Discrete functional contributions of cerebral cortical foci in voluntary swallowing: a functional magnetic resonance imaging (fMRI) “Go, No-Go” study
- First Online:
- Cite this article as:
- Toogood, J.A., Barr, A.M., Stevens, T.K. et al. Exp Brain Res (2005) 161: 81. doi:10.1007/s00221-004-2048-1
- 145 Views
Brain-imaging studies have shown that visually-cued, voluntary swallowing activates a distributed network of cortical regions including the precentral and postcentral gyri, anterior cingulate cortex (ACC), insula, frontoparietal operculum, cuneus and precuneus. To elucidate the functional contributions of these discrete activation foci for swallowing, a “Go, No-Go” functional magnetic resonance imaging (fMRI) paradigm was designed. Brain activation associated with visually-cued swallowing was compared with brain activation evoked by a comparable visual cue instructing the subject not to swallow. Region-of-interest analyses performed on data from eight healthy subjects showed a significantly greater number of activated voxels within the precentral gyrus, postcentral gyrus, and ACC during the “Go” condition compared to the “No-Go” condition. This finding suggests that the precentral gyrus, postcentral gyrus, and ACC contribute primarily to the act of swallowing. In contrast, the numbers of activated voxels within the cuneus and precuneus were not significantly different for the “Go” and “No-Go” conditions, suggesting that these regions mediate processing of the cue to swallow. Together these findings support the view that the discrete cortical foci previously implicated in swallowing mediate functionally distinct components of the swallowing act.
Recent neuroimaging studies have suggested that swallowing is processed within a distributed cortical network (Hamdy et al. 1999a, 1999b; Kern et al. 2001a, 2001b; Martin et al. 2001, 2004; Mosier and Bereznaya 2001, Mosier et al. 1999). This includes the primary sensorimotor cortex, non-primary sensory and motor cortical fields, the insula, and parieto-occipital cortex. It has been hypothesized that these discrete activation foci mediate functionally distinct components of the swallowing act. For example, activation of the primary sensorimotor cortex has been attributed to cortical processing of swallowing motor regulation and execution (Hamdy et al. 1999a; Hamdy et al. 1999b; Mosier et al. 1999; Mosier and Bereznaya 2001; Martin et al. 2001, 2004). Responses within non-primary motor regions, including the anterior cingulate cortex (ACC), have been explained in terms of higher-order motor processing or attention (Hamdy et al. 1999a, 1999b; Martin et al. 2001, 2004; Mosier and Bereznaya 2001). Activation within the parieto-occipital cortex, particularly the cuneus and precuneus, has been attributed to the sensory processing of swallowing (Hamdy et al. 1999b). However, these interpretations of the functional contributions of various brain areas in swallowing have not been fully tested. For example, the parieto-occipital cortex is known to mediate visual and multi-modal sensory processing. Since brain-imaging studies of swallowing have typically examined swallows produced in response to a visual cue, it remains unclear whether activation of the parieto-occipital cortex reported in functional imaging studies of swallowing is related to swallowing per se or to the processing of the cue instructing the subject to swallow.
Therefore, the present study sought to differentiate the cortical areas processing the act of swallowing from those processing the visually-cued experimental task using a “Go, No-Go” fMRI protocol. Processing the visually-cued experimental task would be expected to encompass multiple sub-processes including, but not limited to, visual perception, decoding of the graphemes/words, appreciating the meaning conveyed by the words, and selecting an appropriate response in association with a given visual cue. The “Go, No-Go” paradigm has been applied previously in studies of limb sensorimotor and cognitive processing (Isomura et al. 2003; Liddle et al. 2001; Konishi et al. 1998; Wantanabe et al. 2002). In the present study, brain activation associated with visually-cued swallowing was compared with activation evoked by the presentation of a comparable visual cue instructing the subject not to swallow. Thus, the “Go” and “No-Go” conditions differed in terms of the act of swallowing. It was hypothesized that the “Go” and “No-Go” conditions would evoke differential activation across the multiple cortical regions previously implicated in swallowing control. Specifically, we predicted that the “Go” swallowing condition would evoke activation within the primary and secondary sensorimotor cortices and insula, as well as the parieto-occipital cortex. In contrast, the “No-Go” condition was expected to elicit activation within only the parieto-occipital focus of the previously reported swallowing network.
Eight healthy volunteers, six female and two male (age, 23.8±2.3 yr, mean±SD) participated as subjects. All but one male were right-handed, as measured by the Edinburgh Handedness Inventory (Oldfield 1971). One subject had previous fMRI experience. All subjects gave written informed consent before participating in this study. The study adhered to the MRI safety depositional guidelines established by the United States Food and Drug Administration for clinical scanners, and was approved by the University of Western Ontario Review Board for Health Sciences research involving human subjects.
Each subject participated in two, nine-minute functional imaging runs during a single experimental session. Each functional run consisted of two randomly ordered activation tasks performed in response to visual cues. The visual cues, which were 2 seconds in duration, were back-projected onto a mirror positioned above the subject’s eyes. One of the two activation tasks, the “Go” condition, was a single voluntary saliva swallow performed in response to the visual cue “Do Swallow”. The subject was instructed to swallow his/her saliva once without making exaggerated oral movements to produce extra saliva. The other activation task, the “No-Go” condition, involved the subject making no overt response following the presentation of the visual cue “Don’t Swallow”. The task instruction to the subject was simply to remain at rest. The word “Relax” was displayed between these cues. The inter-stimulus interval was randomly varied between 28 and 32 seconds in order to prevent anticipation of the cues. Thus, each of the two activation tasks was performed eight times during both of the nine-minute functional runs. To ensure that the subject understood the experimental procedures, each subject practiced the activation tasks prior to the functional runs.
Identification of swallowing
Single swallowing trials were verified on the basis of their distinct profiles of laryngeal movement (Logemann et al. 1992; Martin et al. 2001) and respiratory apnea (Preiksaitis et al. 1992). The output signal of a pressure transducer driven from expanding magnetic resonance (MR)-compatible bellows (Siemens, Erlangen, Germany) placed over the subject’s thyroid cartilage recorded laryngeal movements (PowerLab, 4.1.1). Episodes of respiratory apnea were detected using a similar transducer placed around the subject’s ribcage.
A Varian UNITY INOVA 4 Tesla (T) whole-body imaging system (Varian, Palo Alto, CA) equipped with 40 mT/m Siemens Sonata actively shielded whole-body gradients and amplifiers (Siemens, Erlangen, Germany) was used to image all subjects. A whole-head quadrature birdcage radio frequency (RF) coil transmitted and received the MR signal (Barberi et al. 2000). Foam padding was fit snugly between the subject’s head and a Plexiglas head cradle within the head coil in order to immobilize the subject’s head.
Imaging planes for the functional scans were prescribed with the aid of a high resolution [256×256, 28.0 cm field of view (FOV)] sagittal anatomic image with gray/white matter contrast (T1-weighted) acquired using a magnetization-prepared fast low-angle shot (FLASH) imaging sequence [inversion time (TI)=750 ms, echo time (TE)=3.5 ms, repetition time (TR)=8 ms, flip angle=11°, 5-mm slice thickness]. Functional data were collected from 18 contiguous, 5 mm-thick axial slices oriented in a plane approximately parallel to the anterior commisure (AC)-posterior commisure (PC) line and extending from the superior extent of the paracentral lobule to approximately 10 mm below the AC-PC plane. During each functional task described above, blood-oxygenation-level-dependent (BOLD) images (T2*-weighted) were acquired continuously using an interleaved, four segment, echo planar imaging (EPI) sequence (64×64 matrix size, TR=500 ms, TE=10 ms, flip angle=30°, 19.2 cm FOV, volume collection time=2 s). Each image was corrected for physiologic fluctuations using a navigator echo that was collected at the beginning of every image segment (Hu and Kim 1994). At the end of the experimental session, anatomic reference images were acquired along the same orientation as the functional images using a three-dimensional (3-D) FLASH sequence (256×256×128 matrix size, 1.25 mm reconstructed slice thickness, TI=600 ms, TR=10 ms, TE=5.5 ms).
fMRI data analysis
Previous fMRI studies have shown that swallow-related movements of the tongue and jaw occurring immediately outside the imaging FOV can produce false-positive BOLD signal changes by disturbing the magnetic field in nearby imaging slices (Birn et al. 1999). These movements produce both magnitude and phase changes in the complex-valued MRI signal (Martin et al. 2001). In contrast, hemodynamic changes within the microvasculature (the vasculature smaller than intracortical veins) are expected to produce only magnitude changes in the MR signal (Menon 2002). Based on these findings, a motion suppression algorithm that estimates and removes the fraction of the BOLD signal arising from motion by measuring its influence on the phase angle of the complex-valued MRI time series was applied (Menon 2002). Previous work in our lab has demonstrated the effectiveness of this algorithm (Martin et al. 2004). A maximum likelihood estimator was determined based on a linear least-squares fit of the MRI signal phase to the BOLD signal magnitude within each voxel. Baseline drift in the MR time course of each voxel was then removed by applying a band-pass filter.
Subsequent image analyses were performed using BrainVoyager v4.9 (Max Planck Society, Germany; Goebel et al. 1996). The two-dimensional (2-D) functional slices were co-registered with the 3-D anatomic images to create volume-time courses. Anatomic images were aligned with the AC-PC plane and transformed to standard stereotaxic space (Talairach and Tournoux 1988). Small head movements were removed by applying a 3-D motion correction with sinc interpolation using the Levenberg-Marquardt non-linear least squares method to the volume-time courses, to fit six parameters (three translation, three rotation) of each image volume to a reference volume (Press et al. 1992). The data then were spatially smoothed with a Gaussian filter (full width at half maximum=4 mm).
The statistical analysis tested, on a voxel-by-voxel basis, for significant activation associated with the “Do Swallow” and “Don’t Swallow” tasks using multiple regression analyses for each of the eight subjects. A p<0.0003 (uncorrected for the total number of voxels tested) was considered statistically significant for each task. A square-wave function representing the time course of the activation task was convolved with a γ function (δ=1.25; τ=2.5) representing the hemodynamic impulse response (Boynton et al. 1996; Cohen 1997) to generate predictors.
A region-of-interest (ROI) analysis was performed to examine activation within the cortical regions that have been implicated in swallowing control by previous studies (Hamdy et al. 1999a, 1999b; Kern et al. 2001a, 2001b; Martin et al. 2001, 2004; Mosier et al. 1999; Mosier and Bereznaya 2001, Zald et al. 1999). Seven anatomically-defined ROIs were examined. These corresponded to the: (1) precentral gyrus, (2) postcentral gyrus, (3) ACC corresponding to Brodmann area (BA) 24, (4) ACC corresponding to BA 32, (5) precuneus (BA 7), (6) cuneus (BA 17 and 18), and (7) insula/operculum. The insula/operculum ROI was defined with respect to two Talairach volumes: Y=−26 to 4, X=−20 to −44; Z=−6 to 20, and Y=5 to 26, X=−32 to −44; Z=−6 to 20. ROIs within the right-hemisphere and left-hemisphere were defined separately and subsequently combined to examine the two hemispheres together.
The number of activated voxels in each anatomically defined ROI was calculated for each subject and task using a p<0.0003. These values were then submitted to pairwise t-tests to determine whether the number of activated voxels evoked by the “Do Swallow” and “Don’t Swallow” tasks were significantly different across subjects for the ROIs examined. The significance level for the t-tests was set at 0.05.
Single subject analyses
Here D indicates the number of voxels activated by “Do Swallow”, and N indicates the number of voxels activated by “Don’t Swallow”. A Swallow Activation Index of +1 would indicate that activation within a ROI occurred exclusively in the “Do Swallow” condition, whereas an Activation Index of −1 would reflect activation exclusive to the “Don’t Swallow” condition. Because the Swallow Activation Index was designed as a descriptive measure, and because the voxel values used to compute it are themselves obtained statistically, no specific threshold value was assigned to the Swallowing Activation Index.
Verification of swallowing
Areas of activation evoked by the “Do Swallow” condition
Areas of activation evoked by the “Do Swallow” and “Don’t Swallow” conditions
Consistent with this groupwise finding, the single subject Swallow Activation Indices showed that the cuneus was activated by both the “Do Swallow” and “Don’t Swallow” conditions in 7/8 subjects with the “Don’t Swallow” condition activating a greater number of voxels than the “Do Swallow” condition in five subjects. Across subjects, the average value of the activation index for the cuneus was −0.07. Within the precuneus, activation was associated with both conditions in all eight subjects and the “Do Swallow” condition activated a greater number of voxels than the “Don’t Swallow” condition in six subjects. Across subjects, the average value of the activation bias for the precuneus was 0.34. The average activation indices for the cuneus and precuneus were smaller than those for the other ROIs, suggesting relatively more uniform activation during the two tasks. This is in agreement with the finding that there was no significant difference in the number of voxels activated by the “Do Swallow” and the “Don’t Swallow” conditions.
The aim of the present study was to differentiate cortical areas processing the act of swallowing from those processing aspects of the visually-cued experimental context, employing a “Go, No-Go” experimental paradigm. While previous studies have demonstrated the utility of the “Go, No-Go” paradigm in discriminating among the various sub-processes underlying a single task (Isomura et al. 2003; Liddle et al. 2001; Konishi et al. 1998; Wantanabe et al. 2002), our study is the first to apply this paradigm to the study of an oral sensorimotor behavior.
The main finding of the present study was that fMRI-defined activation of the precentral gyrus, postcentral gyrus and ACC and insula were related specifically to the act of swallowing, whereas activation of the cuneus and precuneus were not contingent upon swallowing. These findings suggest that the precentral and postcentral gyri, ACC, and insula mediate the act of swallowing. In contrast, the parieto-occipital foci appear to be involved in processing the experimental context of the cued task.
Role of sensorimotor cortex in swallowing
The present finding, that activation of the precentral gyrus (including BA 4 and 6) and the postcentral gyrus (including BA 3, 2, 1, and 43) is associated specifically with swallowing, is consistent with our previous fMRI finding that autonomic saliva swallowing produced by naïve subjects in the absence of a voluntary task paradigm activates the lateral precentral and postcentral gyri (Martin et al. 2001). Similarly, Kern et al. (2001a) reported activation of the lateral sensory and motor cortices when swallows were evoked by water injected directly into the pharynx. Converging evidence from physiological studies in humans and non-human primates has also implicated the precentral and postcentral gyri in swallowing sensorimotor control. For example, transcranial magnetic stimulation (TMS) applied anterolateral to the motor cortex evokes electromyographic activity within oral and pharyngeal muscles that are activated during swallowing (Hamdy et al. 1999a). In primates, swallowing can be evoked by intracortical microstimulation (ICMS) applied to four areas of the sensorimotor cortex: the face primary motor cortex (MI), face primary somatosensory cortex (SI), BA 6 immediately lateral to face MI, and the frontal operculum (Huang et al. 1989; Martin et al. 1999; Yao et al. 2002). Neuronal recording studies have demonstrated that neurons in the primate face MI fire in relation to swallowing (Lund and Lamarre 1974; Martin et al. 1997; Yao et al. 2002). Furthermore, tongue and jaw movements, as well as pre-swallow bolus transport, are impaired during reversible cold block of the primate face MI and SI (Lin et al. 1998; Narita et al. 2002; Yamamura et al. 2002). Clinical studies in humans have documented swallowing motor impairment following stroke involving the middle cerebral artery territory (Robbins et al. 1993).
Role of anterior cingulate cortex in swallowing
Several studies have reported activation of the ACC in association with swallowing (Hamdy et al. 1999a; Hamdy et al. 1999b; Kern et al. 2001a; Kern et al. 2001b; Martin et al. 2001, 2004; Mosier et al. 1999; Mosier and Bereznaya 2001) and attributed this activation to swallowing movement planning and execution or, alternatively, cognitive and perceptual processes including attention and response selection. However, the role of the ACC in swallowing remains unclear since this brain area has been implicated in sensory, motor, and cognitive processing (Deiber et al. 1999; Devinsky et al. 1995; Downar et al. 2000; Luks et al. 2002; Milham et al. 2003; Paus 2001; Shima et al. 1991). Moreover, studies have examined a variety of swallowing tasks (saliva versus water, autonomic, self-paced voluntary, externally-cued voluntary), and their findings have been inconsistent. For example, there has been disagreement on what region of the ACC is activated by swallowing of a water bolus, with one study reporting activation of the rostral ACC (z=4–8; Hamdy et al. 1999b), and another localizing the activation to the caudal region (z=46; Hamdy et al. 1999a). A previous study by Martin et al. (2001) showed that autonomic swallowing typically evoked activation within the rostral ACC whereas voluntary swallowing elicited activation within the intermediate and caudal regions of the ACC, as defined by Paus et al. (1996).
The present study extends these previous findings in two ways. First, the results indicate that activation associated with voluntary swallowing is localized to the intermediate and caudal regions of the ACC corresponding to BA 24 and 32. Secondly, activation within both foci was significantly greater in the “Do Swallow” condition compared with the “Don’t Swallow” condition. These findings suggest that ACC activation during a voluntary swallowing task is related specifically to the act of swallowing, and cannot be explained in terms of processing of the experimental context, including the task cues, in which swallowing was examined.
Role of parieto-occipital cortex in swallowing
Previous reports of activation of the precuneus and/or cuneus in swallowing have been highly inconsistent both within and across laboratories (Hamdy et al. 1999b; Kern et al. 2001a; Martin et al. 2001, 2004; Mosier et al. 1999). Two PET studies did not identify the precuneus and cuneus as activation foci for water swallowing (Hamdy et al. 1999a; Zald and Pardo 1999), whereas fMRI studies have documented activation of these regions in between 10% and 100% of subjects (Hamdy et al. 1999b; Martin et al. 2001, 2004; Mosier et al. 1999; Kern et al. 2001a, 2001b). Differences across studies in experimental protocols, imaging methodologies, data analyses, and definition of ROIs have likely contributed to the variable findings.
The present study has demonstrated that, unlike the precentral gyrus, postcentral gyrus, and ACC, activation of the precuneus and cuneus was not significantly different for the “Do Swallow” and the “Don’t Swallow” conditions. This suggests that the cuneus and precuneus mediate sub-processes of the swallowing task that fall outside the domain of motor performance.
The cuneus and precuneus are known to be important in mediating a number of perceptual and cognitive processes. For example, the precuneus has been implicated in sequence processing (Catalan et al. 1998; Jenkins et al. 1994), multimodal integration of sensory inputs (Stephan et al. 1995; Grafton et al. 1992), and visuomotor integration. The cuneus is active during a visual texture perception task (Beason-Held et al. 1998) and both the cuneus and precuneus are active during visual searching (Donner et al. 2002). These findings suggest the possibility that activation of the parieto-occipital cortex during visually-cued voluntary swallowing may be due to the visual processing of the cue to swallow. It is also noteworthy, however, that activation within the cuneus and precuneus has been reported not only in association with visually cued swallowing, but also in relation to swallows cued verbally (Mosier et al. 1999), with a tactile cue (Kern et al. 2001a), or by administration of a water bolus (Hamdy et al. 1999b). Therefore, these regions may mediate multiple aspects of cue processing, regardless of the cue modality. Future studies that allow comparison of activation evoked by several different cue presentation modalities are necessary in order to fully resolve the functional role of the precuneus and cuneus in cued voluntary swallowing.
Interpretation of “Go, No-Go” study findings
A “Go, No-Go” paradigm was applied in the present study in an attempt to infer which brain regions are specifically involved in mediating swallowing motor performance in contrast to areas processing other aspects of the cued swallowing task. A potential limitation of this “Go, No-Go” application is the possibility that brain activation associated with uncontrolled variables specifically coupled with either the “Go” or the “No-Go” condition are erroneously attributed to swallowing or the processing of the cued task. For example, one could argue that activation within the parietal cortex seen during the “No-Go” condition may have arisen from either small oral movements, oral sensory processing, or even salivation that occurred as the subject began to focus on not swallowing. However, the laryngeal data showed no evidence of perilaryngeal movement during the “No-Go” condition, arguing against movement being responsible for activation during the “No-Go” condition. With regards to oral sensory processing, one would have expected activation within primary sensory regions, in addition to the activation seen within the precuneus, if oral sensation were the basis for the parietal lobe activation seen in association with the “No-Go” condition. Furthermore, unlike the visual sensory swallowing cue that was consistent across trials and tightly temporally coupled with the onset of the “No-Go” condition, oral sensory processing and salivation would be expected to be more variable across trials and, hence, less likely to be significantly correlated with the discrete events comprising the reference function employed in the present study. We conclude that the “Go, No-Go” paradigm is an effective method of differentiating cortical areas processing the act of swallowing from those processing aspects of the visually-cued experimental context.
This research was supported by a Ontario Ministry Health Career Scientist Award to REM, a Natural Sciences and Engineering Research Council (NSERC) grant (REM), a Canada Research Chair Support to RSM, and a CIHR Maintenance grant (RSM). The authors acknowledge the valuable contributions of Dr. Christopher Thomas in computer programming and data analysis.