A three-arm, randomized controlled trial with a blinded assessor was conducted.

The project was approved by the Research Ethics Committee of Universidade Cidade de São Paulo (UNICID), São Paulo, Brazil (#50548115.8.0000.0064) and was prospectively registered on Clinicaltrials.gov (NCT02759757). All eligible participants were informed about the objectives of the study and were asked to complete an informed consent form. The study was conducted at the Physical Therapy Clinic of the Universidade Cidade de São Paulo.

The present study included participants of both sexes, aged between 18 and 60 years, presenting with chronic non-specific LBP (pain or discomfort with duration of at least three months between the costal margins and the inferior gluteal folds, with or without referred pain to the lower limbs). Patients who had previously used KT for their LBP and/or any other adhesive tape and patients with any contraindication to their use were excluded from the study. Other exclusion criteria were: serious spinal pathologies (fracture, tumor, inflammatory or infectious disease), previous spinal surgery, patients with nerve root compromise, decompensated cardiorespiratory and metabolic diseases, pregnancy, and previous use of KT for any condition.

Initially, an individual physical examination was conducted,6 including collection of sociodemographic (such as age, gender, marital status, and education levels) and anthropometric data (such as weight, height, and body mass index) and assessment of pain intensity using the 11-point Numeric Pain Rating Scale (NPRS).24 Disability was assessed with the Roland-Morris Disability Questionnaire (RMDQ)25 scored from 0 (no disability) to 24 (high disability).

Next, the area to be tested was shaved and cleansed with 70% alcohol. The surface EMG electrodes were applied with T350 double-sided tape, and the electrogoniometer sensors, with T10 double-sided tape. The participants were positioned in slight trunk flexion for the instrumentation: placement of the electrogoniometer and surface EMG electrodes made according to SENIAM26–28 recommendations and placement of the lumbar dynamometry sensors. The EMG electrodes were applied bilaterally on the longissimus, two fingers width from the L1 spinous process and over the iliocostalis at the L2 level (Fig. 1a).29 In addition, an electrogoniometer was positioned between the T12 and S1 vertebrae. One arm of the electrogoniometer was directed caudally and the other cranially with the slightly tensioned spring aligned with the spine (Fig. 1a). The outputs of each channel of the electrogoniometer were previously calibrated according to manufacturer´s instructions to capture the angle variations during data collection.30 Then the participant was positioned in trunk extension, from the position of 30 degrees of trunk flexion measured by a common goniometer (Fig. 1b),31 as well as extension of the cervical spine, elbows, and knees.

Figure 1.

Placement of the EMG electrodes and electrogoniometer (A) Initial position on the dynamometer (B), KT application (C) Missner® surgical tape® application (d).

(0.18MB).

Subsequently, the participants received instructions from the therapist to become familiar with the equipment through the trunk extension movement. From that position, the initial evaluation (pre-intervention) was conducted for a maximum voluntary isometric contraction (MVIC) test, in a sustained position for 10s, with the EMG signal synchronized to the electrogoniometer and dynamometer data.32 The first two seconds and the final two seconds were disregarded to minimize signal contamination, therefore only the central six seconds of data collection were used for analysis.

After this pre-intervention assessment, participants were randomized into three treatment groups: Intervention/KT, Placebo, or Control.

The randomization process was generated by a software program (Excel Office 2010) and performed by a researcher not involved with data collection. Concealed allocation was achieved through the use of sequentially numbered, sealed, and opaque envelopes. After the initial evaluation, the participants were randomly allocated to one of the three groups: Kinesio Taping group; Placebo group, and Control group.

Kinesio Taping Group: Patients allocated to this group received the Tex Gold Kinesio Taping® bandage according to manufacturer´s instructions.17 The bandage was applied in the shape of an “I” in the direction of insertion (S1 region) to the origin (T12 region), with tension of 10–15% (Paper-off), in the trunk flexion position. This procedure was conducted minutes after instrumentation and pre-intervention assessment. KT was applied over the EMG electrodes monitoring the longissimus muscles at the L1 level (portion of the longissimus dorsi; Fig. 1c).

Placebo Group: Patients allocated to this group had a Missner® surgical tape (5cm in width) placed on the longissimus muscle. The form of application adopted generated folds in the skin, and an “I” shape in the trunk flexion position. The tape was applied over the EMG electrodes, bilaterally to the spine towards the insertion at S1 to the origin at T12, minutes after instrumentation (Fig. 1d).

Control Group: The patients allocated to this group did not receive any type of bandage.

Immediately after randomization and application of the intervention, the participants repeated the maximum voluntary isometric contraction (MVIC) test, in a sustained position for 10s, as performed pre-intervention. The same test was again performed 30min post-intervention with the tape remaining in place.

The primary and secondary outcome measures of the study were obtained by an assessor who was unaware of the allocation of patients to their treatment groups by covering their back with a shirt. The primary outcome of the study was the EMG signal of the iliocostalis and longissimus muscles.

The EMG signal was acquired with an eight channel Biometrics DataLog MWX8 wireless data system via SX230–1000 EMG electrodes (Biometrics Ltd., Gwent, UK) with the following specifications: active (single differential), bipolar, reusable surface disk electrodes with eighth order preamplifier and elliptical filter (-60dB at 550Hz), 38mm long, 20mm wide, 0.6mm thick, 10mm diameter discs, 20−450Hz bandpass filter, and third order high pass filter (18dB/octave; 20Hz), CMRR>96dB; noise ratio < 5 μV Input impedance > 1015 Ω Ohms. The connection cables of each channel were previously tagged with the name of the evaluated muscles and the tags remained in place for all data collections. The EMG signal was amplified (K800) and sampled at 1000Hz and data were reported according to the International Society of Electrophysiology and Kinesiology (ISEK) guidelines. A reference electrode (Biometrics R206) was placed on the right wrist.35 In our study, crosstalk was minimized through wireless transmission and the use of surface sensors with amplifiers and internal filters. This choice was based on the small difference in crosstalk when compared to intramuscular electrodes,36 as well as good reliability.37 In addition, the MVIC angle was limited to 30°, minimizing sensor movement. A 20Hz filter was also used, which is above the minimum recommended of 10Hz,38 thus minimizing contamination by up to 47%.39 In the EMG literature, signal normalization is important to eliminate signal variability between evaluations40 and this process depends very much on the type of activity in which signal was collected, e.g., static41 or dynamic condition,42 healthy or unhealthy participants,43 by means of maximum voluntary contraction or not.44 The EMG signal of the right and left iliocostalis and longissimus muscles was captured for signal amplitude analysis (RMS) and Medium Frequency (MF) in the initial evaluation (pre-intervention), after randomization (immediately post-intervention), and 30min post-intervention.

Data collection was performed by three individuals (one therapist and two assessors). The therapist participated in 100% of the EMG data collections. Narayanaswami et al.45 reported the importance of having the same therapist for good reliability with EMG equipment.

The secondary outcome of the study was pain intensity as measured by the 11-point NPRS scored which ranges from 0 (no pain) to 10 (worst possible pain).46 Pain intensity was collected minutes before the initial evaluation (pre-intervention), shortly after randomization (immediately post-intervention), and 30min after randomization (30min post-intervention).

In addition to the collection of the primary and secondary outcomes, RMDQ25,47 was only administered pre-intervention to measure disability at baseline. The questionnaire consists of 24 items that describe activities that patients with LBP may have difficulty performing on a daily basis, where each affirmative answer corresponds to one point on the scale. The final score of the RMDQ was determined by the sum of the values obtained: the higher the score, the greater the disability.25,46 Patients were instructed to complete the statements according to how they felt on the day of the evaluation.

The sample calculation of the study was performed to detect a difference of 2.8 units in RMS of the EMG signal with estimated standard deviation of 2.5 units of the EMG signal.48 Assuming a statistical power of 95%, alpha of 5%, and a possible sample loss of up to 20%, a total of 63 participants (21 participants per group) were included.

The statistical analysis followed intention-to-treat principles (i.e. patients were analyzed into the groups they were originally allocated to).49 The normality of the data was tested by visual inspection of histograms, and the characterization of the participants was presented using descriptive statistical tests. The primary analysis of interest was the difference in EMG signal between the 3 groups. The between-group differences (effects of treatment) and their respective 95% confidence intervals (CI) were calculated by constructing mixed linear models50 using interaction terms (group versus time). Treatment effects were adjusted for baseline values for all outcomes and for multiple comparisons. Missing data were handled using linear mixed models. The analyses were performed using the software SPSS 19.

From a total of 100 potential participants identified between August and December 2016, 63 were considered eligible and were included in the study. Potential participants were excluded due to different ineligibility reasons: age over 60 years (n=25), age under 18 years (n=1), acute non-specific LBP (n=5), previous spinal surgery (n=2), knowledge of the Kinesio Taping® technique (n=2), nerve root involvement (n=1), and knee pain only (n=1). All randomized patients were included in the statistical analysis (Fig. 2). One patient dropped out from the remaining assessments after been assessed at baseline due to personal issues. The characteristics of the included participants in the initial evaluation are described in Table 1. The baseline characteristics of all variables were similar in the three groups.

Figure 2.

Flow diagram of participants throughout the study.

(0.11MB).

Table 2 presents the descriptive data of the primary and secondary variables at all time points. There was no significant change in muscle activity of the iliocostalis and longissimus muscles as measured by EMG amplitude (RMS) and frequency (MF) with the use of Kinesio Taping. Of the 96 statistical comparisons (Table 3), a statistically significant difference in EMG signal was only observed for the RMS max of the right iliocostalis muscle and the medium frequency (MF) of the right longissimus muscle in favor of the Kinesio Taping group compared to the placebo (surgical tape) and control group.