|
nif:isString
|
-
The fMRI study included eight right-handed healthy volunteers (5 males, 3 females, mean age 26.2 years, range 20–34 years), who were either students or physicians and already experienced in participating in psychophysical studies. All participants participated voluntarily in the experiments and none of them were taking medication that might have interfered with the induction of subacute muscle soreness or with pain sensations (i.e. analgesics, anesthetics or other centrally acting drugs). The study adhered to the claims of the Declaration of Helsinki and the protocol was approved by the local ethics committee (Institutional Review Board of the University Erlangen-Nuremberg).
Induction of delayed onset muscle soreness (DOMS): DOMS is a physiological pain state that occurs following a bout of eccentric muscle contractions and peaks in intensity 24-48 hours after exercise [18]. To induce DOMS in the quadriceps muscle group, subjects performed 120–150 up-down step cycles with their right leg. A step cycle consisted of stepping up onto a small platform with the right leg (concentric quadriceps contraction) then stepping back down with the left leg which forces the right leg to bear the body weight (eccentric quadriceps contraction). The height of the platform was adjusted such that the included knee angle was 90° or less when the right foot was placed on the platform, i.e. the eccentric contraction was performed over a long quadriceps muscle length. 36 to 48 hours later, subjects were psychophysically evaluated. Eight of the eleven subjects experienced intense soreness and long lasting pain in their quadriceps muscle upon contraction or physical stimulation and were admitted to the fMRI experiment. None of the subjects reported resting pain.
Experimental pain model and intensity rating: The intensity of the pain was quantified and monitored using the visual analogue scale (VAS) ranging from 0 to 100% as previously described [19]. The rating scale was fed back to the subject as a light bar representing the levels 0–100% of the common VAS scale, whereby “0%” signifies no pain, “100%” marks maximal imaginable pain. For psychophysical measurements, the subjects sat reclined in a dental chair and the paradigm was carried out as later in the fMRI sessions, with the only difference that subjects were asked to operate the VAS with a turning knob to rate the pain as perceived during the contractions and during stimulation with the heavy marble roll. As for the contraction task, the subjects were instructed to perform a maximal isometric contraction of their painful and non-painful quadriceps muscle, but also to refrain from movements of the head as much as possible. Only those subjects who judged the contraction/stimulation with the roll on the sore side as painful, and attributed a rating of at least 15% on average were admitted to the fMRI experiment. The success rate (of achieving painful DOMS with this training method) exceeded 70%, i.e. eight of the eleven subjects that were initially trained by eccentric exercise, had developed painful soreness and were admitted to the scanner to record their brain activation in response to stimulation of the painful muscles. For control purposes, the subjects were asked after the fMRI experiment, to estimate the intensity of pain they had experienced during both self-controlled contraction and physical stimulation. In all subjects the pain ratings confirmed the ratings that were obtained during the psychophysical training session prior to the fMRI measurement.
Shortly before subjects attended the fMRI scanner they were instructed to refrain from movement as much as possible. In the scanner the subjects' head was immobilized by fixing the head binaurally with a thrust device. A mirror was mounted above the eyes to allow the subject to see the light indicating the protocol, which was positioned outside the scanner. Two paradigms were used to evoke DOMS-associated muscle pain in the fMRI environment. The first paradigm involved alternate repeated active contractions of the right (sore) and the left (sound) quadriceps muscle group and was phasic in nature. The stimulus is identical to an isometric contraction and the evoked pain proves to be suited for the requirements of a functional MRI stimulus design that calls for stimuli with fast on- and offset. The whole paradigm consisted of 5 cycles of alternating non-painful (control condition) and painful contractions (stimulus) with eleven interleaved resting conditions (baselines), respectively 24 and 18 seconds (Fig. 1). A red light indicated the stimulation period, during which the subjects were asked to repeatedly contract their right (sore) quadriceps muscle. Indicating control condition, the light switched to green, and the subjects to their healthy, left quadriceps group, maintaining phasic contractions over the period of the non-painful stimulus (control condition). In detail, for the contraction-related task, the subjects were advised to contract their quadriceps muscle either with maximum strength or until maximal tolerable pain was perceived and then to relax and contract the muscle again. One cycle of contraction and subsequent relaxation resulted in approximately 3 seconds and was repeated during the 24 seconds of stimulation until the light switched off to indicate baseline condition. The second paradigm consisted in physical stimulation of the sore areas of the thigh by externally applied localized pressure. It was used to separate brain activation in response to deep muscle pain from activation resulting from muscle contraction. Due to the slow adaptation properties of sensitized nociceptors (in deep tissues), pain sensation in response to tonic locally applied pressure dampens within seconds [20], [21]. In order to obtain a persistent and constant pain sensation during the 24 seconds of stimulation, the pressure was moved constantly within the sore target area of the quadriceps. This was achieved by using a custom-designed marble roll. The marble roll was fixed in a brass handle and did not interfere with the signal of the scanner (Fig. 1a). During the 24 s of stimulation, the experimenter gently rolled the device back and forth over the sore (or contralateral control) target areas. This procedure is referred to as “physical stimulation”. To evoke comparable pain rating and achieve constant stimulation conditions across the group of subjects, the most painful areas of the thigh were marked with tape in each subject prior to the fMRI measurement (tape was placed and stimulation performed through the trousers in order to prevent activation of brain areas due to stimulation of cutaneous cold receptors by the marble roll). Accordingly, the marble roll was held perpendicularly to the surface and rolled only over the marked areas. Furthermore, care was taken to exert pressure only by imposing the 1.75 kg dead load of the roll and no additional force. This way of stimulation produced in all subjects a deep and constant painful feeling with fast on- and offset. Few hours before accessing the fMRI scanner, each subject underwent a training session in order to make the subjects familiar with the stimulation procedure, but also to evaluate intensity and time course of the pain and to find and mark the most painful areas of the thigh with tape.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0047230.g001 Stimulus conditions and experimental paradigm.a. Muscle pain was evoked in a muscle group suffering DOMS during an fMRI measurement by two paradigms: voluntary contraction of the painful muscle (paradigm 1) and painful physical stimulation of the sore muscles with a 1.75 kg heavy marble rolling pin (paradigm 2). b. The scanning procedure included 5 repetitions of the same stimulus combination (5x EPI). A Scout epoque - used to determine the position of the subject's head within the scanner- is followed by the acquisition of the high-resolution anatomical MPRAGE data set. A short EPI sequence is required to accustom the subject to the unfamiliar noise of the following EPI scans. The numbers in the block design (x-axis) are equivalent to scanning frames and one frame corresponds to 3 s acquisition time. The conditions were as follows: 1: resting condition (baseline); 2: contralateral painless quadriceps contraction or painless physical stimulation; 3: ipsilateral painful quadriceps contraction or painful physical stimulation.
Image acquisition, study design and MRI sequence order: Images were collected with a 1.5 Tesla MRI scanner (Magnetom Sonata, Siemens, Erlangen, Germany) using a standard quadrature head coil. Preceding the functional measurements, a high resolution data set of each subject’s brain was acquired using a T1-weighted, 3D gradient-echo pulse sequence (MPRAGE: magnetization prepared rapid acquisition gradient echo; voxel size = 1,0 mm3; TR/TE = 1950 ms/4.38 ms, FOV = 256 mm, 256×256 matrix, 160 slices, FOV = 256 mm, slice thickness = 1 mm). Functional scans were acquired using a blood-oxygen-level-dependent (BOLD) protocol with a T2-weighted gradient echo-planar imaging (EPI) sequence (TR/TE/θ = 3000 ms/60 ms/90°; slice thickness = 4 mm; interslice interval = 1 mm, FOV = 220 mm, 64×64 matrix). 20 axial slices were placed such that the best possible whole brain coverage was obtained, usually from the top of the cortex to the base of the cerebellum. The first three images were discarded to account for spin saturation effects thus eliminating non-equilibrium effects of the magnetization. MRI sequences were measured in the following order: anatomical scout, MPRAGE, and two EPI sequences. The first EPI was short and applied to accustom the subject to the noise. The second EPI consisted of 150 whole brain acquisitions divided into 5 cycles and entailing the protocol as described above. Eight sequences were acquired during the stimulus period time (24 s) and six during the interposed baseline periods (resting condition, 18 s; see Fig.1).
Data processing and statistical analysis: Imaging data analysis, registration, visualisation and statistical analyses were performed with the BrainVoyager QX software package (Brain Innovation B.V., Maastricht; The Netherlands). Pre-processing of the data was performed as previously described [22] and included three-dimensional motion correction, temporal Gaussian smoothing of 4 s, spatial Gaussian smoothing of 4 mm, linear detrending and temporal high pass filtering using 0.01 Hz. The 3D-MPRAGE data set was transformed into the standard stereotactic space [23] and the T2 weighted images were realigned to this high-resolution 3D-data set [24] thus enabling a 3D-reconstruction of the activation maps. Fixed-effect analyses were performed and to search for clusters of activation, we used a block design with two conditions (stimulus vs. control) and an interposed baseline (see Fig.1b). The stimulus sequence pattern was convoluted with a hemodynamic response function to account for the expected delay and devolution of the BOLD signal [25], [26] and served as a basis for the calculation of the correlation coefficients of the respective cluster. Blocks of contraction activity within the stimulation pattern served as independent predictors for a general linear model (GLM). Activation was detected by correlating the time course of the BOLD-signal of each voxel with the predictor pattern of the GLM. Contrasts of interest were calculated by subtraction analysis and t-tests were calculated as differential responses, resulting in statistical maps (painful vs. non-painful states [contraction-/ physical stimulation-induced muscle pain vs. contralateral control]). P-values <5*10−6 (uncorrected) are marked in Table 1. The subtraction analyses were complemented with separate evaluation of the activation pattern of each stimulus versus its control. Analyses were performed at the group level (multi-study). Location of activated clusters was verified using a printed atlas [27] and labelled using Tailarach's nomenclature [23] by means of the Tailarach Daemon (Research Imaging Center, University of Texas Health Center, San Antonio, TX, USA). Table 1 lists clusters with a minimum number of 250 voxels corresponding to 250 mm3. We focussed on clusters located in the following brain areas: frontal and parietal lobe, insula, cingulum, thalamus, basal ganglia and the cerebellum. Processing of the clusters included identification of Brodman's Area (BA; or closest BA in case the centre of the cluster was located in the white matter), determination of spatial centre (coordinates), and their statistical significance (t-value). Psychophysical data were recorded with custom made software.
Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0047230.t001 Brain regions with increased activity during DOMS: comparison between painful voluntary contraction and physical stimulation with a rolling pin. p-values <0.000005 are indicated as “*” (uncorrected); S1: primary somatosensory cortex, M1: primary motor cortex; S/IPL: superior/inferior parietal lobule; STG: superior temporal gyrus; SSA: somatosensory association cortex; SMA: supplementary motor area; ACC: anterior cingulate cortex; PCC: posterior cingulate cortex; CMA: cingulate motor area; BA: Brodmann Areas or closest BA in vicinity; “Pain”, T-values are indicated for Pain = painful condition and Con = control condition and contrast; x, y, z: Talairach coordinates.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}