Autonomic reactivity to sensory stimulation is related to consciousness level after severe traumatic brain injury
Introduction
Severe traumatic brain injury (sTBI) results in high morbidity and mortality rates. A large number of sTBI patients experience long-term or lifelong disabilities, entailing major costs for family and society. In the United States the incidence of sTBI is 8 times as high as that of breast cancer and 34 times that of HIV/AIDS (CDC Injury prevention, 2004). However, research on recovery patterns is scarce, especially when young adults are concerned, who are known to be a group at great risk of sTBI (Finfer and Cohen, 2001, Jennett, 1996).
The clinical recovery pattern after sTBI, including the different levels of consciousness and characteristics of those levels, has been extensively discussed (Andrews, 1996, Giacino, 1997, Multi-Society Task Force on Persistent Vegetative State, 1994a, Zeman, 2001). Immediately after severe brain damage the patient is usually in a coma. Patients who do not awaken from coma within a period of about 4–6 weeks may shift into a vegetative state (VS) (Jennett and Plum, 1972, Multi-Society Task Force on Persistent Vegetative State, 1994a), or die. In a vegetative state patients have sleep–wake cycles, autonomic control of blood pressure and respiration are present, while cognitive functioning and consciousness are absent. For some patients such a vegetative state is the final outcome. A subgroup of patients may shift into a minimally conscious state (MCS) (Giacino et al., 2002), also referred to as a low awareness state (Andrews, 1996). Patients then demonstrate discernible but inconsistent evidence of consciousness. This state is often transient but can also be the permanent outcome. When patients react adequately to the environment and when communication is possible (with or without tools), they are assumed to be conscious. The experience of self and the environment, and the stock of knowledge, thoughts and intentions are then present (Zeman, 2001); however, various cognitive impairments might still affect the patient (Multi-Society Task Force on Persistent Vegetative State, 1994a).
In current practice, decisions concerning consciousness rest principally on clinical observations (Chiappa and Hill, 1998). In the acute phase after TBI, the depth of coma is often determined by means of the Glasgow coma scale (GCS, Teasdale and Jennett, 1974). Complementary diagnostic investigation by means of neurophysiological assessment is often carried out, but this is intended mainly to diagnose the extent of brain damage (haemorrhage, oedema, diffuse swelling, intracranial pressure, epileptic seizures, et cetera). In addition, early neurophysiological methods are sometimes used to predict the clinical outcome (Fischer and Luauté, 2005, Fischer et al., 2004, Guérit et al., 1993, Guérit et al., 1999, Kane et al., 2000, Wardlaw et al., 2002).
In the post-acute phase, observation scales are also used, examining the recovery to consciousness by observing behavioural skills, such as the Western Neuro Sensory Stimulation Profile (WNSSP, Ansell et al., 1989), the Rancho Los Amigos Scale (Hagen et al., 1972), and the Disability Rating Scale (Rappaport et al., 1982). Yet during this phase neurophysiological assessment is not always considered to be important.
The main purpose of the present study was to examine whether the behavioural changes in the post-acute phase of recovery after sTBI are reflected in physiological reactivity. If so, the examination of neurophysiological features and changes within these features could provide more insight into processes and patterns of recovery during the post-acute phase.
Recently, functional neurophysiological reactivity has been demonstrated in VS and MCS using Event Related Potentials (Kotchoubey et al., 2002, Kotchoubey et al., 2005, Neumann and Kotchoubey, 2004), and using fMRI and PET scans (Boly et al., 2004, Jong et al., 1997, Laureys et al., 2004a, Owen et al., 2005, Schiff et al., 2002, Schiff et al., 2005).
It appears that external stimuli (such as sounds) can provoke cortical activity in VS (Kotchoubey et al., 2005, Laureys et al., 2004a, Owen et al., 2005, Schiff et al., 2002). This activity is often limited to the isolated activity in certain ‘cortical islands’ (Menon et al., 1998, Plum et al., 1998, Schiff et al., 1999, Schiff et al., 2002), which are not integrated in the entire network of information processing. Therefore, it is still not certain whether any ability to understand is intact (Robertson and Murre, 1999). Results with PET showed that the brain metabolism in VS is reduced by 50% compared to a healthy brain (Laureys et al., 1999, Laureys et al., 2002). In addition, in VS brain metabolism of different areas is unrelated, presumably because of the disconnection between these areas (Boly et al., 2004, Boly et al., 2005, Laureys et al., 1999, Laureys et al., 2002). Postmortom research showed that in VS often a structurally normal cortex was intact (Adams et al., 2000), however, without any connection to other areas like the thalamus.
In MCS the associative brain areas (secondary and tertiary) are active in response to external stimulation such as sound or pain (Boly et al., 2005). These areas are necessary for the conscious perception of stimuli (Baars et al., 2003). In some studies, it was found that brain activity in MCS in response to sound and pain stimulation was equal to the activity found in a healthy control group (Laureys et al., 2004b, Schiff et al., 2005).
We specifically examined the reactivity of the autonomic nervous system (ANS) to environmental input through different sensory modalities during the recovery from a vegetative state to consciousness. Measurements of the ANS can provide insight into mental activity related to the perception and processing of environmental stimulation, even in the absence of observable behaviour (Öhman et al., 2000). Measuring the function of the ANS in patients with sTBI might be especially informative. According to Plum and Posner (1980), preservation of arousal is required for recovery to consciousness, because conscious behaviour depends on the continuous interaction between cortical systems and the subcortical activating mechanisms (the noradrenergic and cholinergic reticular activation system). Since arousal is mainly mediated by the ANS, we expected that recovery of consciousness is related to, or even dependent on, the functionality and integrity of the ANS.
Spectral analysis of heart rate variability (HRV) and the assessment of skin conductance level (SCL) allowed us to probe the functioning of the sympathetic and parasympathetic branches of the ANS separately. Changes in SCL are influenced primarily by sympathetic elicitation of sweat secretion (Boucsein, 1992). Slow variations of the heart rate mainly reflect the influence of homeostatic control processes, mediated by the sympathetic branch of the ANS (Berntson et al., 1997). More rapid fluctuations reflect processes related to blood pressure control predominantly, but not exclusively, by the sympathetic branch of the ANS (Akselrod et al., 1981). Very fast fluctuations are related to respiratory activity, primarily controlled by the parasympathetic branch of the ANS (Akselrod et al., 1981, Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). In addition, the sympathovagal balance can be examined using HRV (Malliani et al., 1998, Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996).
In the current study, adolescents suffering sTBI were admitted to an ‘Early Intensive Neurorehabilitation Programme’. Since 1987 the Rehabilitation Centre Leijpark offers this programme for children and young adults in a vegetative or minimally conscious state after acquired brain injury. The rationale of this program is that sensory stimulation and an enriched environment leads to better and faster recovery after sTBI (Rosenzweig and Bennett, 1996). The effect of this programme has not yet been demonstrated, however, Eilander et al. (2005a) showed that patients who participated in this programme had a more favourable outcome than predicted by ‘The Multi-Society Task Force on Permanent Vegetative State’ (Multi-Society Task Force on Persistent Vegetative State, 1994b).
In order to learn more about recovery processes during the rehabilitation programme, we studied the activity of the different branches of the ANS during a sensory stimulation protocol (WNSSP) (Ansell et al., 1989).
Several studies reported on HRV and SCL in TBI patients, but virtually all were performed in the acute phase and in adult patients. Comatose TBI patients in the acute phase show very low HRV (Lowensohn et al., 1977). An increase in the sympathovagal balance has been found in patients who had recovered from a comatose state in the acute phase (Hildebrandt et al., 1998). Higher sympathetic and lower parasympathetic activity were paired with better scores on the GCS (8–10). Recently, Su et al. (2005) compared HRV with severity of brain damage. In the more severely brain-damaged patient groups both sympathetic and parasympathetic activity were lower in comparison to less severely brain damaged patient groups, and in comparison to a healthy norm group.
Electrodermal reactivity to auditory stimuli is lower or even absent in sTBI patients in a vegetative state compared to healthy controls (Turkstra, 1995). Higher electrodermal activity can be seen in patients recovering from a vegetative state (Turkstra, 1995), together with higher scores on the GCS (8–10) (Hildebrandt et al., 1998). Only one study reported on HRV in the post-acute phase, in which 4 adult sTBI patients were compared with matched controls (King et al., 1997). These patients showed lower power in all frequency bands of the heart rate spectrum compared to controls.
We are not aware, however, of any study in which longitudinal measurements of HRV and SCL are performed in the post-acute phase during recovery to consciousness. In the present report, the relationship is investigated between the reactivity of the ANS to sensory stimulation, behavioural changes during the recovery to consciousness, and cognitive recovery of the sTBI patients. It was expected that during recovery to consciousness patients would become more aroused during the stimulation sessions. Environmental stimuli normally lead to a higher activity of the sympathetic and a lower activity of the parasympathetic nervous system to environmental stimuli: when sympathetic activity increases, the parasympathetic activity decreases (Berntson et al., 1991). Following this pattern, the sympathovagal balance during environmental stimulation would increase with recovery to consciousness as a consequence of reciprocal sympathetic activation (Berntson et al., 1991).
Section snippets
Patients
Sixteen patients with sTBI who were admitted to the rehabilitation programme between January 2001 and May 2002 were included in the study. Inclusion criteria for participation were: age between 17 and 26 years, no mechanical ventilation, and time between injury and admission no longer than 6 months. Ten (61.1%) were male. Age at the time of the injury ranged from 17.5 to 25 years (M=21.5 years; SD=3.0) (Table 1). Time since injury at admission was at least 4 weeks (M=2.3 months; SD=1.6). The
Behavioural indices of recovery
Sixteen patients participated in the experiment. A maximum of 14 repeated measurements were collected. All patients participated in the first measurement. Fig. 1 presents the amount of patients per measurement and Table 1 presents the number of measurements per patient.
At admission, the patients' averaged LoC score was 3.5 (corresponding to the reflexive vegetative). The LoC score increased to 6.2 (corresponding to the inconsistent minimally conscious state) at discharge. A significant
Discussion
The activity of the ANS in adolescents with sTBI in the post-acute phase was examined during sensory stimulation. The changes in the reactivity of the ANS during each administration of the WNSSP were examined, as well as the longitudinal changes related to recovery in the post-acute phase. We related the longitudinal changes in ANS activity to the recovery of consciousness. The results of our study are clear-cut: changes in autonomic reactivity during recovery were related to recovery to
Acknowledgements
We would like to thank Yvonne Schuttelaars and Sylvia Melisse for their contribution in collecting the data, Prof. Dr. J.K. Vermunt for his advise on the statistical analyses used, and K.L. Mansfield for a critical review of the paper on its use of the English language. This study is part of a larger evaluation study of the rehabilitation programme ‘Early intensive neurorehabilitation for children and young adults in a vegetative or minimally conscious state after severe brain injury’. This
References (75)
- et al.
Brain, conscious experience and the observing self
Trends Neurosci
(2003) - et al.
Evaluation and prognostication in coma
Electroencephalogr Clin Neurophysiol
(1998) - et al.
Fear conditioning in humans: the influence of awareness and autonomic arousal on functional neuroanatomy
Neuron
(2002) - et al.
Severe traumatic brain injury
Resuscitation
(2001) - et al.
The prognostic value of three-modality evoked potentials (TMEPs) in anoxic and traumatic comas
Neurophysiol Clin
(1993) - et al.
ERPs obtained with the auditory oddball paradigm in coma and altered states of consciousness: clinical relationships, prognostic value, and origin of components
Clin Neurophysiol
(1999) - et al.
Persistent vegetative state after brain damage. A syndrome in search of a name
Lancet
(1972) - et al.
Information processing in severe disorders of consciousness: vegetative state and minimally conscious state
Clin Neurophysiol
(2005) - et al.
Impaired effective cortical connectivity in vegetative state: preliminary investigation using PET
Neuroimage
(1999) - et al.
Restoration of thalamocortical connectivity after recovery from persistent vegetative state
Lancet
(2000)