Abstract
Background Little evidence exists regarding parameter selection for hypoalgesia using interferential therapy (IFT).
Objective This study investigated segmental and extrasegmental hypoalgesic effects of different IFT parameter combinations upon experimentally induced pressure pain threshold (PPT) in pain-free volunteers.
Design The participants were randomly assigned to 6 groups: control, placebo, bipolar constant amplitude modulation frequency (AMF), bipolar sweep AMF, quadripolar constant AMF, and quadripolar sweep AMF.
Setting The study was conducted in a university laboratory.
Participants One hundred eighty adults who were healthy and pain-free participated in the study.
Intervention Interferential therapy was delivered to all groups at high, to-tolerance intensity and at high AMF. Stimulation to the dominant forearm was delivered for 30 minutes, with monitoring for a further 30 minutes.
Measurements Pain pressure threshold was measured at the area of first dorsal interosseous muscle of the dominant and nondominant hands (segmental measurements) and over the tibialis anterior muscle (extrasegmental measurement) at baseline and at 10-minute intervals using a pressure algometer. Square root transformed PPT data were analyzed using repeated-measures analysis of variance.
Results There was a significant change in PPT over time, but no significant between-subjects difference in segmental or extrasegmental PPT between any of the IFT groups and the placebo or control group. Thus, IFT delivered in any of these parameter combinations did not significantly affect the PPT of pain-free participants compared with the control or placebo group.
Limitations Success of blinding was not evaluated.
Conclusions This study showed that IFT delivered at high, to-tolerance intensity and high AMF does not produce significant segmental and extrasegmental hypoalgesic effects on PPT in participants who were healthy compared with a control or placebo group. Further research is warranted to investigate the hypoalgesic effect of different IFT parameter combinations and to explain its possible mechanism of action.
Interferential therapy (IFT) is a popular electrotherapeutic modality for pain management.1,2 It is characterized by the interference of 2 medium-frequency currents (ie, 1–10 kHz), which combine to produce a new medium-frequency current whose amplitude is modulated at low frequency (ie, <1 kHz).3 Interferential therapy is applied transcutaneously via electrode pads, either in bipolar application (ie, current is delivered via 2 pads, and amplitude modulation occurs within the stimulator) or quadripolar application (ie, current is delivered via 4 pads, and interference occurs within the tissues). Although the exact mechanism of action of IFT on pain modulation is unknown, selective stimulation of large- or small-diameter afferent fibers, by using different doses of amplitude modulation frequency (AMF), has been proposed as a possible mechanism for achieving analgesia with IFT.4–6 Other theoretical mechanisms of IFT analgesia include activation of the descending pain suppression pathway with release of endogenous opioids, physiological block of nerve conduction, increased circulation, and placebo mechanisms.4–7 Thus, IFT has loosely been categorized as another form of transcutaneous electrical nerve stimulation (TENS), which is traditionally considered to be a low-frequency electrical stimulation technique.7
The main differences between IFT and TENS are the carrier frequency and the AMF. Theoretically, these characteristics of interferential current reduce skin impedance, and penetration of the current to deeper tissues is allowed.3,4 This theory was reinforced by the clinical use of different AMF modes (ie, constant mode, where AMF remains constant over time, or sweep mode, where AMF changes periodically over time) and different modes of IFT application (ie, bipolar or quadripolar). However, the existing research evidence on the effects of different AMF modes and different IFT application modes on pain management is limited, which questions their potential role in IFT stimulation.5,8–14
Experimental studies investigating hypoalgesic effects of different parameter combinations of modalities such as IFT, using established pain models on pain-free participants, can elucidate the mechanisms of action of the modality and its hypoalgesic efficacy.15 However, few studies have been conducted into IFT parameter manipulation, and these studies have not reached consensus on the hypoalgesic effect of different parameter combinations and, therefore, the mechanism of action of IFT.6,13,16–20 Transcutaneous electrical nerve stimulation is a similar modality with a considerable literature base, which has demonstrated that a combination of high frequency (ie, >100 Hz) and high, to-tolerance intensity is effective in changing the pressure pain threshold (PPT) of people who are pain-free through segmental stimulation (ie, in the area of the experimental pain induction), or a combination of both segmental and extrasegmental stimulation (ie, in a different area from that of the experimental pain induction).21 The exact mechanism of action of this TENS parameter combination in pain modulation is not known but may be due to both local and central effects.22,23 A similar combination of high AMF and high, to-tolerance intensity has not been tested in IFT. Given the assumption that TENS and IFT should have comparable physiological mechanisms of action within the tissues, it is plausible to expect similar parameter combinations to have similar outcomes.
Accordingly, this study investigated the segmental and extrasegmental hypoalgesic effects of IFT upon experimentally induced PPT in participants who were healthy using high AMF and to-tolerance intensity parameters. The study also investigated, secondarily, segmental and extrasegmental hypoalgesic effects of different modes of AMF (ie, sweep or constant mode) and different modes of application (ie, bipolar or quadripolar) delivered at high frequency and at to-tolerance intensity. These effects were investigated to determine whether these specific parameters play an important role in IFT stimulation.
Method
A randomized, triple-blinded, placebo-controlled design with repeated measures was used. The experimental groups were 4 active IFT groups with different parameter combinations, a control group, and a placebo group (Tab. 1). The randomization procedure was based on random permuted blocks, with sex as a blocking factor; random number lists were generated by computer. The primary outcome was PPT measured at the dominant and nondominant hands to investigate the segmental hypoalgesic effect (local and dermatomal responses) and at the nondominant leg to investigate the extrasegmental effect (generic response) of the intervention.
Summary of Parameters for the 6 Experimental Groups
Participants
Participants who were healthy and pain-free were recruited from the student and staff population of Keele University. Inclusion criteria were good general health and age between 18 and 50 years. To minimize possible performance bias, only volunteers who had not received electrical nerve stimulation before and were unaware of the sensation produced by the electrical current applied (electrotherapy naive) were included in the study. Exclusion criteria were the contraindications and precautions to electrical nerve stimulation stated by the UK Chartered Society of Physiotherapy, which include: active or suspected malignancy; vascular impairment; sensory loss in the area to be treated; acute dermatological conditions; devitalized tissue; diagnosed neuromuscular, cardiac, or respiratory disorders; epilepsy; metal or any active implant in the area of stimulation; active epiphysis; recent bleeding or hemorrhage; and first 35 weeks of pregnancy.24 Because PPT was the outcome measure, volunteers who were in pain or taking pain medication or who were recovering from illness or an operation were excluded. To ensure no or minimum damage to the tissues, volunteers who bruised easily or had any skin allergies (including those to metal) also were excluded. These inclusion and exclusion criteria were assessed via a screening questionnaire given prior to the experiment by the main researcher, who had the role of coordinator during the experiment. Volunteers also were excluded if they were unable to comprehend the instructions or unable to cooperate with the experimental protocol.
Procedure
Participants were informed of the scope, the procedure, and the dangers of the study at least 24 hours prior to participation. They were assured that their participation was voluntary and they had the right to withdraw from the experiment at any time without giving an explanation and that withdrawing from the experiment would not affect their relationship with the university, the school, or the investigators. Participants were asked to refrain from intense exercise and from consuming coffee, tea, alcohol, or any analgesic medication in the 12 hours prior to the experiment due to the potential confounding influence of these factors on pain threshold.25–30 They were given an information sheet describing the study, and those willing to take part provided written consent.
Participants were seated in a comfortable long-sitting position on a plinth with the forearms resting on tables next to the plinth and the legs in a straight position. The stimulation area was the forearm of the dominant hand, and the electrodes were arranged in a diagonal pattern to permit stimulation of the radial nerve where it becomes superficial (Fig. 1). During the quadripolar application, the electrodes were placed as follows: for the first channel, the first electrode was placed on the medial border of the forearm 1 cm proximal to the styloid process of the ulna, and the second electrode was placed on the lateral border of the forearm 1 cm distal to the elbow crest on the muscle belly of the brachioradialis muscle; for the second channel, the first electrode was placed on the lateral border of the forearm 1 cm proximal to the radial styloid process where the radial nerve becomes superficial, and the second electrode of the second channel was placed on the muscle belly of the extensor carpi ulnaris muscle. For all participants, the forearm was in pronation during the experiment. Prior to stimulation, this area was cleaned with alcohol wipes to create good electrical contact. Stimulation was delivered to the tissues via self-adhesive, single-use, carbon rubber, gel electrode pads (PALS Electrode TPN40 Valutrode [5 × 5 cm], Physio-Med Services, Glossop, Derbyshire, United Kingdom), which were secured with tape for good electrical contact. Based on the computer-generated random number lists, the coordinator assigned participants to groups. The participants were blinded to their group allocation, apart from those in the control group. Accordingly, 4 electrode pads were applied to the dominant forearms of all participants irrespective of their group allocation. During the bipolar application, only the second channel was active. The coordinator was responsible for setting the machine to deliver the current in a bipolar or quadripolar mode and in a constant or sweep AMF mode; the participant was blinded to these settings.
Area of stimulation.
Interferential current was delivered using an Endomed-M 433 two-channel IFT stimulator (Enraf-Nonius BV, Postbus 12080, 3004 GB Rotterdam, the Netherlands), which was calibrated using an oscilloscope prior to the experiment. The IFT literature on the effect of different carrier frequencies on sensory stimulation is sparse and inconclusive.31,32 However, both clinical and experimental studies investigating the effectiveness and efficacy of IFT on pain show consistent use of 4 kHz as the carrier frequency, with other parameters of IFT dose, such as AMF and intensity, being manipulated. Specifically, 8 clinical studies using 4 kHz as the carrier frequency but various combinations of AMF and intensity showed that the IFT intervention decreased pain.33–40 In 3 laboratory studies using 4-kHz as the carrier frequency, significant hypoalgesic effects on ischemic, heat, and cold pain thresholds were produced by the IFT intervention.16,41,42 On this basis, the carrier frequency used in this study was 4 kHz, and the parameters under investigation were high AMF and high, to-tolerance current intensity, combined to form a dose that has not hitherto been tested in the IFT laboratory literature.
All active experimental groups were told to expect a “strong and uncomfortable” buzzing feeling under the electrode pads, which might extend to the dorsal surface of the thumb and the index, middle, and ring fingers. They were told that muscular contractions also were normally expected and that the strong and uncomfortable feeling should not be described as painful at any time, but should be to their tolerance (if a participant could not understand a “strong and uncomfortable” sensation, the intensity of stimulation was increased to a painful level and then decreased to one level lower). The intensity was checked every 2 minutes to maintain the strong and uncomfortable sensation. The swing pattern for both AMF sweep groups was gradual change between boundaries of AMF (ie, 80–110) within 6 seconds. The IFT machine was placed behind the plinth with its screen facing the opposite side of the room so that participants could not see the settings on the machine. Those participants allocated to the placebo group were told they might or might not feel a buzzing sensation under the electrode pads that had been placed on their skin. The IFT machine was turned on, but only the timer was set. The same procedure used for the active experimental groups was followed for the placebo group. Participants allocated to the control group were informed of their group allocation. They had electrodes applied to them to blind the independent rater, and the machine was turned on, but only the timer was set. The success of blinding was not assessed.
Pressure pain threshold was the main outcome measure for this study. Pressure pain threshold is an established experimental model of pain, which is atraumatic to measured tissues (thereby allowing repeated measurements without temporal interaction) and can be easily applied after a short period of training.18 There is evidence that both intrarater and interrater reliability of PPT measurements with handheld algometers in various sites in people who are healthy is moderate to excellent when trained raters are used and clear explanations are given to participants.43–47 For these reasons, PPT has previously been used to investigate the hypoalgesic effects of other electrical stimulation modalities using experimental designs similar to that of the present study and, therefore, was chosen as the experimental model of pain for this study.
Measurement of PPT was performed at 3 sites: (1) at the dorsal surface of the dominant hand, lateral to the midpoint of the second metacarpal bone, in the first dorsal interosseous muscle (corresponding to acupuncture point LI4), (2) the equivalent site on the nondominant hand, and (3) at the midpoint of the muscle belly of the tibialis anterior muscle, contralateral to the stimulation site. These measurement sites were chosen to allow PPT measurements to be taken at the same site as the area of stimulation and within the same dermatomal distribution (ie, first interosseous muscle of the dominant hand; segmental ipsilateral measurement site), at the contralateral site from the area of stimulation but within the same dermatomal distribution (ie, first interosseous muscle of the nondominant hand; segmental contralateral measurement site), and at a site different from the area of stimulation (ie, tibialis anterior muscle; extrasegmental measurement site).
The measurement sites in the area of the first dorsal interosseous muscles of both hands were in the same dermatomal distribution as the superficial branch of the radial nerve (C7), which was the primary target of the electrical stimulation. An increase in the PPT on these sites would suggest that the mechanism of action of IFT is related to a local or peripheral effect explained by pain modulation mechanisms, such as the gating system at the dorsal horn level or the physiological block of nerve fibers at the periphery.4–6,48 These measurement sites have been used previously in similar experimental designs.21,23,48–50 The measurement site in the muscle belly of the tibialis anterior muscle was in a dermatomal distribution different from that of the stimulation area; thus, a change in PPT in this site could suggest a more generic hypoalgesic response, which, in turn, could imply activation of central pain inhibitory mechanisms that involve release of endogenous opiates and activation of the descending pain suppression system.5,6 All measurement sites were chosen so that muscular tissue existed below them to avoid variability in PPT measurements between different tissues.51 The measurement sites were marked as suggested in the literature.52
An independent rater, blind to the treatment allocation of participants, took the PPT measurements with a handheld electronic Somedic type II pressure algometer (Somedic Production AB, Sollentuna, Sweden), which was calibrated before each experimental session. The probe tip area of the algometer was covered with a flat, metal, circular cup (diameter=1 cm) to avoid skin irritation from the rubber tip. The scale range of the algometer for the 1-cm probe tip diameter size was 0 to 2,000 kPa, with increments of 1 kPa. The pressure algometer provided a patient-operated switch that, when activated by the participant, held the PPT value in kilopascals. Participants were instructed to press the switch and say “stop” at the exact moment when the sensation from the applied pressure turned to pain. Demonstrations were allowed prior to the experiment to ensure participants had understood the PPT measuring procedure. The independent rater followed a standardized application of the pressure algometer according to guidelines in the literature.52 Specifically, the independent rater was trained to apply the algometer vertical to the measurement sites at a slow, continuous incremental rate of approximately 10 kPa/s/cm2 and to withdraw the algometer immediately after the participant's indication of PPT.52 Due to the large number of participants, 7 independent raters were recruited and used based on their availability.
The experimental protocol included three 10-minute periods of IFT stimulation followed by three 10-minute periods of no stimulation. Measurement of PPT occurred 7 times: at baseline and after each 10-minute interval until the end of the experiment. Two PPT measurements were taken on each occasion from each measurement site. Each measurement session lasted approximately 3 minutes. For all groups, the total experimental procedure lasted approximately 1 hour 20 minutes. This experimental protocol has been used previously.21,23,48–50
Sample Size
The existing literature on IFT does not provide information on effect sizes for a difference in PPT. However, a 1.5-kg/cm2 (147-kPa) side-to-side difference has been proposed as a criterion to discriminate tender spots from control sites in people who are healthy.53 Accordingly, 147 kPa was interpreted as an important difference in PPT between the active experimental groups and the control and placebo groups.
The standard deviation of PPT was estimated from the first 96 participants, from all 3 measurement sites (ie, in the dominant hand, the nondominant hand, and the leg). The largest PPT standard deviation estimate was for the leg (161 kPa), and this figure was used in the sample size calculation. On the basis of 8 pair-wise comparisons (the 4 active experimental groups versus the placebo and control groups) and assuming a cutoff for statistical significance of P≤.01 (2-tailed), 30 participants were required in each group to detect a standardized effect of 0.913 (147/161), with 80% power. Thus, a total of 180 participants were required, and recruitment continued until this number was reached.
Data Analysis
The data analyzed were the mean values of the 2 PPT measurements taken at each site for each time point. Preliminary analysis showed the residuals were somewhat positively skewed, and a square root transformation, therefore, was applied prior to further analysis.54 The data were analyzed using a repeated-measures analysis of variance (ANOVA) model, with time as the within-subjects factor, group as the between-subjects factor, sex as the blocking factor, a group-time interaction, and baseline PPT values as a covariate.
As the assumption of sphericity was violated (Mauchly test, P<.001), P values for the time factor were calculated for all analyses using the Greenhouse-Geisser correction. Statistical significance was set at P≤.05 (2-tailed) for omnibus tests, but at P≤.01 (2-tailed) for the 8 a priori contrasts between each experimental group and the control and placebo groups in order to control type I error inflation. This approach was preferred to the use of a Bonferroni correction, which is considered unduly conservative when the number of contrasts exceeds approximately 5.55 Effect sizes were calculated, as unbiased estimates of Cohen d (with 95% confidence intervals),56 for the mean difference in PPT across time points between the control group and each of the other groups (the 4 active treatment groups and the placebo group).
A secondary factorial analysis was conducted, using the same repeated-measures ANOVA model, to compare the effect of constant AMF versus sweep AMF and bipolar IFT versus quadripolar IFT. Thus, the 2 active groups that received constant AMF were compared with the 2 active groups that received sweep AMF, and the 2 groups that received bipolar IFT application were compared with the 2 groups that received quadripolar IFT (Fig. 2). This analysis was conducted to determine whether any difference in the effect of AMF mode and IFT mode was masked through their combination within the 4 active treatment groups. The absence of such effect could not be inferred from an absence of difference between the active treatment groups, as the size of each pooled group (n=60) was double that of each of the 4 active treatment groups. Statistical significance was set at P≤.05 (2-tailed) for these analyses. All data analysis was conducted blind to participants' group allocation using SPSS version 18 (SPSS Inc, Chicago, Illinois).
Schematic presentation of pooled groups for factorial analysis. AMF=amplitude modulation frequency.
Results
Descriptive statistics for the baseline characteristics of the participants are summarized in Table 2. Untransformed means and standard deviations for PPT in each group at each time point are shown in Table 3. The ages of participants were similar across the groups; therefore, age was not included in the analyses as a covariate.
Summary of Baseline Characteristics of Participants
Mean (SD) Pressure Pain Threshold Values (Untransformed)
IFT Parameter Combinations
For the baseline-adjusted square root transformed data, the interaction of time and group was not significant for any measurement site (P=.353, P=.717, and P=.685 for the segmental ipsilateral, segmental contralateral, and extrasegmental measurement sites, respectively), showing that the change in PPT scores over time does not follow a different pattern in relation to the individual study groups. Profiles of the adjusted estimated means of the untransformed PPT scores taken at the 3 measurement sites are shown in Figure 3.
Mean untransformed pressure pain threshold (PPT) (in kilopascals) for the segmental ipsilateral measurement site (plot A), the segmental contralateral measurement site (plot B), and the extrasegmental measurement site (plot C). Data are adjusted for baseline values of PPT and sex. The vertical dashed line marks the transition from the stimulation period to the poststimulation period. Key to groups: ●=bipolar sweep, ○=bipolar constant, ■=quadripolar sweep, □=quadripolar constant, ▲=placebo, △=control.
The PPT scores averaged across groups showed a main effect over time points for the segmental ipsilateral, segmental contralateral, and extrasegmental measurement sites (P=.021, P<.0005, and P<.0005, respectively). The main effects of the group factor were nonsignificant for the segmental ipsilateral, segmental contralateral, and extrasegmental measurement sites (P=.211, P=.463 and P=.325, respectively), indicating that, averaging over time points, the means of the groups did not significantly differ. The a priori pair-wise comparisons confirmed the lack of significant difference between any of the active IFT groups and the control and placebo groups. Table 4 shows values of Cohen d for each active IFT group and the placebo group. These values express differences in the mean adjusted untransformed PPT values across time, in relation to the control group. All of the effects lie in the region of a “small” effect for a difference between independent means (ie, d<0.50).56
Differences in Pressure Pain Threshold (PPT) Values (Untransformed, Adjusted for Baseline Values and Sex) in Relation to the Control Group for Each Active Group and the Placebo Group, Expressed as Unbiased Values of Cohen d, With 95% Confidence Intervals (CIs)a
Factorial Analysis
This secondary analysis investigated the effects on PPT of AMF mode (ie, constant AMF or sweep AMF) and IFT mode (ie, quadripolar IFT or bipolar IFT). For the segmental ipsilateral, segmental contralateral, and extrasegmental sites, interactions were nonsignificant for time and AMF mode (P=.635, P=.657, and P=.963, respectively; the upper panes in Fig. 4) and for time and IFT mode (P=.220, P=.550, and P=.574, respectively; the lower panes in Fig. 4), showing that the differences in PPT scores across time were not dependent on the particular mode of AMF or of IFT. Similarly, the between-subjects factor analysis showed that the main effects of AMF mode and IFT mode were nonsignificant (P=.104. and P=.132, respectively, for the segmental ipsilateral site; P=.936 and P=.802, respectively, for the segmental contralateral site; and P=.251 and P=.351, respectively, for the extrasegmental site).
Left pane: segmental ipsilateral effect of mode of amplitude modulation frequency (plot A) and mode of interferential therapy (plot B). Middle pane: segmental contralateral effect of mode of amplitude modulation frequency (plot A) and mode of interferential therapy (plot B). Right pane: extrasegmental effect of mode of amplitude modulation frequency (plot A) and mode of interferential therapy (plot B). Data are untransformed values of pressure pain threshold (PPT) (in kilopascals), adjusted for baseline values and sex. The vertical dashed line marks the transition from the stimulation period to the poststimulation period. Although the placebo and control groups were not included in the corresponding analyses, these groups are shown as reference groups. Key to groups: ●=sweep, ○=constant, ■=quadripolar, □=bipolar, ▲=placebo, △=control.
Summary
Although, averaged across groups, there was a significant increase in PPT scores over time, from the absence of any group × time interactions, it can be concluded that the pattern of change in PPT scores over time was independent of the tested combinations of AMF mode and IFT mode at each of the measurement sites. Furthermore, there was no significant difference in PPT scores, averaged over time, between any of the active IFT groups and the placebo and control groups at these sites. The factorial analysis revealed no significant group × time interactions, and no significant between-group main effects, for the groups defined by AMF mode or for those defined by IFT mode.
Discussion
The findings of this study demonstrated that all groups showed hypoalgesia, but the active IFT groups did not significantly differ from the control and placebo groups. In addition, neither AMF mode nor IFT mode contributed to a significant change in PPT in participants who were pain-free. Furthermore, the magnitude of the effects in the active IFT groups relative to the control group (d≤0.399 in each case) was small.
The lack of a significant hypoalgesic effect of IFT on PPT in this study suggests that IFT, delivered in the investigated doses, is not a potentially effective pain-inhibiting electrotherapeutic modality. This finding might imply either that the investigated IFT doses were incapable of eliciting a hypoalgesic effect on PPT in people who were healthy or that PPT was not an appropriate experimental model of pain to demonstrate the hypoalgesic efficacy of IFT. In relation to the investigated dosage, an extensive search of the IFT literature demonstrated that the combination of high AMF (delivered in either constant or sweep mode) and high, to-tolerance intensity (applied in either a bipolar or quadripolar mode) has not previously been investigated in a similar IFT experimental design.10,11,13,14,18,20,41,42,57,58 In view of the lack of evidence in the IFT literature, and on the assumption that IFT and TENS may activate similar pain inhibitory mechanisms, the investigated doses were drawn from the TENS literature. However, the current waveform of IFT is different from that of TENS, which might explain the inability of similar doses to produce similar hypoalgesic effects when tested in comparable experimental paradigms. The IFT current waveform is sinusoidal and amplitude-modulated at low frequency. However, it has been suggested that within the IFT “beat” formed by the AMF, some phases of the waveform are below the intensity level set, whereas others are suprathreshold.59 Hence, the dose of the effective part of the beat of the current, in terms of intensity and duration, would be uncertain and possibly insufficient to reach the threshold of the afferent fibers to achieve a strong hypoalgesic effect.
Thus, even if the dosage in the present study achieved high, to-tolerance intensity—which theoretically could reach the thresholds of sensory, motor, and probably pain fibers—the duration of the suprathreshold part of the beat might provide insufficient stimulation of the afferent fibers to elicit a hypoalgesic response similar to that observed with TENS when delivered in similar combinations of intensity and frequency. Additionally, the IFT waveform does not include a rest period between pulses, and because it is a medium-frequency current, the bombardment of the afferent fibers continues during their refractory period. Therefore, its mechanism of action might be related to the effects of its carrier frequency on excitable tissues rather than to its low-frequency stimulation.8 The hypoalgesic efficacy of dosages of IFT, including different carrier frequencies, remains to be explored through further experimental research, which ideally would include physiological measurements of fiber responses to the IFT waveform.
Evidence from the IFT experimental literature demonstrates that other IFT dosages, including high AMF and strong and comfortable intensity or low AMF and strong and comfortable intensity, also were found to be ineffective in changing PPT in people who were healthy.57,58 Although limited, this evidence in conjunction with the present findings suggests that IFT, applied at any combination of AMF and intensity, does not significantly influence PPT in people who are healthy. This suggestion, in turn, might imply either that IFT dosage should be manipulated based on parameters other than AMF and intensity (ie, carrier frequency), which would have implications for the pain-inhibitory mechanism of action of IFT, or that PPT is not an adequate experimental model of pain to demonstrate the potential hypoalgesic efficacy of IFT. This latter implication is reinforced by the findings of experimental studies on the hypoalgesic effect of IFT using other experimental models of pain such as heat or ischemic pain, which demonstrated that IFT delivered in various doses produced significant hypoalgesic effects compared with control groups (although the effect sizes were not reported).16,41 Different experimental models of pain induce different sensations,15,60 and this finding might explain different responses to IFT stimulation. Hence, it is possible that the lack of effect of IFT on PPT of people who were healthy demonstrated by the findings of the present study is related to the experimental model used.
Previous data on the effect size of IFT on PPT were lacking, and the power calculation for this study was based on the authors' interpretations of a specific PPT value used as a criterion to discriminate tender spots from control sites in people who were healthy.53 The power calculation, therefore, was based on detecting a large effect size (ie, 0.913; see sample size calculation). Retrospectively, it appears that IFT produces small effects on PPT in people who are healthy; hence, the present study was underpowered to detect such small effects. However, if a small effect on PPT is to be detected with IFT, the importance of this electrotherapeutic modality compared with others such as TENS is questionable. In a similar experimental protocol, with similar electrical stimulation doses, TENS produced large effects on PPT in people who were healthy.21
Findings in the neurobiology of pain suggest that the sensation of mechanically evoked pain—via pressure exerted onto the skin induced by algometry—may activate nociceptive afferents in several tissues, depending on the force applied, the size of the tip of the apparatus, and the technique used.61 It is debatable whether PPT is a function of skin or deep tissue responses, or both, and to what degree cutaneous and deep tissue nociceptors contribute to PPT.51,61–63 There is evidence in the pressure algometry literature that pressure pain sensitivity in the skin influences PPT,62,63 but there also is evidence that responses from deep tissues are evoked, depending on the tool and the force applied.51 However, the literature is not yet conclusive regarding the relative contribution of skin and underlying tissues to PPT. In the present study, electrical stimulation was applied in such a way as to mainly target the radial nerve where the nerve becomes superficial. The segmental PPT measurements were taken from the dorsal webspace of the hands to reflect dermatomal responses, as this area is cutaneously innervated by the superficial branch of the radial nerve (C7). If PPT is partially influenced by the deep tissue nociceptors (ie, first dorsal interosseous muscle), IFT stimulation of the superficial radial nerve could partially affect segmental PPT. In that case, the tactile threshold might be a better reflection of hypoalgesic response. The lack of effect of IFT on PPT found in this study, therefore, might be explained by the different nerve supplies of skin (ie, superficial branch of the radial nerve [spinal nerve C7]) and underlying muscle (ie, deep branch of the ulnar nerve [spinal nerve C8]) in the segmental measurement sites. However, on the basis that the forearm is a small area, the IFT current distribution within the tissues is unpredictable,9,12 and the actual mechanism of action of IFT is unknown, the possibility cannot be excluded that other excitable tissue also was stimulated and that other physiological responses were activated. Thus, the different innervation of skin and underlying tissue of the segmental measurement sites may have had little effect on the main outcome of this study.
The present study investigated the changes produced by specific IFT doses on an experimental model of pain in individuals who were healthy. Thus, extrapolations of the findings to clinical pain would be weak, as experimental pain experience cannot replicate the specific experience of clinical pain.15 Pressure pain threshold is a transient-effect model that does not purport to mimic clinical pain but is helpful in delineating the effect of different doses in laboratory trials preceding clinical work. Also, PPT concerns only mechanical pain induced by pressure, which is an acute feeling different from the sensation of clinical pain.
The findings of the present study contradict those of 2 randomized clinical studies, which assessed the segmental analgesic effectiveness of IFT delivered in similar parameter combinations (ie, high AMF and high, to-tolerance intensity) and reported pain relief as measured by visual analog scale.64,65 However, both studies used suction electrodes, which might produce vasodilatation and increased cutaneous blood flow, which is a possible mechanism of IFT analgesia.4,5,66 The potential placebo effect of IFT cannot be ignored in a clinical study, and neither of these 2 clinical studies assessed the placebo effect by including a sham IFT group in their design. Therefore, the specific therapeutic element of IFT remains to be identified.
A recent systematic review and meta-analysis of clinical trials of IFT in the management of various types of musculoskeletal pain concluded that when IFT was applied alone its effects did not differ from those of placebo or any other standard intervention.67 The authors concluded that—due to the limited number of clinical trials in this area, the heterogeneity across existing studies, and their methodological limitations—no definite conclusions could be made regarding the clinical effectiveness of IFT. Although this is a review of trials in clinical populations, whose findings are not directly comparable to those of the healthy participants in the present study, our findings nonetheless support its conclusion that the hypoalgesic effect of IFT does not differ from that of a placebo condition.
Limitations can be identified in the present study. The study was performed during a single attendance, and PPT was the only outcome measure. Although PPT was assessed using triple blinding (ie, blinding of the participant, the rater, and the researcher who analyzed the data), the success of blinding vis-à-vis the participants and raters was not assessed, and possible lack of success of blinding could have introduced performance bias.
Conclusion
This study showed that IFT delivered at high, to-tolerance intensity and high AMF did not produce significant segmental and extrasegmental hypoalgesic effects on PPT in participants who were healthy compared with a control or placebo group. This finding was the same regardless of the sweep AMF and the number of electrodes used. These new findings suggest that the mechanism of action of IFT for pain relief might not depend on the AMF or the intensity level, and further studies should focus on the role of the carrier frequency in IFT hypoalgesia. The findings of this study also have implications for the role of PPT as an experimental model of pain used to investigate the hypoalgesic effects of IFT in people who are healthy. The measurement properties of PPT should be further investigated to determine the degree of contribution of skin and deep tissues. Further research is needed to clarify the putative hypoalgesic effectiveness of IFT, effective doses, and its mechanism of action.
The Bottom Line
What do we already know about this topic?
Interferential therapy (IFT) is a popular electrostimulation modality used by therapists for the treatment of pain. Interferential therapy is delivered by either 2 or 4 electrode pads placed on the skin. The mechanism of action of IFT and its clinical effectiveness in reducing pain are not known.
What new information does this study offer?
This study investigated different doses of IFT and the effects on experimentally induced (mechanical) pain in otherwise healthy participants. The purpose was to determine whether different doses produce pain relieving effects.
If you're a patient, what might these findings mean for you?
This study showed that IFT delivered at different high-intensity doses does not produce meaningful pain relief under experimental conditions in healthy participants. This finding suggests that further research is still needed to clarify the mechanisms and clinical effectiveness of IFT.
Footnotes
-
All authors provided concept/idea/research design, project management, and consultation (including review of manuscript before submission). Dr Dounavi, Dr Chesterton, and Dr Sim provided writing. Dr Dounavi provided data collection. Dr Sim provided data analysis. Dr Chesterton provided fund procurement. The authors thank Dr Panos Barlas for advice on the design and conduct of this study.
-
The study was approved by the Ethics Committee of the School of Health and Rehabilitation, Keele University.
- Received April 2, 2011.
- Accepted March 26, 2012.
- © 2012 American Physical Therapy Association
References
Issue highlights
