Neuroscience Letters 580 (2014) 169–172
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Trigemino-cervical reflex in spinal cord injury Ays¸egül Gündüz a,∗ , Nurten Uzun a , Nurettin I˙ rem Örnek b , Halil Ünalan b , S¸afak Sahir Karamehmeto˘glu b , Meral E. Kızıltan a a b
Department of Neurology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey Department of Physical Therapy and Rehabilitation, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey
h i g h l i g h t s • • • •
Simultaneous recordings over SCM and SC muscles are first among TCR studies. TCR probability is normal in patients with SCI who used no or low dose oral baclofen. There are findings suggesting enhanced TCR in SCI with no/low dose use of oral baclofen. TCR is not obtained in patients who are using oral baclofen more than 50 mg/day.
a r t i c l e
i n f o
Article history: Received 26 April 2014 Received in revised form 24 July 2014 Accepted 4 August 2014 Available online 12 August 2014 Keywords: Baclofen Brainstem reflexes Spasticity Spinal cord injury Spinal trauma Trigemino-cervical reflex
a b s t r a c t Abnormal enhancement of polysynaptic brainstem reflexes has been previously reported in patients with spinal cord injury (SCI). We aimed to investigate trigemino-cervical reflex (TCR) in SCI since it may reflect alterations in the connections of trigeminal proprioceptive system and cervical motoneurons. Consecutive 14 patients with SCI and 16 healthy subjects were included in this study. All patients were in the chronic phase. TCR was recorded over sternocleidomastoid (SCM) and splenius capitis (SC) muscles by stimulation of infraorbital nerve. We measured onset latency, amplitudes and durations of responses and compared between groups. We obtained stable responses over both muscles after one sided stimulation in healthy volunteers whereas probability of TCR was decreased in patients over both SCM (78.6% vs. 100%, p = 0.050) and SC (71.4% vs. 100%, p = 0.022). The absence of TCR was related to use of oral baclofen (≥50 mg/day). However, when present, responses of SCI group had higher amplitudes and were more persistent. We demonstrated that TCR probability was similar to healthy subjects in SCI patients who used no or low dose oral baclofen. But it had higher amplitudes and longer durations. It was not obtained in only two patients who used oral baclofen more than 50 mg/day. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Spinal cord injury (SCI) leads to functional impairment of almost all extremity and axial muscles below and sometimes at the level of the injury. After the injury, widespread reorganization at all levels including spinal cord, brainstem and cortex develops probably as a compensatory mechanism [1,2]. Alterations of the nervous system after the injury have previously been examined by means of
Abbreviations: ASR, auditory startle response; BR, blink reflex; EMG, electromyography; GABA, ␦-aminobutyric acid; MSA, multisystem atrophy; MSP, masseter silent period; PSP, progressive supranuclear palsy; SC, splenius capitis; SCI, spinal cord injury; SCM, sternocleidomastoid; SD, standard deviations; SSRIs, serotonine reuptake inhibitors; TCR, trigemino-cervical reflex. ∗ Corresponding author. Tel.: +0090 2124143162; fax: +0090 2126330176. E-mail address: [email protected] (A. Gündüz). http://dx.doi.org/10.1016/j.neulet.2014.08.006 0304-3940/© 2014 Elsevier Ireland Ltd. All rights reserved.
different electrophysiological methods. The brainstem circuits which were already examined include blink reflex (BR), jaw jerk, masseter silent period (MSP), and auditory startle response (ASR) [3]. Trigemino-cervical reflex (TCR) is a defensive withdrawal reaction of the head in response to facial stimuli. This reflex originates from the interaction between the trigeminal system and cervical spinal cord motoneurons, specifically those of the neck muscles, namely splenius capitis (SC) [4]. In experimental conditions, TCR is elicited by glabellar tap or electrical stimulation of supraorbital or infraorbital nerves [5–7]. Recordings are generally done from posterior neck muscles using surface or needle electrodes. However, similar responses are also shown to be obtained over sternocleidomastoid (SCM) [7,8] and to some extent over biceps brachii [4]. Its afferent pathway is the trigeminal system, cervical spinal cord (and some bulbar) motoneurons constitute the efferent pathway, and TCR is mediated by various inputs from vestibulospinal
170
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
and tectospinal tracts. Structures in paramedian pontine reticular formation are supposed to be the integrating center of the reflex and have the major role in the generation of TCR. Absence of TCR in progressive supranuclear palsy (PSP) while they were clearly detectable in all the parkinsonism type of multisystem atrophy (MSA-P) patients supports the presence of brainstem generator [9,10]. Baclofen is ␦-aminobutyric acid (GABA) receptor agonist and is commonly used for the treatment of spasticity in SCI. Its inhibitory role provides restoration of inhibitory responses clinically and baclofen restores brainstem reflexes which exaggerate after SCI [3,11] possibly operating at the interneuron level. Cervical motoneurons get inputs from the proprioceptive system extensively [12]. Neck muscles provide orientation of the head and positioning of the visual field during maintenance of posture and movement. They bear a particular role for the maintenance of the body posture along with other axial muscles which are also dense in muscle spindles and which are responsible for the body posture and movement [13,14]. We hypothesize that neck muscles are supposed to undergo changes in SCI to adopt new positioning on earth. As neck muscles are innervated by the most upper cervical spinal cord; they generally survive despite the loss of axial muscle functions. Therefore, in this study, we primarily aimed to investigate alterations of the long latency TCR circuit in patients with SCI. We then analyzed the effect of factors like types of medications, level of SCI and duration of SCI. 2. Subjects and method 2.1. Subjects Consecutive 14 patients with SCI who admitted for rehabilitation to the Department of Physical Therapy and Rehabilitation between June 2010 and June 2011 and 16 age- and sex-matched healthy subjects were included in this study. Mean ages of patients and healthy subjects were 39.0 ± 9.8 and 40.7 ± 12.1 years, respectively.
The filter settings were 3 kHz high cut and 20 Hz low cut. Analysis time was adjusted as 20 ms/div, and amplitude sensitivity was 200 V. Reflex responses were accepted when there was an evident response starting with a sharp negative deflection. We have evaluated long latency TCR. Onset latency, duration and amplitude of raw signals were measured using cursors in both groups. The reflex was identified as an electromyography (EMG) response with an amplitude at least 50 V greater than the background EMG activity. All the measurements were checked by visual inspection. Observation of one definite response was sufficient to categorize as present. The study was approved by the institutional review board, and informed consent was obtained. 3. Statistical analysis Data were pooled to obtain mean values and standard deviations (SD). We firstly compared presence of ipsilateral and contralateral responses over each muscle following right or left sided stimulations separately. We also grouped ipsilateral and contralateral responses regardless of the stimulation side and compared amplitude, onset latency and presence rate between patients and healthy subjects. Since distribution of these groups were non-homogenous, nonparametric, values of two groups were also compared using “Mann Whitney U test”. We compared each clinical finding like types of medications used, level of SCI and duration of disease between the patient groups with and without TCR. We also conducted another analysis for comparison of TCR probability between healthy subjects and SCI patients who were using no or under 50 mg/day baclofen. 4. Results 4.1. Clinical findings
We obtained the informations regarding level of the lesion, types of trauma, duration of SCI and all medications used at the time of investigation from the medical records. Level of the lesion was determined according to neurological examination and radiological findings.
All patients suffered from lesions which developed traumatically at the levels between spinal segments of T2–T12. Encountered traumas were falling off from height or traffic accidents. Duration of the disease ranged from 1 year to 21 years. Nine patients were using oral baclofen with doses between 10 and 90 mg, four were on selective serotonine reuptake inhibitors (SSRIs) and one patient was taking tizanidine. There were three patients using oral baclofen dose of ≥50 mg/day. They used neither intrathecal baclofen nor gabapentine/pregabaline.
2.3. Reflex recordings
4.2. Electrophysiological findings
All electrophysiological recordings were done with surface silver–silver chloride electrodes using Neuropack Sigma MEB5504k, Nihon Kohden Medical, Tokyo, Japan. All subjects were examined under the same conditions: sitting on their own wheelchair (for patients) or armchair (healthy volunteers), in a quiet room. They were asked to remain awake and relaxed. TCR was obtained by percutaneous electrical stimulation (with duration of 0.5 ms) of the infraorbital branch of trigeminal nerve. The intensity of the electrical shocks used for the electrophysiological measurements was adjusted to 3 times above the perceptive threshold and were regarded as painless. We have increased the stimulus intensity until 50 mA when we could not get a response. Recordings were repeated four times and obtained simultaneously over bilateral SCM and SC. The average of all four recordings were used for measurements. Surface cup electrodes were placed at the level of the C3 and C7 vertebrae for SC and 2 cm apart over the midbelly of the muscle for SCM. The ground electrode was set on the neck.
We obtained bilateral responses over both muscles with latencies between 40 and 66 ms in all healthy volunteers with quite stable latency in a given subject. Probability of long latency TCR in control group was 100%. SCM responses seemed polyphasic. Ipsilateral responses had a first positive peak and a second negative peak which was the vice versa for the contralateral responses. As seen in Fig. 1A, right infraorbital stimulation led to a response with first electropositive peak on right SCM and SC. Mean latencies of SCM and SC responses were between 49.4 and 55.3, respectively. Table 1 shows mean latencies, standard deviations and p values. Probability of long latency TCR was reduced in SCI group. There were bilateral responses over SCM in 11 (78.6%) and over SC in 10 (71.4%) patients which were both significantly low compared to healthy subjects (pSCM = 0.05 and pSC = 0.022). Although configurations and mean latencies were quite similar to that seen in healthy subjects, responses of SCI group seemed to have higher amplitudes and to be more persistent (Table 1). Fig. 1 shows TCR
2.2. Clinical findings
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
171
Fig. 1. Trigemino-cervical reflex responses after right infraorbital stimulation in a representative control subject (A) and after left infraorbital stimulation in a 48 year-old male patient with spinal cord injury for 1.5 years (who did not use baclofen) (B). (R, right; L, left; SCM, sternocleidomastoid; SC, splenius capitis)
responses evoked in a representative control subject and in a patient with SCI. We analyzed the factors which led to the loss of long latency TCR and it seemed to be associated with the use of oral baclofen and TCR was specifically absent in patients using baclofen over 50 mg (p = 0.044). Duration of the disease, level of trauma or use of antidepressant drugs and/or tizanidine were not associated. Consequently, we conducted another analysis after excluding patients using oral baclofen therapy higher than 50 mg/day. In this analysis, the probability of reflex was similar to those of healthy subjects (90.9 in SCI patients vs. 100.0 in healthy individuals, p = 0.458). 5. Discussion In this study, we have three main findings: (i) long latency TCR probability was similar to healthy subjects in SCI patients who used no or low dose oral baclofen and it had higher amplitudes and longer durations, (ii) it was not obtained in patients who used oral baclofen more than 50 mg/day, (iii) responses over SCM and SC showed similar changes in SCI. As far as, we know this is the first TCR study which used recordings simultaneously over both SCM and SC muscles providing advantage of interpreting results of these muscles together. In healthy subjects and SCI patients, SCM and SC were triggered at the same intensities of stimulation and the latencies of both muscles were similar. TCR demonstrates the existence functional relationship between the trigeminal system and cervical motoneurons and to some extent bulbar motor nuclei. It is mediated by a pathway Table 1 Comparison of ipsilateral and contralateral TCR responses between patients and healthy subjects. Patients with SCI n = 14 IL SCM latency IL SCM amplitude IL SCM duration CL SCM latency CL SCM amplitude CL SCM duration IL SC latency IL SC amplitude IL SC duration CL SC latency CL SC amplitude CL SC duration
54.3 586.7 82.4 52.6 481.7 91.4 47.7 372.3 66.3 49.2 324.0 70.2
± ± ± ± ± ± ± ± ± ± ± ±
12.7 537.0 36.9 11.7 368.2 38.7 6.8 226.7 25.2 6.1 215.2 35.7
Control n = 16 55.3 402.7 67.4 54.3 410.9 68.5 49.4 223.1 49.1 49.8 230.9 46.9
SCI—spinal cord injury; IL—ipsilateral; SCM—sternocleidomastoid; SC—splenius capitis.
± ± ± ± ± ± ± ± ± ± ± ±
9.3 388.1 22.1 8.2 326.5 29.1 8.6 148.1 8.4 8.1 185.9 8.9
p 0.376 0.100 0.162 0.343 0.406 0.012 0.565 0.014 0.006 0.992 0.023 0.001
CL—contralateral;
comprising the trigeminal nerve, trigeminal spinal tract which projects to motor neurons of SCM and posterior neck muscles. It has two components: short latency response and long latency response [6]. Short latency response is suggested to be mediated by oligo-synaptic pathways and accepted as the correspondence of inhibitory phase of head retraction reflex. The long latency response, on the other hand, is proposed to be a polysynaptic reflex corresponding to the excitatory phase of head retraction reflex. Previously, TCR was shown to be suppressed in all patients with cervicobulbar lesions [15]. It was also absent in PSP while it was clearly detectable in all MSA-P patients [9,10]. This latter finding may be related to the cholinergic neuronal degeneration of the paramedian pontine reticular formation and atrophy of the gigantocellular reticular nucleus and the intermediate reticular zone present in PSP but absent in MSA [9]. Therefore, structures in this level are probably the integrating center of the reflex and have the major role in the generation of TCR. On the other hand, TCR was found to be enhanced in migraine patients [16]. Although the latter study was performed using TCR recovery curve, remarkably and significantly increased amplitudes and more persistent responses may also reflect excitability at some degree. SCI results in extensive functional changes at all levels. Normally, propriospinal system has a specific function in connecting spinal cord segments with each other or with cortical and brainstem structures. A population of propriospinal system neurons located in upper cervical segments (C3 and C4) was suggested to have a critical importance for motor tasks of all limbs as they transmitted corticospinal input and converge inputs from the rubro-, tecto-, and reticulospinal tracts, to lower cervical and even lumbar motoneurons [17]. In any case of incomplete SCI, even with complete transection of corticospinal tract, axons of propriospinal neurons sprout and provide a bridge and an indirect contact between motor neurons. This remodeling forms the basis for functional regain [18]. Increased excitability of the brainstem circuitry in SCI which was demonstrated by Kumru and colleagues [3] is probably secondary to this plasticity. They used BR recovery curve and ASR; however, TCR is a postural reflex and has close functional proximity to C3 and C4 propriospinal system. TCR is not accepted as a part of startle reaction [4] since it did not share any characteristics of the startle response and did not readily habituate, either. Although the evidence is against the functional synergy of TCR and startle reflex, they were hypothesized to share anatomical neuronal pathways due to similar latencies and similar cranio-caudal progression. The most possible candidates in this regard were interneurons located in the reticular formation and the reticulospinal pathways [4,19].
172
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
To assume upright position and maintain balance, signals of visual, vestibular and propriceptive systems are processed to make adjustments of eyes, body, head and limbs. The most important in controlling body position are lateral vestibulospinal reflex and pontine reticulospinal reflex. They promote the development of postural reflexes. Postural reflexes are also under control of motor cortex via medullary reticulospinal pathway and this pathway exerts inhibition on segmental spinal reflexes by way of inhibitory interneurons. The coordinated movement of head and eyes is also essential for equilibrium and depends on functions of connections between vestibular nucleus, superior colliculus, and interstitial nucleus of Cajal and medial longitudinal fasciculus which ends on axons of cervical segments [12]. We suggest that TCR reflects exaggeration of upper cervical propriospinal system after their reliefs from inhibitory neurons in SCI. These changes are probably secondary to adopt alterations on the earth. As patients have difficulties to change position and to gather information about positioning on earth, information coming from head and neck region and positioning of head bears importance in adopting new positioning. Inhibitory mechanisms lose strength not only by levels of neurotransmitters but also by way of some intracellular changes [20]. However, intrathecal baclofen which is a GABA agonist is effective in restoring inhibitory mechanism both clinically and electrophysiologically [3,11] possibly operating at the interneuron level. None of our patients were using intrathecal baclofen; however, moderate to high doses of oral baclofen was able to suppress TCR supporting our hypothesis. GABA is used at the interneuron level to modulate excitation alpha motor neuron. Therefore, inhibition of motor response after administration of GABA agonist is not surprising. Brainstem regions like pedinculopontine nucleus receive GABAerjic input for modulation of serotonergic and dopaminergic activities in animal studies [21]. Baclofen was found to have dual effects on activities of these transmitters which changed according to its dose [21–23]. Thus, it seems reasonable that low and high doses of baclofen have different effects both clinically and electrophysiologically. Since patients suffered from lesions which developed traumatically at the spinal segments between T2 and T12, presence or hyperexcitability of TCR was independent from the level or duration of SCI. However, cervical lesions may cause more heterogeneous results due to the localization of related motor neuron cell. 6. Study limitations There are limitations of our study. We aimed to observe responses over a wider cervical region to analyze the functional changes of cervical muscles after SCI. However, this created a disadvantage of averaging fewer responses. Previous studies generally recorded only SCM or SC responses and were able to average at least 10–15 responses [5–8]. Nevertheless, we were able to obtain evident late responses similar to those reported in the literature. We did not aim to analyze the early responses. However, examining only late responses is also used in literature [9,19]. Second, in the patients who were using more than 50 mg/day oral baclofen and had absent TCR, recording TCR could be performed after cessation of baclofen use or reduction of its dosage. However, we could not investigate because baclofen dosages were not changed during the study period. 7. Conclusion In conclusion, this is the first study investigating TCR over both SCM and SC in SCI and showing not only intrathecal but also oral baclofen inhibits the pathways mediating trigeminal reflexes.
Conflict of interest statement Authors report no conflict of interest. Preliminary results were previously presented as poster in 5th International Meeting of the Brainstem Society, London, United Kingdom, December 09–10, 2010. References [1] B. Calancie, N. Alexeeva, J.G. Broton, S. Suys, A. Hall, K.J. Klose, Distribution and latency of muscle responses to transcranial magnetic stimulation of motor cortex after spinal cord injury in humans, J. Neurotrauma 16 (1999) 49–67. [2] R.K. Shields, S. Dudley-Javoroski, P.D. Oza, Low-frequency H-reflex depression in trained human soleus after spinal cord injury, Neurosci. Lett. 499 (2011) 88–92. [3] H. Kumru, M. Kofler, J. Valls-Solé, E. Portell, J. Vidal, Brainstem reflexes are enhanced following severe spinal cord injury and reduced by continuous intrathecal baclofen, Neurorehabil. Neural Repair 23 (2009) 921–927. [4] M. Serrao, P. Rossi, L. Parisi, A. Perrotta, M. Bartolo, P. Cardinali, G. Amabile, F. Pierelli, Trigemino-cervical-spinal reflexes in humans, Clin. Neurophysiol. 114 (2003) 1697–1703. [5] V. Di Lazzaro, D. Restuccia, R. Nardone, T. Tartaglione, A. Quartarone, P. Tonali, J.C. Rothwell, Preliminary clinical observations on a new trigeminal reflex: the trigemino-cervical reflex, Neurology 46 (1996) 479–485. [6] C. Ertekin, N. Celebisoy, B. Uluda˘g, Trigeminocervical reflexes elicited by stimulation of the infraorbital nerve: head retraction reflex, J Clin. Neurophysiol. 18 (2001) 378–385. [7] F. Sartucci, A. Rossi, B. Rossi, Trigemino cervical reflex in man, Electromyogr. Clin. Neurophysiol. 26 (1986) 123–129. [8] V. Di Lazzaro, A. Quartarone, K. Higuchi, J.C. Rothwell, Short-latency trigeminocervical reflexes in man, Exp. Brain Res. 102 (1995) 474–482. [9] M. Serrao, R. Di Fabio, M. Bartolo, A. Perrotta, C. Tassorelli, G. Coppola, C. Davassi, L. Padua, G. Sandrini, F. Pierelli, The contribution of trigemino-cervical reflexes in distinguishing progressive supranuclear palsy from multiple system atrophy, Clin. Neurophysiol. 122 (2011) 1812–1815. [10] M. Bartolo, M. Serrao, A. Perrotta, C. Tassorelli, G. Sandrini, F. Pierelli, Lack of trigemino-cervical reflexes in progressive supranuclear palsy, Mov. Disord. 23 (2008) 1475–1479. [11] H. Kumru, J. Vidal, M. Kofler, E. Portell, J. Valls-Solé, Alterations in excitatory and inhibitory brainstem interneuronal circuits after severe spinal cord injury, J. Neurotrauma 27 (2010) 721–728. [12] E.E. Benarroch, Spinal and brainstem mechanism of motor control, in: E.E. Benarroch (Ed.), Basic Neurosciences with Clinical Application, Butterworth Heinemann Elsevier, Philadelphia, PA, 2006, pp. 571–626. [13] L.C. Boyd-Clark, C.A. Briggs, M.P. Galea, Muscle spindle distribution, morphology, and density in longus colli and multifidus muscles of the cervical spine, Spine (Phila Pa 1976) 27 (2002) 694–701. [14] V. Gurfinkel, T.W. Cacciatore, P. Cordo, F. Horak, J. Nutt, R. Skoss, Postural muscle tone in the body axis of healthy humans, J. Neurophysiol. 96 (2006) 2678–2687. [15] B. Rossi, S.L. Pasca, F. Sartucci, G. Siciliano, L. Murri, Trigemino-cervical reflex in pathology of the brainstem and of the first cervical cord segments, Electromyogr. Clin. Neurophysiol. 29 (1989) 67–71. [16] M. Serrao, A. Perrotta, M. Bartolo, G. Fiermonte, F. Pauri, P. Rossi, L. Parisi, F. Pierelli, Enhanced trigemino-cervical-spinal reflex recovery cycle in pain-free migraineurs, Headache 45 (2005) 1061–1068. [17] J.R. Flynn, B.A. Graham, M.P. Galea, R.J. Callister, The role of propriospinal interneurons in recovery from spinal cord injury, Neuropharmacology 60 (2011) 809–822. [18] F.M. Bareyre, M. Kerschensteiner, O. Raineteau, T.C. Mettenleiter, O. Weinmann, M.E. Schwab, The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats, Nat. Neurosci. 7 (2004) 269–277. [19] A. Perrotta, M. Serrao, M. Bartolo, L. Valletta, N. Locuratolo, F. Pujia, F. Fattapposta, P. Bramanti, G.A. Amabile, F. Pierelli, L. Parisi, Abnormal head nociceptive withdrawal reaction to facial nociceptive stimuli in Parkinson’s disease, Clin. Neurophysiol. 116 (2005) 2091–2098. [20] P. Boulenguez, L. Vinay, Strategies to restore motor functions after spinal cord injury, Curr. Opin. Neurobiol. 19 (2009) 587–600. [21] K. Takakusaki, K. Obara, T. Nozu, T. Okumura, Modulatory effects of the GABAergic basal ganglia neurons on the PPN and the muscle tone inhibitory system in cats, Arch. Ital. Biol. 149 (2011) 385–405. [22] G. Labouèbe, M. Lomazzi, H.G. Cruz, C. Creton, R. Luján, M. Li, Y. Yanagawa, K. Obata, M. Watanabe, K. Wickman, S.B. Boyer, P.A. Slesinger, C. Lüscher, RGS2 modulates coupling between GABAB receptors and GIRK channels in dopamine neurons of the ventral tegmental area, Nat. Neurosci. 10 (2007) 1559–1568. [23] A. Takahashi, A. Shimamoto, C.O. Boyson, J.F. Debold, K.A. Miczek, Gaba(b) receptor modulation of serotonin neurons in the dorsal raphé nucleus and escalation of aggression in mice, J. Neurosci. 30 (2010) 11771–11780.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
Trigemino-cervical reflex in spinal cord injury Ays¸egül Gündüz a,∗ , Nurten Uzun a , Nurettin I˙ rem Örnek b , Halil Ünalan b , S¸afak Sahir Karamehmeto˘glu b , Meral E. Kızıltan a a b
Department of Neurology, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey Department of Physical Therapy and Rehabilitation, Cerrahpasa School of Medicine, Istanbul University, Istanbul, Turkey
h i g h l i g h t s • • • •
Simultaneous recordings over SCM and SC muscles are first among TCR studies. TCR probability is normal in patients with SCI who used no or low dose oral baclofen. There are findings suggesting enhanced TCR in SCI with no/low dose use of oral baclofen. TCR is not obtained in patients who are using oral baclofen more than 50 mg/day.
a r t i c l e
i n f o
Article history: Received 26 April 2014 Received in revised form 24 July 2014 Accepted 4 August 2014 Available online 12 August 2014 Keywords: Baclofen Brainstem reflexes Spasticity Spinal cord injury Spinal trauma Trigemino-cervical reflex
a b s t r a c t Abnormal enhancement of polysynaptic brainstem reflexes has been previously reported in patients with spinal cord injury (SCI). We aimed to investigate trigemino-cervical reflex (TCR) in SCI since it may reflect alterations in the connections of trigeminal proprioceptive system and cervical motoneurons. Consecutive 14 patients with SCI and 16 healthy subjects were included in this study. All patients were in the chronic phase. TCR was recorded over sternocleidomastoid (SCM) and splenius capitis (SC) muscles by stimulation of infraorbital nerve. We measured onset latency, amplitudes and durations of responses and compared between groups. We obtained stable responses over both muscles after one sided stimulation in healthy volunteers whereas probability of TCR was decreased in patients over both SCM (78.6% vs. 100%, p = 0.050) and SC (71.4% vs. 100%, p = 0.022). The absence of TCR was related to use of oral baclofen (≥50 mg/day). However, when present, responses of SCI group had higher amplitudes and were more persistent. We demonstrated that TCR probability was similar to healthy subjects in SCI patients who used no or low dose oral baclofen. But it had higher amplitudes and longer durations. It was not obtained in only two patients who used oral baclofen more than 50 mg/day. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Spinal cord injury (SCI) leads to functional impairment of almost all extremity and axial muscles below and sometimes at the level of the injury. After the injury, widespread reorganization at all levels including spinal cord, brainstem and cortex develops probably as a compensatory mechanism [1,2]. Alterations of the nervous system after the injury have previously been examined by means of
Abbreviations: ASR, auditory startle response; BR, blink reflex; EMG, electromyography; GABA, ␦-aminobutyric acid; MSA, multisystem atrophy; MSP, masseter silent period; PSP, progressive supranuclear palsy; SC, splenius capitis; SCI, spinal cord injury; SCM, sternocleidomastoid; SD, standard deviations; SSRIs, serotonine reuptake inhibitors; TCR, trigemino-cervical reflex. ∗ Corresponding author. Tel.: +0090 2124143162; fax: +0090 2126330176. E-mail address: [email protected] (A. Gündüz). http://dx.doi.org/10.1016/j.neulet.2014.08.006 0304-3940/© 2014 Elsevier Ireland Ltd. All rights reserved.
different electrophysiological methods. The brainstem circuits which were already examined include blink reflex (BR), jaw jerk, masseter silent period (MSP), and auditory startle response (ASR) [3]. Trigemino-cervical reflex (TCR) is a defensive withdrawal reaction of the head in response to facial stimuli. This reflex originates from the interaction between the trigeminal system and cervical spinal cord motoneurons, specifically those of the neck muscles, namely splenius capitis (SC) [4]. In experimental conditions, TCR is elicited by glabellar tap or electrical stimulation of supraorbital or infraorbital nerves [5–7]. Recordings are generally done from posterior neck muscles using surface or needle electrodes. However, similar responses are also shown to be obtained over sternocleidomastoid (SCM) [7,8] and to some extent over biceps brachii [4]. Its afferent pathway is the trigeminal system, cervical spinal cord (and some bulbar) motoneurons constitute the efferent pathway, and TCR is mediated by various inputs from vestibulospinal
170
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
and tectospinal tracts. Structures in paramedian pontine reticular formation are supposed to be the integrating center of the reflex and have the major role in the generation of TCR. Absence of TCR in progressive supranuclear palsy (PSP) while they were clearly detectable in all the parkinsonism type of multisystem atrophy (MSA-P) patients supports the presence of brainstem generator [9,10]. Baclofen is ␦-aminobutyric acid (GABA) receptor agonist and is commonly used for the treatment of spasticity in SCI. Its inhibitory role provides restoration of inhibitory responses clinically and baclofen restores brainstem reflexes which exaggerate after SCI [3,11] possibly operating at the interneuron level. Cervical motoneurons get inputs from the proprioceptive system extensively [12]. Neck muscles provide orientation of the head and positioning of the visual field during maintenance of posture and movement. They bear a particular role for the maintenance of the body posture along with other axial muscles which are also dense in muscle spindles and which are responsible for the body posture and movement [13,14]. We hypothesize that neck muscles are supposed to undergo changes in SCI to adopt new positioning on earth. As neck muscles are innervated by the most upper cervical spinal cord; they generally survive despite the loss of axial muscle functions. Therefore, in this study, we primarily aimed to investigate alterations of the long latency TCR circuit in patients with SCI. We then analyzed the effect of factors like types of medications, level of SCI and duration of SCI. 2. Subjects and method 2.1. Subjects Consecutive 14 patients with SCI who admitted for rehabilitation to the Department of Physical Therapy and Rehabilitation between June 2010 and June 2011 and 16 age- and sex-matched healthy subjects were included in this study. Mean ages of patients and healthy subjects were 39.0 ± 9.8 and 40.7 ± 12.1 years, respectively.
The filter settings were 3 kHz high cut and 20 Hz low cut. Analysis time was adjusted as 20 ms/div, and amplitude sensitivity was 200 V. Reflex responses were accepted when there was an evident response starting with a sharp negative deflection. We have evaluated long latency TCR. Onset latency, duration and amplitude of raw signals were measured using cursors in both groups. The reflex was identified as an electromyography (EMG) response with an amplitude at least 50 V greater than the background EMG activity. All the measurements were checked by visual inspection. Observation of one definite response was sufficient to categorize as present. The study was approved by the institutional review board, and informed consent was obtained. 3. Statistical analysis Data were pooled to obtain mean values and standard deviations (SD). We firstly compared presence of ipsilateral and contralateral responses over each muscle following right or left sided stimulations separately. We also grouped ipsilateral and contralateral responses regardless of the stimulation side and compared amplitude, onset latency and presence rate between patients and healthy subjects. Since distribution of these groups were non-homogenous, nonparametric, values of two groups were also compared using “Mann Whitney U test”. We compared each clinical finding like types of medications used, level of SCI and duration of disease between the patient groups with and without TCR. We also conducted another analysis for comparison of TCR probability between healthy subjects and SCI patients who were using no or under 50 mg/day baclofen. 4. Results 4.1. Clinical findings
We obtained the informations regarding level of the lesion, types of trauma, duration of SCI and all medications used at the time of investigation from the medical records. Level of the lesion was determined according to neurological examination and radiological findings.
All patients suffered from lesions which developed traumatically at the levels between spinal segments of T2–T12. Encountered traumas were falling off from height or traffic accidents. Duration of the disease ranged from 1 year to 21 years. Nine patients were using oral baclofen with doses between 10 and 90 mg, four were on selective serotonine reuptake inhibitors (SSRIs) and one patient was taking tizanidine. There were three patients using oral baclofen dose of ≥50 mg/day. They used neither intrathecal baclofen nor gabapentine/pregabaline.
2.3. Reflex recordings
4.2. Electrophysiological findings
All electrophysiological recordings were done with surface silver–silver chloride electrodes using Neuropack Sigma MEB5504k, Nihon Kohden Medical, Tokyo, Japan. All subjects were examined under the same conditions: sitting on their own wheelchair (for patients) or armchair (healthy volunteers), in a quiet room. They were asked to remain awake and relaxed. TCR was obtained by percutaneous electrical stimulation (with duration of 0.5 ms) of the infraorbital branch of trigeminal nerve. The intensity of the electrical shocks used for the electrophysiological measurements was adjusted to 3 times above the perceptive threshold and were regarded as painless. We have increased the stimulus intensity until 50 mA when we could not get a response. Recordings were repeated four times and obtained simultaneously over bilateral SCM and SC. The average of all four recordings were used for measurements. Surface cup electrodes were placed at the level of the C3 and C7 vertebrae for SC and 2 cm apart over the midbelly of the muscle for SCM. The ground electrode was set on the neck.
We obtained bilateral responses over both muscles with latencies between 40 and 66 ms in all healthy volunteers with quite stable latency in a given subject. Probability of long latency TCR in control group was 100%. SCM responses seemed polyphasic. Ipsilateral responses had a first positive peak and a second negative peak which was the vice versa for the contralateral responses. As seen in Fig. 1A, right infraorbital stimulation led to a response with first electropositive peak on right SCM and SC. Mean latencies of SCM and SC responses were between 49.4 and 55.3, respectively. Table 1 shows mean latencies, standard deviations and p values. Probability of long latency TCR was reduced in SCI group. There were bilateral responses over SCM in 11 (78.6%) and over SC in 10 (71.4%) patients which were both significantly low compared to healthy subjects (pSCM = 0.05 and pSC = 0.022). Although configurations and mean latencies were quite similar to that seen in healthy subjects, responses of SCI group seemed to have higher amplitudes and to be more persistent (Table 1). Fig. 1 shows TCR
2.2. Clinical findings
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
171
Fig. 1. Trigemino-cervical reflex responses after right infraorbital stimulation in a representative control subject (A) and after left infraorbital stimulation in a 48 year-old male patient with spinal cord injury for 1.5 years (who did not use baclofen) (B). (R, right; L, left; SCM, sternocleidomastoid; SC, splenius capitis)
responses evoked in a representative control subject and in a patient with SCI. We analyzed the factors which led to the loss of long latency TCR and it seemed to be associated with the use of oral baclofen and TCR was specifically absent in patients using baclofen over 50 mg (p = 0.044). Duration of the disease, level of trauma or use of antidepressant drugs and/or tizanidine were not associated. Consequently, we conducted another analysis after excluding patients using oral baclofen therapy higher than 50 mg/day. In this analysis, the probability of reflex was similar to those of healthy subjects (90.9 in SCI patients vs. 100.0 in healthy individuals, p = 0.458). 5. Discussion In this study, we have three main findings: (i) long latency TCR probability was similar to healthy subjects in SCI patients who used no or low dose oral baclofen and it had higher amplitudes and longer durations, (ii) it was not obtained in patients who used oral baclofen more than 50 mg/day, (iii) responses over SCM and SC showed similar changes in SCI. As far as, we know this is the first TCR study which used recordings simultaneously over both SCM and SC muscles providing advantage of interpreting results of these muscles together. In healthy subjects and SCI patients, SCM and SC were triggered at the same intensities of stimulation and the latencies of both muscles were similar. TCR demonstrates the existence functional relationship between the trigeminal system and cervical motoneurons and to some extent bulbar motor nuclei. It is mediated by a pathway Table 1 Comparison of ipsilateral and contralateral TCR responses between patients and healthy subjects. Patients with SCI n = 14 IL SCM latency IL SCM amplitude IL SCM duration CL SCM latency CL SCM amplitude CL SCM duration IL SC latency IL SC amplitude IL SC duration CL SC latency CL SC amplitude CL SC duration
54.3 586.7 82.4 52.6 481.7 91.4 47.7 372.3 66.3 49.2 324.0 70.2
± ± ± ± ± ± ± ± ± ± ± ±
12.7 537.0 36.9 11.7 368.2 38.7 6.8 226.7 25.2 6.1 215.2 35.7
Control n = 16 55.3 402.7 67.4 54.3 410.9 68.5 49.4 223.1 49.1 49.8 230.9 46.9
SCI—spinal cord injury; IL—ipsilateral; SCM—sternocleidomastoid; SC—splenius capitis.
± ± ± ± ± ± ± ± ± ± ± ±
9.3 388.1 22.1 8.2 326.5 29.1 8.6 148.1 8.4 8.1 185.9 8.9
p 0.376 0.100 0.162 0.343 0.406 0.012 0.565 0.014 0.006 0.992 0.023 0.001
CL—contralateral;
comprising the trigeminal nerve, trigeminal spinal tract which projects to motor neurons of SCM and posterior neck muscles. It has two components: short latency response and long latency response [6]. Short latency response is suggested to be mediated by oligo-synaptic pathways and accepted as the correspondence of inhibitory phase of head retraction reflex. The long latency response, on the other hand, is proposed to be a polysynaptic reflex corresponding to the excitatory phase of head retraction reflex. Previously, TCR was shown to be suppressed in all patients with cervicobulbar lesions [15]. It was also absent in PSP while it was clearly detectable in all MSA-P patients [9,10]. This latter finding may be related to the cholinergic neuronal degeneration of the paramedian pontine reticular formation and atrophy of the gigantocellular reticular nucleus and the intermediate reticular zone present in PSP but absent in MSA [9]. Therefore, structures in this level are probably the integrating center of the reflex and have the major role in the generation of TCR. On the other hand, TCR was found to be enhanced in migraine patients [16]. Although the latter study was performed using TCR recovery curve, remarkably and significantly increased amplitudes and more persistent responses may also reflect excitability at some degree. SCI results in extensive functional changes at all levels. Normally, propriospinal system has a specific function in connecting spinal cord segments with each other or with cortical and brainstem structures. A population of propriospinal system neurons located in upper cervical segments (C3 and C4) was suggested to have a critical importance for motor tasks of all limbs as they transmitted corticospinal input and converge inputs from the rubro-, tecto-, and reticulospinal tracts, to lower cervical and even lumbar motoneurons [17]. In any case of incomplete SCI, even with complete transection of corticospinal tract, axons of propriospinal neurons sprout and provide a bridge and an indirect contact between motor neurons. This remodeling forms the basis for functional regain [18]. Increased excitability of the brainstem circuitry in SCI which was demonstrated by Kumru and colleagues [3] is probably secondary to this plasticity. They used BR recovery curve and ASR; however, TCR is a postural reflex and has close functional proximity to C3 and C4 propriospinal system. TCR is not accepted as a part of startle reaction [4] since it did not share any characteristics of the startle response and did not readily habituate, either. Although the evidence is against the functional synergy of TCR and startle reflex, they were hypothesized to share anatomical neuronal pathways due to similar latencies and similar cranio-caudal progression. The most possible candidates in this regard were interneurons located in the reticular formation and the reticulospinal pathways [4,19].
172
A. Gündüz et al. / Neuroscience Letters 580 (2014) 169–172
To assume upright position and maintain balance, signals of visual, vestibular and propriceptive systems are processed to make adjustments of eyes, body, head and limbs. The most important in controlling body position are lateral vestibulospinal reflex and pontine reticulospinal reflex. They promote the development of postural reflexes. Postural reflexes are also under control of motor cortex via medullary reticulospinal pathway and this pathway exerts inhibition on segmental spinal reflexes by way of inhibitory interneurons. The coordinated movement of head and eyes is also essential for equilibrium and depends on functions of connections between vestibular nucleus, superior colliculus, and interstitial nucleus of Cajal and medial longitudinal fasciculus which ends on axons of cervical segments [12]. We suggest that TCR reflects exaggeration of upper cervical propriospinal system after their reliefs from inhibitory neurons in SCI. These changes are probably secondary to adopt alterations on the earth. As patients have difficulties to change position and to gather information about positioning on earth, information coming from head and neck region and positioning of head bears importance in adopting new positioning. Inhibitory mechanisms lose strength not only by levels of neurotransmitters but also by way of some intracellular changes [20]. However, intrathecal baclofen which is a GABA agonist is effective in restoring inhibitory mechanism both clinically and electrophysiologically [3,11] possibly operating at the interneuron level. None of our patients were using intrathecal baclofen; however, moderate to high doses of oral baclofen was able to suppress TCR supporting our hypothesis. GABA is used at the interneuron level to modulate excitation alpha motor neuron. Therefore, inhibition of motor response after administration of GABA agonist is not surprising. Brainstem regions like pedinculopontine nucleus receive GABAerjic input for modulation of serotonergic and dopaminergic activities in animal studies [21]. Baclofen was found to have dual effects on activities of these transmitters which changed according to its dose [21–23]. Thus, it seems reasonable that low and high doses of baclofen have different effects both clinically and electrophysiologically. Since patients suffered from lesions which developed traumatically at the spinal segments between T2 and T12, presence or hyperexcitability of TCR was independent from the level or duration of SCI. However, cervical lesions may cause more heterogeneous results due to the localization of related motor neuron cell. 6. Study limitations There are limitations of our study. We aimed to observe responses over a wider cervical region to analyze the functional changes of cervical muscles after SCI. However, this created a disadvantage of averaging fewer responses. Previous studies generally recorded only SCM or SC responses and were able to average at least 10–15 responses [5–8]. Nevertheless, we were able to obtain evident late responses similar to those reported in the literature. We did not aim to analyze the early responses. However, examining only late responses is also used in literature [9,19]. Second, in the patients who were using more than 50 mg/day oral baclofen and had absent TCR, recording TCR could be performed after cessation of baclofen use or reduction of its dosage. However, we could not investigate because baclofen dosages were not changed during the study period. 7. Conclusion In conclusion, this is the first study investigating TCR over both SCM and SC in SCI and showing not only intrathecal but also oral baclofen inhibits the pathways mediating trigeminal reflexes.
Conflict of interest statement Authors report no conflict of interest. Preliminary results were previously presented as poster in 5th International Meeting of the Brainstem Society, London, United Kingdom, December 09–10, 2010. References [1] B. Calancie, N. Alexeeva, J.G. Broton, S. Suys, A. Hall, K.J. Klose, Distribution and latency of muscle responses to transcranial magnetic stimulation of motor cortex after spinal cord injury in humans, J. Neurotrauma 16 (1999) 49–67. [2] R.K. Shields, S. Dudley-Javoroski, P.D. Oza, Low-frequency H-reflex depression in trained human soleus after spinal cord injury, Neurosci. Lett. 499 (2011) 88–92. [3] H. Kumru, M. Kofler, J. Valls-Solé, E. Portell, J. Vidal, Brainstem reflexes are enhanced following severe spinal cord injury and reduced by continuous intrathecal baclofen, Neurorehabil. Neural Repair 23 (2009) 921–927. [4] M. Serrao, P. Rossi, L. Parisi, A. Perrotta, M. Bartolo, P. Cardinali, G. Amabile, F. Pierelli, Trigemino-cervical-spinal reflexes in humans, Clin. Neurophysiol. 114 (2003) 1697–1703. [5] V. Di Lazzaro, D. Restuccia, R. Nardone, T. Tartaglione, A. Quartarone, P. Tonali, J.C. Rothwell, Preliminary clinical observations on a new trigeminal reflex: the trigemino-cervical reflex, Neurology 46 (1996) 479–485. [6] C. Ertekin, N. Celebisoy, B. Uluda˘g, Trigeminocervical reflexes elicited by stimulation of the infraorbital nerve: head retraction reflex, J Clin. Neurophysiol. 18 (2001) 378–385. [7] F. Sartucci, A. Rossi, B. Rossi, Trigemino cervical reflex in man, Electromyogr. Clin. Neurophysiol. 26 (1986) 123–129. [8] V. Di Lazzaro, A. Quartarone, K. Higuchi, J.C. Rothwell, Short-latency trigeminocervical reflexes in man, Exp. Brain Res. 102 (1995) 474–482. [9] M. Serrao, R. Di Fabio, M. Bartolo, A. Perrotta, C. Tassorelli, G. Coppola, C. Davassi, L. Padua, G. Sandrini, F. Pierelli, The contribution of trigemino-cervical reflexes in distinguishing progressive supranuclear palsy from multiple system atrophy, Clin. Neurophysiol. 122 (2011) 1812–1815. [10] M. Bartolo, M. Serrao, A. Perrotta, C. Tassorelli, G. Sandrini, F. Pierelli, Lack of trigemino-cervical reflexes in progressive supranuclear palsy, Mov. Disord. 23 (2008) 1475–1479. [11] H. Kumru, J. Vidal, M. Kofler, E. Portell, J. Valls-Solé, Alterations in excitatory and inhibitory brainstem interneuronal circuits after severe spinal cord injury, J. Neurotrauma 27 (2010) 721–728. [12] E.E. Benarroch, Spinal and brainstem mechanism of motor control, in: E.E. Benarroch (Ed.), Basic Neurosciences with Clinical Application, Butterworth Heinemann Elsevier, Philadelphia, PA, 2006, pp. 571–626. [13] L.C. Boyd-Clark, C.A. Briggs, M.P. Galea, Muscle spindle distribution, morphology, and density in longus colli and multifidus muscles of the cervical spine, Spine (Phila Pa 1976) 27 (2002) 694–701. [14] V. Gurfinkel, T.W. Cacciatore, P. Cordo, F. Horak, J. Nutt, R. Skoss, Postural muscle tone in the body axis of healthy humans, J. Neurophysiol. 96 (2006) 2678–2687. [15] B. Rossi, S.L. Pasca, F. Sartucci, G. Siciliano, L. Murri, Trigemino-cervical reflex in pathology of the brainstem and of the first cervical cord segments, Electromyogr. Clin. Neurophysiol. 29 (1989) 67–71. [16] M. Serrao, A. Perrotta, M. Bartolo, G. Fiermonte, F. Pauri, P. Rossi, L. Parisi, F. Pierelli, Enhanced trigemino-cervical-spinal reflex recovery cycle in pain-free migraineurs, Headache 45 (2005) 1061–1068. [17] J.R. Flynn, B.A. Graham, M.P. Galea, R.J. Callister, The role of propriospinal interneurons in recovery from spinal cord injury, Neuropharmacology 60 (2011) 809–822. [18] F.M. Bareyre, M. Kerschensteiner, O. Raineteau, T.C. Mettenleiter, O. Weinmann, M.E. Schwab, The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats, Nat. Neurosci. 7 (2004) 269–277. [19] A. Perrotta, M. Serrao, M. Bartolo, L. Valletta, N. Locuratolo, F. Pujia, F. Fattapposta, P. Bramanti, G.A. Amabile, F. Pierelli, L. Parisi, Abnormal head nociceptive withdrawal reaction to facial nociceptive stimuli in Parkinson’s disease, Clin. Neurophysiol. 116 (2005) 2091–2098. [20] P. Boulenguez, L. Vinay, Strategies to restore motor functions after spinal cord injury, Curr. Opin. Neurobiol. 19 (2009) 587–600. [21] K. Takakusaki, K. Obara, T. Nozu, T. Okumura, Modulatory effects of the GABAergic basal ganglia neurons on the PPN and the muscle tone inhibitory system in cats, Arch. Ital. Biol. 149 (2011) 385–405. [22] G. Labouèbe, M. Lomazzi, H.G. Cruz, C. Creton, R. Luján, M. Li, Y. Yanagawa, K. Obata, M. Watanabe, K. Wickman, S.B. Boyer, P.A. Slesinger, C. Lüscher, RGS2 modulates coupling between GABAB receptors and GIRK channels in dopamine neurons of the ventral tegmental area, Nat. Neurosci. 10 (2007) 1559–1568. [23] A. Takahashi, A. Shimamoto, C.O. Boyson, J.F. Debold, K.A. Miczek, Gaba(b) receptor modulation of serotonin neurons in the dorsal raphé nucleus and escalation of aggression in mice, J. Neurosci. 30 (2010) 11771–11780.
