RESEARCH ARTICLE
Ultrasound and Laser as Stand-Alone Therapies for Myofascial Trigger Points: A Randomized, Double-Blind, Placebo-Controlled Study A. Manca1,*, E. Limonta2, G. Pilurzi1,3, F. Ginatempo1, E. R. De Natale1, B. Mercante1, E. Tolu1 & F. Deriu1 1
Department of Biomedical Sciences, University of Sassari, Sassari, Italy
2
Department of Biomedical Sciences for Health, University of Milan, Milan, Italy
3
Department of Clinical and Experimental Medicine, University of Sassari, Italy
Abstract Background and Purpose. Ultrasound (US) and low-level laser therapy (LLLT) are commonly employed for myofascial trigger points (MTP) despite lack of evidence for usage as stand-alone treatments. The aim of the study was to determine, on MTP of the upper trapezius muscle (uTM), the effects of US and LLLT per se, as delivered in accordance with the procedures reported by surveys about their usage among physiotherapists. Methods. Design was set as a double-blind, randomized, placebo-controlled study. Sixty participants with at least one active MTP in uTM (28 women and 32 men; mean age 24.5 ± 1.44 years) were recruited and randomly assigned to one out of five groups: active US (n = 12), placebo US (n = 12), active LLLT (n = 11), placebo LLLT (n = 11) and no therapy (control, n = 14). The participants and outcome assessor were blinded to the group assignment and therapy delivered. Three outcome measures were assessed at baseline, after a 2-week treatment and 12 weeks after the end of the intervention (follow-up): pressure pain threshold, subjective pain on a numerical rating scale and muscle extensibility performing a cervical lateral flexion. All subjects assigned to the intervention groups were treated five times weekly for overall 10 treatments given. Two-way ANOVA was used to compare differences before and after intervention and among groups at each time-point. Results. After the 2-week intervention, all groups showed pressure pain threshold, numerical rating scale and cervical lateral flexion significant improvements (p < 0.05), which were confirmed at the follow-up. When performing multiple comparisons, controls scored significantly less than both the active therapies and placebos, whereas no differences were detected between active therapies and placebos. Conclusions. Ultrasound and LLLT provided significant improvements in pain and muscle extensibility, which were superior to no therapy but not to placebos, thus raising concerns about the suitability, both economically and ethically, of administering such common physical modalities as stand-alone treatments in active MTP of the uTM. Copyright © 2014 John Wiley & Sons, Ltd. Received 31 July 2013; Revised 9 November 2013; Accepted 27 November 2013 Keywords Evidence based practice; Management and Professional issues; Musculoskeletal; Physiotherapy; RCT *Correspondence Department of Biomedical Sciences, University of Sassari, Viale San Pietro 43/b, 07100 Sassari, Italy. E-mail: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1580
Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
Introduction Myofascial trigger points (MTP) in the neck are a common and costly condition, with a point prevalence of 10–18% and a lifetime prevalence of 30%–50% (Gur et al., 2004). Several treatments such as manual therapy, exercise, acupuncture, massage, ice, heat, transcutaneous electrical nerve stimulation, anaesthetics injections, diathermy, ultrasound (US) and low-level laser therapy (LLLT) are currently provided for MTP (Robertson et al., 2001; Alvarez and Rockwell, 2002). Particularly, US and LLLT have achieved popularity and recognition among physicians and physiotherapists as common, non-invasive options of treatment (Beckerman et al., 1992; Draper et al., 2010). Regarding the use of US for MTP in the upper trapezius muscle (uTM), it was reported to induce a significant reduction in pain intensity (Majlesi and Ünalan, 2004), but other studies failed in showing significant effects on pain (Gam et al., 1998) or superiority to placebos (Lee et al., 1997); conflicting evidence for US was also stated by a systematic review (Vernon and Schneider, 2009). Regarding LLLT, reports are generally positive, showing superiority to placebos and effectiveness on pain and functional outcomes in MTP (Vernon and Schneider, 2009). However, when employed in a combined regimen of treatment with exercise and stretching, results are conflicting (Altan et al., 2005). Notably, in the United States, several medical coverage policies do not cover LLLT for any indication because it is considered an experimental/ investigational/unproven intervention (Blue Cross TEC, 2010). A strong disagreement between researchers and clinical practitioners emerges when comparing clinical guidelines (Chartered Society of Physiotherapy, 2006; Australian Physiotherapy Association, 2001; Canadian Physiotherapy Association, 2010) to survey about US and LLLT protocols (Lindsay et al., 1990; Robertson and Spurritt, 1998; Wong et al., 2007). The latter reveals a tendency to their massive utilization as stand-alone therapies in spite of no consensus about their effectiveness (Gam et al., 1993; Robertson and Baker, 2001). Such modus operandi is questionable and generally not recommended by clinical guidelines (see in the previous text) because these therapies are considered complementary to other more established interventions, such as manual therapy and exercise. However, the employment of combined regimens of treatment leads to conflict attributions of effects (Gam et al., 1998; Altan et al., 2005), making it difficult to evaluate the role of each
single therapy. The aim of the present study was to assess the effectiveness of US and LLLT delivery procedures according to protocols commonly used by physiotherapists (Wong et al., 2007; Chow et al., 2009). Firstly, it was evaluated whether a stand-alone therapy with either US or LLLT was effective on MTP in the uTM. Secondly, such protocols were compared with each other, with placebos or no therapy. Finally, all strategies were reassessed after a 12-week follow-up.
Methods Subjects and design Trial design was set as a randomized, double-blind, placebo-controlled, parallel-group study approved by the local Ethical Committee (Bioethics Board of ASL n.1-Sassari, prot. U-553, October 18, 2011; trial data fully available at the Department of Biomedical Sciences). All participants signed a written consent and an information notice regarding experimental procedures before enrolment. Sixty volunteers (28 women and 32 men; mean age 24.5 ± 1.44 years; range 20–30 years) with a spontaneously painful area and a palpable taut band in at least one uTM, disturbing normal daily activity, participated in the study. Subjects were recruited among university students, who are highly exposed to work-related musculoskeletal symptoms (Hupert et al., 2004). Exclusion criteria were the following: to meet general contraindications to electrophysical agents according to established guidelines (Robertson et al., 2001; Chartered Society of Physiotherapy, 2006; Houghton et al., 2010); to be a physiotherapy student or to have already experienced US and LLLT; a numerical rating score for pain lower than 3 out of 10; no impact of the condition under study on normal daily life; concomitant general diseases; diffuse generalized musculoskeletal and joint pain, head, neck or upper back pain due to osteoarticular pathologies; ongoing physiotherapy treatment for the condition under study at the time of enrolling or current use of medications affecting pain and skin sensibility and shoulder and neck surgery within the previous year. The recruited population underwent a clinical examination performed by an orthopaedic physician experienced in musculoskeletal disorders, and MTP were diagnosed as active or latent (Simons, 1995). Selected subjects with at least one active MTP within the uTM underwent functional assessment performed by a physiotherapist experienced in musculoskeletal conditions. The assessment procedure was performed at three different times: Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
before intervention (baseline), after 2 weeks (post) and after 12 weeks (follow-up) from the initial evaluation. Demographic data and participants’ clinical characteristics including age, sex, symptoms duration and referred pain patterns and involved side were recorded.
Stand-alone Therapies in Myofascial Trigger Points
examination, the active MTP (or the most painful one, if more than one was detected) within the uTM was marked with a dermographic pencil to allow a reliable localization. Active ultrasound
Randomization and blinding Participants were randomly assigned to one out of five parallel groups in 1:1:1:1:1 allocation ratio, to receive respectively active US (UsA-group I: n = 12), placebo US (UsP-group II: n = 12), active LLLT (LtA-group III: n = 11), placebo LLLT (LtP-group IV: n = 11) and with the fifth group acting as the control (no therapy: n = 14). No blocking was set for randomization. To generate the allocation sequence, Research Randomizer 3.0 software was employed (Urbaniak and Plous, 2011). Sixty opaque envelopes were prepared and numbered consecutively by AM, whereas participant allocation to envelopes was performed by one of the three physiotherapists employed to deliver treatments. According to the envelope selection, patients were allocated to one of the five groups listed in the previous text. Both outcome assessor and participants were blinded to interventions, whereas the three treating physiotherapists employed could not be blinded when delivering as they had to open each envelope and read the assigned number and group of allocation. No communication occurred between treating physiotherapists and the outcome assessor regarding trial course and participants.
Eme-Medical Ultrasonic 1300, 1–3 MHz US machine was used. Because a dose–response relationship for US has not been identified yet by well-designed randomized controlled trials (Robertson, 2002), delivery parameters were set according to Wong et al. (2007) and Draper et al. (1995): frequency 3 MHz, continuous mode, intensity 1.5 W cm2, duration 12 minutes. US probe (head size: 5 cm2) was applied steady with no pressure over the trigger point. A coupling agent was used to optimize delivery at the probe-skin interface. Because the painful area was limited to a trigger point, there was no need to move the US probe with rotating movements as the ratio between effective radiating area and the size of US head was approximately 1:1. No adverse effects such as overheating occurred during the trial. Placebo ultrasound In group II, sham-US was applied following the same modalities of the active therapy, but the US head cable was disconnected from the working device. Patients could not know whether the cable was connected or not to the machine as the display was also on with the time counting. Active low-level laser therapy
Interventions All interventions and data collection took place at the Physiotherapy Clinic of the University Hospital of Sassari, Italy. Prior to the trials beginning, therapists were instructed on how to deliver both active treatment and placebo, patient’s position and procedures and, also, to maintain an emotional detachment from the participant, avoiding any extra-intervention interaction with him/her. Sessions were performed at the same time of the day and in the same location. The participants were recommended not to use any drugs for the condition under study throughout the 2-week physical therapy or no therapy periods. Groups I–IV were treated five times weekly for overall 10 treatments given in 2 weeks. Patient’s position was set as relaxed high sitting with backrest. During orthopaedic Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A J&S Medical-Cyberlight (J&S, Rome, Italy), Mod. Ls1-Ga/As LLLT machine was used. According to the acknowledged guidelines (WALT, 2004), delivery parameters were: wave-length 904 nM; pulse duration 200 ns; pulse frequency 1953 Hz; peak power 90 mW; average output 30 mW; power density 22.5 mW cm2; treatment time 600 seconds; energy dose 18 J per session; spot size 4 cm2 and treatment frequency five times/week. Laser probe (head size: 4 cm2) was applied steady in skin contact with no pressure over the MTP. Placebo low-level laser therapy In group IV, sham-laser was applied following the same modalities of the active therapy, but the laser probe cable was disconnected from the working device. As for the placebo US, patients could not
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
know whether the cable was connected or not to the working machine. No therapy (Control) Subjects randomly assigned to group V (n = 14) only underwent baseline, post and follow-up evaluations without being given any treatment (active or sham) in between. Outcome measures All groups underwent a baseline, a post-intervention assessment and a follow-up session 12 weeks after the end of the intervention phase. In each session, three outcome measures were set: pressure pain threshold (PPT); pressure-related pain assessed with a numerical rating scale (NRS); upper trapezius extensibility assessed, through cervical goniometry, during an active lateral cervical flexion (cLat-Flex) to the side opposite to the affected muscle. Pressure pain threshold was determined using a hand-held pressure algometer. A dual force gauge mod. FDK-20 (Wagner Instruments, Greenwich-CT) was used (calibration in kg cm2; capacity 10 kg × 100 g; 20 lb × 0.25 lb). The force recorded corresponded to the quantity of pressure required to change subject’s sensation from pressure to pain. According to Fischer (1987), pressure amount was increased at a rate of 1 kg cm2 second, three measurements were taken with a rest time of 20 seconds, and the mean was recorded for data analysis. Immediately after recording PPT, subjects were asked to score on a NRS their pain at the changing point from sensation of pressure to pain, and the mean was retained for data analysis. Lateral cervical flexion was assessed using a gravityreference Myrin goniometer (prod. Patterson Medical, Illinois). Goniometer was fixed to the front of the head with a Velcro® fastening strap. The subject’s position was sitting with backrest and shoulders stabilized by the examiner. As for PPT and NRS, three measurements were taken, and the mean calculated for data analysis. Statistical analysis Data analysis was performed by using SigmaPlot-12 (Systat Software Inc., San Jose, California) by a blinded statistician (EL). A priori power analysis to determine the sample size was performed using G*Power 3
(Faul et al., 2007). By using a power (1-β err. prob.) of 0.80, α = 0.05 and an estimated effect size of 0.5 (Cohen, 1992), the required sample size to detect a 20% difference between treatment groups for the primary outcomes was 11 subjects per group. Shapiro– Wilk test was used to test the normality of distribution for continuous variables, whereas the homogeneity of variances was evaluated by Levene’s test. Two-way (five conditions × three times) ANOVA was used to compare changes in variables before and after intervention (post and follow-up) and among groups, as well as main effects and interactions. When significant differences were detected, a post-hoc Holm–Sidak method multiple comparison procedure was utilized. Significance level was set for a p-value < 0.05. Unless otherwise stated, data are expressed as mean ± standard deviation.
Results Seventy-nine subjects met eligibility criteria for recruitment but 16 of them refused consent, three were excluded due to a NRS score < 3, so that 60 of 79 subjects were enrolled. Recruitment and participant flow are presented in Figure 1. No harm or unintended effects occurred during this trial. At baseline, all groups were statistically homogeneous for age, PPT, NRS, cLat-Flex, symptom duration and gender as shown in Table 1, which reports demographic and clinical characteristics of the sample. Results are shown BY GROUP (Post vs. Pre; Follow-up vs. Pre and Follow-up vs. Post) and detailed with scores and Cohen’s d effect sizes (Cohen, 1992) in Table 2. After the 2-week intervention for groups I–IV and no therapy for group V (Control), each group showed statistically significant changes (p < 0.05) of all outcome variables. In particular, PPT and cLat-Flex increased while NRS decreased. Compared with the baseline, improvements remained statistically significant at follow-up in all groups. Compared with the post, further significant changes of outcome scores were observed in several groups at the follow-up (Table 2). Significances ensued from multiple comparisons among groups at post and followup, performed with a two-way ANOVA, are presented in Figure 2, whereas the main effects (Condition and Time) and interactions (Condition × Time) are reported in Table 3. Regarding PPT within Post, when performing all possible comparisons, no statistically significant differences were detected between active therapies Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
Figure 1. Organization chart
Table 1. Demographic and clinical characteristics of the sample at the baseline assessment Group Variable Age (years) Gender (number and %) 2
PPT (kg cm ) 95% CI NRS (0–10) 95% CI cLat-Flex (°) 95% CI Symptom duration weeks
UsA (n = 12)
UsP (n = 12)
LtA (n = 11)
LtP (n = 11)
Control (n = 14)
24.5 ± 1.7 F: 7 (58.4%) M: 5 (41.6%) 2.13 ± 0.24 1.99–2.27 4.96 ± 1.20 4.28–5.64 31.17 ± 4.39) 28.69–33.65 3.5 ± 1.8
26 ± 0.8 F: 6 (50%) M: 6 (50%) 2.13 ± 0.20 2.02–2.24 4.68 ± 1.31 3.96–5.42 30.75 ± 4.8 28.01–33.49 3.0 ± 1.5
24 ± 2.1 F: 7 (58.4%) M: 5 (41.6%) 2.02 ± 0.34 1.82–2.22 4.49 ± 0.97 3.92–5.06 30.64 ± 4.20 28.16–33.12 4.0 ± 1.0
25.4 ± 0.7 F: 7 (58.4%) M: 5 (41.6%) 2.09 ± 0.27 1.93–2.25 5.15 ± 1.03 4.54–5.76 31.64 ± 5.05 28.66–34.62 3.0 ± 0.9
23 ± 1.91 F: 7 (50%) M: 7 (50%) 2.12 ± 0.20 2.02–2.22 4.60 ± 1.05 4.05–5.15 30.93 ± 5.15 28.23–33.63 3.5 ± 0.8
UsA = active ultrasound; UsP = placebo ultrasound; LtA = active laser therapy; LtP = placebo laser therapy; Cont = no therapy; PPT = Pressure Pain Threshold; NRS = Numerical Rating Scale; cLat-Flex = Cervical Lateral Flexion; CI = confidence interval evaluated by Levene’s test. Values are expressed as mean ± SD.
(UsA; LtA) versus their placebos (UsP; LtP), whereas control group scored significantly less than both active and placebo groups (p < 0.05; d > 0.8). No differences were found among intervention groups (I–IV) at 12-week follow-up, while controls kept on scoring less than any group (p < 0.05; d > 0.5). Regarding NRS within Post, notably LtA scored significantly better than UsA (p = 0.04), UsP (p = 0.03) and Control (p = 0.002) but not than LtP (p = 0.21). Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
Both active (UsA; LtA) and placebo therapies (UsP; LtP) scored better than no therapy (p < 0.05; d > 0.8). Within follow-up, no significant differences were detected among groups with exception of UsA that scored better than Control (p = 0.03) but not than any other intervention group. When comparing LtA versus LtP versus Control for the dependent variable NRS (Table 3), a significant interaction (Condition × Time) was found (F = 2.499; p = 0.048). The effect of different levels of Time depends on the Condition, so that both factors
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points Table 2. Baseline to Post and Follow-up scores by group Group Outcomes
UsA (n = 12)
UsP (n = 12)
LtA (n = 11)
LtP (n = 11)
Control (ref) (n = 14)
2.02 ± 0.34 2.74 ± 0.41* 1.14 (0.25–1.95) , 2.8 ± 0.39§ # 0.49 ( 0.31–1.25)
2.09 ± 0.27 2.69 ± 0.37* 1.06 (0.18–1.86) 2.8 ± 0.35§ 0.51 ( 0.29–1.27)
2.12 ± 0.2 2.35 ± 0.28*
4.49 ± 0.97 2.24 ± 1.01* 1.63 (0.67–2.48) , 1.3 ± 1.01§ # 0.67 (0.16–1.46)
5.15 ± 1.03 2.57 ± 0.87* 1.39 (0.47–2.22) 2.1 ± 1.64§
Ultrasound and Laser as Stand-Alone Therapies for Myofascial Trigger Points: A Randomized, Double-Blind, Placebo-Controlled Study A. Manca1,*, E. Limonta2, G. Pilurzi1,3, F. Ginatempo1, E. R. De Natale1, B. Mercante1, E. Tolu1 & F. Deriu1 1
Department of Biomedical Sciences, University of Sassari, Sassari, Italy
2
Department of Biomedical Sciences for Health, University of Milan, Milan, Italy
3
Department of Clinical and Experimental Medicine, University of Sassari, Italy
Abstract Background and Purpose. Ultrasound (US) and low-level laser therapy (LLLT) are commonly employed for myofascial trigger points (MTP) despite lack of evidence for usage as stand-alone treatments. The aim of the study was to determine, on MTP of the upper trapezius muscle (uTM), the effects of US and LLLT per se, as delivered in accordance with the procedures reported by surveys about their usage among physiotherapists. Methods. Design was set as a double-blind, randomized, placebo-controlled study. Sixty participants with at least one active MTP in uTM (28 women and 32 men; mean age 24.5 ± 1.44 years) were recruited and randomly assigned to one out of five groups: active US (n = 12), placebo US (n = 12), active LLLT (n = 11), placebo LLLT (n = 11) and no therapy (control, n = 14). The participants and outcome assessor were blinded to the group assignment and therapy delivered. Three outcome measures were assessed at baseline, after a 2-week treatment and 12 weeks after the end of the intervention (follow-up): pressure pain threshold, subjective pain on a numerical rating scale and muscle extensibility performing a cervical lateral flexion. All subjects assigned to the intervention groups were treated five times weekly for overall 10 treatments given. Two-way ANOVA was used to compare differences before and after intervention and among groups at each time-point. Results. After the 2-week intervention, all groups showed pressure pain threshold, numerical rating scale and cervical lateral flexion significant improvements (p < 0.05), which were confirmed at the follow-up. When performing multiple comparisons, controls scored significantly less than both the active therapies and placebos, whereas no differences were detected between active therapies and placebos. Conclusions. Ultrasound and LLLT provided significant improvements in pain and muscle extensibility, which were superior to no therapy but not to placebos, thus raising concerns about the suitability, both economically and ethically, of administering such common physical modalities as stand-alone treatments in active MTP of the uTM. Copyright © 2014 John Wiley & Sons, Ltd. Received 31 July 2013; Revised 9 November 2013; Accepted 27 November 2013 Keywords Evidence based practice; Management and Professional issues; Musculoskeletal; Physiotherapy; RCT *Correspondence Department of Biomedical Sciences, University of Sassari, Viale San Pietro 43/b, 07100 Sassari, Italy. E-mail: [email protected]
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pri.1580
Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
Introduction Myofascial trigger points (MTP) in the neck are a common and costly condition, with a point prevalence of 10–18% and a lifetime prevalence of 30%–50% (Gur et al., 2004). Several treatments such as manual therapy, exercise, acupuncture, massage, ice, heat, transcutaneous electrical nerve stimulation, anaesthetics injections, diathermy, ultrasound (US) and low-level laser therapy (LLLT) are currently provided for MTP (Robertson et al., 2001; Alvarez and Rockwell, 2002). Particularly, US and LLLT have achieved popularity and recognition among physicians and physiotherapists as common, non-invasive options of treatment (Beckerman et al., 1992; Draper et al., 2010). Regarding the use of US for MTP in the upper trapezius muscle (uTM), it was reported to induce a significant reduction in pain intensity (Majlesi and Ünalan, 2004), but other studies failed in showing significant effects on pain (Gam et al., 1998) or superiority to placebos (Lee et al., 1997); conflicting evidence for US was also stated by a systematic review (Vernon and Schneider, 2009). Regarding LLLT, reports are generally positive, showing superiority to placebos and effectiveness on pain and functional outcomes in MTP (Vernon and Schneider, 2009). However, when employed in a combined regimen of treatment with exercise and stretching, results are conflicting (Altan et al., 2005). Notably, in the United States, several medical coverage policies do not cover LLLT for any indication because it is considered an experimental/ investigational/unproven intervention (Blue Cross TEC, 2010). A strong disagreement between researchers and clinical practitioners emerges when comparing clinical guidelines (Chartered Society of Physiotherapy, 2006; Australian Physiotherapy Association, 2001; Canadian Physiotherapy Association, 2010) to survey about US and LLLT protocols (Lindsay et al., 1990; Robertson and Spurritt, 1998; Wong et al., 2007). The latter reveals a tendency to their massive utilization as stand-alone therapies in spite of no consensus about their effectiveness (Gam et al., 1993; Robertson and Baker, 2001). Such modus operandi is questionable and generally not recommended by clinical guidelines (see in the previous text) because these therapies are considered complementary to other more established interventions, such as manual therapy and exercise. However, the employment of combined regimens of treatment leads to conflict attributions of effects (Gam et al., 1998; Altan et al., 2005), making it difficult to evaluate the role of each
single therapy. The aim of the present study was to assess the effectiveness of US and LLLT delivery procedures according to protocols commonly used by physiotherapists (Wong et al., 2007; Chow et al., 2009). Firstly, it was evaluated whether a stand-alone therapy with either US or LLLT was effective on MTP in the uTM. Secondly, such protocols were compared with each other, with placebos or no therapy. Finally, all strategies were reassessed after a 12-week follow-up.
Methods Subjects and design Trial design was set as a randomized, double-blind, placebo-controlled, parallel-group study approved by the local Ethical Committee (Bioethics Board of ASL n.1-Sassari, prot. U-553, October 18, 2011; trial data fully available at the Department of Biomedical Sciences). All participants signed a written consent and an information notice regarding experimental procedures before enrolment. Sixty volunteers (28 women and 32 men; mean age 24.5 ± 1.44 years; range 20–30 years) with a spontaneously painful area and a palpable taut band in at least one uTM, disturbing normal daily activity, participated in the study. Subjects were recruited among university students, who are highly exposed to work-related musculoskeletal symptoms (Hupert et al., 2004). Exclusion criteria were the following: to meet general contraindications to electrophysical agents according to established guidelines (Robertson et al., 2001; Chartered Society of Physiotherapy, 2006; Houghton et al., 2010); to be a physiotherapy student or to have already experienced US and LLLT; a numerical rating score for pain lower than 3 out of 10; no impact of the condition under study on normal daily life; concomitant general diseases; diffuse generalized musculoskeletal and joint pain, head, neck or upper back pain due to osteoarticular pathologies; ongoing physiotherapy treatment for the condition under study at the time of enrolling or current use of medications affecting pain and skin sensibility and shoulder and neck surgery within the previous year. The recruited population underwent a clinical examination performed by an orthopaedic physician experienced in musculoskeletal disorders, and MTP were diagnosed as active or latent (Simons, 1995). Selected subjects with at least one active MTP within the uTM underwent functional assessment performed by a physiotherapist experienced in musculoskeletal conditions. The assessment procedure was performed at three different times: Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
before intervention (baseline), after 2 weeks (post) and after 12 weeks (follow-up) from the initial evaluation. Demographic data and participants’ clinical characteristics including age, sex, symptoms duration and referred pain patterns and involved side were recorded.
Stand-alone Therapies in Myofascial Trigger Points
examination, the active MTP (or the most painful one, if more than one was detected) within the uTM was marked with a dermographic pencil to allow a reliable localization. Active ultrasound
Randomization and blinding Participants were randomly assigned to one out of five parallel groups in 1:1:1:1:1 allocation ratio, to receive respectively active US (UsA-group I: n = 12), placebo US (UsP-group II: n = 12), active LLLT (LtA-group III: n = 11), placebo LLLT (LtP-group IV: n = 11) and with the fifth group acting as the control (no therapy: n = 14). No blocking was set for randomization. To generate the allocation sequence, Research Randomizer 3.0 software was employed (Urbaniak and Plous, 2011). Sixty opaque envelopes were prepared and numbered consecutively by AM, whereas participant allocation to envelopes was performed by one of the three physiotherapists employed to deliver treatments. According to the envelope selection, patients were allocated to one of the five groups listed in the previous text. Both outcome assessor and participants were blinded to interventions, whereas the three treating physiotherapists employed could not be blinded when delivering as they had to open each envelope and read the assigned number and group of allocation. No communication occurred between treating physiotherapists and the outcome assessor regarding trial course and participants.
Eme-Medical Ultrasonic 1300, 1–3 MHz US machine was used. Because a dose–response relationship for US has not been identified yet by well-designed randomized controlled trials (Robertson, 2002), delivery parameters were set according to Wong et al. (2007) and Draper et al. (1995): frequency 3 MHz, continuous mode, intensity 1.5 W cm2, duration 12 minutes. US probe (head size: 5 cm2) was applied steady with no pressure over the trigger point. A coupling agent was used to optimize delivery at the probe-skin interface. Because the painful area was limited to a trigger point, there was no need to move the US probe with rotating movements as the ratio between effective radiating area and the size of US head was approximately 1:1. No adverse effects such as overheating occurred during the trial. Placebo ultrasound In group II, sham-US was applied following the same modalities of the active therapy, but the US head cable was disconnected from the working device. Patients could not know whether the cable was connected or not to the machine as the display was also on with the time counting. Active low-level laser therapy
Interventions All interventions and data collection took place at the Physiotherapy Clinic of the University Hospital of Sassari, Italy. Prior to the trials beginning, therapists were instructed on how to deliver both active treatment and placebo, patient’s position and procedures and, also, to maintain an emotional detachment from the participant, avoiding any extra-intervention interaction with him/her. Sessions were performed at the same time of the day and in the same location. The participants were recommended not to use any drugs for the condition under study throughout the 2-week physical therapy or no therapy periods. Groups I–IV were treated five times weekly for overall 10 treatments given in 2 weeks. Patient’s position was set as relaxed high sitting with backrest. During orthopaedic Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A J&S Medical-Cyberlight (J&S, Rome, Italy), Mod. Ls1-Ga/As LLLT machine was used. According to the acknowledged guidelines (WALT, 2004), delivery parameters were: wave-length 904 nM; pulse duration 200 ns; pulse frequency 1953 Hz; peak power 90 mW; average output 30 mW; power density 22.5 mW cm2; treatment time 600 seconds; energy dose 18 J per session; spot size 4 cm2 and treatment frequency five times/week. Laser probe (head size: 4 cm2) was applied steady in skin contact with no pressure over the MTP. Placebo low-level laser therapy In group IV, sham-laser was applied following the same modalities of the active therapy, but the laser probe cable was disconnected from the working device. As for the placebo US, patients could not
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
know whether the cable was connected or not to the working machine. No therapy (Control) Subjects randomly assigned to group V (n = 14) only underwent baseline, post and follow-up evaluations without being given any treatment (active or sham) in between. Outcome measures All groups underwent a baseline, a post-intervention assessment and a follow-up session 12 weeks after the end of the intervention phase. In each session, three outcome measures were set: pressure pain threshold (PPT); pressure-related pain assessed with a numerical rating scale (NRS); upper trapezius extensibility assessed, through cervical goniometry, during an active lateral cervical flexion (cLat-Flex) to the side opposite to the affected muscle. Pressure pain threshold was determined using a hand-held pressure algometer. A dual force gauge mod. FDK-20 (Wagner Instruments, Greenwich-CT) was used (calibration in kg cm2; capacity 10 kg × 100 g; 20 lb × 0.25 lb). The force recorded corresponded to the quantity of pressure required to change subject’s sensation from pressure to pain. According to Fischer (1987), pressure amount was increased at a rate of 1 kg cm2 second, three measurements were taken with a rest time of 20 seconds, and the mean was recorded for data analysis. Immediately after recording PPT, subjects were asked to score on a NRS their pain at the changing point from sensation of pressure to pain, and the mean was retained for data analysis. Lateral cervical flexion was assessed using a gravityreference Myrin goniometer (prod. Patterson Medical, Illinois). Goniometer was fixed to the front of the head with a Velcro® fastening strap. The subject’s position was sitting with backrest and shoulders stabilized by the examiner. As for PPT and NRS, three measurements were taken, and the mean calculated for data analysis. Statistical analysis Data analysis was performed by using SigmaPlot-12 (Systat Software Inc., San Jose, California) by a blinded statistician (EL). A priori power analysis to determine the sample size was performed using G*Power 3
(Faul et al., 2007). By using a power (1-β err. prob.) of 0.80, α = 0.05 and an estimated effect size of 0.5 (Cohen, 1992), the required sample size to detect a 20% difference between treatment groups for the primary outcomes was 11 subjects per group. Shapiro– Wilk test was used to test the normality of distribution for continuous variables, whereas the homogeneity of variances was evaluated by Levene’s test. Two-way (five conditions × three times) ANOVA was used to compare changes in variables before and after intervention (post and follow-up) and among groups, as well as main effects and interactions. When significant differences were detected, a post-hoc Holm–Sidak method multiple comparison procedure was utilized. Significance level was set for a p-value < 0.05. Unless otherwise stated, data are expressed as mean ± standard deviation.
Results Seventy-nine subjects met eligibility criteria for recruitment but 16 of them refused consent, three were excluded due to a NRS score < 3, so that 60 of 79 subjects were enrolled. Recruitment and participant flow are presented in Figure 1. No harm or unintended effects occurred during this trial. At baseline, all groups were statistically homogeneous for age, PPT, NRS, cLat-Flex, symptom duration and gender as shown in Table 1, which reports demographic and clinical characteristics of the sample. Results are shown BY GROUP (Post vs. Pre; Follow-up vs. Pre and Follow-up vs. Post) and detailed with scores and Cohen’s d effect sizes (Cohen, 1992) in Table 2. After the 2-week intervention for groups I–IV and no therapy for group V (Control), each group showed statistically significant changes (p < 0.05) of all outcome variables. In particular, PPT and cLat-Flex increased while NRS decreased. Compared with the baseline, improvements remained statistically significant at follow-up in all groups. Compared with the post, further significant changes of outcome scores were observed in several groups at the follow-up (Table 2). Significances ensued from multiple comparisons among groups at post and followup, performed with a two-way ANOVA, are presented in Figure 2, whereas the main effects (Condition and Time) and interactions (Condition × Time) are reported in Table 3. Regarding PPT within Post, when performing all possible comparisons, no statistically significant differences were detected between active therapies Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points
Figure 1. Organization chart
Table 1. Demographic and clinical characteristics of the sample at the baseline assessment Group Variable Age (years) Gender (number and %) 2
PPT (kg cm ) 95% CI NRS (0–10) 95% CI cLat-Flex (°) 95% CI Symptom duration weeks
UsA (n = 12)
UsP (n = 12)
LtA (n = 11)
LtP (n = 11)
Control (n = 14)
24.5 ± 1.7 F: 7 (58.4%) M: 5 (41.6%) 2.13 ± 0.24 1.99–2.27 4.96 ± 1.20 4.28–5.64 31.17 ± 4.39) 28.69–33.65 3.5 ± 1.8
26 ± 0.8 F: 6 (50%) M: 6 (50%) 2.13 ± 0.20 2.02–2.24 4.68 ± 1.31 3.96–5.42 30.75 ± 4.8 28.01–33.49 3.0 ± 1.5
24 ± 2.1 F: 7 (58.4%) M: 5 (41.6%) 2.02 ± 0.34 1.82–2.22 4.49 ± 0.97 3.92–5.06 30.64 ± 4.20 28.16–33.12 4.0 ± 1.0
25.4 ± 0.7 F: 7 (58.4%) M: 5 (41.6%) 2.09 ± 0.27 1.93–2.25 5.15 ± 1.03 4.54–5.76 31.64 ± 5.05 28.66–34.62 3.0 ± 0.9
23 ± 1.91 F: 7 (50%) M: 7 (50%) 2.12 ± 0.20 2.02–2.22 4.60 ± 1.05 4.05–5.15 30.93 ± 5.15 28.23–33.63 3.5 ± 0.8
UsA = active ultrasound; UsP = placebo ultrasound; LtA = active laser therapy; LtP = placebo laser therapy; Cont = no therapy; PPT = Pressure Pain Threshold; NRS = Numerical Rating Scale; cLat-Flex = Cervical Lateral Flexion; CI = confidence interval evaluated by Levene’s test. Values are expressed as mean ± SD.
(UsA; LtA) versus their placebos (UsP; LtP), whereas control group scored significantly less than both active and placebo groups (p < 0.05; d > 0.8). No differences were found among intervention groups (I–IV) at 12-week follow-up, while controls kept on scoring less than any group (p < 0.05; d > 0.5). Regarding NRS within Post, notably LtA scored significantly better than UsA (p = 0.04), UsP (p = 0.03) and Control (p = 0.002) but not than LtP (p = 0.21). Physiother. Res. Int. (2014) © 2014 John Wiley & Sons, Ltd.
Both active (UsA; LtA) and placebo therapies (UsP; LtP) scored better than no therapy (p < 0.05; d > 0.8). Within follow-up, no significant differences were detected among groups with exception of UsA that scored better than Control (p = 0.03) but not than any other intervention group. When comparing LtA versus LtP versus Control for the dependent variable NRS (Table 3), a significant interaction (Condition × Time) was found (F = 2.499; p = 0.048). The effect of different levels of Time depends on the Condition, so that both factors
A. Manca et al.
Stand-alone Therapies in Myofascial Trigger Points Table 2. Baseline to Post and Follow-up scores by group Group Outcomes
UsA (n = 12)
UsP (n = 12)
LtA (n = 11)
LtP (n = 11)
Control (ref) (n = 14)
2.02 ± 0.34 2.74 ± 0.41* 1.14 (0.25–1.95) , 2.8 ± 0.39§ # 0.49 ( 0.31–1.25)
2.09 ± 0.27 2.69 ± 0.37* 1.06 (0.18–1.86) 2.8 ± 0.35§ 0.51 ( 0.29–1.27)
2.12 ± 0.2 2.35 ± 0.28*
4.49 ± 0.97 2.24 ± 1.01* 1.63 (0.67–2.48) , 1.3 ± 1.01§ # 0.67 (0.16–1.46)
5.15 ± 1.03 2.57 ± 0.87* 1.39 (0.47–2.22) 2.1 ± 1.64§
