JORT Journal

Introduction

Adhesive capsulitis of the glenohumeral joint (AC), also referred to as frozen shoulder, represents one of the most disabling arthropathies of the shoulder [1]. Joint pain and functional disability severely impact the patient’s physical and psychological health, as well as social well-being [2]. Based on etiology, AC can be classified as primary (idiopathic and usually with a gradual onset) or secondary (e.g., resulting from shoulder trauma or surgery) [2]. Although AC pathogenesis is not fully elucidated, intra-articular inflammation and the following reactive fibrosis are the main features, progressively leading to stiffness and reduced range of motion [3]. Like degenerative joint disorders (i.e., osteoarthritis) [4], AC is featured by an increase of immune-inflammatory cells within the joint, including mast cells and macrophages [5,6]. If inadequately controlled by pro-resolving responses, the excessive intra-articular release of immune cell mediators (i.e., cytokines, growth factors, and lytic enzymes) may drive inflammation and fibrosis at the synovial and whole joint level [6-8]. Oxidative stress has long been recognized as a key player in a number of musculoskeletal diseases [9] and is currently considered an important factor in AC pathogenesis [10]. The excessive generation of reactive oxygen and nitrogen species is thought to be responsible not only for cellular damage, but also for initiating the oxidative degradation of biomolecules, like the polysaccharide hyaluronic acid (HA) [11-13]. Indeed, higher levels of low molecular weight HA have been observed in diseased compared to healthy joints [14,15]. It is thus conceivable that, upon injection in the oxidative environment of AC shoulders, HA would be degraded into smaller fragments with negative consequences on intrinsic viscosity and lubricating properties [16]. 

This may account for the limited treatment effect of HA intra-articular injections in AC patients, mainly with respect to the short duration and pain relief [17,18]. Although self-limiting, AC generally resolves within 18 to 30 months [2,19]. The AC patient experiences a long condition of severe functional limitation and pain, accordingly. Indeed, current medication with nonsteroidal anti-inflammatory drugs (oral) and corticosteroids (either oral or intra-articular injections), provide pain relief in the short-term, while long-term outcomes are still unsatisfactory [20]. Oxidantprotected HA may thus be of help, similar to what was previously suggested for rheumatoid arthritis [21]. Adelmidrol (ADM) is the international nonproprietary name of an azelaic acid derivate and an analogue of palmitoylethanolamide (PEA) [22]. PEA is a naturally occurring lipid amide exerting a local, autacoid, antiinjury function via mast cell down-modulation, as originally speculated by the Nobel laureate Rita Levi Montalcini [23] and later repeatedly confirmed [24-27]. ADM was shown to increase the endogenous levels of PEA [28], the latter being reduced in the course of joint diseases [29]. Recent evidence supports the in vivo antioxidant activity of ADM in animal models of colitis [30], ischemia/reperfusion injury [31] and pulmonary fibrosis [32], but the antioxidant protective effect has not yet been investigated at the molecular and cellular level. Recent preclinical and clinical studies have shed some light on the benefit of a new intra-articular formulation based on high molecular weight HA (1%) and ADM (2%) in osteoarthritis [33-35].

The aim of the present study was twofold. First, to gain new information on the antioxidant effect of ADM and its ability to protect HA against oxidative depolymerization. Second, to report our clinical experience on two patients with primary and secondary AC respectively, managed through a multimodal approach including tailored physical therapy and intra-articular injections of HA and ADM.

Material and Methods

In vitro Chemical Study

Materials

Sodium hyaluronate and Adelmidrol solution of the same standard quality as the commercially available formulation (Hyadrol^®, factory code 6301) was kindly donated by Epitech Group S.p.A., Milan, Italy. Hyadrol^® is a steam sterilized solution for intra-articular use in the form of ready-to-administer prefilled syringes and contains high molecular weight HA (1%, molecular weight 1300-2000 kDa), ADM (2%), sodium chloride, potassium dihydrogen phosphate, sodium hydrate, EDTA disodium and water. A similar solution was prepared with the exception of ADM, and thereafter referred to as solution A. Sodium chloride, potassium dihydrogen phosphate, sodium hydrate and EDTA were supplied by ACEF S.p.A. (Fiorenzuola d’Arda, Piacenza, Italy). Copper sulphate and hydrogen peroxide (30%) were purchased from Merck KGaA (Darmstadt, Germany). High molecular weight sodium hyaluronate (molecular weight 1500 kDa, batch # PH14202C) obtained from bacterial fermentation was provided by HTL Biotechnology (Javene, France). 

Hyaluronan Oxidative Degradation

Three aliquots of solution A (100 mL each) were collected (i.e., A1, A2 and A3). A1 was conserved as such (control solution). A volume of 170 µL CuSO₄ (0.1M) and 25 µL of EDTA (0.22 mM) was added to A2 and A3, stirred for 30 s and incubated at 37°C for 24 h, as previously described [36]. ADM (2 g) was then added to solution A2, in order to achieve the final 2% concentration of Hyadrol^®. A volume of 1 mL H₂O₂ (30%) was added to both A2 and A3, followed by 30-s stirring. In this condition, hydrogen peroxide (H₂O₂) reacts with cupric ions (Cu²⁺) generating hydroxyl radicals responsible of HA oxidative degradation [36].

Rotational Viscometry

Degradation of HA was assessed by the changes of the dynamic viscosity [37]. A rotational viscometer was used (Brookfield RVDV-II+P, Brookfield Engineering Laboratories, Inc., Middleboro, MA, USA). Dynamic viscosity of the sample solutions was measured at 37°C for up to 4h and expressed in centipoise (cP), with one cP being one millipascal-second (mPa·s). The viscometer was equipped with spindle N. 3, rotation speed 50 rpm.

In vitro Cell Study

Culturing and Treatment

The monocyte/macrophage J774-A1 cells (8x10⁵ cells) were plated into 24 well tissue culture plates and grown to confluence (48h at 37°C). Subsequently, cells were washed twice in 1X phosphate buffered saline (PBS) and cultured in F-12/ FCS 1% for the period of the experiment. After 6h pre-treatment with Adelmidrol (0, 10, 50, 100 μM) cells were stimulated with hydrogen peroxide (H₂O₂, 200 μM) for 10 min. At the end of each experiment, the assays described below were performed.

Cell Protection against Oxidative Damage

To assess cells viability, cells were incubated at 37°C with 0.2 mg/mL MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) for 1 h and the resulting solution was measured for optical density (OD) at 550 nm (OD550) using a microplate reader, as previously described [38].

Measurement of Nitrite levels

Total nitrite level, as an indicator of nitric oxide (NO) synthesis, was measured in the supernatant. Briefly, nitrate in the medium was reduced to nitrite by incubation with nitrate reductase (670 mU/mL) and β-nicotinamide adenine dinucleotide 3-phosphate (160 mM) at room temperature for 3 h. Total nitrite concentration was then measured using the Griess reaction by adding 100 μl Griess reagent (0.1% (w/v) N-(1-naphthyl) ethylenediamine dihydrochloride in H₂O and 1% (w/v) sulfanilamide in 5% (v/v) concentrated H₃PO₄; volume 1:1) to the 100 μl sample. OD550 was measured using an enzyme-linked immunosorbent assay microplate reader (Tecan, Männedorf, Switzerland). Nitrite concentrations were calculated by comparison with OD550 of standard solutions of sodium nitrite prepared in H₂O, as previously described [39].

Determination of intracellular ROS

A chemically reduced, acetylated form of fluorescein (i.e., carboxy-H2DCFDA dye) was used as an indicator for reactive oxygen species (ROS) in cells, as previously described [40,41]. Briefly, cells were trypsinized after treatment and then washed twice with 1x washing buffer. Cells were then incubated with carboxy-H2DCFDA (10 µM final concentration) at 37°C in the dark for 30 min. The fluorescence microplate reader detected the light emission.

Statistical analysis

All values from the in vitro cell study are expressed as mean ± standard error of the mean (SEM). The results were analyzed by one-way analysis of variance (ANOVA) followed by a Bonferroni post-hoc test for multiple comparisons. A P-value of less than 0.05 was considered significant.

Case Presentation

Case 1

A 57-year-old woman visited the Functional Recovery and Rehabilitation Unit at the Belluno Hospital because of pain worsening at the left shoulder. Pain was long-lasting (more than 3 months) and intermittent at rest and at night, severely interfering with sleep. Supraspinatus tendinopathy was suspected on musculoskeletal ultrasound and confirmed with magnetic resonance (Figure 1).

Figure 1: MR of the left shoulder. Supraspinatus tendinopathy, with tendon alteration and thickening, in sagittal (A), frontal (B) and coronal projections (C) is evident.

No further abnormalities were displayed, not even on X-ray investigation. At clinical evaluation, the patient presented with severe stiffness of the shoulder, internal rotation limited to 10-15 degrees, no external rotation and severe pain at shoulder movement in any direction, i.e., 9 out of 10 on the Numerical Rating Scale (NRS). Abnormal scapular motion with moderate dyskinesia was also evident. Diagnosis of primary adhesive capsulitis with supraspinatus tendinopathy was made. For antalgic purposes, a single intra-articular infiltration with triamcinolone hexacetonide (40 mg\mL) and lidocaine (20 mg\mL) was performed. A prescription for an initial 10-session physical therapy (PT) was issued, in order to reduce pain and preserve shoulder mobility and function. The patient attended a 40-minute PT session every other day, consisting of active assisted mobilization and passive gentle traction of the left shoulder (“pompage”). Active assisted mobilization was performed during the acute phase to promote soft tissue stretching and joint lubrication (by stimulating hyaluronic acid production), as well as to restore proprioception and ROM. Passive PT was carried out through progressive mobilization techniques tailored to the patient’s response. Low-intensity continued ultrasound (LICUS) was also used for both pain and stiffness relieving purposes (41). In each PT session, LICUS lasted 12 minutes at 1 MHz, 1.5-1.7 W/cm² intensity range (HC SOUND, Elettronica Pagani, Paderno Dugnano, Milan, Italy).

Case 2

A 71-year-old woman was referred to the Functional Recovery and Rehabilitation Unit of the Belluno Hospital because of complications after a proximal humerus fracture (Figure 2A).

Figure 2: Orthogonal radiographs of the left shoulder - (A) Acute proximal humeral fractures at the surgical neck (black arrows) and greater tuberosity level (red arrow). (B) Evidence of initial consolidation after 30-day shoulder brace immobilization.

The fracture was caused by a fall from standing height and was conservatively managed with a shoulder brace for 30 days.

Thereafter, an acceptable consolidation was seen at radiographic examination (Figure 2B) and the patient was referred to our Rehabilitation Unit. At the first rehabilitation visit, the patient held her left arm in marked adduction and internal rotation. An extensive hematoma in the left upper limb was also present. Pain at NRS was scored 9 and ROM was very limited in either abduction, internal rotation, external rotation, or flexion. Post-traumatic / post-immobilization shoulder stiffness was initially diagnosed. A 10-session PT on an every-other-day regimen was recommended. Each session lasted 40 minutes and consisted of passive PT and magnetic field therapy. Passive PT was performed through relaxation and alignment techniques to preserve the patient’s psychophysical well-being and correct poor posture responsible for functional limitations. Passive mobilization to increase joint mobility was also performed. Moreover, 25-minute magnetic field therapy per session was delivered at 30 Gauss and 100 Hz (MT- Chinesport Italia, Udine - Italy) with the aim of speeding up consolidation.

Treatment Monitoring

For both patients, treatment effect was monitored at each control visit by measuring:

(i) pain severity, assessed by an 11-point scale (0 meaning no pain and 10 meaning the worst pain imaginable), i.e., the Numeric Rating Scale (NRS) [42]; (ii) upper limb function, assessed through the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire, ranging from 0 (no disability) to 100 (most severe disability) [43]. DASH measures impairment, activity limitations and participation restriction [44], allowing the patient to rate difficulty and interference with activities of daily living (ADL).

Results

Adelmidrol Protects Hyaluronan against Oxidative Degradation

The kinetics of the hyaluronan oxidative degradation was monitored by changing the dynamic viscosity of the solution. The experimental oxidative system (CuSO₄ + H₂O₂) effectively induced HA degradation as shown by the fast decrease of dynamic viscosity in solution A3 (Figure 3, red curve). As evident from the blue curve in Figure 3, the oxidative depolymerization of HA was greatly counteracted by ADM. At 30 min, more than 50% inhibition of oxidative degradation was observed.

Figure 3: Time-dependent changes in dynamic viscosity of HA solutions exposed to oxidative degradation initiated by cupric ions (Cu²⁺) and hydrogen peroxide (H₂O₂). In 30 min, viscosity dropped from 450 to 0 cP without the addition of ADM (red curve, A3 solution), while it was higher than 50% of the original value with the addition of ADM (blue curve, A2 solution). Grey dotted line represents 50% viscosity. Black curve represents the control HA solution, unexposed to oxidative degradation (A1 solution). Data represent the mean ± SEM of three independent experiments. See Materials and Methods for further details.

Adelmidrol protects macrophages against oxidative damage and decreases free radical production

As shown in Figure 4A, pre-treatment with ADM showed a dose-dependent protective effect against H₂O₂ oxidative stress, preserving cell viability. Notably, the lowest concentration (10 μM) was significantly effective and was used in the following experiments. Cell stimulation with H₂O₂ increased the levels of NO₂^-, with the effect being significantly counteracted by ADM (Figure 4B). Similar results were observed for the intracellular ROS production (Figure 4C).

Figure 4: Antioxidant activity of Adelmidrol on H₂O₂-stimulated monocyte/macrophage J774-A1 cells. The antioxidant activity was measured as (A) cells protection against H₂O₂^-induced cell death, (B) nitrite release and (C) intracellular ROS production. The lowest dose of Adelmidrol that was effective in counteracting H₂O₂-induced cell death (i.e., 10 μM) was used in the nitrite (B) and ROS assays (C). * P < 0.001 vs CTR; ° P < 0.05 vs H₂O₂. CRT = control; ADM = Adelmidrol.

Case Report

Case 1

At the first check-up (i.e., after 4-week PT), a slight improvement in night pain was reported (NRS 8). Stiffness and pain upon active and passive movements were still severe. The DASH score was 67.5. A further 10-session PT (same type, duration, and frequency as the previous one) was planned. Briefly, manual capsule stretching, antalgic maneuvers and massage to reduce muscle contractures were initially performed. Next, hold-relax and contract-relax proprioceptive neuromuscular facilitation techniques were performed to regain shoulder muscle tone and proprioception. In order to synergize with PT effect, high molecular weight HA (1%) protected by ADM (2%) against oxidative degradation (Hyadrol^®, 2 mL) was injected and two more injections were planned fortnightly. At the second check-up, two weeks after the third intra-articular injection, the patient showed a marked improvement in pain (NRS 1) and motor function, with a total recovery of the physiological ROM. The rotator cuff specific tests (i.e., lift-off, Napoleon, painful arc, Yocum, and Jobe’s tests) returned negative results. Scapular dyskinesia was no longer present. The DASH questionnaire revealed a significant improvement in ADL (score 2.5). At the telephone check-up, 6 months after the last visit, the patient reported complete motor relief, with only minimal pain (NRS 3). Pain changes are summarized in Figure 5.

 Figure 5: Change in pain severity during and after (6-month follow-up) the combined PT and intra-articular management of case 1. Arrows indicate when intra-articular HA+ADM injections were performed.

Case 2

After the first PT regimen, the patient still scored pain as quite severe (NRS 7). Ten more sessions of active-assisted PT (40 minutes each) were planned, with active exercises combined with magnetic field therapy (same regimen as the initial protocol) in order to improve consolidation. During each PT session, antalgic maneuvers, techniques to reduce muscle contractures and active exercises were performed for improving shoulder muscle tone and proprioception. At the end of the second PT rehabilitation phase, left shoulder mobility and ROM were improved, albeit the patient still complained severe stiffness and pain on rotation and overhead motion (NRS 6). The DASH questionnaire scored 15.8. A diagnosis of secondary AC was then made. An intra-articular course of three fortnightly injections with high molecular weight HA 1% and 2% ADM (Hyadrol^®, 2 mL) was initiated in combination with a third 10-session activeassisted PT. Each 40-minute session consisted in capsule stretching, mobilization techniques and, in the last phase, active exercises. The aim was multifaceted: to strengthen the rotator cuff, promote the recovery of ROM, drain periarticular oedema, facilitate venous return, and prevent intra-articular adhesions. Magnetic field therapy was no longer performed, since consolidation was considered complete (no pain or motion at the fracture site). At the last check-up (15 days after the third and last intra-articular injection), the patient achieved a total recovery of active ROM, and an improvement in ADL (DASH score 2.5), in the absence of pain (NRS 0). At the telephone check-up, 6 months later, the patient reported persistence of the acquired relief on movements and remained pain free (NRS 0). Pain changes along time (i.e., during rehabilitation management and after 6-month follow-up) are summarized in Figure 6.