Newrolmage 195 (2019) 396-408 (Content lists available a ScenceDisect Rend =) Neurolmage wv journal homepage: ww: elsevier com/locatelneuroimage Neurobiological mechanisms of TENS-induced analgesia W.W. Peng’, Z.Y. Tang”, F.R, Zhang", H. Li’, ¥.2. Kong", G.D. lannetti *, L. Hue"? 1 Cote of Peel and Sl, Steen Unive Sheren, Chia GAS Rey aeray of Mena ea ae of Pola, Be Ce “beproene of Pcl Ut of Cae Adon of Sows, ing, na "neath Cone obra Cae Nese ng Namal Urs, Din, Cine ‘ Neraiewe gud Birr, bt alae Teel, Roe al “pyre Renton Pystng and Psy, ow Cag tai, Ln, * bgt ofan Mapas Te Key Cel Specialy Pe Moin, The Send Aint Hop of Cages Maden, Cano, Chia AnTicLe Ineo Apsraact iow ‘rin bin by edad somatsencry pate anus forte wieprnd we of Tascaneans “cso eee tnton “Hectic Nerve Stimulation TENS) rele pain-Dwo man ype of TENS produce nage a anal mdse = high-frequency (~-S0-100 Hz) and low-intensity ‘conventional’ TENS, and low-frequency (~2-A Hz) and high- in inten ‘seupuncure le TENG. However, TINS eer in umanpatpant etd ling the ue Aa erty (0) ono wheter theaalgestmechanlae ied nana modi are ald in hema re, we acd sh song one nd the neuobsogical eflecofcoventonl an tomes ee" TENS nO an vlter To cet aaige elec TENS we ecored the prepa and bean responses elicited by radiant heat laser pulses ha arivate selectively A¥ and C cutaneous nociceptors Totest whether TENS bas along lasting eet on brain state we recorded spontaneous elcrocrtcal oscilaions "The anager efecto conventional TENS was ranma when nociceptive smal were delivered Romaopicaly, ‘0 the same hand that received the TENS. tn contrast, ‘acupuncture’ TENS produced & spatally-ifse nagesc eect, coupled with longlasting changes both inthe ate of the primary sensorimotor cortex (S1/ [Map cd inthe fnetionsl connectivity Between SI/A and the mia prefrontl corte, a cre teion inthe descending pin inhibitory system. These results demonstrate tat conventional’ and ‘acapanrure-ke TENS. 1. Introduetion suggested to be one ofthe crucial determinants of the duration and type of analgesic effect provided by TENS (Johason and Martinson, 2007). Ie is well4nown that eubbing the skin over a bruised area inhibits pain. Yet, the physiological mechanisms of touch-indued anslgesia remain unclear (Braz etal, 2014 Mendel, 2014), The analgesi effect of tactile stimolation constitutes the rationale for wsing Transcutaneous Hleeeial Nerve Stimulation (TENS) (Sluka and Walsh, 2002), the de livery of electrical stimuli that activate peripheral somatosensory afer ents, to relieve both acute and chronic pain (Inui eta, 2006; Rakel etl 2014), Despite being widely offered asa treatment for pain Rake eta. 2014), theres no conclusive evidence that TENS is effective in a number ‘of clinical conditions, and a great deal of confusion about the efficacy of ‘TENS reigns (Bergeron-Vezina etal, 2015; Rake et al, 2014. ‘he frequency ané the intensity of electrical pulses has been ‘mal addres lcsychacce (LH, * thee suthor cont buted equally 2 shared senior authorship TENS is typeally delivered using two distinc sets of stimulus parameters (1) high Frequency (~50-100 H2) and low intensity “conventional TENS, evoking a comfortable, nonpainful tingling sensations (Leonard (al, 2010), and (2) low frequency (2-4 Ha) and high intensity “aeu- puncturedike” TENS, evoking (olerable but painful sensations (Han, 2003). Considering this diversity of stimulus parameters, itis evident that despite being bot labeled using the same acronym, the two types of TENS have profounely diferent effects on the nervous system. Perhaps ‘unsurprisingly, evidence from animal studies suggests that “conven tional” and “acupuncture-like” TENS engage diferent analgesic mecha- nisms (Radhakrishnan et 2, 2003). “Conventional” TENS is usually related tothe gate control theory that high-frequency and low-intensity “arreaponding suthor. Key Laboratory of Mental Heath, lite of Psychology, Chinese Academy of Sciences, Bejing, 10010, Chins. Received 14 November 2018 Reveved in cevised frm 22 March 2019; Accepted 30 March 2019 1055-8119/0 2019 The Author. Published by Elsevier Ine. Thi is an open access ale under the CC BY Hesse (tp//cealivcommons.onleenses/by/4.07.WoW Pega stimulation of large-diameter A afferents results in a segmental inhibition (ofthe transmission of nocieeptive information atthe dorsal hoe level (elzack and Wall, 1965). On the other hand, “acupunctutesike” TENS (esantana etal, 2009; Kalra etal, 2001; lake nd Walsh, 2003) Is more related t0 the diffase noxious inhibitory control (DNIC) phenom ‘non (Le Bars etal, 1992); a strong noxious input causes the release of ‘endogenous opioids inthe periaqueductal gray (PAG) and rostral ventral ‘medulla (RVM), which in turn resultsin a diffuse descending inhibition of nociception. Despite the evidence that the two types of TENS have analgesic effects ‘mediated by distinct mechanisms in animal pain modes (Radhakrishnan tal, 2003; Radhakrishnan and Sika, 2003; Sluka and Walsh, 2003), ‘whether TENS produces analgesia in healthy participants (Basset al. 2006; Hergeron-Verina et aly 2015; Liebano et al, 2011) and chronic ‘in patents (Ezz0 eta, 2000; Oosterhof eal, 2008) remains contro versal. This question is particularly relevant when considering the Inkerent eifferenes between species aswell asthe fat ht animal TENS. cflcts are typically detected in anaesthetized or “spinal” preparations (@iackburn-Miunro, 2004; Garrison and Foreman, 1996; Hu et 3, 20159 ‘Mogi, 2009). In this seudy we explored the neurophysiological and perceptual ef fects of sham and active TENS delivered at either high or low-frequency in a population of 80 healthy human volunteers. While recording both spontaneous and stimulus-evoked brain activites, we tested whether TENS affected subjective ratings of pain inensiy and unpleasantness elicited by nociceptive stimulation, a well as whether TENS affected the transient electrocotea responses elicited bythe same nociceptive simul ‘We also explored whether TENS had a long-lasting effet on brain state, indexed using ongoing electrocorialosilations 2. Materials and methods 21. Subjects ‘total of 80 heathy, pan See volunteers who never had TENS before {all rightchanded; 40 female; mean age 20.5 1.8 years age range 18-27 years) were receuted through loal advertisement, Subjects were med cally screened, and excluded if they had peripheral and central nervous system disease, cardiac pacemaker, chronic pain, or if they were under ‘any type of pain medication, After the screening, subjects were told that the aim of the study was “to investigate the neurophysiological and perceptual effects of some electrical stimulation delivered to the skin’ All subjects gave thei written informed consent prior to testing and were aid for their participation. Fxperimentel procedures were approved by the local ethics committee 2.2, Sensory stimulation and experimental design -lectrical stimuli for both active and sham TENS were generated by @ constant current electrical simulator (Sanxia technique Ine, China), and delivered through a par of surface round eletrodes (diameter: 16mm; Interelectrode distance: 3 em) placed over the radial nerve atthe wrist, tether on the let or on the right side, Four diferent sets of stimulus parameters were used in four experimental groups, as follows. Group 1: high frequency ative TENS; Group 2 low-frequency active TENS; Group 3 high frequency sham TENS; Group 4: low-frequency sham TENS. Each subject was randomly assigned to one of the four groups. Age and sex ‘were matched between groups. High-frequency (100H2) and low- frequency (4 Hz) TENS (both active and sham) consisted ofa series of succeeding constanturrent square-wave pulses (0.2ms duration for ‘each pulse). The duration ofthe TENS session was the same inthe active land sham TENS, However, in the active TENS the stimulation lasted for ‘30 min, ehereas in the sham TENS the stimulation lasted for 45. Inthe highfrequency active TENS the stimslus intensity was individvally adjusted to elicit a strong but non-painful tingling sensation. In the low frequency ative TENS the stimulus intensity was individually adjusted to earings 195 (2019) 996-408 elicit a tolerable painful sensation (Bergeron-Vezina ct al, 2015)-In both high-frequency ad low-frequency sham TENS the stimu intensity was Individually adjusted tothe thresholds for detecting the electrical stn ‘lus. Importantly, given that the TENS sensation habituates ater dhe rt occurrences of the stimulus Pantaleo et al, 2011), in all eanditions (ctive and sham, high-frequency and low frequency) the stimulus i tensity was determined as follows ist, the desired perceptual outcome ‘of each condition (the strong but non-painal tingling sensation in the high-frequency TENS, the tolerable painful sensation in the low-frequency TENS, and so on) was obtained in three consecutive a tempts. Second, the stimulus intensity used inthe actual experiment vas ‘the highest ofthe chee attempts for each condition. Before the TENS, participants were given the following instruction: “You will now receive ‘30min of TENS. This can result in sensations that vary greatly among, subjects, and range from intense sensations to mich weaker sensation, for even no sensation at all Before and after the TENS procedure, we recorded the brain responses elicited by nociceptivespecificradiantheat stimull generated by an. infrared neodymium yttrium aluminum perovskite laser with a wave Jength of 1.34 jm and a pulse duration of 4ms (Electronic Engincerng, aly). At this wavelength and pulse duration, laser stimuli atvate direcly nociceptive ceminals in the most superficial skin layers in a synchronized fashion (lane tal, 2004). A He ~ Ne laser pointed to the area to be stimulated, The laser beam was transmitted via en optic ‘Aber and its dlameter was set at approximately 7 mm by focusing Tenses, laser pulses were delivered toa squared area (4 x4 em) onthe dorsum, of each hand (ie, both ipsilateral and contralateral to the TENS side), ‘After each stimuli, the beam target was sified Dy at least Lem in a :andom direction within the squared area, to avoid nocicepto fatigue oF sensitization, Ina preliminary session, the laser energy was individually determined by increasing the stimulus energy in steps of 0.25, until a rating of 7 out of 10 was obtained on a numerical rating sale (NRS) zanging fom 0 (no pain) to 10 (pain as bad as it could be) ‘ placebo-controlled experimental design was used (Pig. 1). Subjects were randomly assigned to one of the four experimental groups (20 subjects per group). The experiment consisted of thee sessions, sepa rated by aS-min break Pre-TENS (10 min), TENS (30 min), and Post- TENS (~10 min). In both Pre-TENS and Pos-TENS sessions, 20 noci= ceptve laser stimuli of dential energy were delivered (0 the dorsum of Doth hands (10 stimuli per band). The interseimulus interval varied randomly between 18 and 20s, The order of stimulated hand was pseudorandomized, with the constraint that no more than two stimuli could be delivered tothe same hand. In haf ofthe subjects of each group, the firs timulas was delivered onthe left hand; in the ater half che first stimulus was delivered on the right hand. Approximately 35 after each laser stimulus, subjects were asked to verbally ceport both pain intensity (with the same NRS used inthe preliminary energy determination) and pain unpleasantnes (using an NRS ranging from 0 [not unpleasant} to 10, {maximally unpleasant), Tt As important to mention that setive and sham TENS induced diferent sensations, and therefore within-subject design would 1 bhave been blinded with respect tothe experimental conditions Thus, we decided to adopt instead a betweer-subect, plcebo-controlled experi- ‘mental design in which subjects were randomly assigned to one of four cxperimental groups. In addition, data from different subjects were collected at different time periods, and subjects were not able to ‘communicate with each other about the sensation felt during the experiment. Sil the fact that active and sham TENS produced different sensations does no allow ust completly eule out the possible influence ‘of unspecific stimulation effets on the observed results 2.3, BBG data collection Subjects seated in a comfortable chat in silent room which tem perature was maintained between 26 aad 26°C. They were instructed (0 focus on the stimuli, keep their eyes open, and gaze ata fixation point on