Efficacy and Tolerability of Peritendinous Hyaluronic Acid in Patients with Supraspinatus Tendinopathy: a Multicenter, Randomized, Controlled Trial

Tendinopathy is a common injury in athletic populations, secondary to overuse [1, 2]. However, it may also occur in the overall population as a result of repetitive or excessive loading, and abnormal or unusual movements [3]. Also, metabolic- and hormone-related clinical conditions such as diabetes, menopause, and adiposity have been identified as systemic risk factors for tendinopathies [4–6]. Although the incidence of tendinopathies in the overall population has not been established, different authors have reported a rising trend associated with increasing interest in sport activities in high-income societies [1, 2, 7].

The first-line pharmacological treatment for tendinopathies is most often based on non-steroidal anti-inflammatory drugs (NSAIDs), which are useful for pain control but may cause patients to ignore warning symptoms, resulting in further damage to the affected tendon [3, 8]. Intratendinous injections of corticosteroids are also common in the standard management of tendinopathies and appear to successfully reduce inflammation and pain at short-term; however, the risk-benefit ratio of their use in the treatment of tendinopathies is currently controversial [7, 9]. Owing to the limitations of pharmacological treatments in modifying the structure of the tendon, the management of tendinopathies usually includes other non-pharmacological interventions with proven benefits in tendon recovery such as relative rest, cold, ultrasound, and physical therapy, specifically eccentric style exercises and stretching to prevent stiffness [1, 3]. Other non-pharmacological interventions proposed include shock wave therapy, iontophoresis, sclerotherapy, and nitric oxide patches; however, results are not very consistent and multicenter trials are needed to confirm their efficacy. Among all of the physical therapeutic interventions, eccentric strengthening has shown remarkably good therapeutic outcomes in the treatment of various tendinopathies such as those affecting the patellar, Achilles, supraspinatus, and wrist tendons [1, 10–14].

Among the recent advances in the management of tendon disorders, the peritendinous administration of hyaluronic acid (HA) has shown promising results in clinical trials including patients with tennis elbow [15], patellar tendinopathy [16], Achilles tendinopathy [17, 18], and various disorders involving tendons in the rotator cuff [19–22]. HA is the primary component of synovial fluid and provides the joint with lubrication and shock absorption [23]. Although the mechanisms of action in the treatment of tendinopathies are not well established, peritendinous injection of HA has shown to reduce tendon adhesion, provide mechanical protection, and upregulate the vascular endothelial growth factor and type IV collagen, resulting in the acceleration of tendon healing [24–26]. In the particular case of tendinopathies involving the rotator cuff, the use of HA injections led to a significant improvement of shoulder function in all published trials [19, 20, 22, 27, 28]. However, the superiority of HA over other interventions in scales assessing pain is controversial, and few studies have evaluated the efficacy of HA in the treatment of lesions affecting a particular tendon. In this regard, some authors have highlighted the need for further randomized trials to better establish which grade of lesions may benefit most from peritendinous injection of HA [27].

In this parallel-group, randomized, controlled trial, we investigated the efficacy and safety of peritendinous injection of HA in patients with persistent supraspinatus tendinopathy. To this end, we compared the therapeutic outcome of treatment with HA as an adjuvant to physical therapy with that of physical therapy as sole therapeutic intervention.

In this multicenter, parallel-group, randomized trial including patients with supraspinatus tendinopathy, we found that both physical therapy alone and physical therapy combined with subacromial HA injections had good tolerability and resulted in significant pain reduction and successful functional recovery. When compared to physical therapy only, patients treated with physical therapy and HA returned significantly earlier to work and needed fewer rehabilitation sessions.

The study groups were well balanced regarding the baseline demographic and clinic characteristics, including the prevalence of the affected shoulder, which was mostly the right one, as expected for an overuse injury. NSAID consumption was greater in the HA group; however, no significant differences arose between the two groups. Likewise, we did not observe significant differences in the baseline scores of the assessed scales, except the TSK, which was significantly higher in the HA group. The TSK has been little used in the assessment of HA treatment efficacy and measures the perceived threat of movement and expected pain in injured patients. In this regard, the fear of an injected treatment (not existing in the control group, as no placebo treatment was administered) could explain the discrepancy between the different TSK score and the homogeneity in other baseline measures of pain, such as the VAS score or NSAID consumption. Overall, the baseline scores of the TSK, VAS, and ADL scales revealed a moderate severity of the tendinopathy in our study patients.

After 90 days of follow-up, patients in both groups experienced a significant improvement in the VAS, ADL, and TSK scores, irrespective of the intervention received. This result is consistent with previous studies proving that treatment based solely on eccentric strengthening is sufficient for mid-term improvement in pain and function [1]. When compared with physical therapy only, we found that the addition of treatment with subacromial HA significantly reduced the number of rehabilitation sessions and the treatment days needed for recovery. The reduction in the TSK score during the follow-up period was also greater and occurred earlier in the HA group, although the baseline differences between scores in each group might have influenced the differences observed during follow-up. In line with the better therapeutic outcome observed in the HA group, both patients and physicians rated significantly higher the efficacy of the combined treatment of physical therapy and subacromial HA than physical therapy only. In the case of physician perception, combined treatment was rated significantly higher than physical therapy from the first visit (day 7).

In light of the differences observed between study groups in the indirect measures of efficacy, we would also expect more remarkable differences between groups in treatment outcomes related to pain and the ability to perform ADL. However, NSAID consumption and the assessment of the VAS and ADL scores did not reveal significant differences between patients treated with physical therapy only and those treated with physical therapy and subacromial HA injections. Of note, as a clinical study performed on a medical device, the doses of concomitant medicines were not recorded, and therefore, no dose adjustment was performed in the comparative analysis. Considering the proven efficacy of physical therapy in the treatment of tendinopathies of the rotator cuff, it seems reasonable that a larger sample size would provide clearer differences between groups. Nevertheless, the superiority of HA treatment in pain improvement is unclear, and results in previous studies assessing the efficacy of HA injections in patients with tendinopathies affecting the rotator cuff are inconsistent. Merolla et al. found that HA monotherapy significantly reduced pain in more visits than physical therapy only [19], whereas Sengul et al. reported similar pain scores in patients receiving both interventions [22]. Similar discrepancies have been found when investigating the efficacy of a combined treatment using HA injections and physical therapy. Thus, Meloni et al. concluded that rehabilitation exercises and HA injections were significantly more efficacious in pain reduction than the same exercises accompanied with placebo injections containing sodium chloride [20]. On the other hand, Ozgen et al. compared the efficacy of HA injections with that of physical therapy and found no differences in pain improvement in most of the follow-up visits [28]. Of note, the baseline VAS scores reported by Ozgen et al. in the study sample were remarkably small (below 1), which may have contributed to narrow the differences between groups. On the other hand, the ADL score has been barely assessed in trials investigating the efficacy of injected HA in patients with tendinopathies of the rotator cuff. In our study, all participants received clear instructions not to undertake physical labor with the affected arm until recovery. Such indication, not included in pivotal trials assessing the reliability of the ASES score [30, 31], allowed us to limit the bias related to the different jobs and activities performed by each patient; however, it was likely to influence the overall ADL score, as patients may have limited their day-to-day activities, regardless of their functional recovery and pain.

In line with safety data previously reported in the literature [19, 20, 22], both interventions were safe and well tolerated, with a 12% rate of adverse events rate—all of them mild—, mostly reported by patients in the control group. Furthermore, both patients and physicians rated tolerability over 2.5 points on a 4-point Likert scale during the entire follow-up. Perceived tolerability progressively increased throughout the follow-up period, and no significant differences were observed between the study groups.

The main limitation of our study is the lack of control groups. Unlike other studies, aimed at investigating the efficacy of HA injections alone, our objective was to compare the combined treatment with the gold-standard for the management of tendinopathies. The presence of a placebo group with non-active peritendinous injections might have limited possible biases, particularly in patient-reported efficacy. Accordingly, the inclusion of a sham physical therapy would have yielded more extreme differences between groups. However, including more than two intervention groups would have dramatically reduced the size of each group. In fact, the final sample size (84 patients) was larger than that in most studies assessing the efficacy of HA injections in tendinopathies of the rotator cuff [27], but it still limited the observation of significant differences in some scales. Along the same lines, the open-label HA administration may introduce a bias in both the physician and patient efficacy assessment. Finally, due to the sample size limitations, all centers were considered as a whole and no cluster analyses were performed to explore the center bias.

Despite the limitations of our study, the results obtained in most of the assessed scales are consistent and support the use of subacromial HA injections as adjuvant treatment to physical therapy in the management of supraspinatus tendinopathy. In our experience, treatment based only on physical therapy was sufficient to reduce pain and improve function. In addition, the combination of physical therapy with subacromial HA reduced the number of rehabilitation sessions and recovery time. These findings suggest that combined treatment with physical therapy and HA may not only increase patients’ quality of life but it might also benefit the healthcare system and society by reducing rehabilitation costs and time off work.