Management of Adhesive Capsulitis: An Annotated Bibliography

Introduction

 Adhesive capsulitis(AC) of the shoulder is a pathological process where excessive fibrous tissue are formed across the glenohumeral joint causing restriction of the shoulder range of motion (ROM) which then leads to subsequent pain and dysfunction (Le, Lee, Nazarian, & Rodriguez, 2017). AC is a common musculoskeletal malady in primary care. AC affects up to 5% of the population with the propensity for women and age of onset usually around 40 to 60 years old (Kitridis, Tsikopoulos, Bisbinas, Papaioannidou, & Givissis, 2019). Most commonly AC causes are divided into idiopathic if the aetiology is unknown, or secondary if attributed to rotator-cuff pathology, diabetes, or trauma (Mezian & Chang, 2018).  It is a challenging, yet fascinating scope as evidenced by the myriad of treatment options since it is a difficult condition to treat. Most often this condition is not covered by the Accident Compensation Corporation (ACC) since it is more often than not,  non-traumatic (Cadogan & Mohammed, 2016). It is important that a differential diagnosis is postulated in the early stages. Emphasis is placed on the clinical acumen of the clinician to best manage AC since there are multiple treatment modalities. The chronicity of the condition which can take up to a few years can often be frustrating. The three widely accepted sequential phases of AC are pain, freezing and thawing/recovery stages (Ewald, 2011). As for all the facets of treatment, treatment targets anti-inflammation and anti-adhesion (Sun, Lu, Zhang, Wang, & Chen, 2016). Despite all the medical technological armamentarium, we still lack adequate pain artillery and patient functioning management skill.  This annotated bibliography examines some of the evidence available for the management of AC pain. This paper includes seven articles which encompasses systematic reviews, Randomised Controlled Trials (RCTs), a qualitative study and a weaker evidence-based retrospective study.

Kitridis, D., Tsikopoulos, K., Bisbinas, I., Papaioannidou, P., & Givissis, P. (2019). Efficacy of                                                        

pharmacological therapies for adhesive capsulitis of the shoulder: A systematic review and network meta-analysis. The American journal of sports medicine, 47(14), 3552-3560.

Kitridis et al (2019)  presents a systematic review and meta-analysis looking at high quality studies. The eligibility criteria is reasonable with 30-studies selected. Kitridis studied numerous pharmacological interventions in the management of AC from analgesia to more invasive corticosteroid injections (CSI). The primary outcome of the study was pain intensity reduction which was measured by self-administered questionnaires to patients with follow-up periods > 6 months. The conclusion from the study was CSI, either administered alone or after-distention of the shoulder capsule, provided a clinically meaningful improvement in the short-term. Although this is a robust study with a large sample size and a reasonable follow-up period,  there are several limitations; inconsistencies in some of the analyses, numerous interventions which diminish the statistical power of the result, variability in the diagnosis of AC, not taking placebo effect(s) into account (natural disease progression), missing pre-study ROM deficit outcome and almost half of these studies did not report adverse effects. In conclusion, this study does comply with the current best practice in New Zealand(NZ) amongst physicians. If a CSI is to be offered, it should be carried out during the early stages of AC as the role of CSI diminishes in the later stages

Koh, K. H. (2016). Corticosteroid injection for adhesive capsulitis in primary care: a systematic review of randomised clinical trials. Singapore medical journal, 57(12), 646.

Author Koh (2016) provides a systematic review comparing CSI with placebo, non-steroidal anti-inflammatories (NSAID) and physiotherapy in the management of AC. This was a rigorous study using the Jaded scale with only 10-RCTs included from a broad search using a reviewer selection.  Primary outcomes included pain using visual analog scale (VAS), ROM and shoulder function. This study’s finding has corroborated the results of Kitridis et al (2019), i.e. CSI provides benefit in the short term. This study was also well constructed in terms of examining secondary outcomes which included, among others, the adverse events. Due to the strict criteria in the literature search, some studies may have been missed which, in turn, may lead to bias. There was variability in the inclusion criteria of the studies. Most of the studies utilised co-interventions with home-exercise programmes which included passive ROM and pendulum-exercises within pain limits. Only two studies have no co-interventions. The follow-up period for some of these studies are short. Furthermore, the difficulty in blinding patients, e.g. with injections vs physiotherapy could lead to bias. Despite some of the limitations, the RCTs chosen are of decent standard and provide some evidence for the use of CSI.

*Page, M. J., O’Connor, D. A., Malek, M., Haas, R., Beaton, D., Huang, H., . . . Shea, B. (2019). Patients’ experience of shoulder disorders: a systematic review of qualitative studies for the OMERACT Shoulder Core Domain Set. Rheumatology, 58(8), 1410-1421.

Page and colleagues present a systematic review of qualitative studies by searching a wide range of clinical databases. Two authors independently screened these studies and employed a robust Critical Appraisal Skills Programme checklist. Eight studies met the criteria. The authors used the illustrative quotes from patients to form seven main themes: pain, function, participation restriction, sleep disturbance, cognitive dysfunction, and emotional distress. These were semi-structured focus-group method. The main message conveyed by the authors is that most clinical trials tend to look at pain outcome to evaluate interventions. Other measures such as work absence are usually ignored. The main strength of this study is that the method was pre-specified clearly. Some limitations identified are selection bias by the authors, risk of bias in reporting, and the study included a few “umbrella terms” shoulder conditions (not just AC). In summary, there is lack of clarity in best practice from the current research where clinicians might be oblivious and concentrating on reducing pain which may not be helpful in addressing the elephant in the room, in contrast to the Kitridis(2019) and Koh(2016) et al’s studies. There is still room for further research to explore experiences people contend with in shoulder pain.

Prestgaard, T., Wormgoor, M. E., Haugen, S., Harstad, H., Mowinckel, P., & Brox, J. I. (2015). Ultrasound-guided intra-articular and rotator interval corticosteroid injections in adhesive capsulitis of the shoulder: a double-blind, sham-controlled randomized study. Pain, 156(9), 1683‐1691. doi:10.1097/j.pain.0000000000000209

Prestgaard et al present a double-blind, sham-controlled, intention-to-treat RCT. The jury is still out in terms of CSI optimal dose and needle placement. The primary aim of this study was to measure pain reduction comparing ultrasound-guided intra-articular injection and combined intra-articular and rotator-interval injection. Pain reduction differences was evident up to 12 weeks in both the CSI groups but not at week 26. There were no significant differences between the CSI groups at any time. The statistical methods used to measure the data is quite robust and the exclusion criteria is thorough. The study technique is carefully planned with two injections of equal volume, number, and needle placement in all three groups. The injections were administered by two well-experienced physicians who were not part of the result interpretation, ergo minimizing bias. Outcome measures were well-validated. The author acknowledged that expectation could have been a possible confounder. The study does not include psychological factors and work absenteeism which Page (2019) et al stressed.  This study agrees with the aforementioned systematic reviews by Koh (2016) and Kitridis (2019) et al, i.e. CSI is of benefit in the early stages of AC. The benefit of CSI wears-off during the latter stages of AC.

Sharma, S. P., Bærheim, A., Moe-Nilssen, R., & Kvåle, A. (2016). Adhesive capsulitis of the shoulder, treatment with corticosteroid, corticosteroid with distension or treatment-as-usual; a randomised controlled trial in primary care. BMC musculoskeletal disorders, 17(1), 232.

This was an intention-to-treat prospective randomized trial by Sharma and colleagues (2016) done in Norway. 106 patients were randomised into three groups; CSI with Lidocaine (IS), CSI and additional saline as distension (ISD) and treatment-as-usual (TAU) as the control group. There were no statistically significant differences between both intervention groups. Inversely there were statistical difference between the intervention groups and TAU group. The follow-up rate was good. This was a sound study, in terms of inclusion/exclusion criteria and generalizability. Nevertheless, this study might be a trap to readers in terms of bias. The injections were administered by the author himself. Deciphering this paper shows some baseline characteristics discrepancy where the ISD group had four times more previous trauma to the shoulder, higher analgesia use in the IS group and the TAU group has double the number of smokers. Although the injections done were by the posterior approach, 62% patients from the injection groups guess the right intervention group. There was no standardization in the injection volume. The conclusion from this study is similar to Kitridis (2019) and Koh’s (2016) et al’s findings. Albeit caution should be exercised when interpreting this study as the clinician was unblinded to the patients.

Sun, Y., Lu, S., Zhang, P., Wang, Z., & Chen, J. (2016). Steroid injection versus physiotherapy for patients with adhesive capsulitis of the shoulder: a PRIMSA systematic review and meta-analysis of randomized controlled trials. Medicine, 95(20).

Sun et al (2016) provide a systematic review in line with the Preferred Reporting Items for Systematic Reviews and Meta-analyses(PRISMA) check list (Moher, Liberati, Tetzlaff, Altman, & Group, 2009).  Nine RCTs are included assessing the effectiveness CSI and physiotherapy in AC. The outcomes looked at were clear using Shoulder Pain and Disability Index (SPADI) and secondary outcomes of pain relief, passive external rotation, and adverse effects. Inclusion criteria is unambiguous.  Statistical analyses employed are suitable. The study did not report any superiority of steroid injection over physiotherapy for functional improvement. There is a good follow-up period at three points. Using the PRISMA checklist, scores ranged from 3 to 8 out of 10 suggesting heterogeneity of studies. Patients was also included at different stages of AC. Co-interventions in the studies are not elaborated on. There is no cost-effective analysis in terms of one CSI versus physiotherapy sessions.  Given the nature of this study question, risk of bias arises with both clinicians and patients are not blinded to interventions.  We can postulate CSI is safe and both CSI and physiotherapy have similar benefits at long term. One CSI in the early stage for pain management might be reasonable in AC in alignment with Kitridis et al’s study (2016).

**Yuan, X., Zhou, F., Zhang, L., Zhang, Z., & Li, J. (2018). Analgesic effect of extracorporeal shock wave treatment combined with fascial manipulation theory for adhesive capsulitis of the shoulder: a retrospective study. BioMed research international, 2018.

The prevailing view of the myofascial system as a 3-dimensional continuum has led research into fascial manipulation (FM).  Yuan and colleagues (2018) present a retrospective study comparing extracorporeal shock wave therapy (ESWT) combined with the theory of FM to traditional local ESWT (L-ESWT). 34 patients were included. All patients underwent five sessions over a period of five weeks. Statistical analysis is standard with SPSS software and ANOVA one-way analysis of variance employed. Pain scores and ROM were recorded at baseline, after the first treatment and at the fifth treatment. Numerical rating scale showed statistically significant improvement in both groups with ESWT- FM group showing larger improvement. The paper concluded that ESWT-FM provided faster pain relief and a slight improvement in function compared to L-ESWT.  The baseline clinical characteristics were well presented with similarity in both groups. The inclusion/exclusion criteria stated were ambiguous. The author acknowledged this retrospective design study cannot produce the same high caliber evidence as a double-blind RCT. The sample size was small. Besides Kitridis (2019) paper, there is not much literature comparing ESWT-FM with interventions such as CSI and physiotherapy. In conclusion, further research in ESWT-FM is required with high quality study design and larger patient population.

Conclusion

There are still a few differing schools of thought in terms of clinical diagnosis and universally effective treatment option for AC (Le et al., 2017). There is a clinical realm to educate our clinicians and patients the complexity of AC. Our knowledge about neural circuits and pain mechanisms remain rudimentary. The traditional biomedical approach to treat the underlying pathology is invariably the holy grail of most clinicians. Although this may be plausible, this is not always the case in AC.  As we face more pain patients in our aging population, it is imperative we continue researching and develop novel coping mechanisms to help patients with AC. Patients with shoulder disorders may need to contend with several disruptive experiences in life and societal burden. Although, this annotated bibliography further reinforces the use of CSI  in the early stage of AC to reduce the occurrence of synovitis and limiting development of capsular fibrosis, we are still left without a definitive treatment guideline (Koh, 2016). Physiotherapy in the painful phase might be a difficult task for most patients. Accordingly, a concerted action of prescribing CSI to ameliorate initial pain, with physiotherapy might be advocated to restore ROM. It would be fair to say that AC is a self-limiting disease and its natural progression tends to lead to pain reduction and improvement in ROM. For now, combined initial CSI and physiotherapy might confer greater improvements in pain scores and ROM than either treatment alone for the treatment of AC.