Short communication Effects of electrical cutaneous stimulation of hallux on walking patterns in adults Bhavini Haria a , Jeroen Bergmann b , Christopher Smith a, * a King’s College London, London SE1 1UL, United Kingdom b Imperial College, SW7 2AZ, United Kingdom 1. Introduction Experimental studies of tripping have previously centred on either the sudden appearance of an obstacle in the pathway [1] or the sudden induction of body movement [2,3]. These primarily test the sensory inputs from the visual and proprioceptive systems, respectively. However, another sensory component is the impact of the foot against an unexpected object. The purpose this study is to simulate an impact by using a single-shock electrical stimulus applied to the big toe at controlled phases of the walking cycle and to measure the effect of this on the walking performance. The withdrawal reflex is not only multisegmental, as initially described by Sherrington [4], but depends on the location of the noxious stimulus. For example the ankle dorsiflexes with distal stimulation of the sole but plantarflexes after stimulation of the heel end [5]. Spaich et al. [6] have extended this to show that the withdrawal reflex also depends on gait cycle: an early (60–120 ms) electromyography (EMG) response to sole electrical stimulation was maximal in mid-swing, but a dominant late EMG (120– 200 ms) was largest during the foot-flat phase. Likewise stimula- tion of the sural nerve at the ankle, inducing sensation at the heel of the foot, increased flexion during the swing phase but produced an inhibition during stance [7,8]. However, stimulation of the sole might well be expected to produce complex effects via plantar responses [9]. Effects of stimulation of the toe have not been reported but electrical stimulation of the index finger [10] produced predominately an inhibition of all movements with latency and duration of around 50 ms. 2. Methods Eight healthy male subjects (age range 18–65) walked on a treadmill at 5 km/h for about 20 min. All subjects were fit and were informed that electrical stimulation would occur. The experimental protocols were approved by the local ethical committee (BDM/08/09-64). Anterior–posterior angular velocity of the lower limbs was measured by miniature gyroscopes (IDG300, Sparkfun, USA) and sampled digitally (12-bit, 100 Hz). For the first five minutes of walking there was no stimulation. The start of the stride cycle was defined by the time of the positive- going transition of angular velocity and used to control in real time the electrical stimulus presented at one of ten preselected fractions (0%, 10%, 20%, etc.) of the stride time. The stimulus reoccurred at a random stride between the third and sixth stride, a balance between the needs to avoid (a) predictability, (b) fatigue of the subjects and (c) interference with the timescale of the observed effect (Fig. 2B). At least 20 stimulated strides at each stride fractional time were averaged. The difference between the gyroscope angular velocity signals for control strides and stimulated strides was analysed for both amplitude and phase variations. Measurement, control and analysis were performed within a single locally written program using Delphi (Fig. 1). The stimulus was a single shock of 150 mA 0.1 ms (Digitimer D57) applied to the distal skin of the big toe using a small electrode (20 mm  20 mm, Blue Sensor BS3400) with the anodic electrode (60 mm  40 mm) attached to the ankle. The stimulus amplitude was chosen to provide a moderate level of discomfort to the subject, such as one might experience when hitting the toe on an obstacle. Video filming of the experiment was used to confirm the relation of the stimulus time, identified by the flash from the stimulator (Fig. 1), to the stride position. The preselected positions were: 10% = toe off, 20% = early swing, 30% = mid-swing, 40% = late swing, 50% = heel contact, 70% = foot flat, 80% = mid-stance, and 90% = heel off. In subsequent analysis of the data 20 traces of angular velocity were averaged in synchronism to the stimulus and compared to 20 traces without stimulation (Fig. 2). Inspection showed that the predominant effect of stimulation was a change in phase rather than amplitude of the stride, so the moment by moment phase shift was calculated by an algorithm taking 30 centre-weighted sample points at a time and sweeping these against the control trace to find the time of maximum Gait & Posture 33 (2011) 718–720 ARTICLE INFO Article history: Received 16 September 2010 Received in revised form 2 January 2011 Accepted 6 January 2011 Keywords: Motor control Falls Withdrawal reflex ABSTRACT Single-shock electrical stimulation of the big toe results in a 20 ms advancement of the walking cycle, lasting for 1–2 strides, when the shock is applied in early swing. During stance there is a retardation of the cycle but of similar magnitude. However unlike the complex responses reported for electrical stimulation of the sole of the foot, toe stimulation is best described as a simple withdrawal response. ß 2011 Elsevier B.V. All rights reserved. * Corresponding author at: King’s College London, School of Biomedical Sciences, Guy’s Campus, London Bridge, London SE1 1UL, United Kingdom. Tel.: +44 7970713507. E-mail addresses: christopher.smith@kcl.ac.uk, ichs@scube.co.uk (C. Smith). Contents lists available at ScienceDirect Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost 0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2011.01.005