Importance of immobilization after intraarticular glucocorticoid treatment for elbow synovitis: A randomized controlled study


  • Tomas Weitoft,

    Corresponding author
    1. Center for Research and Development, County Council of Gävleborg/Uppsala University, Uppsala, and Gävle Hospital, Gävle, Sweden
    • Section of Rheumatology, Division of Internal Medicine, Gävle Hospital, 801 87 Gävle, Sweden
    Search for more papers by this author
    • Dr. Weitoft has received speaking fees (less than $10,000) from Meda AB.

  • Catarina Forsberg

    1. Falu Hospital, Falun, Sweden
    Search for more papers by this author

  • identifier: NCT00972530.



To investigate whether better treatment results might be achieved with postinjection rest following intraarticular glucocorticoid treatment for elbow synovitis.


Ninety patients with rheumatoid arthritis and elbow synovitis were treated with 20 mg of intraarticular triamcinolone hexacetonide and randomized to either a 48-hour immobilization in a triangular sling (n = 46) or to normal activity without restrictions (n = 44). The primary end point was relapse of synovitis. In addition, pain, function according to a self-assessment questionnaire (the Patient Rated Elbow Evaluation), and range of movement were followed for 6 months.


Thirteen relapses occurred in the rest group and 8 in the active group (results not significant). The secondary measures showed no statistically significant differences between the groups.


Immobilization procedures do not improve the outcome of intraarticular glucocorticoid therapy for elbow synovitis. In future guidelines for such treatment of the upper extremity, normal activity after the injection should be recommended.


For decades, intraarticular glucocorticoid injections have been a cornerstone in the treatment of arthritis. Despite this large amount of clinical experience, there is still no consensus about optimal injection routines. One point of debate concerns what recommendations should be given to patients after the steroid injection.

Previous studies have shown that a 24–48-hour postinjection rest improves the outcome of intraarticular glucocorticoid treatment for knee synovitis (1, 2). In contrast, a 48-hour immobilization of the wrist after injection treatment did not improve the result (3). The discrepancy might be explained by differences in joint size, anatomy, or the position for loading. Obviously, results obtained in studies on a specific joint cannot be generalized, and further studies on different joints and their response to postinjection rest are needed.

Similar to the wrist, the elbow has no weight-bearing position, but the anatomy and range of movement are more similar to those of the knee, and the joint size is in-between the two. The effects of postinjection rest can therefore not be predicted, and in the present study we asked whether immobilization procedures may improve the outcome of intraarticular glucocorticoid treatment for elbow synovitis. To our knowledge, the question has never been investigated before and the findings may be important for the future guidelines of intraarticular glucocorticoid treatment for elbow synovitis.

Patients and Methods

Patients fulfilling the 1987 American College of Rheumatology (formerly the American Rheumatism Association) criteria for rheumatoid arthritis (RA) (4) who had clinical signs of elbow synovitis (swelling, tenderness, and reduced range of movement) and attended the outpatient rheumatology departments in Gävle, Uppsala, and Falun were invited to participate in the study. Patients in functional class IV according to the Steinbrocker criteria (5) were excluded, as well as patients receiving oral glucocorticoid treatment corresponding to >7.5 mg of prednisolone or intraarticular treatment in the joint in the past 3 months.

The injections were performed with a lateral technique using a 0.8 × 40–mm needle. After synovial fluid aspiration, 20 mg of triamcinolone hexacetonide was injected in the joint. The patients were randomized to either 48-hour immobilization with the arm resting in a triangular sling (mitella) or normal activity without any restrictions.

The joint function and pain were evaluated before injection, after 1 week, after 3 months, and after 6 months using a self-assessment questionnaire, the Patient Rated Elbow Evaluation (6), which contains 5 questions on pain symptoms (maximum score 50) and 11 questions on functional impairment (maximum score 110). The range of movement was measured with an Össur goniometer (Össur Nordic). At inclusion, the patient's disability was assessed using the Swedish version of the Health Assessment Questionnaire (7) and the patient's disease activity was assessed using the serum level of C-reactive protein and the Disease Activity Score in 28 joints (8). The patients were followed for 6 months, and if signs and symptoms recurred the patients were told to contact the rheumatology department. In such cases, the elbow was reexamined and if a relapse could be confirmed, the duration of effect was recorded and if needed the patient was offered another injection.

For a comparison of baseline data between the groups, the Mann-Whitney U test, Student's t-test, and chi-square test were used when appropriate. A chi-square test was used to test for significance between the proportions of relapses in the groups at the end of the study. To test for differences compared with baseline of the mean change of the secondary parameters after 1 week, 3 months, and 6 months, a t-test for independent variables was used. P values less than 0.05 were considered significant and all tests were 2-tailed. The statistical analysis was made using the statistical computer software program SPSS, version 16.0 (SPSS). The study was approved by the Committee of Ethics, Faculty of Medicine, Uppsala University, and informed consent was obtained from all patients.


A total of 90 patients entered the study. One patient did not attend the 1 week visit, 3 patients did not attend the 3 month visit, and 4 did not attend the 6 month visit. The latter patients received a phone call to ensure that no symptoms from the treated elbow had recurred.

A total of 46 patients were randomized to the immobilization group and 44 patients to the activity group. The baseline patient characteristics are presented in Table 1. During the observation period, 13 patients in the immobilization group and 8 in the activity group had a relapse of arthritis (not significant).

Table 1. Patient characteristics at baseline*
 Activity group (n = 44)Rest group (n = 46)P
  • *

    Values are the mean ± SD unless otherwise indicated. ns = not significant (P > 0.05); HAQ = Health Assessment Questionnaire; DAS28 = Disease Activity Index in 28 joints; CRP = C-reactive protein; NSAID = nonsteroidal antiinflammatory drug; DMARD = disease-modifying antirheumatic drug; anti-TNF = anti–tumor necrosis factor; PREE = Patient Rated Elbow Evaluation.

  • Mann-Whitney U test.

  • Chi-square test.

  • §

    Student's t-test.

Age, median (range) years64 (17–85)63 (25–86)ns
Sex, no. of men/women6/3812/34ns
Disease duration, median (range) years24 (1–54)11 (1–41)ns
HAQ score, median (range)§1.63 (0–2.88)1.25 (0.13–2.38)ns
DAS28 score, median (range)§4.66 (2.37–7.71)4.45 (2.05–6.37)ns
Serum CRP level, median (range) mg/liter§16 (1–194)12 (1–115)ns
Treatment, no. of patients   
Arthritis in dominant side, no. of patients2226ns
Successful synovial fluid aspiration, no. of patients2621ns
Maximum elbow extension, degrees§156 ± 10.6156 ± 13.2ns
Elbow pain, PREE score§31 ± 10.631 ± 11.5ns
Elbow function, PREE score§57.5 ± 27.657 ± 23.8ns

Both groups improved significantly compared with baseline. The treatment response of elbow pain, function, and mobility for patients without clinical relapses were similar in the groups. The results of the measurements are presented in Table 2.

Table 2. Measurement results: change from baseline*
 Change at 1 weekChange at 3 monthsChange at 6 months
Rest groupActive groupRest groupActive groupRest groupActive group
  • *

    Values are the mean ± SD. PREE = Patient Rated Elbow Evaluation.

Mobility, degrees14 ± 12.912 ± 10.315 ± 20.214 ± 11.418 ± 22.415 ± 11.3
Function, PREE score26 ± 23.127 ± 23.426 ± 27.530 ± 26.631 ± 27.129 ± 26.3
Pain, PREE score14 ± 13.216 ± 10.314 ± 15.117 ± 12.017 ± 13.916 ± 12.7


The main finding in this investigation was that elbow immobilization does not improve the outcome of intraarticular glucocorticoid treatment. In accordance with our previous study of wrist injections, there was no favorable response in the immobilization group. In fact, there were more relapses occurring in this group, indicating that postinjection rest of the upper extremity should be avoided. These findings are in contrast with the beneficial effects of postinjection rest for knee synovitis and suggest that the weight-bearing position of the joint is an important factor for a favorable outcome of immobilization procedures.

In our previous investigations on knees (9) and wrists (3), we found a relapse rate of 23–24% within 6 months. Surprisingly, the figure observed in this elbow study was equal and may therefore represent the normal proportion of relapse after intraarticular triamcinolone hexacetonide treatment in RA.

This is the first study focusing on the outcome of intraarticular glucocorticoid treatment for elbow synovitis. This study confirms that this therapy is very effective in reducing pain and increasing function and joint mobility. However, elbow synovitis often occurs without (or with only mild) symptoms. Despite obvious clinical signs, the only presenting symptom could be limits in joint extension, which are not always associated with discomfort or disability. This may explain the difficulties in patient recruitment seen in the present study. Another reason may be that the introduction of biologic agent treatments has reduced the overall need for glucocorticoid injections. Furthermore, patients who traveled by car often declined to participate in this investigation because of the risk for randomization to immobilization, which would cause difficulties with driving home again. To our knowledge, this is the largest elbow injection study to date, but an even larger patient group would have been ideal.

Despite a careful randomized allocation, there was a nonsignificant trend to a shorter disease duration (P = 0.074) and less impaired general function (P = 0.064) in immobilized patients at baseline. However, if these factors had influenced the results, a better outcome after rest would have been expected; in contrast, there were more relapses found in this group. This further supports our conclusion that normal activity is preferred before rest.

This investigation had an intent-to-treat analysis and there was no control of compliance for the use of mitella in the immobilization group. A low compliance rate could be a source of possible error. On the other hand, this is also the reality if joint rest is recommended in the outpatient clinical practice.

In summary, because neither wrists nor elbows respond with a better outcome after postinjection rest, we conclude that patients with intraarticular glucocorticoid treatment of joints of the upper extremity should not be given advice to rest after the injection.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Weitoft had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Weitoft, Forsberg.

Acquisition of data. Weitoft, Forsberg.

Analysis and interpretation of data. Weitoft.


Meda AB has generously given financial support, but has not in any other way (not study design, data collection, data analysis, writing of the manuscript, or approval of the manuscript) been involved in the study.


We wish to thank Shirani Jayawardena, Sven Tegmark, Esmail Karim, Ann Knight, and Jörgen Lysholm for great assistance in patient recruiting, and Britt-Inger Nilsson, Åsa Christensen in Gävle, and Britta Eriksson in Uppsala for helping us with elbow measurements.