Management of lumbar zygapophysial (facet) joint pain

Table 1 Modified grading of qualitative evidence with best evidence synthesis for diagnostic accuracy and therapeutic interventions

Level I	Evidence obtained from multiple relevant high quality randomized controlled trials
	or
	Evidence obtained from multiple high quality diagnostic accuracy studies
Level II	Evidence obtained from at least one relevant high quality randomized controlled trial or multiple relevant moderate or low quality randomized controlled trials
	or
	Evidence obtained from at least one high quality diagnostic accuracy study or multiple moderate or low quality diagnostic accuracy studies
Level III	Evidence obtained from at least one relevant moderate or low quality randomized controlled trial study
	or
	Evidence obtained from at least one relevant high quality non-randomized trial or observational study with multiple moderate or low quality observational studies
	or
	Evidence obtained from at least one moderate quality diagnostic accuracy study in addition to low quality studies
Level IV	Evidence obtained from multiple moderate or low quality relevant observational studies
	or
	Evidence obtained from multiple relevant low quality diagnostic accuracy studies
Level V	Opinion or consensus of large group of clinicians and/or scientists.

Source: Manchikanti et al[82].

Table 2 Quality appraisal of the diagnostic accuracy of lumbar facet joint nerve block diagnostic studies

	Manchikanti et al[23]	Pang et al[25]	Schwarzer et al[22]	Schwarzer et al[84]	Manchikanti et al[86]	Manchikanti et al[79]	Manchikanti et al[85]	Manchikanti et al[93]	Manchikanti et al[94]	Manchikanti et al[88]	Manchikanti et al[82]	Manchikanti et al[90]	Manchukonda et al[91]	Manchikanti et al[92]	Manchikanti et al[95]	DePalma et al[24]
(1) Was the test evaluated in a spectrum of subjects representative of patients who would normally receive the test in clinical practice?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(2) Was the test performed by examiners representative of those who would normally perform the test in practice?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(3) Were raters blinded to the reference standard for the target disorder being evaluated?	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N
(4) Were raters blinded to the findings of other raters during the study?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(5) Were raters blinded to their own prior outcomes of the test under evaluation?	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N
(6) Were raters blinded to clinical information that may have influenced the test outcome?	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N	N
(7) Were raters blinded to additional cues, not intended to form part of the diagnostic test procedure?	Y	N	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(8) Was the order in which raters examined subjects varied?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(9) Were appropriate statistical measures of agreement used?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(10) Was the application and interpretation of the test appropriate?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(11) Was the time interval between measurements suitable in relation to the stability of the variable being measured?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
(12) If there were dropouts from the study, was this less than 20% of the sample	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Total	9/12	8/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12	9/12

Y: Yes; N: No. Source: Lucas et al[80].

Table 3 Methodological quality assessment of randomized trials of lumbar facet joint interventions utilizing Cochrane review criteria

	Manchikanti et al[102]	Carette et al[103]	Fuchs et al[104]	Nath et al[105]	van Wijk et al[97]	van Kleef et al[99]	Tekin et al[96]	Civelek et al[108]	Dobrogowski et al[98]	Cohen et al[109]	Ribeiro et al[107]	Moon et al[100]	Lakemeier et al[101]	Yun et al[106]
Randomization adequate	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Concealed treatment allocation	Y	Y	N	Y	Y	Y	Y	Y	U	N	Y	Y	Y	Y
Patient blinded	Y	Y	Y	Y	Y	Y	Y	N	Y	N	Y	Y	Y	N
Care provider blinded	Y	Y	N	Y	Y	Y	Y	N	Y	U	N	Y	N	N
Outcome assessor blinded	N	Y	Y	Y	Y	Y	Y	U	U	U	N	Y	N	N
Drop-out rate described	Y	Y	N	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
All randomized participants analyzed in the group	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y
Reports of the study free of suggestion of selective outcome reporting	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Groups similar at baseline regarding most important prognostic indicators	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Co-intervention avoided or similar in all groups	Y	N	N	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Compliance acceptable in all groups	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	N	N	Y
Time of outcome assessment in all groups similar	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y
Score	11/12	11/12	8/12	12/12	12/12	12/12	12/12	9/12	10/12	8/12	10/12	9/12	9/12	9/12

Y: Yes; N: No; U: Unclear. Source: Furlan et al[78].

Table 4 Methodological quality assessment of randomized trials utilizing Interventional Pain Management techniques - Quality Appraisal of Reliability and Risk of Bias Assessment criteria

	Manchikanti et al[102]	Carette et al[103]	Fuchs et al[104]	Nath et al[105]	van Wijk et al[97]	van Kleef et al[99]	Tekin et al[96]	Civelek et al[108]	Dobrogowski et al[98]	Cohen et al[109]	Ribeiro et al[107]	Moon et al[100]	Lakemeier et al[101]	Yun et al[106]
I Trial design and guidance reporting
(1) Consort or spirit	3	3	3	3	2	2	2	2	2	3	2	2	2	2
II Design factors
(2) Type and design of trial	2	3	2	3	3	3	3	2	2	2	2	2	2	2
(3) Setting/physician	2	1	1	3	3	3	2	2	2	2	2	2	1	1
(4) Imaging	3	3	2	3	3	3	3	3	2	3	3	3	3	3
(5) Sample size	3	2	2	1	1	1	1	1	1	3	2	2	2	2
(6) Statistical methodology	1	1	1	1	1	1	1	1	1	1	1	1	1	1
III Patient factors
(7) Inclusiveness of Population
For facet joint interventions:	2	1	0	2	1	1	1	0	1	1	0	2	1	0
(8) Duration of pain	2	2	1	2	2	3	2	0	2	0	0	2	2	0
(9) Previous treatments	2	0	0	0	0	0	1	2	0	0	0	2	2	1
(10) Duration of follow-up with appropriate interventions	3	2	1	2	2	2	2	2	1	0	1	1	2	1
IV Outcomes
(11) Outcomes assessment criteria for significant improvement	4	4	2	4	4	2	2	2	2	0	2	2	2	1
(12) Analysis of all randomized participants in the groups	2	2	2	2	2	2	2	2	2	2	2	0	2	2
(13) Description of drop out rate	2	1	2	2	2	2	2	2	2	2	2	2	2	2
(14) Similarity of groups at baseline for important prognostic indicators	2	2	2	2	2	2	2	2	2	2	2	2	2	2
(15) Role of co-interventions	1	0	0	1	1	1	1	1	1	1	1	1	1	1
V Randomization
(16) Method of Randomization	2	2	2	2	2	2	2	2	2	2	2	2	2	2
VI Allocation concealment
(17) Concealed treatment allocation	2	2	1	2	2	2	2	2	2	0	2	2	2	2
VII Blinding
(18) Patient blinding	1	1	0	1	1	1	1	0	1	0	1	1	1	0
(19) Care provider blinding	1	1	0	1	1	1	1	0	1	0	0	1	0	0
(20) Outcome assessor blinding	0	1	0	0	1	0	0	0	0	0	0	1	0	0
VIII Conflicts of interest
(21) Funding and sponsorship	2	3	1	2	0	3	2	0	0	2	2	2	2	1
(22) Conflicts of interest	3	3	1	3	0	3	2	0	0	2	3	3	3	0
Total	45	40	26	42	36	40	37	28	29	28	32	32	38	37

Source: Manchikanti et al[79].

Table 5 Characteristics of studies assessing the accuracy of diagnostic facet joint injections and nerve blocks in the lumbar spine

Study/methods	Participants	Intervention(s)	Outcome measures	Comments	Results
Methodological quality scoring					Prevalence with 95%CI and criterion standard	False-positive rate with 95%CI
Pang et al[25], 1998	100 consecutive adult patients with chronic low back pain with undetermined etiology were evaluated with spinal mapping	Single block was performed by injecting 2% lidocaine into facet joints	Verbal analog scale	This is the first study evaluating application of diagnostic blocks in the diagnosis of intractable low back pain of undetermined etiology with facet joint disease in potentially 48% of patients with a single block	Single block 90% pain relief	NA
Prospective, single block			Pain mapping		Only facet joint pain = 24%
8/12			90% pain relief		Lumbar nerve root and facet disease = 24%
					Total = 48%
Schwarzer et al[22], 1994	176 consecutive patients with chronic low back pain after some type of injury	Zygapophysial joint nerve blocks or intraarticular injections were performed with either 2% lignocaine or 0.5% bupivacaine	At least 50% pain relief concordant with the duration of local anesthetic injected	First study of evaluation of controlled prevalence and false-positive rates	50% pain relief	38% (95%CI: 30%-46%)
Prospective, controlled diagnostic blocks					15% (95%CI: 10%-20%)
9/12
Schwarzer et al[84], 1995	63 patients with low back pain lasting for longer than 3 mo underwent computed tomography and blocks of the zygapophysial joints	Patients underwent a placebo injection followed by intraarticular zygapophysial joint injections with 1.5 mL of 0.5% bupivacaine	At least 50% reduction in pain maintained for minimum of 3 h	This study shows that computed tomography has no place in the diagnosis of lumbar zygapophysial joint pain, with an impure placebo design	50% pain relief	NA
Randomized, impure placebo, controlled diagnostic blocks					40% (95%CI: 27%-53%)
9/12
Manchikanti [95], 2010	491 patients with chronic low back pain undergoing evaluation for facet joint pain with 80% pain relief and 181 patients with 50% pain relief	Controlled diagnostic blocks of lumbar facet joint nerves with 1% preservative-free lidocaine and 0.25% preservative-free bupivacaine 1 mL	At least 80% pain relief with the ability to perform previously painful movements	Higher prevalence than with 50% pain relief, but still higher than double block algorithmic approach	50% pain relief	50% pain relief
Retrospective, controlled diagnostic blocks					61% (95%CI: 53%-81%)	17% (95%CI: 10%-24%)
9/12					80% pain relief	80% pain relief
					31% (95%CI: 26%-35%)	42% (95%CI: 35%-50%)
Manchikanti et al[85], 2000	200 consecutive patients with chronic low back pain were evaluated	Controlled diagnostic blocks with 1% lidocaine and 0.25% bupivacaine were injected over facet joint nerves with 0.4 to 0.6 mL	75% pain relief with ability to perform previously painful movements	The study showed that the clinical picture failed to diagnose facet joint pain	75% pain relief	37% (95%CI: 32%-42%)
Prospective, controlled diagnostic blocks					42% (95%CI: 35%-42%)
9/12
DePalma et al[24], 2011	A total of 156 patients with chronic low back pain were assessed for the source of chronic low back pain including discogenic pain, facet joint pain, and sacroiliac joint pain	Dual controlled diagnostic blocks with 1% lidocaine for the first block with 0.5% bupivacaine for the second	Concordant relief with 2 h for lidocaine and 8 h for bupivacaine with 75% pain relief as the criterion standard	This is the third study evaluating various structures implicated in the cause of low back pain with controlled diagnostic blocks[23,25]	75% pain relief	NA
Retrospective, controlled diagnostic blocks					31% (24%-38%)
9/12
Manchikanti et al[93], 2001	Prevalence study in 100 patients with 50 patients below age of 65 and 50 patients aged 65 or over was assessed	Controlled diagnostic blocks	75% pain relief with ability to perform previously painful movements was utilized as the criterion standard	This study showed higher prevalence of facet joint pain in the elderly compared to the younger age group in contrast to the latest study by Manchikanti et al[117] which showed no differences	75% pain relief	< 65 yr = 26% (95%CI: 11%-40%)
Prospective, controlled diagnostic blocks					< 65 yr = 30% (95%CI: 17%-43%)	> 65 yr = 33% (95%CI: 14%-35%)
					> 65 yr = 52% (95%CI: 38%-66%)
9/12
Manchikanti et al[94], 2001	100 patients with low back pain were evaluated. Patients were divided into 2 groups, group I was normal weight and group II was obese	Facet joints were investigated with diagnostic blocks using lidocaine 1% initially followed by bupivacaine 0.25%, at least 2 wk apart	A definite response was defined as relief of at least 75% in the symptomatic area	This study showed no significant difference between obese and non-obese individuals either with prevalence or false-positive rate of diagnostic blocks in chronic facet joint pain	75% pain relief	Non-obese individuals = 44% (95%CI: 26%-61%)
Prospective, controlled diagnostic blocks
9/12					Prevalence:	Obese individuals = 33% (95%CI: 16%-51%)
					Non-obese individuals = 36% (95%CI: 22%-50%)
					Obese individuals = 40% (95%CI: 26%-54%)
Manchikanti et al[23], 2001	120 patients were evaluated with chief complaint of chronic low back pain to evaluate relative contributions of various structures in chronic low back pain. All 120 patients underwent facet joint nerve blocks	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	80% pain relief with ability to perform previously painful movements	This study evaluated all the patients with low back pain, even with suspected discogenic pain	80% pain relief	47% (95%CI: 35%-59%)
Prospective, controlled diagnostic blocks					40% (95%CI: 31%-49%)
9/12
Manchikanti et al[86], 1999	120 patients with chronic low back pain after failure of conservative management were evaluated	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	Concordant pain relief with 75% or greater criterion standard with ability to perform previously painful movements	This was the first study performed in the United States in the heterogenous population as previous studies were performed in only post-injury patients	75% pain relief	41% (95%CI: 29%-53%)
Prospective, controlled diagnostic blocks					45% (95%CI: 36%-54%)
9/12
Manchikanti et al[79], 2014	180 consecutive patients with chronic low back pain were evaluated after having failed conservative management	Controlled diagnostic blocks with 1% lidocaine and 0.25% bupivacaine with or without Sarapin and/or steroids	75% pain relief with ability to perform previously painful movements	This study showed no significant difference if the steroids were used or not	75% pain relief	25% (95%CI: 21%-39%)
Prospective, controlled diagnostic blocks					36% (95%CI: 29%-43%)
9/12
Manchikanti et al[88], 2003	At total of 300 patients with chronic low back pain were evaluated to assess the difference based on involvement of single or multiple spinal regions	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	80% pain relief with ability to perform previously painful movements	This study shows a higher prevalence when multiple regions are involved	80% pain relief	Single region:
Prospective, controlled diagnostic blocks					Single region:	17% (95%CI: 10%-24%)
					21% (95%CI: 14%-27%)	Multiple regions:
9/12					Multiple regions:	27% (95%CI: 18%-36%)
					41% (95%CI: 33%-49%)
Manchikanti et al[82], 2014	120 consecutive patients with chronic low back pain and neck pain were evaluated to assess involvement of facet joints as causative factors	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	80% pain relief with ability to perform previously painful movements	The results are similar to involvement of multiple regions with a prevalence of 40% as illustrated in another study	80% pain relief	30% (95%CI: 20%-40%)
Prospective, controlled diagnostic blocks					40% (95%CI: 31%-49%)
9/12
Manchikanti et al[90], 2004	500 consecutive patients with chronic, non-specific spinal pain were evaluated of which 397 patients suffered with chronic low back pain	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	80% pain relief with ability to perform previously painful movements	Largest study performed involving all regions of the spine	80% pain relief	27% (95%CI: 22%-32%)
Prospective, controlled diagnostic blocks					31% (95%CI: 27%-36%)
9/12
Manchukonda et al[91], 2007	500 consecutive patients with chronic spinal pain were evaluated of which 303 patients were evaluated for chronic low back pain	Controlled diagnostic blocks with 1% lidocaine followed by 0.25% bupivacaine	80% pain relief with ability to perform previously painful movements	Second largest study performed involving all regions of the spine	80% pain relief	45% (95%CI: 36%-53%)
Retrospective, controlled diagnostic blocks					27% (95%CI: 22%-33%)
9/12
Manchikanti et al[92], 2007	A total of 117 consecutive patients with chronic non-specific low back pain were evaluated, after lumbar surgical interventions, with postsurgery syndrome and continued axial low back pain	Controlled, comparative, local anesthetic blocks with 1% lidocaine and 0.25% bupivacaine	80% relief as the criterion standard with ability to perform previously painful movements	Lower prevalence in postsurgery patients	80% pain relief	49% (95%CI: 39%-59%)
Prospective, controlled diagnostic blocks					16% (95%CI: 9%-23%)
9/12

NA: Not applicable.

Table 6 Study characteristics of randomized controlled trials of lumbar radiofrequency neurotomy, facet joint nerve blocks, and intraarticular injections

Study/study characteristic/methodological quality Scoring	No. of patients and selection criteria	Control	Interventions	Outcome measures	Time of measurement	Reported results	Strengths	Weaknesses	Conclusion/comments
Radiofrequency neurotomy
Civelek et al[108], 2012	100 patients with chronic low back pain with failed conservative therapy and strict selection criteria; however, without diagnostic blocks	Facet joint nerve block with local anesthetic and steroids in 50 patients	Conventional radiofrequency neurotomy at 80 °C for 120 s in combination with high dose local anesthetic and steroids, in 50 patients	Visual Numeric Pain Scale, North American Spine Society patient satisfaction questionnaire, Euro-Qol in 5 dimensions and ≥ 50% relief	1, 6 and 12 mo	At one year, 90% of patients in the radiofrequency group and 69% of the patients in the facet joint nerve block group showed significant improvement compared to 92% and 75% at 6-mo follow-up	Randomized relatively large number of patients with 50 in each group	No diagnostic blocks were performed. High dose steroids and local anesthetics were utilized in both groups	Efficacy was shown even without diagnostic blocks, both for facet joint nerve blocks and radiofrequency neurotomy
Randomized, active-control trial
Quality scores:
Cochrane = 9/12
IPM-QRB = 28/48
Cohen et al[109], 2010	151 chronic low back pain, 51 patients with no diagnostic block, 50 patients a single diagnostic block, 50 patients in double diagnostic block	Radiofrequency neurotomy in patients without diagnostic blocks	Conventional radiofrequency neurotomy at 80 °C for 90 s in all patients; however, in 2 groups with either a single block paradigm or a double block paradigm testing for positive results	Greater than 50% pain relief coupled with a positive global perceived effect persisting for 3 mo	3 mo	Denervation success rates in groups 0, 1 and 2 were 33%, 39% and 64%, respectively	Multicenter, randomized controlled trial with or without diagnostic blocks	Authors misinterpreted cost-effectiveness without consideration of many factors reported	Results showed efficacy when double diagnostic blocks were utilized
Randomized, double-blind, control trial
Quality scores:
Cochrane = 8/12
IPM-QRB = 28/48
Nath et al[105], 2008	40 patients with chronic low back pain for at least 2 yr with 80% relief of low back pain after controlled medial branch blocks. The patients were randomized into an active and a control group	Controlled sham lesion in 20 patients in the control group	The 20 patients in the active group received conventional lumbar facet joint radiofrequency neurolysis at 85 °C for 60 s. The 20 patients in the control group received sham treatment without radiofrequency neurolysis of the lumbar facet joints	Numeric Rating Scale, global functional improvement, reduced opioid intake, employment status	6 mo	Significant reduction not only in back and leg pain; functional improvement; opioid reduction; and employment status in the active group compared to the control group	Randomized, double-blind trial after the diagnosis of facet joint pain with triple diagnostic blocks	Short-term follow-up with small number of patients	Efficacy of radiofreqency neurotomy was shown compared to local anesthetic injection and sham lesioning
Randomized, double-blind, sham control trial
Quality scores:
Cochrane = 12/12
IPM-QRB = 42/48
Tekin et al[96], 2007	60 patients with chronic low back pain randomized into 3 groups with 20 patients in each group. Single diagnostic block of facet joint nerves with 0.3 mL of lidocaine 2% with 50% or greater relief	Sham control with local anesthetic injection	Either pulsed radiofrequency (42 °C for 4 min) or conventional radiofrequency neurotomy (80 °C for 90 s) in 20 patients in each group	Visual analog scale and Oswestry Disability Index	3, 6 and 12 mo	Visual analog scale and Oswestry Disability Index scores decreased in all groups from 3 procedural levels. Decrease in pain scores was maintained in the conventional radiofrequency group at 6 mo and one year. However, in pulsed radiofrequency group, the improvement was significant only at 6 mo, but not 1 yr	Randomized, double-blind, controlled trial comparing control, pulsed radiofrequency, and conventional radiofreqency neurotomy. Authors also utilized a parallel needle placement approach	Small sample size with a single block and 50% relief as inclusion criteria. Authors have not described the significant improvement percentages	Efficacy with conventional radiofreqency neurotomy up to one year, whereas efficacy with local anesthetic block with sham control radiofrequency neurotomy and pulsed radiofrequency neurotomy at 6 mo only
Randomized, active and sham, double-blind controlled trial
Quality scores:
Cochrane = 12/12
IPM-QRB = 37/48
van Wijk et al[97], 2005	81 patients with chronic low back pain were evaluated with radiofreqency neurotomy with 41 patients in the control group with at least 50% relief for 30 min with a single block with intraarticular injection of 0.5 mL lidocaine 2%	Sham lesion procedure after local anesthetic injection	40 patients received conventional radiofrequency lesioning at 80 °C for 60 s and 41 patients received sham lesioning	Pain relief, physical activities, analgesic intake, global perceived effect, short-form-36, quality of life measures	3 mo	Global perceived effect improved after radiofrequency facet joint denervation. The visual analog scale in both groups improved. The combined outcome measures showed no difference between radiofreqency facet joint denervation (27.5% vs 29.3% success rate)	Double-blind, sham control, randomized trial	Poor selection with a single diagnostic block of 50% pain reduction even though 17.5% of the patients tested positive. Further, authors described that the needle was positioned parallel; however, the radiographic figures illustrate the needle was being positioned perpendicularly rather than parallel to the nerve	Lack of efficacy with methodological deficiencies and a short-term follow-up
Randomized, double-blind, sham control trial
Quality scores:
Cochrane = 12/12
IPM-QRB = 36/48
Dobrogowski et al[98], 2005	45 consecutive patients with chronic low back pain judged to be positive with controlled diagnostic blocks	Injection of saline in patients after conventional radiofrequency (85 °C for 60 s) neurotomy to evaluate postoperative pain	Conventional radiofrequency neurotomy at 85 °C for 60 s, followed by injection of either methylprednisolone or pentoxifylline	Visual analog scale, minimum of 50% reduction of pain intensity, patient satisfaction score	1, 3, 6 and 12 mo	Greater than 50% of reduction of pain intensity was observed in 66% of the patients 12 mo later. There was no difference in the long-term outcomes	Randomized, active control trial	Very small study evaluating effectiveness of radiofrequency neurotomy and postoperative pain	Radiofrequency neurotomy effective with or without steroid injection after neurolysis
Randomized, active control trial
Quality scores:
Cochrane = 10/12
IPM-QRB = 29/48
van Kleef et al[99], 1999	31 patients with a history of at least one year of chronic low back pain randomly assigned to one of 2 treatment groups. Single diagnostic block with 50% relief	Sham control of radiofrequency after local anesthetic injection in 16 patients	The 15 patients in the conventional radiofrequency treatment group received an 80 °C radiofrequency lesion for 60 s	Visual analog scale, pain scores, global perceived effect, Oswestry Disability Index	3, 6 and 12 mo	After 3, 6 and 12 mo, the number of successes in the lesion and sham groups was 9 of 15 (60%) and 4 of 16 (25%), 7 of 15 (47%) and 3 of 16 (19%), and 7 of 15 (47%) and 2 of 16 (13%) respectively. There was a statistically significant difference	Double-blind, randomized, sham controlled trial	A single block with a small sample with inclusion criteria of 50% pain relief to enter the study. The study has been criticized that electrodes were placed at an angle to the target nerve, instead of parallel (51A)	Efficacy shown in a small sample with a single diagnostic block
Randomized, double-blind, sham control trial
Quality scores:
Cochrane = 12/12
IPM-QRB = 40/48
Moon et al[100], 2013	82 patients were included with low back pain with 41 patients in each group either with a parallel placement of the needle or perpendicular placement of the needle	An active control trial with needle placement with perpendicular approach	41 patients in each group were treated with radiofrequency (80 °C for 90 s) after appropriate diagnosis of facet joint pain with dual diagnostic blocks with 50% relief as the criterion standard. The needle was positioned either utilizing a discal approach or perpendicular or utilizing tunnel vision approach with parallel placement of the needle	NRS, ODI	1 and 6 mo	Patients in both groups showed a statistically significant reduction in NRS and Oswestry disability index scores from baseline to that of the scores at 1 and 6 mo (all P < 0.0001, Bonferroni corrected)	Randomized, double-blind, controlled trial. The major strength is that authors have proven that parallel approach may not be the best as has been described. Diagnosis of facet joint pain by dual blocks	Active controlled trial without placebo group. Short-term follow-up	Positive results in an active controlled trial, in a relatively short-term follow-up of 6 mo, with positioning of the needle either with distal approach (perpendicular placement or tunnel vision) with parallel placement of the needle with some superiority with perpendicular approach. This trial abates any criticism of needle positioning one way or the other and the traditional needle positioning appears to be superior to parallel needle placement
Prospective, randomized, comparative study
Quality scores:
Cochrane = 9/12
IPM-QRB = 38/48
Lakemeier et al[101], 2013	56 patients were randomized into 2 groups with 29 patients receiving intraarticular steroid injections and 27 patients receiving radiofrequency denervation after the diagnosis was made with intraarticular injection of local anesthetic with a single block	Intraarticular injection of local anesthetic and steroid	Radiofrequency neurotomy for 90 s at 80 °C	Roland-Morris questionnaire, VAS, ODI, analgesic intake	6 mo	Pain relief and functional improvement were observed in both groups. There were no significant differences between the 2 groups for pain relief and functional status improvement	Lack of placebo group. Relatively short-term follow-up	Randomized, double-blind trial with single diagnostic block with intraarticular injection	Both groups showed improvement. Effectiveness at 6 mo in both groups with intraarticular injection or radiofrequency neurotomy
Randomized, double-blind, active controlled trial
Quality scores:
Cochrane = 9/12
IPM-QRB = 37/48
Lumbar facet joint nerve blocks
Civelek et al[108], 2012	100 patients with chronic low back with failed conservative therapy and strict selection criteria; however, without diagnostic blocks	Blocks of facet joint nerves with local anesthetic and steroids	Conventional radiofrequency neurotomy at 80 °C for 120 s in combination with high dose local anesthetic and steroids	Visual Numeric Pain Scale, North American Spine Society patient satisfaction questionnaire, Euro-Qol in 5 dimensions and ≥ 50% relief	1, 6 and 12 mo	At the end of one year, 90% of patients in the radiofrequency group and 69% of the patients in the facet joint nerve block group showed significant improvement vs 92% and 75% at 6-mo follow-up	Randomized active-control trial with relatively large number of patients with 50 in each group	No diagnostic blocks were performed. High dose steroids and local anesthetics were provided in both groups	Results showed efficacy even without diagnostic blocks, both for facet joint nerve blocks and radiofrequency neurotomy
Randomized, active-control trial
Quality scores:
Cochrane = 9/12
IPM-QRB = 28/48
Manchikanti et al[102], 2010	120 patients with chronic low back pain of facet joint origin treated with therapeutic lumbar facet joint nerve blocks Double diagnostic blocks with 80% relief	Local anesthetic only	Total of 120 patients with 60 patients in each group with local anesthetic alone or local anesthetic and steroids. Both groups were also divided into 2 categories each with the addition of Sarapin	Numeric Rating Scale, Oswestry Disability Index, employment status, and opioid intake	3, 6, 12, 18 and 24 mo	Significant pain relief was shown in 85% in local anesthetic group and 90% in local anesthetic with steroids group at the end of the 2 yr study period in both groups, with an average of 5-6 total treatments	Randomized trial with relatively large proportion of patients with 2-yr follow-up, with inclusion of patients diagnosed with controlled diagnostic blocks	Lack of placebo group	Effectiveness demonstrated with facet joint nerve blocks with local anesthetic with or without steroids
Randomized, double blind, active control trial
Quality scores:
Cochrane = 11/12
IPM-QRB = 45/48
Lumbar intraarticular injections
Carette et al[103], 1991	Patients with chronic low back pain who reported immediate relief of their pain after injection of local anesthetic into the facet joints. Single diagnostic blocks with 50% relief were randomly assigned to receive injections under fluoroscopic guidance	Intraarticular injection of isotonic saline	Injection of either sodium chloride or methylprednisolone into the facet joints (49 for isotonic saline and 48 for sodium chloride). Only one injection was provided	Visual Analog Scale, McGill Pain Questionnaire, mean sickness impact profile	1, 3 and 6 mo	After 1 mo, 42% of the patients in the methylprednisolone group and 33% in the sodium chloride group reported marked or very marked improvement. At the 6 mo evaluation, 46% in the methylprednisolone group and 15% in the placebo group showed sustained relief. Revised statistics showed 22% improvement in active group and 10% in control group	Well-performed randomized, double-blind controlled trial	Only single block was applied and patients were treated with steroids without local anesthetic with only one treatment and expected 6 mo of relief	The authors concluded that results were negative in an active-control trial with injection of either sodium chloride solution or steroid into the facet joints after diagnosis with a single block
Randomized, double blind, impure placebo or active- control trial
Quality scores:
Cochrane = 11/12
IPM-QRB = 40/48
Fuchs et al[104], 2005	60 patients with chronic low back pain were included with patients randomly assigned into 2 groups. No diagnostic blocks	Active-control study with no control group	Intraarticular injection of hyaluronic acid vs glucocorticoid injection	VAS, Rowland-Morris Questionnaire, ODI, low back outcomes score, short form-36	3 and 6 mo	Patients reported lasting pain relief, better function, and improved quality of life with both treatments	Randomized, active-control, double-blind study	Relatively small sample of patients with 6 mo follow-up without a placebo group, without diagnostic blocks	Undetermined (clinically inapplicable) results with high number of injections during a 6-mo period
Randomized, double-blind, active-control trial
Quality scores:
Cochrane = 8/12
IPM-QRB = 26/48
Ribeiro et al[107], 2013	60 patients with a diagnosis of facet joint syndrome randomized into experimental and control groups	Triamcinolone acetonide intramuscular injection of 6 lumbar paravertebral points	Intraarticular injection of 6 lumbar facet joints with triamcinolone hexacetonide	Pain visual analogue scale, pain visual analogue scale during extension of the spine, Likert scale, improvement percentage scale, Roland-Morris, 36-Item Short Form Health Survey, and accountability of medications taken	1, 4, 12 and 24 wk	Both groups showed improvement with no statistical difference between the groups. Improvement "percentage" analysis at each time point showed significant differences between the groups at week 7 and week 12. Improvement percentage was > 50% at all times in the experimental group with intraarticular steroids; however, significant difference was noted at 24 wk	Randomized, double-blind controlled trial	Diagnostic blocks were not employed, thus, many patients without facet joint pain may have been included in this trial	Overall intraarticular steroids showed positive effectiveness for 24 wk compared to intramuscular steroids provided in a double-blind manner
Randomized, double-blind, active control
Quality scores:
Cochrane = 10/12
IPM-QRB = 32/48
Lakemeier et al[101], 2013	56 patients were randomized into 2 groups receiving intraarticular steroid injections or radiofrequency denervation after the diagnosis was made with intraarticular injection of local anesthetic with a single block	Intraarticular injection of local anesthetic and steroid in 29 patients	Radiofrequency neurotomy for 90 s at 80 °C in 27 patients	Roland-Morris questionnaire, VAS, ODI, analgesic intake	6 mo	Pain relief and functional improvement were observed in both groups. There were no significant differences between the 2 groups for pain relief and functional status improvement	Lack of placebo group. Relatively short-term follow-up	Randomized, double-blind trial with single diagnostic block with intraarticular injection	Both groups showed improvement. Effectiveness of both modalities at 6 mo in both groups
Randomized, double-blind, active controlled trial
Quality scores:
Cochrane = 9/12
IPM-QRB = 37/48
Yun et al[106], 2012	57 patients with facet syndrome were assigned to 2 groups with 32 patients in the fluoroscopy group and 25 patients in the under ultrasonography group without diagnostic blocks	Intraarticular injection of lidocaine and triamcinolone under fluoroscopic guidance	Intraarticular injection of lidocaine and triamcinolone under ultrasonic guidance	VAS, physician’s and patient’s global assessment (PhyGA, PaGA), modified ODI	1 wk, 1 and 3 mo	Each group showed significant improvement from the facet joint injections. However at a week, a mo, and 3 mo after injections, no significant differences were observed between the groups	Randomized trial	Short-term follow-up with no diagnostic blocks, thus increasing the potential for inclusion of patients without facet joint pain. The aim of study mainly was to confirm if ultrasonic imaging was appropriate	The study showed positive results in both groups with intraarticular steroid injections with a short-term follow-up whether performed under ultrasonic guidance or fluoroscopy
Randomized, active controlled trial
Quality scores:
Cochrane = 9/12
IPM-QRB = 26/48

IPM-QRB: Interventional Pain Management Techniques - Quality Appraisal of Reliability and Risk of Bias Assessment; VAS: Visual analog scale; ODI: Oswestry Disability Index.