Effectiveness of Shortwave Diathermy for Subacromial Impingement Syndrome and Value of Night Pain for Patient Selection

A Double-Blinded, Randomized, Placebo-Controlled Trial

Abstract

Objective 

The aim of this study was to investigate the effectiveness of short wave diathermy (SWD) in patients with subacromial impingement syndrome.

Design 

In this double-blinded, randomized, placebo-controlled trial, 57 patients (aged 35–65 yrs) were classified into night pain positive (NP[+]) (n = 28) and night pain negative (NP[−]) (n = 29) groups. Both groups were randomly assigned to SWD (NP[+], n = 14; NP[−], n = 14) and sham (NP[+], n = 15; NP[−], n = 14) subgroups. Visual analog scale, Constant-Murley Scale (CS), and Shoulder Disability Questionnaire (SDQ) scores were used for evaluation.

Results 

There was only a significant difference in pain with activity at 1-mo (mean difference [MD], −1.65; 95% confidence interval, −3.01 to −0.28]) and 2-mo evaluations (MD, −2.1; 95% confidence interval, −3.51 to −0.69) between SWD versus sham groups. In the NP(+) SWD group, the CS pain score was significantly higher than in the NP(+) sham group at all evaluations after treatment. At 1 mo, the NP(−) SWD group showed significantly better pain, strength, total CS, and SDQ scores than the NP(−) sham group. At 2 mos, the pain, range of motion, strength, and total CS and SDQ scores were better in the NP(−) SWD group than in the NP(−) sham group (P < 0.05).

Conclusions 

Short wave diathermy is more effective in subacromial impingement syndrome without NP.

Shoulder pain is highly prevalent within general population, and shoulder impingement syndrome (SIS) is a common cause of shoulder pain.¹ The underlying mechanism of SIS is compression of the rotator cuff tendons, subacromial bursa, and other soft tissues (e.g., long biceps tendon) between the humeral head and the undersurface of the acromion, acromioclavicular joint, or the coracoacromial arch.^2,3 Although nonoperative and operative treatment strategies are available for SIS, many patients recover with nonsteroidal anti-inflammatory drugs, physical therapy, activity modification, and corticosteroid injections.^4,5

Short wave diathermy (SWD) is one of the deep heating modalities used in the treatment of musculoskeletal disorders.⁶ In our clinical practice, we have observed that night pain (NP) is an important musculoskeletal symptom and might be related to local inflammation.^7–9 Previous reports have demonstrated that proinflammatory cytokines, including tumor necrosis factor, interleukin (IL) 1α, IL-1β, IL-6, IL-8, transforming growth factor-β, and basic fibroblast growth factor are present at high levels in the subacromial bursa of patients with rotator cuff disease.^10–12 Okamura et al. suggested that intra-articular cytokines in the shoulder joint might be associated with resting shoulder pain in patients with rotator cuff disease.¹² Many studies have reported the use of deep heat for the treatment of SIS and adhesive capsulitis.^13–15 Detailed review of these studies revealed that deep heat was used for these pathologies without consideration of NP. We have observed that patients without NP seemed to benefit greatly from deep heating modalities; therefore, we hypothesized that NP could be a factor in the choice of treatment. The latest Cochrane reviews also showed conflicting results with regard to the use of deep heating modalities for adhesive shoulder capsulitis.¹⁶ No randomized controlled study has examined the effect of NP in treatment of SIS. Therefore, we aimed to evaluate the effectiveness of SWD in patients with SIS and to determine the effect of NP on treatment.

MATERIALS AND METHODS

A randomized, prospective, double-blinded, placebo-controlled trial was conducted. The study evaluated 98 patients aged 35–65 yrs, with SIS diagnosed according to history and clinical evaluation using Neer, Hawkins-Kennedy, and painful arc shoulder impingement tests. A careful examination of the neck and upper limbs, conventional shoulder radiography and magnetic resonance imaging of neck and shoulder were performed to rule out abnormalities of the cervical spine and other shoulder pathologies. Detailed routine laboratory tests (complete blood count, blood glucose level, serum thyroid-stimulating hormone level, erythrocyte sedimentation rate, and C-reactive protein level) were performed. Data included patient age, sex, occupation, body mass index, hand dominance, duration of pain, and presence of NP and medication history. A subacromial injection test was performed in 72 patients for differential diagnosis. The subacromial space just under the acromion was injected with 5 mL of 2% lidocaine using a 21-G needle via an anterior approach. All patients were injected by the same specialist. A positive subacromial injection test was indicated by 80% relief of pain and almost total improvement in passive and/or active range of motion (ROM), 30 mins after the injection.^17,18 Patients who had (1) concomitant shoulder pathologies such as adhesive capsulitis, calcific tendinitis, full-thickness tears of the rotator cuff tendons, osteoarthritis of the acromioclavicular joint, dislocations, acute traumatic conditions, etc.; (2) previous applications of physiotherapy and injection of hyaluronic acid and/or corticosteroid during the preceding 6 mos; (3) cervical pain or other conditions such as fibromyalgia confusing the clinical picture; (4) malignancy; (5) neurologic, motor, and/or sensory deficit in the upper extremity; (6) pregnancy; (7) an open wound near the shoulder; (8) presence of an implanted cardiac pace maker; (9) and local anesthetic allergy were excluded from the study. One patient with uncontrolled diabetes mellitus, three with calcific tendinitis, four with cervical discopathy, one with a risk of pregnancy, five who underwent subacromial corticosteroid injection during the preceding 6 mos, two with a history of shoulder trauma, and ten who refused the subacromial injection test were excluded.

Written informed consent was obtained from all patients, and all procedures were conducted in accordance with the Helsinki Declaration of 1975, and approved by the local institutional clinical research ethical committee (2012/15).

Treatment Protocol

The study had a randomized, placebo-controlled, double-blinded design. A sample size of 27 subjects was determined for NP positive (NP[+]) and NP negative (NP[−]) groups, and follow-up loss of 10% was assumed. Sixty patients were separated into NP(+) and NP(−) groups, with 30 patients per group. Each group was also randomly separated into SWD and sham subgroups, with 15 patients per group. Patients in the NP(+) SWD (n = 14) and NP(−) SWD (n = 14) subgroups received 27.12 MHz of continuous SWD, as a 20-min daily session, 5 days per week for 2 wks, for a total of ten sessions. The physiotherapist applied SWD ITO SW-180 device. The affected shoulder was positioned between 2 condenser electrodes, with 20 cm between the anterior and posterior electrodes with the patient seated on a wooden chair. Patients in the NP(+) sham (n = 15) and NP(−) sham groups (n = 14) underwent treatment with the same protocol. The physiotherapist applied sham SWD by starting the device but without applying heat. The same physiotherapist applied all SWD or sham treatments. All groups were instructed in a conservative treatment program, Codman pendulum exercises (five sets daily, 5 mins each set), cold pack application (cold pack gel, three times daily, 20 mins per session) on the affected shoulder, restriction of daily activities requiring the overhead use of hands, and sleeping on the affected shoulder, and oral meloxicam 15 mg daily was administered for 2 wks. The same physiotherapist supervised the exercise program and cold pack application. Patients were requested to continue the Codman pendulum exercises and cold pack application for 2 wks after SWD or sham treatment.

Randomization

Randomization was performed with a computer-generated list of numbers, and group allocation was concealed. A physiatrist not involved with data collection performed random number generation using Microsoft Excel for Windows. The assignment of subjects into groups (NP[+] SWD, NP[−] SWD, NP[+] sham, NP[−] sham) was performed randomly using opaque sealed envelopes. One physiotherapist who was blinded to the randomization process provided all SWD or sham treatments. A second blinded physiotherapist supervised the exercise protocols. Finally, all assessments were made by another physiatrist who was blinded to the group assignments and interventions.

Outcome Evaluation

Patient evaluation was performed at treatment initiation and at 2 wks and 1 and 2 mos after treatment initiation. All patients were evaluated by the same physiatrist who was blinded to the randomization process and treatment protocols. All groups were evaluated for shoulder pain with visual analogue scale (VAS) for rest, activity, and sleep-disturbing NP.¹⁹ Patients were instructed to indicate the severity of pain on a 10-point scale, on which 0 meant no pain, 5 meant moderate pain, and 10 meant intolerable pain. Functional shoulder status in all groups was evaluated with the total Constant-Murley Scale (CS).²⁰ The CS evaluates overall shoulder function on a 100-point scale. The higher the CS score, the better the functional status of the shoulder. Subsections of this analysis score up to 15 points for shoulder pain, 20 points for activities of daily living, 40 points for active ROM, and 25 points for strength. Activities of daily living were evaluated with the Shoulder Disability Questionnaire (SDQ), which was translated into Turkish and validated by Ozsahin et al.²¹ The SDQ is a pain-based disability questionnaire and contains 16 items describing common situations that may induce symptoms in patients with shoulder disorders. All items refer to the preceding 24 hrs. A final score is calculated by dividing the number of positively scored items by the total number of applicable items and subsequently multiplying the score by 100, resulting in a final score between 0 (no disability) and 100 (severe disability).²²

STATISTICAL ANALYSIS

Calculations of the estimated sample size were based on a previous study by Galace de Freitas et al.²³ assuming that α is 0.05 and power is 0.80. Sample sizes of 27 subjects were determined for both the NP(+) and NP(−) groups, and follow-up loss of 10% was assumed. With allowance for dropouts, 60 subjects were recruited for this study. G*Power was used for this calculation. Number Cruncher Statistical System 2007 and Power analysis and Sample Size 2008 (Statistical Software, Kaysville, UT) were used for data analysis. Student t test was used to compare the study data for average, standard deviation, median, frequency, and ratio for quantitative data and parameters that showed a normal distribution. The Mann-Whitney U test was used to compare groups that did not show a normal distribution. The Fisher-Freeman-Halton test, Fisher exact test, and the Yates Continuity correction test were used in the comparison of qualitative data. A P < 0.05 was considered significant.

RESULTS

Sixty-four patients with a positive subacromial injection test were asked to participate in the study. Four patients declined to participate. Two patients disagreed with the treatment program, and one patient was unwilling to continue treatment because of increased pain. Seven were excluded, leaving a final sample of 57 participants. A flow diagram of the patient allocation is presented in Figure 1 (Checklist, CONSORT 2012 Checklist, Supplemental Digital Content, https://links.lww.com/PHM/A492).

FIGURE 1: Flow diagram of patient selection process.

No significant difference was found between the NP(+) and NP(−) groups or subgroups for age, sex, body mass index, work status, and education. Other than sex and work status, no significant difference was found between NP(+) SWD and sham groups for disease-related variables of pain duration, dominant hand, affected shoulder, etiology of pain, and chronic disease. There were more women and homemakers in the sham group (see Table, Supplemental Digital Content 1, https://links.lww.com/PHM/A477, which demonstrates the comparison of SWD and sham groups according to demographic parameters and disease-related variables).

There was no significant difference between pain scores measured by the VAS, total and subsection parameters of CS, and SDQ at baseline evaluation of all groups (P > 0.05).

Primary Outcomes

When we compared SWD versus sham treatment groups according to VAS scores, there was only a significant difference in pain with activity at the 1- and 2-mo evaluations (P < 0.05). In the SWD group, VAS scores with activity were lower than those in the sham group (see Table, Supplemental Digital Content 2, https://links.lww.com/PHM/A478, which demonstrates the comparison of SWD and sham treatment groups according to VAS scores for rest, activity, and NP). The CS pain scores after 2 wks, and at 1- and 2-mo evaluations, total CS scores at 1- and 2-mo evaluations, and CS activities of daily living scores at the 2-mo evaluation were higher in the SWD group than in the sham group (P < 0.05) (see Table, Supplemental Digital Content 3, https://links.lww.com/PHM/A479, which demonstrates the comparison of SWD and sham treatment groups according to the total CS and subsection parameters scores). The SDQ scores showed a significant difference between SWD and sham groups at 1-mo and 2-mo evaluations with lower SDQ scores in the SWD group (P < 0.05) (see Table, Supplemental Digital Content 4, https://links.lww.com/PHM/A480, which demonstrates the comparison of SWD and sham treatment groups according to SDQ scores).

Secondary Outcomes

In the NP(+) group, there were no significant differences between SWD and sham groups according to VAS scores with rest and activity pain at 2 wks and at 1- and 2-mo after treatment initiation and VAS scores with NP at 2 wks and at 2 mos after treatment initiation (P > 0.05). There was a statistically significant decrease in VAS for NP scores at 1 mo in a comparison of SWD and sham treatment (P < 0.05) (Table 1). In the NP(−) group, there was no statistically significant difference between SWD and sham treatment according to VAS scores with rest and activity at 2 wks and 1 mo after treatment initiation. Short wave diathermy was superior to sham treatment at 2 mos with both rest and activity pain (P < 0.05) (Table 2) (see Figures, Supplemental Digital Content 5, 6, 7, 8, 9, and 10, https://links.lww.com/PHM/A481, https://links.lww.com/PHM/A482, https://links.lww.com/PHM/A483, https://links.lww.com/PHM/A484, https://links.lww.com/PHM/A485, https://links.lww.com/PHM/A486, which demonstrate the VAS scores for rest, activity, and night pain in patients with or without NP).

TABLE 1: Comparison of SWD and sham treatment groups according to VAS scores for rest, activity, and night pain in NP(+) group

TABLE 2: Comparison of SWD and sham treatment groups according to VAS scores for rest, activity, and night pain in NP(−) group

In the NP(+) group, there were no statistically significant differences between SWD and sham treatment groups according to the total CS and subsection parameters at all evaluations (P > 0.05); however, CS pain scores showed statistically significant improvements with SWD compared with that in the sham treatment group (P < 0.05) (Table 3) (see Figure, Supplemental Digital Content 11, https://links.lww.com/PHM/A487, which demonstrates the total CS scores in patients with NP). In the NP(−) group, the total CS scores at all evaluations, the CS pain and strength scores at 1 mo and 2 mos, and the CS ROM scores after 2 wks showed significant improvements with SWD compared with those with sham treatment (P < 0.05) (Table 3) (see Figure, Supplemental Digital Content 12, https://links.lww.com/PHM/A488, which demonstrates the total CS scores in patients without NP).

TABLE 3: Comparison of SWD and sham treatments according to the total CS and subsection parameters in NP(+) and NP(−) groups

Based on the SDQ, in the NP(−) groups, there was a statistically significant difference between SWD and sham treatment groups (P < 0.05) (Table 4; Fig. 2). The SDQ scores decreased in the SWD group at all posttreatment evaluations. There were no significant differences between SWD and sham treatment in NP(+) groups at all evaluations (P > 0.05) (Table 4; Fig. 3).

TABLE 4: Comparison of SWD and sham treatment groups according to SDQ in NP(+) and NP(−) groups

FIGURE 2: Shoulder Disability Questionnaire scores in patients without NP. This figure shows the SDQ scores in patients without NP before treatment, at 2 wks and at 1 mo, and 2 mos after treatment initiation. The vertical axis shows the SDQ scores.

FIGURE 3: Shoulder Disability Questionnaire scores in patients with NP. This figure shows the SDQ scores in patients with NP before treatment, at 2 wks and at 1 mo, and 2 mos after treatment initiation. The vertical axis shows the SDQ scores.

Comparison of NP(+) SWD and NP(−) SWD groups showed that VAS scores for resting, night, and activity pain were lower in NP(−) SWD groups at all evaluations except for resting pain after 2 wks of treatment, and activity pain before treatment (see Table, Supplemental Digital Content 13, https://links.lww.com/PHM/A489, which demonstrates the comparison of NP(−) SWD and NP(+) SWD treatment groups according to VAS scores for rest, activity, and NP). The CS showed a significant difference between NP(−) SWD and NP(+) SWD groups after 2 wks of treatment, and at 1- and 2-mo evaluations (P < 0.05), except for the CS pain scores at the 1-mo evaluation (P > 0.05) (see Table, Supplemental Digital Content 14, https://links.lww.com/PHM/A490, which demonstrates the comparison of NP(−) SWD and NP(+) SWD treatment groups according to the total CS and subsection parameter scores). There was a significant difference between NP(−) SWD and NP(+) SWD groups at all evaluations (P < 0.05), with SDQ scores lower in the NP(−) group (see Table, Supplementary Digital Content 15, https://links.lww.com/PHM/A491, which demonstrates the comparison of NP(−) SWD and NP(+) SWD treatment groups according to SDQ scores).

DISCUSSION

Shoulder impingement syndrome is a common cause of shoulder pain and loss of function.¹ Physical therapy modalities aim to increase blood flow and cellular metabolism and stimulate healing in tendons.^5,23,24 Short wave diathermy is one of the most preferred deep-heating modalities and typically uses electromagnetic radiation at 27.17 MHz in either continuous or pulsed mode. Short wave diathermy has thermal and nonthermal effects on tissues. Increased tissue temperature causes arteriolar and capillary dilatation followed by increased blood flow to the tissue. This leads to an increase in cellular metabolism and increased tendon and muscle flexibility.²⁴ In this way, it reduces muscle spasm and raises the pain threshold.^6,25

Short wave diathermy is frequently used in the management of shoulder pathologies, particularly frozen shoulder and SIS. However, there is no randomized controlled study on the therapeutic effects on SIS in the literature.^15,26 Leung and Cheing¹⁵ compared the effects of SWD and superficial heating in patients with frozen shoulder. They found greater improvement in pain relief, activities of daily living, and ROM with addition of SWD to stretching exercises. Pajareya et al.²⁷ studied patients with frozen shoulder, and found that SWD enhanced the therapeutic effects of nonsteroidal anti-inflammatory drugs and exercise after 3 wks, but there was no significant additional effect at 12 wks.

Many clinical trials have evaluated the effects of therapeutic ultrasound (US) for deep-heat therapy in patients with SIS; however, studies investigating the effects of SWD in SIS are lacking, despite its common clinical use. Yazmalar et al.¹³ conducted a randomized controlled study to evaluate the efficacy of therapeutic US on pain, disability, anxiety, depression, sleep quality, and quality of life in patients with SIS. They randomly divided the patients with SIS into two groups, which are continuous US and sham US groups. Transcutaneous electrical nerve stimulation and an exercise program were also added for both groups. They found that US does not offer any benefits for SIS. Desmeules et al.¹⁴ conducted a systematic review on the efficacy of therapeutic US in adults with rotator cuff tendinopathy. They stated that therapeutic US was not superior to placebo intervention in terms of pain reduction and functional improvement. When provided in conjunction with exercise, US therapy was not superior to exercise alone in terms of pain reduction and functional improvement. In another study, which compared the short-term effectiveness of high-intensity laser therapy and US therapy in the treatment of SIS, there was greater reduction in pain and improvement in functionality and muscle strength in the affected shoulder after high-intensity laser therapy therapy.²⁸ Calis et al²⁹ compared US, laser, and exercise therapies in the treatment of SIS. They found that US and laser treatments were not superior to each other in the treatment of SIS. The results of a recent Cochrane review of electrotherapy modalities for rotator cuff disease support these contradictory results.³⁰ The conflicting results in these trials with deep heat might be due to methodological deficits. In particular, the NP symptoms in SIS may increase with the severity of the bursitis or tendinitis. Therefore, application of deep-heat therapy may not be effective. Based on this hypothesis, the patients were separated according to groups with and without NP, to determine the effect of NP on the treatment response. The aim was to identify patients with SIS who were most likely to benefit from deep-heating modalities.

In this study, NP(+) groups showed statistically significant improvement over baseline values in VAS scores for rest, night, and activity pain, in the total CS and subsection parameters, and in the SDQ. However, there was no significant difference between the SWD and sham treatment groups in the aforementioned parameters other than for the CS pain score, the SWD group was superior to the sham treatment group. The results of VAS and CS pain scores may have differed from each other for two reasons. First, the CS pain score evaluates shoulder pain in one dimension. Activity pain, NP, and rest pain are recorded, and the most severe pain is used for scoring. Second, the CS pain score uses a narrow scale (15 indicates no pain; 10, mild pain; 5, moderate pain; and 0, severe pain) in comparison with VAS scores (scores range from 0, with no pain, to 10, with severe pain).²⁰

In the NP(−) groups, statistically significant improvements over the baseline values were also obtained for all evaluation parameters. In comparison with the sham group, the SWD group showed superior results for all parameters except for the CS activities of daily living scores at the 2-mo evaluation. The SWD group also showed superior results for parameters of CS pain, CS strength, CS total score, and SDQ at 1-mo evaluation. These results suggest that baseline cold pack application, Codman pendulum exercises, and oral nonsteroidal anti-inflammatory drugs are the main reasons for an early decrease in pain scores and improvements of shoulder functions. Short wave diathermy may provide long-term benefit in NP(−) patients whose pain does not have an inflammatory character. These outcomes emphasize the importance of identifying the NP symptoms in SIS. This basic history may provide evidence for the inflammatory status of the disease and guide the choice of a better treatment modality.

There are limitations in this study. The results with the small sample size require replication with larger samples and longer follow-up periods. It could be better to analyze the magnetic resonance imaging findings to support the effect of NP symptoms on treatment of SIS. We did not investigate acromion morphology in these patients. The use of acromion morphology might have yielded different results.

In conclusion, addition of 27.12 MHz of continuous SWD (as a 20-min daily session, 5 days per week for 2 wks, for a total of 10 sessions) to conventional therapies provides long-term benefits when compared with sham treatment, in terms of VAS scores for rest and activity pain, and CS and SDQ scores in SIS patients without NP. However, there was no convincing evidence that SWD treatment is of additional benefit in NP(+) SIS. Therefore, in the case of SIS, NP symptoms should be evaluated to use deep heat more effectively in clinical practice.