- Email: [email protected]
Changes in excitability of corticomotor inputs to the trunk muscles during experimentally-induced acute low back pain
Neuroscience 181 (2011) 127–133
CHANGES IN EXCITABILITY OF CORTICOMOTOR INPUTS TO THE TRUNK MUSCLES DURING EXPERIMENTALLY-INDUCED ACUTE LOW BACK PAIN H. TSAO, K. J. TUCKER AND P. W. HODGES*
adaptation is argued to stabilise and protect the spine from further stress and injury (Hodges and Moseley, 2003). As similar differential changes in deep and superficial muscle activity are found in healthy individuals with experimentally-induced acute LBP (Arendt-Nielsen et al., 1996; Zedka et al., 1999; Hodges et al., 2003), it is argued that nociceptive input can directly influence neural inputs to the trunk muscles. However, exactly how LBP drives changes in the motor system remains poorly understood. Trunk muscles can be controlled by descending inputs from the motor cortex via rapid-conducting corticospinal pathways to spinal motoneurones (Plassman and Gandevia, 1989; Ferbert et al., 1992; Lissens et al., 1995). Notably, people with recurrent or chronic LBP show changes in corticomotor input to the trunk muscles. For instance, Strutton et al. (2005) reported an increased threshold to elicit motor responses in the back extensor muscles with chronic LBP. More recently, our group observed changes in organisation and excitability of corticomotor pathways to the deep abdominal muscle, transversus abdominis (TrA; Tsao et al., 2008a). However, both of those studies involved investigations of changes in a single muscle and in people with recurrent episodic LBP. It remains unclear whether acute nociceptive stimulation in the low back can induce immediate changes in corticomotor drive to the trunk muscles, or whether changes, if present, differ between the deep and superficial trunk muscles. Several studies on hand and jaw muscles have demonstrated altered corticomotor excitability during experimentally-induced muscle pain. In general, most studies report that the amplitude of motor responses evoked by magnetic brain stimulation is reduced in the painful muscle when subjects are at rest (Valeriani et al., 1999; Le Pera et al., 2001). One study showed increased evoked responses. However, as subjects in that study were required to actively maintain a pre-set level of motor activity, it was argued that increased activation with pain served to compensate for reduced baseline excitability due to pain (Del Santo et al., 2007). However, the organisation of control of the trunk muscles by the motor system differs somewhat from that of the limb and jaw muscles. For example, trunk muscles receive greater bilateral drive compared to the distal muscles of the limbs (Carr et al., 1985; Marsden et al., 1999) and although jaw muscles receive largely bilateral drive, their activity is controlled by corticobulbar rather than corticospinal inputs (Nordstrom, 2007). Therefore, it is difficult to translate findings from studies on the limb or jaw muscles directly to the trunk muscles.
Centre for Clinical Research Excellence in Spinal Pain, Injury and Health, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia 4072
Abstract—Acute low back pain (LBP) is associated with differential changes in motor coordination of deep and superficial trunk muscles. Whether this is related to differential changes in excitability of descending corticomotor inputs remains unclear and was investigated in nine healthy individuals. Fine-wire i.m. electrodes were inserted bilaterally into deep (transversus abdominis (TrA)) and superficial abdominal muscles (obliquus externus abdominis (OE)), and surface electrodes were placed bilaterally over obliquus internus abdominis (OI), rectus abdominis (RA) and lumbar erector spinae (LES) muscles. Corticomotor excitability was assessed as amplitude of motor evoked potentials (MEPs) to transcranial magnetic stimulation (TMS) at a range of stimulator intensities, at rest and during voluntary abdominal contractions. Pain was induced by injection of hypertonic saline into interspinous ligaments of the lumbar spine. Corticomotor excitability was examined before, during and after the induction of LBP. During pain, amplitude of TrA MEPs to contralateral cortical stimulation was reduced, whereas amplitudes of OE and LES MEPs contralateral and ipsilateral to the stimulated cortex were increased. The findings highlight differential changes in excitability of corticomotor inputs to trunk muscles during acute LBP. Further work is required to reveal whether such changes involve spinal and/or supraspinal centres and their consequence for spine control. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: low back pain, motor cortex, corticospinal excitability, trunk muscles, experimental pain.
Low back pain (LBP) is associated with altered motor coordination of trunk muscles during postural and functional tasks (Hodges and Moseley, 2003; van Dieen et al., 2003). Although inter-subject and inter-task variability exists, most studies highlight differential changes between deep and superficial trunk muscles. For instance, in people with recurrent LBP, activity of deep muscles is reduced (Danneels et al., 2002) or delayed (Hodges and Richardson, 1996; MacDonald et al., 2009), whereas activity of superficial trunk muscles is often increased (ArendtNielsen et al., 1996; Radebold et al., 2001). This latter *Corresponding author. Tel: ⫹61-7-3365-4567; fax: ⫹61-7-3365-4567. E-mail address: [email protected] (P. W. Hodges). Abbreviations: EMG, electromyography; LBP, low back pain; LES, lumbar erector spinae; MEPs, motor evoked potentials; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; RMS, root-mean-square; TMS, transcranial magnetic stimulation; TrA, transversus abdominis.
0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2011.02.033
127
128
H. Tsao et al. / Neuroscience 181 (2011) 127–133
The current study aimed to investigate the effects of experimentally-induced LBP on changes in excitability of the motor cortex and descending corticospinal pathways to the trunk muscles in healthy individuals. Given that differential changes between the deep and superficial trunk muscles have been previously observed with pain, we hypothesised that changes in corticomotor excitability may also differ between the deep and superficial trunk muscles.
EXPERIMENTAL PROCEDURES Participants Nine healthy volunteers were recruited (five males, four females, age 25⫾4 [mean⫾SD] years, height 176⫾10 cm, weight 68⫾9 kg). Subjects were excluded if they had any history of back and/or leg pain, epilepsy or family history of epilepsy, major neurological, respiratory, orthopaedic or circulatory disorders, previous spinal or abdominal surgery, pregnancy in the last 2 years, had taken any analgesic medication in the past 12 months, or if they had undertaken any form of abdominal or back muscle exercise programs in the past 12 months. The study was approved by the Institutional Medical Ethics Committee and all procedures conformed to the Declaration of Helsinki.
Electromyography Following informed consent, electromyographic (EMG) activity of the deep abdominal muscle, TrA, and the more superficial abdominal muscle, obliquus externus abdominis (OE), were recorded bilaterally using bipolar fine-wire intramuscular (i.m.) electrodes. This involved pairs of Teflon-coated stainless steel wires (75 m diameter) with 1 mm of insulation removed from end and bent back by 1 and 2 mm, respectively. These were threaded into a hypodermic needle and inserted with ultrasound guidance (Hodges and Richardson, 1996). In addition, pairs of surface electrodes (Ag/AgCl disc electrode, 10 mm diameter, 20 mm inter-electrode spacing, Grass Telefactor, RI, USA) were placed bilaterally over the lumbar erector spinae (LES; 5 cm lateral to the L2 spinous process (Cholewicki et al., 1997)), rectus abdominis (RA; 2 cm lateral to umbilicus (Ng et al., 1998)) and obliquus internus abdominis (OI; anteromedial to anterior inferior iliac spine (Ng et al., 1998)). No subject reported any discomfort or pain during the experiment that was associated with fine-wire placement. A reference electrode was placed over the lateral ribs. EMG data were amplified 2000 times, band-pass filtered between 20 and 1000 Hz, and sampled at 2 kHz using a Power 1401 Data Acquisition System with Spike 2 software (Cambridge Electronic Design, Cambridge, UK).
Procedures Subjects were positioned comfortably in prone with pillows under their legs and arms by their side. This position was chosen as it allowed easy access to interspinous space without having to move the subject. Motor cortical excitability of the trunk muscles was examined before and during experimentally-induced LBP and ⬃30 min after the cessation of induced-pain (when reports of pain had returned to zero). Excitability of corticomotor projections to the trunk muscles were investigated using transcranial magnetic stimulation (TMS). A single-pulse monophasic MagStim 2002 (The Magstim Company, Carmarthenshire, Wales, UK) was used to stimulate the motor cortex at rest and during submaximal activity. As it was difficult and impractical for naïve subjects to activate and maintain a pre-set motor activity for all the trunk muscle recorded, we elected to standardise voluntary contractions of the abdominal muscles through pre-set level of TrA activation. To determine the
level of voluntary contraction during motor cortical stimulation, EMG activity during maximum voluntary contraction (MVC) for TrA was recorded. Subjects performed three repetitions of a forced expiratory manoeuvre with verbal encouragement, each lasting at least 3 s (Ninane et al., 1992). The root-mean-square (RMS) EMG of the left and right TrA was plotted online using 200 ms windows and the peak RMS EMG over a 1 s period was calculated. A target level of ⬃10% TrA activation was set and subjects achieved this level of contraction through an abdominal brace, which has been shown to activate TrA with other abdominal muscles (Urquhart et al., 2005). Feedback of the intensity of TrA contractions was provided via a monitor although subjects were instructed not to focus on any specific muscle(s). A 7 cm figure-of-eight coil was used to identify to the optimal location to elicit motor evoked potentials (MEPs) in the test muscles contralateral to the stimulated cortex. The coil was oriented ⬃45° (to induce current in the brain in an anteromedial direction) and set at suprathreshold intensity (⬃70 –100% maximum stimulator output) during ⬃10% voluntary contraction of TrA. This coil provides better focality of stimulation and is more suited for identification of the optimal location (Cohen et al., 1990; Brasil-Neto et al., 1992). In most subjects, the optimal location was found to be ⬃2 cm anterior and lateral to the vertex, and is consistent with previous studies of TrA (Tsao et al., 2008a,b) and other trunk muscles (Strutton et al., 2004; O’Connell et al., 2007). We elected to stimulate this optimal location (determined with the figure-ofeight coil) with the double-cone coil. Although the induced current in the motor cortex differs with coils shape/configuration, the figure-of-eight provided a standardised method to identify the optimal location. As the same scalp site was used to stimulate the motor cortex before, during and after pain for each participant, any inaccuracy in identification of the optimal site based on the figureof-eight coil is unlikely to have influenced the results. A 9 cm double-cone coil was used to stimulate the motor cortex. The excitability of corticomotor pathways was evaluated by constructing a recruitment curve with the amplitude of motor evoked potentials (MEPs) measured across stimulator intensities at rest and during submaximal abdominal contractions (Ridding and Rothwell, 1997). This was positioned over the optimal location (largest MEP for a given stimulator output) with the coil oriented such that induced currents in the motor cortex flowed in an anterior direction. The 9 cm double-cone coil produces a stronger magnetic field and can evoke consistent MEPs at rest and in trunk muscles ipsilateral to the stimulated cortex (Tsao et al., 2008a). Five stimuli were delivered at each 10% interval from 50% to 90% of maximum stimulator output at rest, and from 30% to 80% of maximum stimulator output during voluntary abdominal muscle activation to 10% MVC for TrA. This procedure was repeated over both hemispheres.
Experimental pain Experimentally-induced LBP was elicited via two consecutive 0.3 ml bolus injections of hypertonic saline solution into the interspinous ligaments between the 4th and 5th (L4/5), and the 3rd and 4th (L3/4) lumbar spinous processes. Injections delivered into different spinal levels induced pain of sufficient duration to complete measures of corticomotor excitability and motor coordination. We did not inject into the same location twice as pilot trials showed that a second injection into the same area yielded shorterlasting and less-intense LBP. This injection method was first described by Kellgren (1939). Recent work from our team has shown pain from this injection produces central back pain that has several advantages over muscle injection (Tsao et al., 2010). First, the pain from hypertonic saline injections of the interspinous ligament is longer lasting. Second, ligament pain does not attenuate with trunk muscle contractions compared to muscle pain. Third, this procedure avoids injection into the muscles under investigation. The interspinous spaces of L4/5 and L3/4 were identified by palpation and if required, confirmed with real-time ultrasound imag-
H. Tsao et al. / Neuroscience 181 (2011) 127–133
129
Fig. 1. (A) Injection of hypertonic saline into L4/5 interspinous ligament. (B) Location and area of experimentally-induced pain. (C) Pain intensity plotted against time for each individual averaged across hypertonic saline injections at L4/5 and L3/4 interspinous ligament.
ing. The needle (27 G⫻25 mm) was introduced into the interspinous space angled at ⬃30° in a cranial direction to the skin until a gentle tap was felt against the spinous process of the upper spinal level (Fig. 1A). At this point, 0.3 ml of hypertonic saline (5% NaCl) was introduced. Subjects rated the intensity of the pain using an electronic visual analogue scale (VAS) anchored with 0 (‘no pain’) and 10 (‘worst pain imaginable’). The location and area of pain was marked on a body chart by the participant at the end of the experiment.
Data analysis Data were imported and analysed using Matlab 7 (The Mathsworks, MA, USA). For TMS evoked responses, EMG data were full-wave rectified, averaged for each stimulator intensity at rest and during abdominal activation, and the onset and offset of MEPs that were clearly discernible from background EMG activity were visually identified. To remove any potential for bias, data were presented in random order without reference to the identity of the muscle, intensity of TMS or whether the trials were before, during or after pain. In trials with identifiable MEPs, RMS EMG of MEPs (from onset to offset) was calculated and grouped into responses elicited contralateral and ipsilateral to the stimulated hemisphere. In trials in which no MEPs were identified, the RMS EMG of myoelectric activity was calculated from 15 to 35 ms (which corresponded to the time range of expected MEPs) following TMS stimulus. RMS EMG amplitudes for MEPs were normalised to the peak response across trials and across both hemispheres at rest and during activity, and plotted as a recruitment curve against the stimulator intensity. The overall excitability of corticomotor pathways was also represented by calculating the area under the recruitment curve. The background EMG activity for each trial was also calculated as the amplitude of RMS EMG from 55 to 5 ms before the stimulation.
Statistical analysis Statistica 7 (Statsoft, OK, USA) was used for statistical analysis, with significance level set to P⫽0.05. To examine whether pain
influenced corticomotor excitability of the trunk muscles, normalised amplitudes of trunk muscle MEPs were compared between Conditions (pre, pain vs. post), Muscles and Intensities (% maximum stimulator output) at rest and during abdominal activation using repeated measures analysis of variance (ANOVA). The area under the recruitment curve and background EMG activity was also compared between Conditions, Muscle and Activity (rest vs. activity) using repeated-measures ANOVA. The onset of MEPs was also compared between Muscles and Activity (rest vs. abdominal activation). Post-hoc analyses were undertaken using Duncan’s multiple range test.
RESULTS Injection of hypertonic saline into the interspinous ligament induced central back pain in all subjects (Fig. 1B). The average (SD) of the peak pain intensity experienced during TMS stimulation was 4.3 (0.8) cm on the 10 cm VAS. The average pain duration from the onset of injection until the cessation of pain was 11.9 (2.6) min for each injection (or ⬃23.8 min across both injections). Fig. 1C shows pain intensity (averaged each min) against time, averaged across both injections. Notably, for the pain condition, all TMS stimulation was completed before the cessation of pain. Corticomotor excitability Figs. 2 and 3 show recruitment curves at rest and during abdominal activation. As expected, the peak MEP amplitudes were 2.6 –5.1 times greater during voluntary abdominal brace than that at rest. During pain, reduced amplitude of TrA MEPs contralateral to the stimulated cortex was observed at rest from 70% to 90% maximum stimulator output (Interaction Pain⫻Intensity⫻Muscles, P⬍0.001; post hoc P⬍0.048), and during abdominal activity from
130
H. Tsao et al. / Neuroscience 181 (2011) 127–133
P⬍0.036) and from 70% to 80% maximum stimulator output during voluntary contractions (Fig. 3; post-hoc P⬍0.013). Similarly, amplitude of MEPs ipsilateral to the stimulated cortex was increased at 90% maximum stimulator output at rest (post-hoc P⬍0.004) and at 80% maximum stimulator output during voluntary contractions (posthoc P⬍0.043). Area under the recruitment curve for OE was also increased, although this was only observed for the rest condition (Fig. 4; post-hoc P⬍0.047). Responses of LES contralateral and ipsilateral (from 80% to 90% maximum stimulator output) to the stimulated cortex and the area under the recruitment curve were also increased, but only for the rest condition (MEPs: post-hoc P⬍0.034; Area under curve: post-hoc P⬍0.048). Amplitude of LES MEPs (post-hoc P⬎0.19) or area under LES recruitment curve (post-hoc P⬎0.07) did not change during voluntary contraction of the abdominal muscles. OI and RA MEP amplitude did not change at rest (post-hoc P⬎0.091) or during voluntary abdominal contraction (post-hoc P⬎0.080). Similarly, there was no
Fig. 2. Amplitude of motor evoked potentials (MEPs) at rest expressed as recruitment curves from 50% to 90% maximum stimulator output before (white circles), during (black circles) and after induced back pain (grey circles). Values are expressed as root-mean-square (RMS) EMG amplitude normalised to peak response. Mean and 95% confidence intervals are shown for responses ipsilateral (left panel) and contralateral (right panel) to the stimulated cortex. * indicates significant difference in MEP amplitude between pain and pre-/postpain (P⬍0.05). TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
60% to 80% maximum stimulator output (Interaction Pain⫻Intensity⫻Muscles, P⬍0.001; post-hoc P⬍0.003). This is consistent with reduced area under the recruitment curve at rest (Fig. 4A; Interaction Pain⫻Muscles, P⬍0.001; post-hoc P⬍0.008) and during voluntary activation (Interaction Pain⫻Muscles, P⬍0.001; post-hoc P⬍0.001; Fig. 4B). In contrast, there was no change in the amplitude of TrA MEPs ipsilateral to the stimulated cortex at rest (post-hoc P⬎0.21) and during activity (post-hoc P⬎0.11), and no change was observed for area under the recruitment curve (rest: post-hoc P⬎0.32; voluntary activation: post-hoc P⬎0.15). Pain was associated with increased amplitude of OE MEPs contralateral to the stimulated cortex from 80% to 90% maximum stimulator output at rest (Fig. 2; post-hoc
Fig. 3. Amplitude of motor evoked potentials (MEPs) during abdominal brace expressed as recruitment curves from 30% to 80% maximum stimulator output before (white circles), during (black circles) and after induced back pain (grey circles). Values are expressed as rootmean-square (RMS) EMG amplitude normalised to peak response. Mean and 95% confidence intervals are shown for responses ipsilateral (left panel) and contralateral (right panel) to stimulated cortex. * indicates significant difference in MEP amplitude between pain and pre-/post-pain (P⬍0.05). TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
H. Tsao et al. / Neuroscience 181 (2011) 127–133
Fig. 4. Areas under the recruitment curves at rest (A) and during voluntary abdominal contractions (B). Mean and 95% confidence interval are shown for responses ipsilateral and contralateral to the stimulated hemisphere before (white circles), during (black circles) and after pain (grey circles). * P⬍0.05. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
change in area under the curve for these muscles (posthoc P⬎0.16). Fig. 5 shows representative data from the right sided trunk muscles for a single subject. MEP onset The onset of MEPs for all abdominal muscles contralateral to the stimulated cortex was shorter than that ipsilateral to the stimulated cortex (Fig. 6; Interaction
131
Fig. 6. Mean latency to onset (and 95% confidence interval) of motor evoked potentials (MEPs) for trunk muscles at rest (black circles) and during abdominal contractions (white circles) ipsilateral and contralateral to the stimulated hemisphere. Note for all muscles except the lumbar erector spinae (LES; which was not activated by the task), the onset of MEPs during abdominal contractions was faster than that at rest and the latency of MEPs contralateral to the stimulated cortex was faster than that ipsilateral to the stimulated hemisphere. * P⬍0.05. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis.
Muscles⫻Activity, P⬍0.001; post-hoc for all abdominal muscles, P⬍0.048). The shorter latency to MEP onset for contralateral responses is consistent with activation of the monosynaptic crossed-corticospinal pathway, whereas the longer latency to MEP onset for ipsilateral responses is consistent with activation of polysynaptic uncrossed-corticospinal pathways (Ziemann et al., 1999; Strutton et al., 2004). However, LES MEP onset did not differ between sides (post-hoc P⬎0.38). As expected, the onset of abdominal muscle MEPs during activity was shorter than that at rest (Main effect Activity, P⬍0.001; post-hoc P⬍0.001). LES MEP onset was sim-
Fig. 5. Representative data from right (R) trunk muscles of a single subject at rest before, during and after pain. Dotted vertical lines represent time of stimulation. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae. EMG calibration 0.2 mV.
132
H. Tsao et al. / Neuroscience 181 (2011) 127–133
ilar between rest and abdominal muscle activation (posthoc P⬎0.097). Background activity There was no change in background activity for any muscle between pre-pain, pain and post-pain conditions (Main effect for Conditions: P⫽0.62). This implies that any paininduced changes in corticomotor excitability are unlikely to be due to differences in background activity. As expected, the amplitude of background activity for TrA, OE and OI was significantly greater during abdominal contractions than at rest (Interaction between Activity and Muscle: P⬍0.001; post-hoc P⬍0.037). Background activity of RA and LES did not increase during voluntary abdominal muscle contractions (post-hoc P⬎0.32).
DISCUSSION The findings suggest that acute LBP is associated with differential changes in corticomotor excitability of the trunk muscles. The deep abdominal muscle, TrA, showed reduced corticomotor excitability contralateral to the stimulated cortex. In contrast, the more superficial trunk muscles including OE and LES demonstrated increased excitability to cortical stimulation over both hemispheres during pain. The results demonstrate the potential for rapid plasticity of corticomotor pathways with short-term pain and importantly, highlight that pain does not induce a stereotypical change in corticomotor excitability. Rather, the response varies between muscles. Acute LBP is commonly associated with differential changes in muscle recruitment between the deep and superficial trunk muscles (Hodges and Moseley, 2003; van Dieen et al., 2003). Here we show that these behavioural findings are mirrored by differential changes in corticomotor excitability. As background activity was similar between pain and no pain conditions, modulation of corticomotor excitability during pain could not be explained by changes in background activity. Increased corticomotor drive to the superficial trunk muscles is somewhat consistent with the aim of the CNS to “protect the part” from further injury/pain (Hodges and Moseley, 2003). That is, increased excitability of these pathways and thus potential for increased activity of the superficial trunk muscles can contribute to splinting and reduced mobility of the spine (Hodges et al., 2006). As a result, the contribution of the deeper trunk muscles such as TrA could be perceived as redundant during pain. This is consistent with findings of reduced corticomotor drive to the deep trunk muscles, at least on the side contralateral to the stimulated cortex. The data showed increased excitability of corticomotor inputs to the LES with acute LBP, but this was only observed at rest. Findings of increased excitability contrast previous findings by Strutton et al. (2005), which showed reduced excitability (i.e, increased motor threshold to induce a response to TMS) of the LES in patients with chronic LBP. One possible explanation for this discrepancy could be differences between experimentally-induced acute LBP and more persistent clinical LBP. Alternatively,
in the study by Strutton et al. (2005), subjects were required to activate their back muscles to a pre-set intensity, whereas in the current study, subjects performed an abdominal brace and standardised activation to TrA, not the LES. For practical reasons, we did not control the level of activity of other trunk muscles, but no systematic increase in LES activity was recorded during this task. This may have increased inter-subject variability of MEPs during activity, although this is unlikely to have influenced the current findings as similar background activities were found between pain and no pain conditions. Future work would be required to further verify the influence of pain on corticomotor excitability of LES muscle during voluntary activation of these muscles. No change in corticomotor excitability of TrA was observed on the ipsilateral side to the stimulated cortex. The onset of MEPs for all abdominal muscles (TrA, OE, OI and RA) contralateral to the stimulated cortex was faster than that ipsilateral to the stimulated cortex. This suggests at least for the initial part of the response, contralateral responses were evoked via the rapid-conducting crossedcorticomotor pathways, whereas ipsilateral responses were mediated via slower cortico-brainstem-spinal tracts (Ziemann et al., 1999; Strutton et al., 2004). Our previous finding suggests that motor threshold of ipsilateral responses was reduced in people with chronic LBP, but this was only evident over the more excitable hemisphere (Tsao et al., 2008b). As the current study did not evaluate motor threshold and instead grouped ipsilateral responses from stimulation of both hemispheres, this may explain the lack of changes in ipsilateral responses during pain. Our previous study in patients with chronic LBP did not observe a correlation between changes in motor excitability (as measured by motor threshold) and changes in motor coordination (Tsao et al., 2008b). Rather, changes in motor coordination were found to be associated with changes in the organisation of intra-cortical networks at the motor cortex. Due to the length of time required to map the motor cortex using TMS and the short-duration of experimentallyinduced LBP, mapping was not conducted in the current study. Furthermore, corticomotor excitability may be associated with other temporal and spatial parameters or even other functional tasks which were not examined in the current study. Further work is needed to unravel the functional relevance of pain-associated changes in corticomotor excitability of inputs to the trunk muscles. Changes in motor coordination of the trunk muscles with LBP, which may be useful in the short term to protect the body from further injury, can persist following resolution of symptoms and have been argued to contribute to the recurrence and chronicity of pain episodes (Hodges and Moseley, 2003). Therefore, findings of rapid alteration in corticomotor excitability may highlight the importance of early motor rehabilitation following acute LBP and injury. However the exact nature of the training to change motor control may depend on the mechanism and site of changes in corticomotor excitability. TMS evoked responses can be affected by changes in excitability of cortical cells, motoneurones and any interneurons (Iscoe, 1998). As the
H. Tsao et al. / Neuroscience 181 (2011) 127–133
exact site of pain-induced changes in TMS-evoked responses cannot be determined in the current study, future studies are needed to further investigate spinal and/or supraspinal mechanisms. Acknowledgments—We like to thank Ms Leanne Hall and Ms Rachel Park for their assistance with data collection. Funding was provided by the National Health and Medical Research Council of Australia.
REFERENCES Arendt-Nielsen L, Graven-Nielsen T, Svarrer H, Svensson P (1996) The influence of low back pain on muscle activity and coordination during gait: a clinical and experimental study. Pain 64:231–240. Brasil-Neto JP, Cohen LG, Panizza M, Nilsson J, Roth BJ, Hallett M (1992) Optimal focal transcranial magnetic activation of the human motor cortex: effects of coil orientation, shape of the induced current pulse, and stimulus intensity. J Clin Neurophysiol 9:132– 136. Carr JH, Shepherd RB, Nordholm L, Lynne D (1985) Investigation of a new motor assessment scale for stroke patients. Phys Ther 65:175–180. Cholewicki J, Panjabi MM, Khachatryan A (1997) Stabilizing function of trunk flexor-extensor muscles around a neutral spine posture. Spine 22:2207–2212. Cohen LG, Roth BJ, Nilsson J, Dang N, Panizza M, Bandinelli S, Friauf W, Hallett M (1990) Effects of coil design on delivery of focal magnetic stimulation. Technical considerations. Electroencephalogr Clin Neurophysiol 75:350 –357. Danneels LA, Coorevits PL, Cools AM, Vanderstraeten GG, Cambier DC, Witvrouw EE, De Cuyper HJ (2002) Differences in multifidus and iliocostalis lumborum activity between healthy subjects and patients with sub-acute and chronic low back pain. Eur Spine J 11:13–19. Del Santo F, Gelli F, Spidalieri R, Rossi A (2007) Corticospinal drive during painful voluntary contractions at constant force output. Brain Res 1128:91–98. Ferbert A, Caramia D, Priori A, Bertolasi L, Rothwell JC (1992) Cortical projection to erector spinae muscles in man as assessed by focal transcranial magnetic stimulation. Electroencephalogr Clin Neurophysiol 85:382–387. Hodges PW, Cholewicki J, Coppieters MW, MacDonald D (2006) Trunk muscle activity is increased during experimental back pain, but the pattern varies between individuals. In: International Society for Electrophysiology and Kinesiology, Torino, Italy. Hodges PW, Moseley GL (2003) Pain and motor control of the lumbopelvic region: effect and possible mechanisms. J Electromyogr Kinesiol 13:361–370. Hodges PW, Moseley GL, Gabrielsson A, Gandevia SC (2003) Experimental muscle pain changes feedforward postural responses of the trunk muscles. Exp Brain Res 151:262–271. Hodges PW, Richardson CA (1996) Inefficient muscular stabilization of the lumbar spine associated with low back pain: a motor control evaluation of transversus abdominis. Spine 21:2640 –2650. Iscoe S (1998) Control of abdominal muscles. Prog Neurobiol 56:433– 506. Kellgren JH (1939) On the distribution of pain arising from deep somatic structures with charts of segmental pain areas. Clin Sci 4:35. Le Pera D, Graven-Nielsen T, Valeriani M, Oliviero A, Di Lazzaro V, Tonali PA, Arendt-Nielsen L (2001) Inhibition of motor system excitability at cortical and spinal level by tonic muscle pain. Clin Neurophysiol 112:1633–1641. Lissens MA, De Muynck MC, Decleir AM, Vanderstraeten GG (1995) Motor evoked potentials of the abdominal muscles elicited through
133
magnetic transcranial brain stimulation. Muscle Nerve 18:1353– 1354. MacDonald D, Moseley GL, Hodges PW (2009) Why do some patients keep hurting their back? Evidence of ongoing back muscle dysfunction during remission from recurrent back pain. Pain 142: 183–188. Marsden JF, Farmer SF, Halliday DM, Rosenberg JR, Brown P (1999) The unilateral and bilateral control of motor unit pairs in the first dorsal interosseous and paraspinal muscles in man. J Physiol (Lond) 521:553–564. Ng JK-F, Kippers V, Richardson CA (1998) Muscle fibre orientation of abdominal muscles and suggested surface EMG electrode positions. Electromyogr Clin Neurophysiol 38:51–58. Ninane V, Rypens F, Yernault JC, De Troyer A (1992) Abdominal muscle use during breathing in patients with chronic airflow obstruction. Am Rev Respir Dis 146:16 –21. Nordstrom MA (2007) Insights into the bilateral cortical control of human masticatory muscles revealed by transcranial magnetic stimulation. Arch Oral Biol 52:338 –342. O’Connell NE, Maskill DW, Cossar J, Nowicky AV (2007) Mapping the cortical representation of the lumbar paravertebral muscles. Clin Neurophysiol 118:2451–2455. Plassman BL, Gandevia SC (1989) Comparison of human motor cortical projections to abdominal muscles and intrinsic muscles of the hand. Exp Brain Res 78:301–308. Radebold A, Cholewicki J, Polzhofer GK, Greene HS (2001) Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 26:724 –730. Ridding MC, Rothwell JC (1997) Stimulus/response curves as a method of measuring motor cortical excitability in man. Electroencephalogr Clin Neurophysiol 105:340 –344. Strutton PH, Beith I, Theodorou S, Catley M, McGregor AH, Davey NJ (2004) Corticospinal activation of internal oblique muscles has a strong ipsilateral component and can be lateralized in man. Exp Brain Res 158:474 – 479. Strutton PH, Theodorou S, Catley M, McGregor AH, Davey NJ (2005) Corticospinal excitability in patients with chronic low back pain. J Spinal Disord Tech 18:420 – 424. Tsao H, Galea MP, Hodges PW (2008a) Concurrent excitation of the opposite motor cortex during transcranial magnetic stimulation to activate the abdominal muscles. J Neurosci Methods 171:132–139. Tsao H, Galea MP, Hodges PW (2008b) Reorganization of the motor cortex is associated with postural control deficits in recurrent low back pain. Brain 131:2161–2171. Tsao H, Tucker KJ, Coppieters MW, Hodges PW (2010) Experimentally induced low back pain from hypertonic saline injections into lumbar interspinous ligament and erector spinae muscle. Pain 150:167–172. Urquhart DM, Hodges PW, Allen TJ, Story IH (2005) Abdominal muscle recruitment during a range of voluntary exercises. Man Ther 10:144 –153. Valeriani M, Restuccia D, Di Lazzaro V, Oliviero A, Profice P, Le Pera D, Saturno E, Tonali PA (1999) Inhibition of the human primary motor area by painful heat stimulation of the skin. Clin Neurophysiol 110:1475–1480. van Dieen JH, Cholewicki J, Radebold A (2003) Trunk muscle recruitment patterns in patients with low back pain enhance the stability of the lumbar spine. Spine 28:834 – 841. Zedka M, Prochazka A, Knight B, Gillard D, Gauthier M (1999) Voluntary and reflex control of human back muscles during induced pain. J Physiol (Lond) 520:591– 604. Ziemann U, Ishii K, Borgheresi A, Yaseen Z, Battaglia F, Hallett M, Cincotta M, Wassermann EM (1999) Dissociation of the pathways mediating ipsilateral and contralateral motor-evoked potentials in human hand and arm muscles. J Physiol (Lond) 518:895–906.
(Accepted 11 February 2011)
CHANGES IN EXCITABILITY OF CORTICOMOTOR INPUTS TO THE TRUNK MUSCLES DURING EXPERIMENTALLY-INDUCED ACUTE LOW BACK PAIN H. TSAO, K. J. TUCKER AND P. W. HODGES*
adaptation is argued to stabilise and protect the spine from further stress and injury (Hodges and Moseley, 2003). As similar differential changes in deep and superficial muscle activity are found in healthy individuals with experimentally-induced acute LBP (Arendt-Nielsen et al., 1996; Zedka et al., 1999; Hodges et al., 2003), it is argued that nociceptive input can directly influence neural inputs to the trunk muscles. However, exactly how LBP drives changes in the motor system remains poorly understood. Trunk muscles can be controlled by descending inputs from the motor cortex via rapid-conducting corticospinal pathways to spinal motoneurones (Plassman and Gandevia, 1989; Ferbert et al., 1992; Lissens et al., 1995). Notably, people with recurrent or chronic LBP show changes in corticomotor input to the trunk muscles. For instance, Strutton et al. (2005) reported an increased threshold to elicit motor responses in the back extensor muscles with chronic LBP. More recently, our group observed changes in organisation and excitability of corticomotor pathways to the deep abdominal muscle, transversus abdominis (TrA; Tsao et al., 2008a). However, both of those studies involved investigations of changes in a single muscle and in people with recurrent episodic LBP. It remains unclear whether acute nociceptive stimulation in the low back can induce immediate changes in corticomotor drive to the trunk muscles, or whether changes, if present, differ between the deep and superficial trunk muscles. Several studies on hand and jaw muscles have demonstrated altered corticomotor excitability during experimentally-induced muscle pain. In general, most studies report that the amplitude of motor responses evoked by magnetic brain stimulation is reduced in the painful muscle when subjects are at rest (Valeriani et al., 1999; Le Pera et al., 2001). One study showed increased evoked responses. However, as subjects in that study were required to actively maintain a pre-set level of motor activity, it was argued that increased activation with pain served to compensate for reduced baseline excitability due to pain (Del Santo et al., 2007). However, the organisation of control of the trunk muscles by the motor system differs somewhat from that of the limb and jaw muscles. For example, trunk muscles receive greater bilateral drive compared to the distal muscles of the limbs (Carr et al., 1985; Marsden et al., 1999) and although jaw muscles receive largely bilateral drive, their activity is controlled by corticobulbar rather than corticospinal inputs (Nordstrom, 2007). Therefore, it is difficult to translate findings from studies on the limb or jaw muscles directly to the trunk muscles.
Centre for Clinical Research Excellence in Spinal Pain, Injury and Health, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane, Australia 4072
Abstract—Acute low back pain (LBP) is associated with differential changes in motor coordination of deep and superficial trunk muscles. Whether this is related to differential changes in excitability of descending corticomotor inputs remains unclear and was investigated in nine healthy individuals. Fine-wire i.m. electrodes were inserted bilaterally into deep (transversus abdominis (TrA)) and superficial abdominal muscles (obliquus externus abdominis (OE)), and surface electrodes were placed bilaterally over obliquus internus abdominis (OI), rectus abdominis (RA) and lumbar erector spinae (LES) muscles. Corticomotor excitability was assessed as amplitude of motor evoked potentials (MEPs) to transcranial magnetic stimulation (TMS) at a range of stimulator intensities, at rest and during voluntary abdominal contractions. Pain was induced by injection of hypertonic saline into interspinous ligaments of the lumbar spine. Corticomotor excitability was examined before, during and after the induction of LBP. During pain, amplitude of TrA MEPs to contralateral cortical stimulation was reduced, whereas amplitudes of OE and LES MEPs contralateral and ipsilateral to the stimulated cortex were increased. The findings highlight differential changes in excitability of corticomotor inputs to trunk muscles during acute LBP. Further work is required to reveal whether such changes involve spinal and/or supraspinal centres and their consequence for spine control. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: low back pain, motor cortex, corticospinal excitability, trunk muscles, experimental pain.
Low back pain (LBP) is associated with altered motor coordination of trunk muscles during postural and functional tasks (Hodges and Moseley, 2003; van Dieen et al., 2003). Although inter-subject and inter-task variability exists, most studies highlight differential changes between deep and superficial trunk muscles. For instance, in people with recurrent LBP, activity of deep muscles is reduced (Danneels et al., 2002) or delayed (Hodges and Richardson, 1996; MacDonald et al., 2009), whereas activity of superficial trunk muscles is often increased (ArendtNielsen et al., 1996; Radebold et al., 2001). This latter *Corresponding author. Tel: ⫹61-7-3365-4567; fax: ⫹61-7-3365-4567. E-mail address: [email protected] (P. W. Hodges). Abbreviations: EMG, electromyography; LBP, low back pain; LES, lumbar erector spinae; MEPs, motor evoked potentials; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; RMS, root-mean-square; TMS, transcranial magnetic stimulation; TrA, transversus abdominis.
0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2011.02.033
127
128
H. Tsao et al. / Neuroscience 181 (2011) 127–133
The current study aimed to investigate the effects of experimentally-induced LBP on changes in excitability of the motor cortex and descending corticospinal pathways to the trunk muscles in healthy individuals. Given that differential changes between the deep and superficial trunk muscles have been previously observed with pain, we hypothesised that changes in corticomotor excitability may also differ between the deep and superficial trunk muscles.
EXPERIMENTAL PROCEDURES Participants Nine healthy volunteers were recruited (five males, four females, age 25⫾4 [mean⫾SD] years, height 176⫾10 cm, weight 68⫾9 kg). Subjects were excluded if they had any history of back and/or leg pain, epilepsy or family history of epilepsy, major neurological, respiratory, orthopaedic or circulatory disorders, previous spinal or abdominal surgery, pregnancy in the last 2 years, had taken any analgesic medication in the past 12 months, or if they had undertaken any form of abdominal or back muscle exercise programs in the past 12 months. The study was approved by the Institutional Medical Ethics Committee and all procedures conformed to the Declaration of Helsinki.
Electromyography Following informed consent, electromyographic (EMG) activity of the deep abdominal muscle, TrA, and the more superficial abdominal muscle, obliquus externus abdominis (OE), were recorded bilaterally using bipolar fine-wire intramuscular (i.m.) electrodes. This involved pairs of Teflon-coated stainless steel wires (75 m diameter) with 1 mm of insulation removed from end and bent back by 1 and 2 mm, respectively. These were threaded into a hypodermic needle and inserted with ultrasound guidance (Hodges and Richardson, 1996). In addition, pairs of surface electrodes (Ag/AgCl disc electrode, 10 mm diameter, 20 mm inter-electrode spacing, Grass Telefactor, RI, USA) were placed bilaterally over the lumbar erector spinae (LES; 5 cm lateral to the L2 spinous process (Cholewicki et al., 1997)), rectus abdominis (RA; 2 cm lateral to umbilicus (Ng et al., 1998)) and obliquus internus abdominis (OI; anteromedial to anterior inferior iliac spine (Ng et al., 1998)). No subject reported any discomfort or pain during the experiment that was associated with fine-wire placement. A reference electrode was placed over the lateral ribs. EMG data were amplified 2000 times, band-pass filtered between 20 and 1000 Hz, and sampled at 2 kHz using a Power 1401 Data Acquisition System with Spike 2 software (Cambridge Electronic Design, Cambridge, UK).
Procedures Subjects were positioned comfortably in prone with pillows under their legs and arms by their side. This position was chosen as it allowed easy access to interspinous space without having to move the subject. Motor cortical excitability of the trunk muscles was examined before and during experimentally-induced LBP and ⬃30 min after the cessation of induced-pain (when reports of pain had returned to zero). Excitability of corticomotor projections to the trunk muscles were investigated using transcranial magnetic stimulation (TMS). A single-pulse monophasic MagStim 2002 (The Magstim Company, Carmarthenshire, Wales, UK) was used to stimulate the motor cortex at rest and during submaximal activity. As it was difficult and impractical for naïve subjects to activate and maintain a pre-set motor activity for all the trunk muscle recorded, we elected to standardise voluntary contractions of the abdominal muscles through pre-set level of TrA activation. To determine the
level of voluntary contraction during motor cortical stimulation, EMG activity during maximum voluntary contraction (MVC) for TrA was recorded. Subjects performed three repetitions of a forced expiratory manoeuvre with verbal encouragement, each lasting at least 3 s (Ninane et al., 1992). The root-mean-square (RMS) EMG of the left and right TrA was plotted online using 200 ms windows and the peak RMS EMG over a 1 s period was calculated. A target level of ⬃10% TrA activation was set and subjects achieved this level of contraction through an abdominal brace, which has been shown to activate TrA with other abdominal muscles (Urquhart et al., 2005). Feedback of the intensity of TrA contractions was provided via a monitor although subjects were instructed not to focus on any specific muscle(s). A 7 cm figure-of-eight coil was used to identify to the optimal location to elicit motor evoked potentials (MEPs) in the test muscles contralateral to the stimulated cortex. The coil was oriented ⬃45° (to induce current in the brain in an anteromedial direction) and set at suprathreshold intensity (⬃70 –100% maximum stimulator output) during ⬃10% voluntary contraction of TrA. This coil provides better focality of stimulation and is more suited for identification of the optimal location (Cohen et al., 1990; Brasil-Neto et al., 1992). In most subjects, the optimal location was found to be ⬃2 cm anterior and lateral to the vertex, and is consistent with previous studies of TrA (Tsao et al., 2008a,b) and other trunk muscles (Strutton et al., 2004; O’Connell et al., 2007). We elected to stimulate this optimal location (determined with the figure-ofeight coil) with the double-cone coil. Although the induced current in the motor cortex differs with coils shape/configuration, the figure-of-eight provided a standardised method to identify the optimal location. As the same scalp site was used to stimulate the motor cortex before, during and after pain for each participant, any inaccuracy in identification of the optimal site based on the figureof-eight coil is unlikely to have influenced the results. A 9 cm double-cone coil was used to stimulate the motor cortex. The excitability of corticomotor pathways was evaluated by constructing a recruitment curve with the amplitude of motor evoked potentials (MEPs) measured across stimulator intensities at rest and during submaximal abdominal contractions (Ridding and Rothwell, 1997). This was positioned over the optimal location (largest MEP for a given stimulator output) with the coil oriented such that induced currents in the motor cortex flowed in an anterior direction. The 9 cm double-cone coil produces a stronger magnetic field and can evoke consistent MEPs at rest and in trunk muscles ipsilateral to the stimulated cortex (Tsao et al., 2008a). Five stimuli were delivered at each 10% interval from 50% to 90% of maximum stimulator output at rest, and from 30% to 80% of maximum stimulator output during voluntary abdominal muscle activation to 10% MVC for TrA. This procedure was repeated over both hemispheres.
Experimental pain Experimentally-induced LBP was elicited via two consecutive 0.3 ml bolus injections of hypertonic saline solution into the interspinous ligaments between the 4th and 5th (L4/5), and the 3rd and 4th (L3/4) lumbar spinous processes. Injections delivered into different spinal levels induced pain of sufficient duration to complete measures of corticomotor excitability and motor coordination. We did not inject into the same location twice as pilot trials showed that a second injection into the same area yielded shorterlasting and less-intense LBP. This injection method was first described by Kellgren (1939). Recent work from our team has shown pain from this injection produces central back pain that has several advantages over muscle injection (Tsao et al., 2010). First, the pain from hypertonic saline injections of the interspinous ligament is longer lasting. Second, ligament pain does not attenuate with trunk muscle contractions compared to muscle pain. Third, this procedure avoids injection into the muscles under investigation. The interspinous spaces of L4/5 and L3/4 were identified by palpation and if required, confirmed with real-time ultrasound imag-
H. Tsao et al. / Neuroscience 181 (2011) 127–133
129
Fig. 1. (A) Injection of hypertonic saline into L4/5 interspinous ligament. (B) Location and area of experimentally-induced pain. (C) Pain intensity plotted against time for each individual averaged across hypertonic saline injections at L4/5 and L3/4 interspinous ligament.
ing. The needle (27 G⫻25 mm) was introduced into the interspinous space angled at ⬃30° in a cranial direction to the skin until a gentle tap was felt against the spinous process of the upper spinal level (Fig. 1A). At this point, 0.3 ml of hypertonic saline (5% NaCl) was introduced. Subjects rated the intensity of the pain using an electronic visual analogue scale (VAS) anchored with 0 (‘no pain’) and 10 (‘worst pain imaginable’). The location and area of pain was marked on a body chart by the participant at the end of the experiment.
Data analysis Data were imported and analysed using Matlab 7 (The Mathsworks, MA, USA). For TMS evoked responses, EMG data were full-wave rectified, averaged for each stimulator intensity at rest and during abdominal activation, and the onset and offset of MEPs that were clearly discernible from background EMG activity were visually identified. To remove any potential for bias, data were presented in random order without reference to the identity of the muscle, intensity of TMS or whether the trials were before, during or after pain. In trials with identifiable MEPs, RMS EMG of MEPs (from onset to offset) was calculated and grouped into responses elicited contralateral and ipsilateral to the stimulated hemisphere. In trials in which no MEPs were identified, the RMS EMG of myoelectric activity was calculated from 15 to 35 ms (which corresponded to the time range of expected MEPs) following TMS stimulus. RMS EMG amplitudes for MEPs were normalised to the peak response across trials and across both hemispheres at rest and during activity, and plotted as a recruitment curve against the stimulator intensity. The overall excitability of corticomotor pathways was also represented by calculating the area under the recruitment curve. The background EMG activity for each trial was also calculated as the amplitude of RMS EMG from 55 to 5 ms before the stimulation.
Statistical analysis Statistica 7 (Statsoft, OK, USA) was used for statistical analysis, with significance level set to P⫽0.05. To examine whether pain
influenced corticomotor excitability of the trunk muscles, normalised amplitudes of trunk muscle MEPs were compared between Conditions (pre, pain vs. post), Muscles and Intensities (% maximum stimulator output) at rest and during abdominal activation using repeated measures analysis of variance (ANOVA). The area under the recruitment curve and background EMG activity was also compared between Conditions, Muscle and Activity (rest vs. activity) using repeated-measures ANOVA. The onset of MEPs was also compared between Muscles and Activity (rest vs. abdominal activation). Post-hoc analyses were undertaken using Duncan’s multiple range test.
RESULTS Injection of hypertonic saline into the interspinous ligament induced central back pain in all subjects (Fig. 1B). The average (SD) of the peak pain intensity experienced during TMS stimulation was 4.3 (0.8) cm on the 10 cm VAS. The average pain duration from the onset of injection until the cessation of pain was 11.9 (2.6) min for each injection (or ⬃23.8 min across both injections). Fig. 1C shows pain intensity (averaged each min) against time, averaged across both injections. Notably, for the pain condition, all TMS stimulation was completed before the cessation of pain. Corticomotor excitability Figs. 2 and 3 show recruitment curves at rest and during abdominal activation. As expected, the peak MEP amplitudes were 2.6 –5.1 times greater during voluntary abdominal brace than that at rest. During pain, reduced amplitude of TrA MEPs contralateral to the stimulated cortex was observed at rest from 70% to 90% maximum stimulator output (Interaction Pain⫻Intensity⫻Muscles, P⬍0.001; post hoc P⬍0.048), and during abdominal activity from
130
H. Tsao et al. / Neuroscience 181 (2011) 127–133
P⬍0.036) and from 70% to 80% maximum stimulator output during voluntary contractions (Fig. 3; post-hoc P⬍0.013). Similarly, amplitude of MEPs ipsilateral to the stimulated cortex was increased at 90% maximum stimulator output at rest (post-hoc P⬍0.004) and at 80% maximum stimulator output during voluntary contractions (posthoc P⬍0.043). Area under the recruitment curve for OE was also increased, although this was only observed for the rest condition (Fig. 4; post-hoc P⬍0.047). Responses of LES contralateral and ipsilateral (from 80% to 90% maximum stimulator output) to the stimulated cortex and the area under the recruitment curve were also increased, but only for the rest condition (MEPs: post-hoc P⬍0.034; Area under curve: post-hoc P⬍0.048). Amplitude of LES MEPs (post-hoc P⬎0.19) or area under LES recruitment curve (post-hoc P⬎0.07) did not change during voluntary contraction of the abdominal muscles. OI and RA MEP amplitude did not change at rest (post-hoc P⬎0.091) or during voluntary abdominal contraction (post-hoc P⬎0.080). Similarly, there was no
Fig. 2. Amplitude of motor evoked potentials (MEPs) at rest expressed as recruitment curves from 50% to 90% maximum stimulator output before (white circles), during (black circles) and after induced back pain (grey circles). Values are expressed as root-mean-square (RMS) EMG amplitude normalised to peak response. Mean and 95% confidence intervals are shown for responses ipsilateral (left panel) and contralateral (right panel) to the stimulated cortex. * indicates significant difference in MEP amplitude between pain and pre-/postpain (P⬍0.05). TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
60% to 80% maximum stimulator output (Interaction Pain⫻Intensity⫻Muscles, P⬍0.001; post-hoc P⬍0.003). This is consistent with reduced area under the recruitment curve at rest (Fig. 4A; Interaction Pain⫻Muscles, P⬍0.001; post-hoc P⬍0.008) and during voluntary activation (Interaction Pain⫻Muscles, P⬍0.001; post-hoc P⬍0.001; Fig. 4B). In contrast, there was no change in the amplitude of TrA MEPs ipsilateral to the stimulated cortex at rest (post-hoc P⬎0.21) and during activity (post-hoc P⬎0.11), and no change was observed for area under the recruitment curve (rest: post-hoc P⬎0.32; voluntary activation: post-hoc P⬎0.15). Pain was associated with increased amplitude of OE MEPs contralateral to the stimulated cortex from 80% to 90% maximum stimulator output at rest (Fig. 2; post-hoc
Fig. 3. Amplitude of motor evoked potentials (MEPs) during abdominal brace expressed as recruitment curves from 30% to 80% maximum stimulator output before (white circles), during (black circles) and after induced back pain (grey circles). Values are expressed as rootmean-square (RMS) EMG amplitude normalised to peak response. Mean and 95% confidence intervals are shown for responses ipsilateral (left panel) and contralateral (right panel) to stimulated cortex. * indicates significant difference in MEP amplitude between pain and pre-/post-pain (P⬍0.05). TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
H. Tsao et al. / Neuroscience 181 (2011) 127–133
Fig. 4. Areas under the recruitment curves at rest (A) and during voluntary abdominal contractions (B). Mean and 95% confidence interval are shown for responses ipsilateral and contralateral to the stimulated hemisphere before (white circles), during (black circles) and after pain (grey circles). * P⬍0.05. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae.
change in area under the curve for these muscles (posthoc P⬎0.16). Fig. 5 shows representative data from the right sided trunk muscles for a single subject. MEP onset The onset of MEPs for all abdominal muscles contralateral to the stimulated cortex was shorter than that ipsilateral to the stimulated cortex (Fig. 6; Interaction
131
Fig. 6. Mean latency to onset (and 95% confidence interval) of motor evoked potentials (MEPs) for trunk muscles at rest (black circles) and during abdominal contractions (white circles) ipsilateral and contralateral to the stimulated hemisphere. Note for all muscles except the lumbar erector spinae (LES; which was not activated by the task), the onset of MEPs during abdominal contractions was faster than that at rest and the latency of MEPs contralateral to the stimulated cortex was faster than that ipsilateral to the stimulated hemisphere. * P⬍0.05. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis.
Muscles⫻Activity, P⬍0.001; post-hoc for all abdominal muscles, P⬍0.048). The shorter latency to MEP onset for contralateral responses is consistent with activation of the monosynaptic crossed-corticospinal pathway, whereas the longer latency to MEP onset for ipsilateral responses is consistent with activation of polysynaptic uncrossed-corticospinal pathways (Ziemann et al., 1999; Strutton et al., 2004). However, LES MEP onset did not differ between sides (post-hoc P⬎0.38). As expected, the onset of abdominal muscle MEPs during activity was shorter than that at rest (Main effect Activity, P⬍0.001; post-hoc P⬍0.001). LES MEP onset was sim-
Fig. 5. Representative data from right (R) trunk muscles of a single subject at rest before, during and after pain. Dotted vertical lines represent time of stimulation. TrA, transversus abdominis; OE, obliquus externus abdominis; OI, obliquus internus abdominis; RA, rectus abdominis; LES, lumbar erector spinae. EMG calibration 0.2 mV.
132
H. Tsao et al. / Neuroscience 181 (2011) 127–133
ilar between rest and abdominal muscle activation (posthoc P⬎0.097). Background activity There was no change in background activity for any muscle between pre-pain, pain and post-pain conditions (Main effect for Conditions: P⫽0.62). This implies that any paininduced changes in corticomotor excitability are unlikely to be due to differences in background activity. As expected, the amplitude of background activity for TrA, OE and OI was significantly greater during abdominal contractions than at rest (Interaction between Activity and Muscle: P⬍0.001; post-hoc P⬍0.037). Background activity of RA and LES did not increase during voluntary abdominal muscle contractions (post-hoc P⬎0.32).
DISCUSSION The findings suggest that acute LBP is associated with differential changes in corticomotor excitability of the trunk muscles. The deep abdominal muscle, TrA, showed reduced corticomotor excitability contralateral to the stimulated cortex. In contrast, the more superficial trunk muscles including OE and LES demonstrated increased excitability to cortical stimulation over both hemispheres during pain. The results demonstrate the potential for rapid plasticity of corticomotor pathways with short-term pain and importantly, highlight that pain does not induce a stereotypical change in corticomotor excitability. Rather, the response varies between muscles. Acute LBP is commonly associated with differential changes in muscle recruitment between the deep and superficial trunk muscles (Hodges and Moseley, 2003; van Dieen et al., 2003). Here we show that these behavioural findings are mirrored by differential changes in corticomotor excitability. As background activity was similar between pain and no pain conditions, modulation of corticomotor excitability during pain could not be explained by changes in background activity. Increased corticomotor drive to the superficial trunk muscles is somewhat consistent with the aim of the CNS to “protect the part” from further injury/pain (Hodges and Moseley, 2003). That is, increased excitability of these pathways and thus potential for increased activity of the superficial trunk muscles can contribute to splinting and reduced mobility of the spine (Hodges et al., 2006). As a result, the contribution of the deeper trunk muscles such as TrA could be perceived as redundant during pain. This is consistent with findings of reduced corticomotor drive to the deep trunk muscles, at least on the side contralateral to the stimulated cortex. The data showed increased excitability of corticomotor inputs to the LES with acute LBP, but this was only observed at rest. Findings of increased excitability contrast previous findings by Strutton et al. (2005), which showed reduced excitability (i.e, increased motor threshold to induce a response to TMS) of the LES in patients with chronic LBP. One possible explanation for this discrepancy could be differences between experimentally-induced acute LBP and more persistent clinical LBP. Alternatively,
in the study by Strutton et al. (2005), subjects were required to activate their back muscles to a pre-set intensity, whereas in the current study, subjects performed an abdominal brace and standardised activation to TrA, not the LES. For practical reasons, we did not control the level of activity of other trunk muscles, but no systematic increase in LES activity was recorded during this task. This may have increased inter-subject variability of MEPs during activity, although this is unlikely to have influenced the current findings as similar background activities were found between pain and no pain conditions. Future work would be required to further verify the influence of pain on corticomotor excitability of LES muscle during voluntary activation of these muscles. No change in corticomotor excitability of TrA was observed on the ipsilateral side to the stimulated cortex. The onset of MEPs for all abdominal muscles (TrA, OE, OI and RA) contralateral to the stimulated cortex was faster than that ipsilateral to the stimulated cortex. This suggests at least for the initial part of the response, contralateral responses were evoked via the rapid-conducting crossedcorticomotor pathways, whereas ipsilateral responses were mediated via slower cortico-brainstem-spinal tracts (Ziemann et al., 1999; Strutton et al., 2004). Our previous finding suggests that motor threshold of ipsilateral responses was reduced in people with chronic LBP, but this was only evident over the more excitable hemisphere (Tsao et al., 2008b). As the current study did not evaluate motor threshold and instead grouped ipsilateral responses from stimulation of both hemispheres, this may explain the lack of changes in ipsilateral responses during pain. Our previous study in patients with chronic LBP did not observe a correlation between changes in motor excitability (as measured by motor threshold) and changes in motor coordination (Tsao et al., 2008b). Rather, changes in motor coordination were found to be associated with changes in the organisation of intra-cortical networks at the motor cortex. Due to the length of time required to map the motor cortex using TMS and the short-duration of experimentallyinduced LBP, mapping was not conducted in the current study. Furthermore, corticomotor excitability may be associated with other temporal and spatial parameters or even other functional tasks which were not examined in the current study. Further work is needed to unravel the functional relevance of pain-associated changes in corticomotor excitability of inputs to the trunk muscles. Changes in motor coordination of the trunk muscles with LBP, which may be useful in the short term to protect the body from further injury, can persist following resolution of symptoms and have been argued to contribute to the recurrence and chronicity of pain episodes (Hodges and Moseley, 2003). Therefore, findings of rapid alteration in corticomotor excitability may highlight the importance of early motor rehabilitation following acute LBP and injury. However the exact nature of the training to change motor control may depend on the mechanism and site of changes in corticomotor excitability. TMS evoked responses can be affected by changes in excitability of cortical cells, motoneurones and any interneurons (Iscoe, 1998). As the
H. Tsao et al. / Neuroscience 181 (2011) 127–133
exact site of pain-induced changes in TMS-evoked responses cannot be determined in the current study, future studies are needed to further investigate spinal and/or supraspinal mechanisms. Acknowledgments—We like to thank Ms Leanne Hall and Ms Rachel Park for their assistance with data collection. Funding was provided by the National Health and Medical Research Council of Australia.
REFERENCES Arendt-Nielsen L, Graven-Nielsen T, Svarrer H, Svensson P (1996) The influence of low back pain on muscle activity and coordination during gait: a clinical and experimental study. Pain 64:231–240. Brasil-Neto JP, Cohen LG, Panizza M, Nilsson J, Roth BJ, Hallett M (1992) Optimal focal transcranial magnetic activation of the human motor cortex: effects of coil orientation, shape of the induced current pulse, and stimulus intensity. J Clin Neurophysiol 9:132– 136. Carr JH, Shepherd RB, Nordholm L, Lynne D (1985) Investigation of a new motor assessment scale for stroke patients. Phys Ther 65:175–180. Cholewicki J, Panjabi MM, Khachatryan A (1997) Stabilizing function of trunk flexor-extensor muscles around a neutral spine posture. Spine 22:2207–2212. Cohen LG, Roth BJ, Nilsson J, Dang N, Panizza M, Bandinelli S, Friauf W, Hallett M (1990) Effects of coil design on delivery of focal magnetic stimulation. Technical considerations. Electroencephalogr Clin Neurophysiol 75:350 –357. Danneels LA, Coorevits PL, Cools AM, Vanderstraeten GG, Cambier DC, Witvrouw EE, De Cuyper HJ (2002) Differences in multifidus and iliocostalis lumborum activity between healthy subjects and patients with sub-acute and chronic low back pain. Eur Spine J 11:13–19. Del Santo F, Gelli F, Spidalieri R, Rossi A (2007) Corticospinal drive during painful voluntary contractions at constant force output. Brain Res 1128:91–98. Ferbert A, Caramia D, Priori A, Bertolasi L, Rothwell JC (1992) Cortical projection to erector spinae muscles in man as assessed by focal transcranial magnetic stimulation. Electroencephalogr Clin Neurophysiol 85:382–387. Hodges PW, Cholewicki J, Coppieters MW, MacDonald D (2006) Trunk muscle activity is increased during experimental back pain, but the pattern varies between individuals. In: International Society for Electrophysiology and Kinesiology, Torino, Italy. Hodges PW, Moseley GL (2003) Pain and motor control of the lumbopelvic region: effect and possible mechanisms. J Electromyogr Kinesiol 13:361–370. Hodges PW, Moseley GL, Gabrielsson A, Gandevia SC (2003) Experimental muscle pain changes feedforward postural responses of the trunk muscles. Exp Brain Res 151:262–271. Hodges PW, Richardson CA (1996) Inefficient muscular stabilization of the lumbar spine associated with low back pain: a motor control evaluation of transversus abdominis. Spine 21:2640 –2650. Iscoe S (1998) Control of abdominal muscles. Prog Neurobiol 56:433– 506. Kellgren JH (1939) On the distribution of pain arising from deep somatic structures with charts of segmental pain areas. Clin Sci 4:35. Le Pera D, Graven-Nielsen T, Valeriani M, Oliviero A, Di Lazzaro V, Tonali PA, Arendt-Nielsen L (2001) Inhibition of motor system excitability at cortical and spinal level by tonic muscle pain. Clin Neurophysiol 112:1633–1641. Lissens MA, De Muynck MC, Decleir AM, Vanderstraeten GG (1995) Motor evoked potentials of the abdominal muscles elicited through
133
magnetic transcranial brain stimulation. Muscle Nerve 18:1353– 1354. MacDonald D, Moseley GL, Hodges PW (2009) Why do some patients keep hurting their back? Evidence of ongoing back muscle dysfunction during remission from recurrent back pain. Pain 142: 183–188. Marsden JF, Farmer SF, Halliday DM, Rosenberg JR, Brown P (1999) The unilateral and bilateral control of motor unit pairs in the first dorsal interosseous and paraspinal muscles in man. J Physiol (Lond) 521:553–564. Ng JK-F, Kippers V, Richardson CA (1998) Muscle fibre orientation of abdominal muscles and suggested surface EMG electrode positions. Electromyogr Clin Neurophysiol 38:51–58. Ninane V, Rypens F, Yernault JC, De Troyer A (1992) Abdominal muscle use during breathing in patients with chronic airflow obstruction. Am Rev Respir Dis 146:16 –21. Nordstrom MA (2007) Insights into the bilateral cortical control of human masticatory muscles revealed by transcranial magnetic stimulation. Arch Oral Biol 52:338 –342. O’Connell NE, Maskill DW, Cossar J, Nowicky AV (2007) Mapping the cortical representation of the lumbar paravertebral muscles. Clin Neurophysiol 118:2451–2455. Plassman BL, Gandevia SC (1989) Comparison of human motor cortical projections to abdominal muscles and intrinsic muscles of the hand. Exp Brain Res 78:301–308. Radebold A, Cholewicki J, Polzhofer GK, Greene HS (2001) Impaired postural control of the lumbar spine is associated with delayed muscle response times in patients with chronic idiopathic low back pain. Spine 26:724 –730. Ridding MC, Rothwell JC (1997) Stimulus/response curves as a method of measuring motor cortical excitability in man. Electroencephalogr Clin Neurophysiol 105:340 –344. Strutton PH, Beith I, Theodorou S, Catley M, McGregor AH, Davey NJ (2004) Corticospinal activation of internal oblique muscles has a strong ipsilateral component and can be lateralized in man. Exp Brain Res 158:474 – 479. Strutton PH, Theodorou S, Catley M, McGregor AH, Davey NJ (2005) Corticospinal excitability in patients with chronic low back pain. J Spinal Disord Tech 18:420 – 424. Tsao H, Galea MP, Hodges PW (2008a) Concurrent excitation of the opposite motor cortex during transcranial magnetic stimulation to activate the abdominal muscles. J Neurosci Methods 171:132–139. Tsao H, Galea MP, Hodges PW (2008b) Reorganization of the motor cortex is associated with postural control deficits in recurrent low back pain. Brain 131:2161–2171. Tsao H, Tucker KJ, Coppieters MW, Hodges PW (2010) Experimentally induced low back pain from hypertonic saline injections into lumbar interspinous ligament and erector spinae muscle. Pain 150:167–172. Urquhart DM, Hodges PW, Allen TJ, Story IH (2005) Abdominal muscle recruitment during a range of voluntary exercises. Man Ther 10:144 –153. Valeriani M, Restuccia D, Di Lazzaro V, Oliviero A, Profice P, Le Pera D, Saturno E, Tonali PA (1999) Inhibition of the human primary motor area by painful heat stimulation of the skin. Clin Neurophysiol 110:1475–1480. van Dieen JH, Cholewicki J, Radebold A (2003) Trunk muscle recruitment patterns in patients with low back pain enhance the stability of the lumbar spine. Spine 28:834 – 841. Zedka M, Prochazka A, Knight B, Gillard D, Gauthier M (1999) Voluntary and reflex control of human back muscles during induced pain. J Physiol (Lond) 520:591– 604. Ziemann U, Ishii K, Borgheresi A, Yaseen Z, Battaglia F, Hallett M, Cincotta M, Wassermann EM (1999) Dissociation of the pathways mediating ipsilateral and contralateral motor-evoked potentials in human hand and arm muscles. J Physiol (Lond) 518:895–906.
(Accepted 11 February 2011)