Abstract
Objectives
This meta-analysis compares change in wrist pain following ultrasound-guided (US-guided) intra-articular glucocorticoid injections with change in pain after palpation-guided injections in persons with inflammatory arthritis or osteoarthritis.
Methods
Data sources included MEDLINE, Cochrane, BIOSIS, CINAHL, ACR/AHRP abstracts, and ClinicalTrials.gov. Studies that assessed change in wrist pain with direct comparison of US-guided and palpation-guided injections were included in the meta-analysis. Subject-level data was sought from authors of all relevant studies.
Primary outcome was mean change in wrist pain from baseline to 1–6 week follow-up by visual analog scale (VAS). Mean difference in VAS was calculated for comparative studies. Secondary outcome was proportion attaining Minimal Clinically Important Improvement (MCII), defined as VAS reduction ≥20%. Odds ratios (ORs) of MCII were calculated for comparative studies. Mean differences in VAS and ORs for MCII for comparative studies were combined using fixed and random effects meta-analysis.
Results
Ten studies were eligible, and adequate data was available from 4 studies with direct comparison of US-guided and palpation-guided treatment arms. The difference in mean VAS reduction (US-guided minus palpation-guided) ranged from −0.2 to 1.3, with a combined estimate of 1.0 (95% CI 0.3, 1.7). OR for MCII in comparative studies ranged from 1.0 to 12.4, with a combined OR of 3.2 (95% CI 1.2, 8.5) in favor of ultrasound.
Conclusions
US-guided glucocorticoid injections to the wrist result in greater reductions in pain, and greater likelihood of attaining MCII than palpation-guided injections at 1–6 weeks follow-up.
Keywords: Ultrasound, Injection, Wrist
Intra-articular (IA) injections are a standard therapy for a variety of inflammatory and non-inflammatory joint conditions. Use of IA injections is supported by both American College of Rheumatology and European League Against Rheumatism recommendations for treatment of rheumatoid arthritis (RA) [1,2].
IA injections have traditionally been given using palpation of relevant landmarks to identify the joint, but palpation-guided injections generally have lower accuracy than those guided by ultrasound (US) [3]. For this reason, US has become commonly used not only to aid in diagnosis and localization of synovitis, but also to guide intra-articular injections. In a survey of ACR members, 21% reported current use of US with patients, and 26% referred patients to radiologists for US procedures [4].
Despite the popularity of US, few studies have directly compared the accuracy of IA injections given by palpation of landmarks with those given using US-guidance. Cunnington found that injections using US-guidance were accurate more often than those given using palpation, and that accurate injections led to a trend toward greater improvement in pain scores [5]. Similarly, improvement in outcome was associated with accuracy of injections of the shoulder in a study by Eustace et al. [6] and in a study of injections to multiple joints in a study by Jones et al. (though statistically non-significant) [7].
The wrist is commonly affected by rheumatoid arthritis, and often injected for symptomatic relief. Though studies on the efficacy of US-guided injections have included subjects who received wrist injections, such studies have lacked power to determine joint-specific effects. More precise knowledge on the efficacy of US-guided injections and any benefit of US over palpation-guidance requires a larger numbers of subjects.
Because US-guidance leads to greater accuracy of IA injections, and prior studies have lacked power to demonstrate joint-specific effects, we hypothesized that meta-analysis would demonstrate that US-guidance leads to reduction in pain following injection as compared to palpation-guidance. We performed meta-analysis of individual patient data from existing prospective studies that assessed pain relief following US-guided intra-articular glucocorticoid injections of the wrist as compared with standard, palpation-guided injections for arthritis in adults.
METHODS
We searched for studies using research databases, hand-search of reviews and references, search of ACR abstracts, recommendations from experts, and personal communication with investigators. Databases included MEDLINE, Cochrane, BIOSIS, CINAHL, and ClinicalTrials.gov. ACR/AHRP meeting abstracts from 2006 to 2010 were searched using the online search engine. MEDLINE searching was performed using MeSH terms “ultrasonography” OR “ultrasonics” AND “injections” in combination with diagnostic MeSH terms “osteoarthritis” OR “arthritis, rheumatoid”, “spondylarthritis”, “gout”, “arthritis”, or “synovitis”, with limits for “humans”. Searches were closed on June 13, 2011.
We sought original prospective or retrospective studies or case series of living adults who had an inflammatory arthritis, crystalline arthritis/arthropathy, or osteoarthritis, and were treated with a single intra-articular injection of a long-acting glucocorticoid preparation either by palpation/clinical exam or using ultrasound guidance. Studies of ultrasound-guided injections were judged as relevant if they used US to guide the injection, but not if US was used solely for diagnosis, assessment of disease, and/or therapy. We included only long-acting glucocorticoid preparations (triamcinolone acetonide and hexacetonide and methylprednisolone) as shorter-acting preparations may not have an effect lasting several weeks. We included studies that assessed subject-reported wrist pain at rest prior to injection, and 6 days to 6 weeks following the injection. Eligible outcomes included patient rating of pain on 0–10 cm or 0–100 mm visual analog scale (VAS) or Health Assessment Questionnaire (HAQ), with preference for VAS measures. We assessed outcome measure at 1–6 weeks, as 2–4 weeks has been demonstrated to be the time frame likely to reveal clinical effect [8–10].
Injections to carpometacarpal joints, carpal tunnel and other tendon sheaths were excluded.
Citations from database queries were evaluated for relevance, and if relevant, the full manuscript/publication was reviewed (Fig. 1). Study investigators were contacted if there was insufficient data in a given publication for inclusion in the meta-analysis (missing standard deviation of mean, joint/structure injected, etc.). Subject-level data was requested from investigators of all studies to allow for pooled analysis of individual patient data. Database queries, citation review, and manuscript review were performed independently by 2 investigators (MD and SG), and disagreement on eligibility of citation or manuscript was resolved by discussion or review by a third investigator (EK). Data was extracted independently by 2 investigators for statistical analysis.
Figure 1.
Summary of search results and study selection.
Individual patient-level data was pooled for calculations of mean values for rest pain before and after injection, with standard deviations. The primary outcome was change in outcome measure between baseline measurement and 1–6 weeks follow-up. Secondary outcome was proportion of subjects in each group who had a minimally clinically important improvement (MCII), as defined by an improvement in VAS by 20% or more [11]. For studies which used a 0–100 mm VAS scale, the scores were rescaled to a 0–10 cm scale for data pooling. For studies with more than 1 eligible follow-up time, the follow-up closest to the midpoint of the eligible window (3 weeks) was used.
Study quality was assessed in 8 aspects: indication of clear inclusion criteria, clear exclusion criteria, randomization to treatment arm, blinding of subject and outcome assessor to treatment arm, indication of experience/training of the clinician who performed injections, detailed description of injection technique/approach and equipment, indication of drug(s) and dosage specific to the wrist, and description of post-injection activity restrictions. For studies with a single treatment arm, randomization and blinding were not assessed. Quality assessment was used for description of the studies, but not used as a basis for study inclusion.
Statistical Analysis
Patient-level data were obtained from authors of all relevant studies, including wrist-specific baseline and follow-up pain scores by VAS or HAQ. Subjects with VAS pain score of zero at baseline were excluded from all analyses due to inability to demonstrate improvement. Change in VAS pain from baseline to follow-up on a scale of 0–10 was the primary outcome, and subjects with a reduction of 20% or more were considered to have experienced minimal clinically important improvement (MCII).
Mean and 95% confidence interval for change in VAS pain was calculated for each US-guided and palpation-guided injection arm by study. Mean difference in change in VAS was calculated for studies with direct comparison of US- and palpation-guided arms. The proportions of subjects attaining MCII were calculated for each group with exact binomial 95% confidence intervals. Odds ratios (OR) of MCII attainment were calculated for comparative studies unless there were cells with counts of zero in the 2-way table of treatment by MCII.
Mean differences in VAS pain were combined across studies using inverse-variance weighting and random effects meta-analysis. Due to the presence of zero cell counts this method could not be used to combine the OR for MCII attainment. Instead, the fixed-effects Mantel–Haenszel method was used to provide a combined OR as it can accommodate some zero cell counts. Due to the potential for ORs to overestimate effect, we also calculated combined relative risk (RR) using the Mantel–Haenszel method and fixed-effect inverse-variance. Heterogeneity in the mean differences of VAS pain change was measured using the I2 statistic and tested using the Cochran Q-test. Heterogeneity in the ORs for MCII attainment was tested using the Breslow–Day test. In a sensitivity analysis, subjects with baseline VAS values less than 3 were excluded from the MCII calculations. All calculations were performed using SAS version 9.1.
RESULTS
In review of 1516 citations, 74 were selected for further review, and, of these, 10 were eligible for inclusion in this study. Adequate data was available for 7, which were included in subsequent analyses (Fig. 1). Reasons for inadequate data included loss of the data due to natural disaster (1), request after the period for which the trial data was maintained (1), and lack of response from study authors (1).
Study and Subject Characteristics
Four studies had direct comparison of US-guided and palpation-guided injections [5,12–14], and 3 studies had a single eligible treatment arm [15–17]. Six studies included a VAS 0–10 cm scale for rest pain and 1 included a VAS 0–100 mm. Triamcinolone acetonide (TA) was used in 4 studies (dose 20–60 mg), and triamcinolone hexacetonide (TH) was used in 2 studies (10–30 mg). The mean equivalent dose of TH was 22.7 mg (±9.23) and 26.5 mg (±5.92) among subjects treated with palpation-guided and US-guided injections, respectively. Mean follow-up time was 2.28 weeks (±1.21) and 2.51 weeks (±0.875), respectively. Two subjects were excluded due to baseline VAS scores of zero (1 US-guided and 1 palpation-guided).
Demographic and diagnosis information was available for 38% of subjects in the study (Table 1). Of those subjects for whom demographic information was available, 88% (palpation-guided) and 84% (US-guided) of subjects were female, and mean age was 54 years (range 26–80, palpation-guided) and 47 years (range 20–82, US-guided). Rheumatoid arthritis was the most common diagnosis, and was the indication for injection in 83% and 64% of those treated with palpation-guided and US-guided injections, respectively.
Table 1.
Characteristics of Studies Included in Meta-analysis
| Study Author | Drug and Dose (mg) |
Follow-up Time (Weeks) |
n | % Female |
Mean Age (Range) |
Diagnoses |
|---|---|---|---|---|---|---|
| Cunnington et al. [5] | TA 40 | 2 | 29 | 75.9 | 57.4 (30–87) | RA, PsA, Inflammatory arthritis |
| Filippucci et al. [15] | TA 20–30 | 2 | 13 | 79.6 | 56.0 (30–82) | RA, CPPD, ReA, PsA |
| Lopes et al. [16] | TH 30 | 4 | 36 | NA | NA | RA |
| Luz et al. [12] | TH 30 | 4 | 60 | 93.3 | 49.2 (NA) | RA |
| Sibbitt et al. [14] | TA 60 | 2 | 30 | 88.5 | 48.3 (26–80) | RA, SSc, OA, ReA, SLE, Gout |
| Sibbitt et al. [13] | TA 60 | 2 | 83 | 91.6 | 49.3 (20–75) | RA, SSc, PsA, SLE, OA |
| Weitoft and Rönnblom [17] | TH 10 | 1 | 59 | 68.9 | 61.0 (28–86) | RA |
TA = triamcinolone acetonide, TH = triamcinolone hexacetonide. RA = rheumatoid arthritis, SSc = systemic sclerosis, OA = osteoarthritis, ReA = reactive arthritis, SLE = systemic lupus erythematosus, PsA = psoriatic arthritis, CPPD = calcium pyrophosphate deposition disease, NA = not available.
In 6 of the 7 studies, the credentials of the clinician performing US-guided injections were indicated: 2 were rheumatologists, 3 rheumatology fellows, and 1 orthopedist. The specific approach to US-guided wrist injection was described in all 4 comparative studies: all 4 used direct US visualization of the needle, 3 used an in-plane approach at the level of the radiocarpal joint (1 with needle tip oriented proximally, 1 oriented distally, and 1 not indicated), and 1 study used an out-of-plane approach (needle perpendicular to the probe axis) at the midcarpal joint. An example of the transverse in-plane approach at the level of the scapholunate interval is available as Supplementary Figure S1.
Studies with more than 1 treatment arm generally met 6 or more of the 8 quality measures (1 met 5). The 2 studies with a single treatment arm met 3 and 5 of the 6 possible measures, respectively. Inclusion criteria varied, but generally required stable treatment for RA (4 of 7 studies) and a specific description of wrist synovitis (6 of 7 studies). The most common items omitted from publications were aspects of the injection approach (4 studies), blinding to treatment allocation (3), and description of post-injection activity restrictions (3).
Change in VAS
Mean change in VAS in the 4 comparative studies ranged from 2.7 to 5.7 in the palpation-guided injection arms, and from 3.8 to 5.8 in US-guided injection arms (Table 2). The mean difference in VAS change (US–palpation) ranged from −0.2 to 1.3, and was non-significant at the 0.05 level for 3 of the 4 studies. A forest plot depicting mean difference in VAS change (Fig. 2) shows the values for individual studies, and the combined estimate of 1.0 (95% CI 0.3, 1.7). Because the CI for this pooled difference measure excludes zero, this indicates a statistically significant improvement in VAS pain reduction with US-guided injection. There was no heterogeneity detected in these study results with an I2 value of 0 and a Q-test p-value of 0.62.
Table 2.
Mean Reduction in VAS Pain Scores Among Comparative Studies
| Palpation-Guided | Ultrasound-Guided | Difference (Ultrasound–Palpation) |
|||
|---|---|---|---|---|---|
| Study | n | Mean VAS Improvement (SD) | n | Mean VAS Improvement (SD) | Mean (95% CI) |
| Cunnington et al. [5] | 15 | 3.7 (2.4) | 14 | 4.9 (2.9) | 1.2 (−0.8, 3.1) |
| Luz et al. [12] | 30 | 2.7 (1.8) | 30 | 3.8 (2.5) | 1.1 (−0.0, 2.2) |
| Sibbitt et al. [14] | 15 | 5.7 (2.9) | 15 | 5.5 (2.4) | −0.2 (−2.1, 1.7) |
| Sibbitt et al. [13] | 37 | 4.5 (2.8) | 46 | 5.8 (2.5) | 1.3 (0.1, 2.4) |
| Combined | 1.0 (0.3, 1.7) | ||||
SD = standard deviation, CI = confidence interval.
Figure 2.
Forest plot of mean differences in VAS change.
Minimal Clinically Important Improvement
The proportion of subjects achieving MCII in the 4 comparative studies ranged from 78.4% to 86.7% in the palpation-guided injection arms, and from 92.9% to 100% in the US-guided arms (Table 3). The OR for achievement of MCII among subjects treated with US-guided injections relative to palpation-guided injection ranged from 1.0 to 12.4 in the 3 studies for which OR could be calculated. For the Sibbitt et al. [14] study, OR could not be calculated due to all the US-guided subjects achieving MCII.
Table 3.
Proportion of Subjects and Odds Ratios of Attaining Minimal Clinically Important Improvement in Pain Score by Study
| Palpation-Guided | Ultrasound-Guided | Comparison | |||
|---|---|---|---|---|---|
| Study | MCII/n | Percentage MCII (95% CI) |
MCII/n | Percentage MCII (95% CI) |
Odds Ratio (95% CI) |
| Cunnington et al. [5] | 13/15 | 86.7 (59.5, 98.3) | 13/14 | 92.9 (66.1, 99.8) | 2.0 (0.2, 24.9) |
| Luz et al. [12] | 26/30 | 86.7 (69.3, 96.2) | 26/30 | 86.7 (69.3, 96.2) | 1.0 (0.2, 4.4) |
| Sibbitt et al. [14] | 13/15 | 86.7 (59.5, 98.3) | 15/15 | 100.0 (78.2,100.0) | – |
| Sibbitt et al. [13] | 29/37 | 78.4 (61.8, 90.2) | 45/46 | 97.8 (88.5, 99.9) | 12.4 (1.5,104.5) |
| Combined | 3.1 (1.2, 803) | ||||
For the Sibbitt et al. [14] study, OR was unable to be calculated due to all of the subjects in the US-guided injection arm attaining MCII.
Using the fixed-effect Mantel–Haenszel method for combining ORs, the combined OR for attaining MCII in case of US-guided injections relative to palpation-guided injections was 3.1 (95% CI 1.2, 8.3), indicating statistically significant benefit of ultrasound (p = 0.01). The Breslow–Day test for heterogeneity of the study ORs was not significant (p = 0.16) indicating that there was no between-study variation requiring the use of a random effects model. When subjects with a baseline VAS pain below 3.0 on the 10-point scale were excluded, the results were not affected in a substantial manner (OR 3.2, 95% CI 1.2, 8.5).
The RR for achievement of MCII among subjects treated with US-guided injection relative to palpation-guided injection ranged from 1.0 to 1.25 in the 4 comparative studies. Using the fixed-effects Mantel–Haenzel method of combining relative risks, the combined RR was 1.13 (95% CI 1.02, 1.25). The estimate of between-study variability was 0, also giving an I2 value of 0, indicating no between-study heterogeneity. Because there was no between-study variation detected, the random effect combined RR was also 1.13 (95% CI 1.02, 1.25).
DISCUSSION
US has gained popularity among rheumatologists, but there remains justifiable reluctance to adopt US among those who believe outcomes with palpation-guided injections are adequate and those who have concern about increased costs with US. Previous expert recommendations for the use of US-guidance in performing intraarticular injections have focused on joints with challenging anatomy; however studies evaluating the efficacy of US-guided injections have generally lacked the power to show significant reductions in pain in a specific joint.
The radiocarpal joint is generally thought to be a superficial joint with easily palpable landmarks. Nonetheless, this meta-analysis demonstrated a benefit of US-guidance over palpation-guided injections to the wrist, even when compared to palpation-guided injections by experienced physicians. Though accuracy was not the focus of the studies included in this meta-analysis, the most likely cause for the benefit of US-guided injections we observed is improved accuracy of needle and drug placement with use of US-guidance. Another possible explanation is that US-guided injections cause less needle trauma to the synovial tissues and thus less hemorrhage and leakage of blood proteins into joints, which would also tend to result in less inflammation and better outcome for US-guided injections [14].
Subjects who received palpation-guided injections generally had improvements in VAS pain scores, and more than 83% overall achieved MCII, providing some support for the conservative view that palpation-guided injections have acceptable outcomes. However, in the US-guided group, over 94% achieved MCII, potentially preventing 11% of subjects from requiring a second injection, with all its associated costs. This translates to a number of patients that need to be treat of 9; if 9 patients were treated with US-guided injections, 1 could be prevented from not achieving clinically important improvement.
This study was limited by the paucity of specific trials addressing the subject. Despite the rising popularity of ultrasound in rheumatology practice, relatively few studies addressed this important issue with head-to-head comparisons of US- and palpation-guided injections. There was insufficient data available to assess several other factors that may potentially predict response to injection, such as the specific injection approach for the wrist, and unique patient characteristics.
CONCLUSION
This meta-analysis is the first to assess the efficacy of US-guided wrist injections using combined data from all of the available relevant studies. We demonstrated a statistically and clinically significant benefit of US-guidance over palpation-guided injections in reduction of wrist pain and achievement of MCII. Research into the outcomes of US-guidance is critical for establishing the fact-based role of this technology in medicine and education, and for providing the necessary data to permit funding agencies to support appropriate reimbursement. Thus, the present study represents 1 of the first necessary steps to develop an outcomes-based approach to the use of US-guidance.
Though these results show a clear benefit of US-guidance in these important pain outcomes among patients with arthritis, further research is needed to address specific patient or provider characteristics or injection techniques that predict outcome, and to assess the joint-specific cost effectiveness of US-guided injections.
Supplementary Material
ACKNOWLEDGMENTS
Thanks to all authors who shared individual patient data, including Karine Rodrigues da Luz, Thomas Weitoft, and Roberta Lopes. Thanks to Diana Vradii and Naomi Amudala for their help with the injection images.
Footnotes
APPENDIX A
SUPPLEMENTARY MATERIAL
Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.semarthrit.2012.09.006.
REFERENCES
- 1.American College of Rheumatology. Subcommittee on Rheumatoid Arthritis Guidelines. Guidelines for the management of rheumatoid arthritis: 2002 update. Arthritis Rheum. 2002;46:328–346. doi: 10.1002/art.10148. [DOI] [PubMed] [Google Scholar]
- 2.Combe B, Landewe R, Lukas C, Bolosiu HD, Breedveld F, Dougados M, et al. EULAR recommendations for the management of early arthritis: report of a task force of the European Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT) Ann Rheum Dis. 2007;66(1):34–45. doi: 10.1136/ard.2005.044354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gilliland CA, Salazar LD, Borchers JR. Ultrasound versus anatomic guidance for intra-articular and periarticular injection: a systematic review. Phys Sportsmed. 2011;39(3):121–131. doi: 10.3810/psm.2011.09.1928. [DOI] [PubMed] [Google Scholar]
- 4.Samuels J, Abramson SB, Kaeley GS. The use of musculoskeletal ultrasound by rheumatologists in the United States. Bull NYU Hosp Joint Dis. 2010;68(4):292–298. [PubMed] [Google Scholar]
- 5.Cunnington J, Marshall N, Hide G, Bracewell C, Isaacs J, Platt P, et al. A randomized, double-blind, controlled study of ultrasound-guided orticosteroid injection into the joint of patients with inflammatory arthritis. Arthritis Rheum. 2010;62(7):1862–1869. doi: 10.1002/art.27448. [DOI] [PubMed] [Google Scholar]
- 6.Eustace JA, Brophy DP, Gibney RP, Bresnihan B, FitzGerald O. Comparison of the accuracy of steroid placement with clinical outcome in patients with shoulder symptoms. Ann Rheum Dis. 1997;56:59–63. doi: 10.1136/ard.56.1.59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jones A, Regan M, Ledingham J, Pattrick M, Manhire A, Doherty M. Importance of placement of intra-articular steroid injections. Br Med J. 1993;307:1329–1330. doi: 10.1136/bmj.307.6915.1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bellamy N, Campbell J, Robinson V, Gee T, Bourne R, Wells G. Intraarticular corticosteroid for treatment of osteoarthritis of the knee. Cochrane Database Syst Rev. 2006;19(2) doi: 10.1002/14651858.CD005328.pub2. CD005328. [DOI] [PubMed] [Google Scholar]
- 9.Firestein GS, Paine MM, Littman BH. Gene expression (collagenase, tissue inhibitor of metalloproteinases, complement, and HLA-DR) in rheumatoid arthritis and osteoarthritis synovium. Quantitative analysis and effect of intraarticular corticosteroids. Arthritis Rheum. 1991;34(9):1094–1105. doi: 10.1002/art.1780340905. [DOI] [PubMed] [Google Scholar]
- 10.Boesen M, Boesen L, Jensen KE, Cimmino MA, Torp-Pedersen S, Terslev L, et al. Clinical outcome and imaging changes after intraarticular (IA) application of etanercept or methylprednisolone in rheumatoid arthritis: magnetic resonance imaging and ultrasound-Doppler show no effect of IA injections in the wrist after 4 weeks. J Rheumatol. 2008;35(4):584–591. [PubMed] [Google Scholar]
- 11.Tubach F, Ravaud P, Beaton D, Boers M, Bombardier C, Felson DT, et al. Minimal clinically important improvement and patient acceptable symptom state for subjective outcome measures in rheumatic disorders. J Rheumatol. 2007;34(5):1188–1193. [PMC free article] [PubMed] [Google Scholar]
- 12.Luz KR, Furtado RN, Nunes CC, Rosenfeld A, Fernandes AR, Natour J. Ultrasound-guided intra-articular injections in the wrist in patients with rheumatoid arthritis: a double-blind, randomised controlled study. Ann Rheum Dis. 2008;67(8):1198–1200. doi: 10.1136/ard.2007.084616. [DOI] [PubMed] [Google Scholar]
- 13.Sibbitt WL, Jr, Band PA, Chavez-Chiang NR, Delea SL, Norton HE, Bankhurst AD. A randomized controlled trial of the cost-effectiveness of ultrasound-guided intra-articular injection of inflammatory arthritis. J Rheumatol. 2011;38(2):252–263. doi: 10.3899/jrheum.100866. [DOI] [PubMed] [Google Scholar]
- 14.Sibbitt WL, Jr, Peisajovich A, Michael AA, Park KS, Sibbitt RR, Band PA, et al. Does sonographic needle guidance affect the clinical outcome of intraarticular injections? J Rheumatol. 2009;36(9):1892–1902. doi: 10.3899/jrheum.090013. [DOI] [PubMed] [Google Scholar]
- 15.Filippucci E, Farina A, Carotti M, Salaffi F, Grassi W. Grey scale and power Doppler sonographic changes induced by intraarticular steroid injection treatment. Ann Rheum Dis. 2004;63(6):740–743. doi: 10.1136/ard.2003.007971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lopes RV, Furtado RN, Parmigiani L, Rosenfeld A, Fernandes AR, Natour J. Accuracy of intra-articular injections in peripheral joints performed blindly in patients with rheumatoid arthritis. Rheumatology (Oxford) 2008;47(12):1792–1794. doi: 10.1093/rheumatology/ken355. [DOI] [PubMed] [Google Scholar]
- 17.Weitoft T, Rönnblom L. Randomised controlled study of postinjection immobilisation after intra-articular glucocorticoid treatment for wrist synovitis. Ann Rheum Dis. 2003;62(10):1013–1015. doi: 10.1136/ard.62.10.1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.