could not be accessed/injected (three in the KA group, one in the HA group). Five subjects experienced transient or short-term pain increase (three in KA and two in HA). The procedure triggered onset of rheumatoid arthritis in one subject because of a new allergic reaction to Synvisc (was excluded from the study). One subject (HA) died because of unrelated reasons after completing the study.

Back to Top | Article Outline

Results of Primary Data Analyses

The primary outcome measures of pain (VAS) and functional disability (PDQ) in the two groups, compared using the t test (Mann–Whitney U test for nonnormally distributed data), revealed no statistically significant group differences (Table 2). The VAS scores were (KA and HA, respectively) the following: baseline: 70 (15)/74 (10) (P = 0.418); 1 mo: 58 (29)/45 (25) (P = 0.25); 3 mos: 58 (29)/56 (25) (P = 0.85); and 6 mos: 59 (28)/63 (24) (P = 0.71). The PDQ scores were (KA and HA, respectively) the following: baseline: 100 (23)/102 (28) (P = 0.85); 1 mo: 77 (30)/74 (34) (P = 0.80); 3 mos: 87 (26)/74 (36) (P = 0.32); and 6 mos: 96 (25)/79 (25) (P = 0.62). Overall percent improvement (global satisfaction) at 6 mos was not significantly different between the groups: KA (51 [35]) and HA (42 [33], P = 0.51). Values are reported as means and standard deviation, if they passed the test for normality. If values did not pass the test for normality, they were reported as median and quartile values.



Within-group analysis of the changes in primary outcome measures over time revealed the following findings. Analyzing the pain score (VAS) changes over time, we found that the only statistically significant change noted was in the HA group in the pain score, which showed statistically significant improvement only at the 1-mo time point (70 [20] to 45 [25], P = 0.008) (Table 3). There were no statistically significant within-group differences in pain scores in the KA group, at any time point. Analyzing the functional disability scores (PDQ) over time, we found that patients in the KA group showed statistically significant improvement only at the 1-mo time point (100 [23] to 77 [30], P = 0.009) and patients in the HA group showed significant improvements from baseline (102 [28]) to the following three time points: 1 mo (74 [34], P = 0.002), 3 mos (74 [36], P = 0.037), and 6 mos (79, median = 52–99.5, P < 0.001) (Table 4).





Back to Top | Article Outline

Results of Secondary Data Analyses (Only Summary of Results Presented for Sake of Brevity)

Using a general linear model multivariate analysis of variance with four multivariate tests (Pillai's Trace, Wilks' Lambda, Hotelling's Trace, and Roy's Largest Root), we studied multiple baseline variables (pain duration, age, body mass index, baseline pain, baseline PDQ) and group assignment (KA or HA). This analysis revealed that group assignment (KA or HA) was not significant (P > 0.05) in determining outcome (VAS or PDQ).

However, the two covariates of significance were initial pain rating (VAS, P = 0.011) and initial PDQ score (PDQ, P = 0.036). Between-subjects effects were significant for the total change in VAS score (partial η2 = 0.362). Regression coefficient (adjusted R2) for this change in total VAS was 0.388.

We then looked at the cohort as a whole (regardless of group assignment) and analyzed the total change in outcomes for the 6-mo period of the study. We found (paired samples test) that there was a statistically significant mean change in PDQ of 18 (SE = 5.69) (99% confidence interval = 3.040–31.680, P < 0.004). Controlling for duration of LBP, the total VAS score also showed strong positive change, but total PDQ change did not improve much.

Back to Top | Article Outline


This comparative effectiveness study showed that in comparison with a commonly performed, clinical standard, treatment of corticosteroid injection of the lumbar ZJT, HA injection did not reveal any statistically significant difference in pain relief or functional improvement, either short term or long term. The outcomes were similar in patients in both HA and KA groups. However, within-group analysis revealed that patients in our study in both the treatment group (HA) and control group (KA) attained short-term functional improvement. In addition, patients in the HA group attained long-term functional improvement and short-term pain improvement as well.

Our study's results were similar to a study reported by Fuchs et al. in 2005,23 in which they compared intra-articular HA with triamcinolone injections. Much like our study, they found that reduction in pain and gain of function, as measured by multiple questionnaires, were evident in both treatment arms. Again, similar to our study, they found that the reduction in pain and gain of function in the HA group was equal to or better than the triamcinolone group and that the positive effects lasted longer in the sodium hyaluronate arm of the study.23 Although we used Synvisc, a higher molecular weight hyaluronan, instead of sodium hyaluronate, these results suggest that multiple hyaluronan products may be effective in treatment of lumbar facet joint arthropathy.

In contrast, the study by Cleary et al.22 did not find any significant improvement after HA injections, in pain or function (Oswestry Disability Index) at 6 wks in the 13 patients that they studied. Their study could have benefitted from a larger cohort and lacked randomization or a control group, making it difficult to determine the true effect of viscosupplementation in these patients. The authors themselves suggest that variability in dosing regimens may have led to the lack of improvement because there are no dosing guidelines in place.

DePalma et al.7 studied intra-articular Synvisc injections to treat LBP. Unlike our study, they used a successful diagnostic block with local anesthetic to identify facet-related LBP. Based on the VAS, the Oswestry Disability Index, SF-36, finger to floor distance, tolerance, and satisfaction, they found statistically significant improvement from baseline up to the 6-mo mark in many outcomes, but the effects were modest and most disappeared at 12 mos.7 It is possible that a larger sample size may have better born out the effects of Synvisc and suggests that future studies should also examine the efficacy of Synvisc at 1 yr.

Our study consisted of fewer female participants compared with the previously mentioned three studies, a common scenario in clinical research conducted at a VA medical center. The average onset of LBP in our study is earlier in our study population, which can be attributed to military service-connected injuries and is frequently seen in veterans. DePalma et al.7 injected up to 3 ml of Synvisc using three 1-ml injections into the same joint for multiple weeks. Cleary injected 2 ml of HA per facet. Fuchs injected 3 ml of HA using three 1-ml injections into the same joint for 3 wks. Given the size of most facet joints, it is common practice to inject a small amount of injectate when performing a facet injection. The theory of facetogenic pain was confirmed by Hirsch by injecting a large volume of fluid into the joint to induce pain.1 We chose to inject 1 ml of HA so as to not overfill the facet joint. Cleary et al.22 used a clinical diagnosis similar to ours to classify facetogenic pain, whereas others used diagnostic blocks to confirm the etiology of facetogenic pain. Diagnostic blocks are useful to more specifically identify facetogenic pain compared with clinical criteria, but because of practical considerations and to more closely mimic practice patterns in our facility, we did not perform diagnostic blocks in our study, unlike the previously cited studies.

An additional, common method of treatment of lower back pain due to ZJT arthropathy is radiofrequency ablation of the medial branches of the dorsal rami. Recent comparisons of pulsed radiofrequency ablation versus intra-articular corticosteroid injection of the lumbar facet joint showed reduction of lower back pain in both study groups with better pain relief at 2 wks and 1 mo in the corticosteroid cohort, with the statistically significant difference in improvement of pain disappearing by 3 mos.24 Both groups had reduction of pain at 6 mos in more than 45% of participants, but the study would have benefitted from a control group and from longer follow-up to examine the long-term effectiveness of these treatments.24 With our study suggesting equivalency in the treatment effects of ZJT corticosteroid injections and HA injections, it is worthy of speculation whether intra-articular HA would be equivalent to or better than radiofrequency ablation in patients with ZJT arthropathy-mediated pain.

Although HA is Food and Drug Administration approved only for treating osteoarthritis of the knee joint,10 studies have suggested HA viscosupplementation to be effective in several joints affected by arthropathy, including the hip,11,12 TMC joint of thumb,13 ankle,14 shoulder,15 and sacroiliac joint.25

Joint injections with HA is generally thought to be a relatively safe procedure with few adverse effects.10,26 Our study, though not designed primarily to study safety of treatment, did yield an interesting outcome of one participant in the HA group experiencing symptoms of rheumatoid arthritis for the first time, confirmed by testing. Per our review, it appears that this is the first case of its kind. The injection of corticosteroid, on the other hand, has significant possible adverse effects, in particular cortisol suppression, hyperglycemia, and osteoporosis, although with less risk than systemic corticosteroids.27

The variability of the results of the previously mentioned studies into the treatments of lumbar pain caused by facet arthropathy suggests that while treatment with HA may hold potential, further studies are needed to establish a solid evidence-based foundation on which to make a strong clinical recommendation. This falls in line with the results of our study of facet viscosupplementation with Synvisc.

Back to Top | Article Outline

Study Limitations

Our study fell short of the target power of more than 80%, translating to the possibility that there is greater than acceptable chance that there may have been a statistically significant difference between the treatment groups that our study was unable to discern. The small number of women included in our study made it difficult to analyze the effect of sex on outcomes. Our study benefitted from including patients that fit previously defined criteria for ZJT injection and double-blind randomization but could have benefitted from a larger cohort and longer follow-up to at least 12 mos.

Another limitation of the study was relying solely on clinical criteria to diagnose facetogenic pain. Research standards recommend diagnostic confirmation with dual comparative blocks. We concede the possibility that some of our study subjects, who all had symptoms and signs consistent with lumbar facet syndrome, may not have met the recommended research criteria of dual comparative block-confirmed facet pain. This discrepancy may limit the generalizability of our study's findings to “dual comparative blocks-confirmed” facet pain. However, using clinical criteria to diagnose facet pain makes this study more applicable to the widely used clinical standard of diagnosing lumbar facet syndrome.28

Within-group analysis, especially in the HA group, showed statistically significant short- and long-term functional improvement and short-term pain improvement. However, these findings must be interpreted with caution, because any outcomes measured in within-group analysis, which resembles an uncontrolled cohort, cannot be unequivocally attributed to the intervention used, such as Synvisc or triamcinolone (statistically significant short-term functional benefit was observed in the triamcinolone group). A third group, with sham injection serving as a control intervention, would have helped differentiate the effect of Synvisc and Kenalog from placebo effect and would have strengthened the case for attribution of outcome changes noted to the treatment (Kenalog vs. Synvisc) given. Concurrent treatments such as use of non-steroidal anti-inflammatory drugs and physical therapy, both of which are commonly employed in the treatment of chronic LBP, may account for some of the improvements seen in both groups in the study. Although participants were instructed not to make any changes to their concurrent treatments during their study participation, we cannot rule out a confounding effect from these treatments.

Secondary data analyses performed in this study revealed significant contributory effects of baseline variables such as duration of pain, intensity of pain, and level of functional disability, all of which were relatively high in our study population. Although these characteristics are largely reflective of the population treated in our facility, it may not be reflective of the patients in all facilities and practices. The service-connected disability compensation system used by the VA can provide secondary gain incentives to veterans to continue to report symptoms. Therefore, the generalizability of our study's findings may be limited to patient populations with similar baseline characteristics. On the other hand, secondary data analysis of the entire cohort also revealed that there was significant functional improvement in PDQ scores and pain scores (when duration of LBP was controlled). These findings reflect optimism that even in a population with chronic LBP and high baseline levels of pain and disability, significant and long-term clinical effects are attainable with ZJT injections of either HA or KA.

No intergroup differences were observed when comparing overall satisfaction at 6 mos. This observation along with the main results of this study that there were no statistically significant intergroup differences between Synvisc and triamcinolone groups strongly suggests that HA is at least noninferior to corticosteroid in the management of lumbar ZJT pain. However, this also suggests that further investigation of HA in the treatment of lumbar pain secondary to facet arthropathy with randomized, double-blinded, sham-controlled studies is warranted.

Back to Top | Article Outline


No statistical intergroup differences in pain relief or functional improvement were found when comparing triamcinolone and hyaluronate injections in treating patients with chronic LBP likely caused by or having a component of lumbar ZJT arthropathy. Within-group comparisons of hyaluronate injections revealed statistically significant short- and long-term functional improvement and short-term pain improvement compared with baseline levels. Within-group comparison of triamcinolone injections only revealed statistically significant short-term functional benefit and no significant short- or long-term pain improvement compared with baseline levels.

Back to Top | Article Outline


The authors thank Dr. Ravid Avraham, key member of the research team, Ms. Desirae Ransom, research coordinator, and Mr. David Rezaei, research pharmacist, whose efforts and contributions were responsible for the successful completion of this research project.

Back to Top | Article Outline


1. Hirsch C, Ingelmark BE, Miller M: The anatomical basis for low back pain. Studies on the presence of sensory nerve endings in ligamentous, capsular and intervertebral disc structures in the human lumbar spine. Acta Orthop Scand 1963;33:1–17
2. Gellhorn AC, Katz JN, Suri P: Osteoarthritis of the spine: the facet joints. Nat Rev Rheumatol 2013;9:216–24
3. Binder DS, Nampiaparampil DE: The provocative lumbar facet joint. Curr Rev Musculoskelet Med 2009;2:15–24
4. Stephens MB, Beutler AI, O'Connor FG: Musculoskeletal injections: a review of the evidence. Am Fam Physician 2008;78:971–6
5. Carette S, Marcoux S, Truchon R, et al: A controlled trial of corticosteroid injections into facet joints for chronic low back pain. N Engl J Med 1991;325:1002–7
6. Marks RC, Houston T, Thulbourne T: Facet joint injection and facet nerve block: a randomised comparison in 86 patients with chronic low back pain. Pain 1992;49:325–8
7. DePalma MJ, Ketchum JM, Queler ED, et al: Prospective pilot study of painful lumbar facet joint arthropathy after intra-articular injection of hylan G-F 20. PM R 2009;1:908–15
8. Moreland LW: Intra-articular hyaluronan (hyaluronic acid) and hylans for the treatment of osteoarthritis: mechanisms of action. Arthritis Res Ther 2003;5:54–67
9. Gomis A, Pawlak M, Balazs EA, et al: Effects of different molecular weight elastoviscous hyaluronan solutions on articular nociceptive afferents. Arthritis Rheum 2004;50:314–26
10. Sun SF, Hsu CW, Lin HS, et al: Comparison of single intra-articular injection of novel hyaluronan (HYA-JOINT Plus) with Synvisc-one for knee osteoarthritis: a randomized, controlled, double-blind trial of efficacy and safety. J Bone Joint Surg Am 2017;99:462–71
11. Conrozier T, Vignon E: Is there evidence to support the inclusion of viscosupplementation in the treatment paradigm for patients with hip osteoarthritis? Clin Exp Rheumatol 2005;23:711–6
12. Migliore A, Massafra U, Frediani B, et al: HyalOne® in the treatment of symptomatic hip OA - data from the ANTIAGE register: seven years of observation. Eur Rev Med Pharmacol Sci 2017;21:1635–44
13. Velasco E, Ribera MV, Pi J: Single-arm open-label study of Durolane (NASHA nonanimal hyaluronic acid) for the treatment of osteoarthritis of the thumb. Open Access Rheumatol 2017;9:61–6
14. Murphy EP, Curtin M, McGoldrick NP, et al: Prospective evaluation of intra-articular sodium hyaluronate injection in the ankle. J Foot Ankle Surg 2017;56:327–31
15. Legre-Boyer V: Viscosupplementation: techniques, indications, results. OTSR 2015;101(1 Suppl):S101–8
16. Helbig T, Lee CK: The lumbar facet syndrome. Spine (Phila Pa 1976) 1988;13:61–4
17. Mooney V, Robertson J: The facet syndrome. Clin Orthop Relat Res 1976:149–56
18. Eisenstein SM, Parry CR: The lumbar facet arthrosis syndrome. Clinical presentation and articular surface changes. J Bone Joint Surg Br 1987;69:3–7
19. Bogduk N: Lumbar dorsal ramus syndrome. Med J Aust 1980;2:537–41
20. Jackson RP, Jacobs RR, Montesano PX: 1988 Volvo award in clinical sciences. Facet joint injection in low-back pain. A prospective statistical study. Spine 1988;13:966–71
21. Schwarzer AC, Wang SC, O'Driscoll D, et al: The ability of computed tomography to identify a painful zygapophysial joint in patients with chronic low back pain. Spine 1995;20:907–12
22. Cleary M, Keating C, Poynton AR: Viscosupplementation in lumbar facet joint arthropathy: a pilot study. J Spinal Disord Tech 2008;21:29–32
23. Fuchs S, Erbe T, Fischer HL, et al: Intraarticular hyaluronic acid versus glucocorticoid injections for nonradicular pain in the lumbar spine. J Vasc Interv Radiol 2005;16:1493–8
24. Lakemeier S, Lind M, Schultz W, et al: A comparison of intraarticular lumbar facet joint steroid injections and lumbar facet joint radiofrequency denervation in the treatment of low back pain: a randomized, controlled, double-blind trial. Anesth Analg 2013;117:228–35
25. Srejic U, Calvillo O, Kabakibou K: Viscosupplementation: a new concept in the treatment of sacroiliac joint syndrome: a preliminary report of four cases. Reg Anesth Pain Med 1999;24:84–8
26. O'Hanlon CE, Newberry SJ, Booth M, et al: Hyaluronic acid injection therapy for osteoarthritis of the knee: concordant efficacy and conflicting serious adverse events in two systematic reviews. Syst Rev 2016;5:186
27. Habib GS: Systemic effects of intra-articular corticosteroids. Clin Rheumatol 2009;28:749–56
28. DePalma MJ, Ketchum JM, Saullo TR: Etiology of chronic low back pain in patients having undergone lumbar fusion. Pain Med 2011;12:732–9
29. Saravanakumar K, Harvey A: Lumbar zygapophyseal (facet) joint pain. Rev Pain 2008;2:8–13
30. Manchikanti L, Hirsch JA, Kaye AD, et al: Cervical zygapophysial (facet) joint pain: effectiveness of interventional management strategies. Postgrad Med 2016;128:54–68

Zygapophyseal Joint; Hyaluronic Acid; Triamcinolone; Patient Outcome Assessment

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.