# Specific motor cortex hypoexcitability and hypoactivation in COPD patients with peripheral muscle weakness | BMC Pulmonary Medicine | Full Text

Forty COPD patients and 22 healthy controls, aged between 40 and 80 years, were recruited for the study (Fig. 1). The COPD patients were recruited and tested at their entrance in two French pulmonary rehabilitation centers (Cliniques du Souffle La Vallonie, Lodève, and Les Clarines, Riom-ès-Montagne, France) between 2012 and 2014. The healthy controls were recruited through an ad in a local newspaper within the same period. The participation criteria for the COPD patients were a diagnosis of COPD with forced expiratory volume in the 1st second (FEV1) between 30 and 80% of the theoretical values (GOLD 2 and 3), with no exacerbation or weight loss in the month preceding the study. The non-inclusion criteria were the same for patients and controls: inability to give written consent, inability to perform the experimental maneuvers, impaired visual function, use of drugs known to impact brain function (GABA agonist, Z-drugs, tricyclic antidepressants, melatoninergic antidepressants, selective serotonin/noradrenalin reuptake inhibitors and opioid receptor agonists), chronic current or past alcohol abuse (> 14 units of alcohol per week), mental disorder and neurologic or neuromuscular disease. For the diagnosis of peripheral muscle weakness, the isometric maximal quadriceps torque (QMVC) of each participant was expressed as a percentage of predicted values obtained from the national isometric muscle strength database consortium [33]. Patients with QMVC below 80% of predicted values were then assigned to the muscle weakness group (COPDMW) and the others to the non-muscle weakness group (COPDNoMW) [34]. Healthy controls with peripheral muscle weakness were excluded from the analyses (n = 2). All participants gave written consent. Procedures were approved by the local ethics committee (CPP Sud-Est VI, Clermont-Ferrand, number AU980) and complied with the principles of the Declaration of Helsinki for human experimentation.

Both patients and controls underwent plethysmography (V6200 Autobox, Sensormedics Corp., Yorba Linda, CA, USA). Measurements included forced vital capacity (FVC) and FEV1. The presence of persistent airflow obstruction and thus COPD was defined by a postbronchodilatator FEV1/FVC ratio < 0.7 [35]. The FEV1 values were expressed as a percentage of predicted value [36].

#### Blood gas analyses

Measurement of blood gases (PaO2 and PaCO2) collected from the radial artery was performed in resting patients while they breathed room air, using a blood gas analyzer (ABL 825, Radiometer Medical, Bronshoj, Denmark).

### Neuromuscular tests

#### Experimental design

After determination of the dominant leg [37], the participants were comfortably seated on a dedicated ergometer for knee extensor testing (Quadriergoforme, Aleo Industrie, Salome, France) equipped with a strain gauge torque sensor (Captels, Saint Mathieu de Treviers, France). The hip and the knee angle were set at 90°. The pelvis and the proximal extremity of the patella were securely attached to the chair in order to minimize movements of adjacent muscles. All the experimental manoeuvers of the protocol were done on the ergometer and in the same body position (including stimulations at rest). The participants were systematically familiarized to the experimental procedures the day before the protocol through a physical training session. This session included transcranial magnetic and femoral nerve stimulation recruitment curves, followed by 3 maximal voluntary contractions and several submaximal voluntary contractions at 30 and 50% of MVC lasting 5 s or until the targets were correctly reached, with superimposed transcranial magnetic and femoral nerve stimulations.

#### Evaluation of isometric maximal quadriceps torque

Isometric maximal quadriceps torque of the dominant leg was assessed as the highest torque value recorded during the protocol. Participants were verbally encouraged during each contraction to ensure maximal personal implication. QMVC was expressed in Nm and as a percentage of predicted values obtained from the national isometric muscle strength database consortium [33]. The maximal electrically evoked torque (quadriceps peak twitch, QPt) was assessed at rest as the highest twitch response induced by IMmax femoral nerve stimulation (IMmax determination is described in the following paragraph).

#### Evaluation of peripheral and spinal excitability by femoral nerve stimulation

The femoral nerve stimulation was applied to assess peripheral and spinal excitability. A constant-current and high voltage stimulator (DS7AH, Digitimer, Hertforshire, UK) was used. Rectangular monophasic pulses of 500 μs were used to ensure optimal activation of deeper muscle fibers [38] and to enable the appearance of H-waves [39]. The anode, a self-adhesive electrode (10 × 5 cm), was placed over the greater trochanter. The cathode, a ball electrode covered with damp foam, was placed over the participant’s femoral triangle (Scarpa), 3 to 5 cm below the inguinal ligament. To determine optimal location, the cathode was moved by small amounts while delivering pulses at 50 mA until the highest M-wave response was obtained over the vastus medialis with the smallest possible response over the antagonist biceps femoris. Then markers were set over the participant to maintain the cathode position. A recruitment curve was performed at rest to determine the intensities at which the highest M-wave (Mmax) and H-reflex (Hmax) were obtained. One pulse was delivered on the femoral nerve every 10 s, with the intensity beginning at 50 mA and increasing by 10 mA until no further increase in twitch mechanical response and M-wave amplitude occurred. The intensity used during the protocol was set as 10% above the intensity at which Mmax was elicited (supramaximal intensity noted IMmax). IMmax was used to evoke M-wave at rest (Mmax) and during maximal voluntary contraction to deliver double twitch pulses (doublet) at 100 Hz. Subsequently to the IMmax determination, the intensity at which the maximum Hmax was obtained was carefully sought. This intensity was used to evoke H-reflex at rest (Hmax). Peripheral and spinal excitability were defined as the highest Mmax and Hmax recorded during the protocol, respectively. Hmax was normalized with respect to Mmax (Hmax/Mmax) to avoid potential bias due to peripheral excitability differences. Mmax and Hmax latencies were defined as the time between the stimulation onset and the evoked potential onset.

#### Evaluation of corticospinal excitability by transcranial magnetic stimulation

Single transcranial magnetic stimulation (TMS) pulses of 1-ms duration were delivered over the motor cortex using a Magstim 200 (Magstim Co., Whitland, UK). During the settings, TMS pulses were delivered during isometric submaximal voluntary contraction at 10% of the maximal quadriceps torque (facilitation). The figure-of-eight coil was held over the contralateral motor cortex at the optimum scalp position to elicit motor evoked potential (MEP) responses in the contralateral vastus medialis muscle. The contralateral motor cortex was first localized using the 10–10 EEG system (C3 point for right limb stimulation, C4 point for left limb stimulation). Then, the coil was moved by small amounts until the highest MEP response on the vastus medialis was obtained with suprathreshold stimuli, with the smallest possible response over the antagonist biceps femoris, in order to determine the optimal coil location. If significant activation of the antagonist biceps femoris muscle was noted, the coil was slightly moved, until its activation was minimized. Then markers were positioned over the participant and over the coil to maintain the coil location. After that, a recruitment curve was performed during voluntary contraction, at 10% of the maximal quadriceps torque, in order to determine the maximal intensity (noted IMep) [40]. One pulse was delivered every 10 s with increasing intensity in steps of 2% until the highest response was obtained. At least three pulses were delivered at each intensity level to check for reproducibility. The maximal intensity was defined as the intensity at which the highest MEP amplitude was obtained over the vastus medialis. This was then used during the protocol to elicit MEP responses during maximal voluntary contractions in order to assess corticospinal excitability and primary motor cortex activation. If a participant reached the maximum stimulator output without evidence of maximal MEP response (i.e., no evidence of plateau of the MEP amplitude before reaching the maximal output), the data were excluded from the analyses.

Corticospinal excitability was assessed during maximal voluntary contractions by the highest amplitude of the MEP induced by IMEP with respect to peripheral excitability (MEP/Mmax). The silent period duration was measured as the time between the MEP onset and the return of voluntary EMG activity. The central motor conduction time was calculated from the delay between stimulus artifact and the MEP onset.

#### Evaluation of voluntary activation with femoral nerve and transcranial magnetic stimulation

The voluntary activation was assessed by peripheral nerve stimulation (VAperipheral) and transcranial magnetic stimulation (VAcortical).

VAperipheral was calculated according to the twitch interpolation technique (4). A supramaximal doublet was delivered during the force plateau of the maximal voluntary contraction (superimposed doublet) and 2 s after relaxation (control doublet). VAperipheral was calculated as the ratio between the twitch-like increment in torque induced by the supramaximal doublet during maximal voluntary contraction and after relaxation:

$${\mathrm{VA}}_{\mathrm{cortical}}\left(\%\right)=\left[1-\left(\mathrm{superimposed}\ \mathrm{twitch}/\mathrm{estimated}\ \mathrm{resting}\ \mathrm{twitch}\right)\right]\times 100$$

VAcortical was calculated by stimulating the motor cortex during the quadriceps contractions according to the method described by Sidhu et al. [19]. The estimated resting twitch was calculated from the curve-response relationship obtained by plotting the twitch-like increment in torque induced by the transcranial magnetic pulses delivered during the last two maximal voluntary contractions, as well as those obtained during submaximal voluntary contractions at 30 and 50% of QMVC. When no linear relationship could be obtained between the voluntary force and the twitch-like increment in torque (r < 0.9), the data were excluded from the analyses [41]. VAcortical was calculated as the ratio between the highest twitch-like increment in torque induced by the TMS pulses during maximal voluntary contractions and the estimated resting twitch:

$${\mathrm{VA}}_{\mathrm{cortical}}\ \left(\%\right)=\left[1-\left(\mathrm{superimposed}\ \mathrm{twitch}/\mathrm{estimated}\ \mathrm{resting}\ \mathrm{twitch}\right)\right]\times 100$$

#### EMG activity

The surface EMG activity of the vastus medialis, rectus femoris and biceps femoris was recorded throughout the protocol with Biopac technology (Biopac MP100, Biopac Systems, Santa Barbara, CA, USA). Bipolar, silver chloride, square surface electrodes with a 9-mm diameter were used (Contrôle Graphique Médical, Brie-Compte-Robert, France). In order to minimize impedance (< 5 kΩ), the skin was shaved, abraded, and cleaned with alcohol. Two electrodes were set at the middle belly of the vastus medialis, rectus femoris and long head of the biceps femoris muscles of the dominant leg with an interelectrode distance of 2 cm. The reference electrode was placed on the opposite patella. The EMG signal was band-pass-filtered (10–500 Hz), amplified (× 1000) and recorded at a sample frequency of 4096 Hz.

The participants performed four maximal voluntary contractions of the knee extensors, each separated by 2 min of recovery (Fig. 2). They were asked to maintain maximal effort for at least 4 s. During the first two maximal voluntary contraction maneuvers, a double pulse at 100 Hz was delivered over the femoral nerve (superimposed doublet) during the force plateau and 2 s after relaxation (control doublet). During the last two maximal voluntary contraction maneuvers, a single TMS pulse at IMep was delivered over the motor cortex to elicit MEPs during the force plateau. Three single pulses at IMmax or Hmax intensity separated by 10 s were delivered twice between maximal voluntary contractions to elicit Mmax and Hmax at rest, respectively. The time interval between Mmax and Hmax stimuli was between 30 and 40s. If any pre-stimulus voluntary activity was observed, the involved stimuli were discarded. After the maximal voluntary contractions, three submaximal voluntary contractions (SVC) with visual feedback were performed at 50 and 30% of QMVC. A single TMS pulse at IMep was delivered during the force plateau of each SVC to elicit superimposed twitch responses at 30 and 50% of QMVC.

Experimental design. QMVC: Quadriceps voluntary contractions at maximal (100% of QMVC) or submaximal (50 and 30% of QMVC) intensity. Superimposed and control doublets, maximal M-waves (Mmax), and maximal H-waves (Hmax) were delivered via electrical stimulation over the femoral nerve. Motor evoked potentials (MEP) were delivered over the motor cortex via transcranial magnetic stimulation

All statistical analyses except slope comparisons were performed using Statistica software (StatSoft, Inc., version 6.0, Tulsa, OK, USA). All data were examined for normality using a Shapiro-Wilk test. Differences between the pooled COPD patients and the healthy controls were studied using unpaired t-tests for parametric data and the non-parametric Mann-Whitney U test otherwise. Differences between the COPDMW and COPDNoMW groups and healthy controls were tested using a one-way between-subject analysis of variance (ANOVA), unless when no data were available for healthy controls (i.e. blood gas analyses and comorbidities), with the 2 groups of patients compared using unpaired t-test instead. The underlying assumptions of ANOVA were checked using a Levene test (homogeneity of the variance). When the ANOVA F ratio was significant (p < 0.05), the means were compared by a Studentized Newman-Keuls (SNK) post-hoc test. Analysis of covariance (ANCOVA) was used with 1) QMVC as the criterion variable and QPt as the covariate (adjusted maximal voluntary strength), 2) VAcortical as the criterion variable and PaO2 as the covariate. Bivariate regression analyses were performed using the Pearson coefficient. The slopes and Y-intercepts of the relationships between QMVC and QPt were compared for differences between the three groups using a specific ANOVA procedure of the Statgraphics Centurion XVII statistical package. Data are presented mean ± standard error (SE).

This content was originally published here.