CORR Insights®: Is the Anterior Injection Approach Without Ultrasound Guidance Superior to the Posterior Approach for Adhesive Capsulitis of the Shoulder? A Sequential, Prospective Trial

Where Are We Now?

Adhesive capsulitis, also known as frozen shoulder, is a chronic condition that often presents with marked limitation in both passive and active shoulder ROM. It’s estimated that adhesive capsulitis affects up to 5% of the population, with a predilection for women. The nondominant shoulder is affected in more than 50% of patients with the condition, and the contralateral shoulder is subsequently affected in 30% of those with the diagnosis [5].

Adhesive capsulitis is idiopathic, but there are several risk factors associated with its development. Endocrinopathies such as diabetes mellitus, hypothyroidism, hyperthyroidism, rheumatoid arthritis, and cardiovascular diseases have been implicated in the development of adhesive capsulitis. The pathophysiology of adhesive capsulitis may involve both inflammatory and fibrotic pathways, which result in glenohumeral joint capsular contracture with loss of volume [1, 6].

Clinical manifestations of adhesive capsulitis vary depending on the stage of the disease [8]. However, the hallmark of the disease is global loss of active and passive shoulder ROM. Stage 1, the inflammatory stage, is characterized by an intense, painful inflammatory reaction. This is followed by Stage 2, the frozen stage, during which the shoulder becomes stiff. The third stage, the thawing stage, includes a gradual improvement in shoulder motion. The duration of adhesive capsulitis varies, and typically ranges from 1 to 3 years. Before diagnosing adhesive capsulitis, the clinician may obtain both plain radiographs and advanced imaging in order to exclude other potential diagnoses such as osteoarthritis or inflammatory arthropathies.

Treatment management for Stage 1 often consists of NSAIDs, supervised physical therapy, and intraarticular corticosteroid injections, though several studies recommend early intervention with intra-articular corticosteroid injections in the early inflammatory or frozen stages of adhesive capsulitis [2, 10], and this may be more efficacious than other approaches [10, 14, 16]. Sun et al. [14] demonstrated that intraarticular corticosteroids improve function, decrease pain, increase ROM, and may last as long as 26 weeks. However, it appears that intraarticular corticosteroids do not shorten the course of the disease. In a meta-analysis by Wang et al. [15] on four randomized trials, intraarticular corticosteroid injections demonstrated better pain relief when compared with placebo up to 8 weeks, but there were no differences from 9 months through 2 years. However, intraarticular corticosteroids resulted in superior glenohumeral ROM at all time points up to 2 years [15]. Other studies have compared intraarticular with oral corticosteroids. In a randomized trial by Lorbach et al. [7], intraarticular corticosteroid injections demonstrated superior results in shoulder scores up to 12 months and ROM up to 8 weeks compared with oral corticosteroids, but they found no differences in pain scores. The location of intraarticular injection has also been another point of debate. In two randomized studies, there was no difference in pain relief when comparing intraarticular corticosteroids with subacromial injections [9, 12]. In terms of corticosteroid dosage, a triple-armed randomized trial by Yoon et al. [17] found no difference in outcomes in pain and passive ROM between 40 mg of triamcinolone versus 20 mg when used in intraarticular corticosteroid injections.

The current sequential prospective study by Rijs et al. [11] demonstrated an effective technique for intraarticular corticosteroid injections for adhesive capsulitis without the aid of imaging. The authors studied 95 patients comparing a new anterior approach versus posterior approach for intraarticular injections using anatomic landmarks by two experienced surgeons. Their accuracy was 94% compared to 78% for the posterior approach. The study demonstrates that the landmark-based approach recommended by the authors has comparable accuracy to ultrasound techniques. Ultrasound has been used for intraarticular corticoid injections to increase the accuracy of the injection. However, ultrasound machines can be costly, and many offices do not have ultrasound machines. It can also be time consuming either by increasing time of procedure or referral to radiology for the intraarticular injections. The anterior approach in the current study would help shift away from ultrasound-guided injections and move to landmark-based injections.

Where Do We Need To Go?

The authors demonstrated the accuracy and the cost effectiveness of their technique, but there are still questions that need to be answered in future studies. We must address the importance of accuracy and whether this affects the clinical effectiveness of intraarticular corticosteroid injections in patients with adhesive capsulitis of the shoulder. An accurate corticosteroid injection should have a significant local effect. In cases with an inaccurate injection, it may still be close enough to elicit a similar effect or the possibility of a secretory effect on the inflammation of the capsule. Cho et al. [4] found that ultrasound was much more accurate than anatomic-based injection. However, they found no major differences in pain and outcome between both techniques. Other studies have questioned the location of the injection [3, 4, 13]. Chen et al. [3] showed intraarticular injections had improved outcome for pain relief compared to subacromial injection but no difference in function. Finally, a randomized control trial that used three different locations (rotator interval, intraarticular, and subacromial) showed that injections in the rotator interval yielded the best results followed by intraarticular injections for pain and passive ROM. In the subacromial group, there was a decrease in pain but a loss in passive ROM [13]. Pathological changes in the rotator interval occur in the early phase of adhesive capsulitis and it may explain the significant improvement with an early corticosteroid injection [6, 13]. There should be follow-up studies to determine the most efficacious location. Accuracy of the injection is not important until we know the exact location to inject.

Also, the authors mentioned that both surgeons in the current study were experienced with performing both the anterior and posterior approach for glenohumeral injections. It would be of benefit to determine whether less-experienced surgeons performing these techniques can achieve similar accuracy [11].

How Do We Get There?

Future prospective studies could be designed with surgeons who have not used this approach to determine the accuracy and learning curve among them. This would be important for inexperienced surgeons to determine their accuracy and if they should use an ultrasound-guided technique. The more accurate, the less likely an ultrasound would be needed. Like previous study designs [13] that used a posterior approach and demonstrated the location of the injection with ultrasound imaging and then determined the efficacy of each location, a larger prospective study could be designed with several different approaches and determine location again with imaging techniques. The purpose of the study would be to determine the most efficacious location for injection: subacromial, capsular (rotator interval), or inarticular. If the outcomes of decreased pain and improved ROM are similar, then a landmark-based approach would be best. It would save time and cost of imaging techniques.

As mentioned, a previous study [4] demonstrated that accuracy was not important since the results were the same even when there was a difference in accuracy. Future studies could determine whether the accuracy of the placement of the needle correlates with clinical outcomes.

Once we know the most efficacious location, then the next step would be to determine how accurate we need to be for the best clinical outcomes. A future propective study could determine the exact location of the needle with ultrasound and it’s distance from the intended location. We could then determine the clinical outcomes of an accurate corticosteroid injection that is having a local effect versus an inaccurate injection possibly having a secretory effect on inflammation [4].