End-tidal CO2: An important parameter for a correct interpretation in functional brain studies using speech tasks
Highlights
► Effects of PaCO2 in functional brain studies have as yet been neglected. ► PaCO2 changed significantly and substantially during speech exercises. ► Inverse changes in blood circulation and oxygenation correlated to PaCO2. ► Previously inverse fNIRS/negative BOLD signals had not been attributed to PaCO2. ► PaCO2 is still underestimated as a potential confounder in functional brain studies.
Introduction
Previous studies showed that guided rhythmic speech exercises in the context of arts speech therapy (AST) cause changes in heart rate variability (Bettermann et al., 2002), and cardiorespiratory interactions (Cysarz et al., 2004) as well as blood circulation and oxygenation in the brain and muscle (Wolf et al., 2011a, Wolf et al., 2011b). In particular we demonstrated that during AST a decrease in cerebral hemodynamics and oxygenation occurred in the prefrontal cortex (PFC) (Wolf et al., 2011a, Wolf et al., 2011b). A similar effect in the PFC was found in a fNIRS-study by Fallgatter et al., (1998) involving a task with reading aloud. Hashimoto and Sakai (2003) in a fMRI study found an inverse (blood oxygen level-dependent) BOLD response, corresponding to an increase in the concentration of deoxyhemoglobin ([HHb]), in the PFC and occipital regions during two tasks involving reading aloud visually presented sentences.
These fNIRS and fMRI findings were surprising, since normally one would expect an increase in tissue oxygen saturation (StO2) and the concentrations of oxyhemoglobin ([O2Hb]) and total hemoglobin [tHb] as well as a decrease in the concentration of deoxyhemoglobin ([HHb]) due to increased brain activity [e.g. (Wolf et al., 2002)]. We hypothesized that a decrease in the partial pressure of carbon dioxide in the arterial blood (PaCO2) during speaking could account for the observed effects present in the fNIRS-derived signals (Wolf et al., 2011a, Wolf et al., 2011b) since it is known that a hyperventilation-induced decrease in PaCO2 (hypocapnia) causes a reduction in cerebral blood flow (CBF) by cerebral vasoconstriction (i.e. decrease in the cerebral arteriole diameter) (Poeppel et al., 2007, Szabo et al., 2011).
There is a significant and relevant effect of respiration-induced changes in PaCO2 on the fNIRS-derived signals, e.g. (Klaessens et al., 2003) and the fMRI-derived BOLD signal (Birn et al., 2006, Birn et al., 2009, Chang and Glover, 2009, Kastrup et al., 1998, Kastrup et al., 1999a, Kastrup et al., 1999b, Kastrup et al., 1999c, Kemna and Posse, 2001, Madjar et al., 2012, Mäkiranta et al., 2004, Posse et al., 1997, Posse et al., 2001, Rostrup et al., 2000, Stillman et al., 1995, Thomason and Glover, 2008, Van de Ven et al., 1999, Vesely et al., 2001, Wise et al., 2004, Wise et al., 2007). For example, a breath-hold of 30 s duration causes an average BOLD increase of 3–5% (Kastrup et al., 1999a, Kastrup et al., 1999b, Kastrup et al., 1999c) and hyperventilation for 30 s is associated with a BOLD decrease of 3.6–4.7% (gray matter) and approx. 1.5% (white matter) (Mäkiranta et al., 2004). Concerning the PaCO2 influence of fNIRS-derived signals it was shown for example that hypocapnia causes a reduction of 2.1% in the cerebral tissue oxygenation index (TOI) and hypercapnia (i.e. an elevated CO2 pressure in the blood) an increase of 2.6% in TOI (Tisdall et al., 2009). In addition, Tisdall et al. (2009) showed that changes in PaCO2 are the second most important factor affecting the value of TOI.
Surprisingly the impact of PaCO2 changes particularly caused by speech tasks on the fNIRS- and fMRI-derived signals has not been previously investigated. Past fNIRS-based studies involving speech exercises, e.g. visual object naming, reading, talking, performing of a verbal fluency task, etc. did not measure PaCO2 changes (Cannestra et al., 2003, Fallgatter et al., 1998, Herrmann et al., 2003, Herrmann et al., 2004, Herrmann et al., 2006, Holtzer et al., 2011, Hull et al., 2009, Kahlaoui et al., 2012, Kakimoto et al., 2009, Kono et al., 2007, Liu et al., 2008, Lo et al., 2009, Sakatani et al., 1999, Schecklmann et al., 2008, Schecklmann et al., 2010, Suda et al., 2010). PaCO2 measurements during fMRI studies are also not included during speech tasks. In general, only newer studies on the role of PaCO2 fluctuations in the generation of low frequency oscillations in the BOLD signal (Birn et al., 2009, Chang and Glover, 2009, Wise et al., 2004, Wise et al., 2007) include the measurement and analysis of PaCO2 changes during the experiment. Duong et al. (2000) performed the monitoring of PaCO2 in a fMRI study involving visual stimulation and found that a stable PaCO2 during the experiment is critical for obtaining a reproducible and consistent evoked BOLD response.
Consequently, the questions are how speech affects PaCO2, whether this can explain decreases in cerebral oxygenation found in previous fNIRS studies (Fallgatter et al., 1998, Wolf et al., 2011a, Wolf et al., 2011b) as well as fMRI studies (Hashimoto and Sakai, 2003), and whether it may be crucial to include this parameter in future functional brain studies — especially when involving speech tasks.
These questions are in particular important to address since previous studies showed that a substantial amount of the measured hemodynamic and oxygenation changes using fNIRS during a functional brain-activation task can be attributed to systemic (non-neuronal) changes, like transient changes in mean blood pressure (MBP) and scalp blood flow (Kirilina et al., 2012, Minati et al., 2011, Tachtsidis et al., 2009). Since (i) fNIRS is sensitive to intra- and extracranial hemodynamic and oxygenation changes in parallel and (ii) both intra- and extracranial hemodynamics and oxygenation are influenced by systemic hemodynamic changes (Minati et al., 2011), the separation of the intra- and extracranial component of fNIRS signals as well as the characterization of systemic influences is crucial for a proper interpretation of signals measured with fNIRS. As explained below, we have employed an approach based on multi-distance frequency domain fNIRS, which strongly reduces the influence of superficial tissue.
To test our hypothesis about the relevance of the impact of PaCO2 changes in speech studies, the aim of the present study was to measure and analyze cerebral hemodynamics/oxygenation using fNIRS, and the partial pressure of end-tidal carbon dioxide (PETCO2), a reliable and accurate estimate of PaCO2 (Dager et al., 1995, Dubois et al., 1952, Nunn and Hill, 1960, Robbins et al., 1990, Young et al., 1991), during 3 different speech tasks and a control task.
Section snippets
Subjects, experimental protocol and instrumentation
24 healthy subjects (13 men, 11 women, mean age: 22 ± 64 years) participated in this study. The study was carried out as a controlled and randomized crossover trial. The design of the study was in accordance with the Declaration of Helsinki; the approval was obtained from the Ethical Committee of the Canton of Zurich. The participants were German/Swiss German native speakers who had no previous knowledge of AST, and were asked not to smoke, eat or consume any stimulants (such as caffeine or other
Results
An example of a typical recorded change of PETCO2 and StO2 during a speech task is given in Fig. 2. Fig. 3 shows the measured changes in PETCO2, StO2, [O2Hb], [HHb] and [tHb] for the right and left PFC and the 4 different tasks. In the following subsection, the significant (p < 0.05) changes of the signals recorded are reported. A change caused by the task was considered significant when a significant change happened in interval 2 or interval 4 or in both.
Measured changes in end-tidal CO2, cerebral hemodynamics and oxygenation
The observed decreases in StO2, [O2Hb] and [tHb] during the 3 speech tasks are in agreement with previous fNIRS studies of our group (Wolf et al., 2011a, Wolf et al., 2011b), and fNIRS studies (Hashimoto and Sakai, 2003) involving speech tasks, whereby in (Wolf et al., 2011b) a decrease in StO2 could be observed but not in [tHb]. This might be explained by the smaller power due to the smaller sample size (N = 17) of this study (Wolf et al., 2011b). The observed increase in [HHb] and the decrease
Conclusion
In conclusion, we found that changes in breathing (hyperventilation) during speech tasks led to hypocapnia, which resulted in reduced cerebral hemodynamics and oxygenation. We highlight that the measurement of PETCO2 during functional speech tasks seems to be important for a reliable and correct interpretation of the measured parameters using fNIRS. We recommend that PETCO2 changes be measured and analyzed in future fNIRS studies in general. This conclusion is also true for fMRI and other
Acknowledgment
We thank all subjects and the arts speech therapist Andrea Klapproth for their participation in this study, Reto Kofmehl for help in statistics, Rachel Folkes for proofreading of the manuscript, and the numerous participants of the International Society on Oxygen Transport to Tissue (ISOTT) conferences 2010, 2011 and the conference of the Optical Society of America 2012 for their stimulating discussions about CO2 and cerebral hemodynamics/oxygenation.
References (135)
- et al.
The variability of human BOLD hemodynamic responses
NeuroImage
(1998) Effects of speech therapy with poetry on heart rate rhythmicity and cardiorespiratory coordination
Int. J. Cardiol.
(2002)Separating respiratory-variation-related fluctuations from neuronal-activity related fluctuations in fMRI
NeuroImage
(2006)Dynamic models of BOLD contrast
NeuroImage
(2012)- et al.
Relationship between respiration, end-tidal CO2, and BOLD signals in resting-state fMRI
NeuroImage
(2009) Between-brain coherence during cooperative n-back task performance: a functional near-infrared spectroscopy study
Behav. Brain Res.
(2012)- et al.
Individual differences in emotion processing
Curr. Opin. Neurobiol.
(2004) Reduction of excitability (“inhibition”) in the ipsilateral primary motor cortex is mirrored by fMRI signal decreases
NeuroImage
(2002)- et al.
Frontal activation during a verbal-fluency task as measured by near-infrared spectroscopy
Brain Res. Bull.
(2003) Cerebral oxygenation changes in the prefrontal cortex: effects of age and gender
Neurobiol. Aging
(2006)
Extension of mental preparation positively affects motor imagery: a functional near-infrared spectroscopy study
Cortex
Trial-to-trial variability differentiates motor imagery during observation between low versus high responders: a functional near-infrared spectroscopy study
Behav. Brain Res.
Between-brain connectivity of premotor cortical signals during stimulus- versus self-paced imitation measured by fNIRS
NeuroImage
A temporal comparison of BOLD, ASL, and NIRS hemodynamic responses to motor stimuli in adult humans
NeuroImage
The relationship between BOLD signal and autonomic nervous system functions: implications for processing of “physiological noise”
Magn. Reson. Med.
Contribution of NIRS to the study of prefrontal cortex for verbal fluency in aging
Brain Lang.
Regional variability of cerebral blood oxygenation response to hypercapnia
NeuroImage
Behavioral correlates of negative BOLD signal changes in the primary somatosensory cortex
NeuroImage
A functional MRI study of simple arithmetic—a comparison between children and adults
Cogn. Brain Res.
Effect of respiratory CO2 changes on the temporal dynamics of the hemodynamic response in functional MR imaging
NeuroImage
Is optical imaging spectroscopy a viable measurement technique for the investigation of the negative BOLD phenomenon? A concurrent optical imaging spectroscopy and fMRI study at high field (7 T)
NeuroImage
The physiological origin of task-evoked systemic artefacts in functional near infrared spectroscopy
NeuroImage
Multiple-time replicability of near-infrared spectroscopy recording during prefrontal activation task in healthy men
Neurosci. Res.
Task-related BOLD responses and resting-state functional connectivity during physiological clamping of end-tidal CO2
NeuroImage
BOLD-contrast functional MRI signal changes related to intermittent rhythmic delta activity in EEG during voluntary hyperventilation—simultaneous EEG and fMRI study
NeuroImage
Dissociating prefrontal and parietal cortex activation during arithmetic processing
NeuroImage
Characterizing the hemodynamic response: effects of presentation rate, sampling procedure, and the possibility of ordering brain activity based on relative timing
NeuroImage
Intra- and extra-cranial effects of transient blood pressure changes on brain near-infrared spectroscopy (NIRS) measurements
J. Neurosci. Methods
fMRI signal decreases in ipsilateral primary motor cortex during unilateral hand movements are related to duration and side of movement
NeuroImage
Spontaneous low frequency oscillations of cerebral hemodynamics and metabolism in human adults
NeuroImage
Analysis of oxygen metabolism implies a neural origin for the negative BOLD response in human visual cortex
NeuroImage
Focal frontal (de)oxyhemoglobin responses during simple arithmetic
Int. J. Psychophysiol.
Regional differences in the CBF and BOLD responses to hypercapnia: a combined PET fMRI study
NeuroImage
Cerebral hemodynamics measured with simultaneous PET and near-infrared spectroscopy in humans
Brain Res. Bull.
Effects of aging on language-activated cerebral blood oxygenation changes of the left prefrontal cortex: near infrared spectroscopy study
J. Stroke Cardiovasc. Dis.
Negative BOLD signal changes in ipsilateral primary somatosensory cortex are associated with perfusion decreases and behavioral evidence for functional inhibition
NeuroImage
Functional near-infrared spectroscopy: a long-term reliable tool for measuring brain activity during verbal fluency
NeuroImage
Influence of muscle activity on brain oxygenation during verbal fluency assessed with functional near-infrared spectroscopy
Neurosci. Res.
Functional MRI cerebral activation and deactivation during finger movement
Neurobiology
Sensory inputs from whisking movements modify cortical whisker maps visualized with functional magnetic resonance imaging
Cereb. Cortex
How to conduct studies with neonates combining near-infrared imaging and electroencephalography
Advances in Experimental Biology and Medicine
Negative BOLD-fMRI signals in large cerebral veins
J. Cereb. Blood Flow Metab.
The respiration response function: the temporal dynamics of fMRI signal fluctuations related to changes in respiration
NeuroImage
Negative blood oxygen level dependence in the rat: a model for investigating the role of suppression in neurovascular coupling
J. Neurosci.
Negative BOLD fMRI response in the visual cortex carries precise stimulus-specific information
PLoS One
The effect of basal vasodilation on hypercapnic and hypocapnic reactivity measured using magnetic resonance imaging
J. Cereb. Blood Flow Metab.
The brain's default network
Ann. NY Acad. Sci.
Functional assessment of Broca's area using near infrared spectroscopy in humans
NeuroReport
MRI measurement of the BOLD-specific flow-volume relationship during hypercapnia and hypocapnia in humans
NeuroImage
Noninvasive determination of the optical properties of adult brain: near-infrared spectroscopy approach
J. Biomed. Opt.
Cited by (102)
Body fat predictive of acute effects of exercise on prefrontal hemodynamics and speed
2024, NeuropsychologiaCerebral, systemic physiological and behavioral responses to colored light exposure during a cognitive task: A SPA-fNIRS study
2024, Behavioural Brain ResearchAn Approach to Neuroimaging Interpersonal Interactions in Mental Health Interventions
2022, Biological Psychiatry: Cognitive Neuroscience and NeuroimagingBody, Brain, Behavior: Three Views and a Conversation
2022, Body, Brain, Behavior: Three Views and a ConversationUsing functional near-infrared spectroscopy to study word production in the brain: A picture-word interference study
2021, Journal of Neurolinguistics