Besides physiologic mechanisms, the role of psychological factors in the perception of breathlessness has been recognized,>> but research on this topic is still at the beginning. To the present time, negative emotions have been shown to be associated predominantly with decreased accuracy of dyspnea percep-tion. Furthermore, a repressive-defensive coping style might be related to blunted symptom percep-tion,> but some findings have not been fully conclusive. Psychopathologic characteristics such as hypochondriasis or somatoform tendencies have not been shown to influence the sensing airflow limitations, and there is at present no conclusive evidence that distinct personality profiles predispose a person to the inaccurate perception of dyspnea.> However, an important influence might arise from learning processes or contextual factors and lead to the overperception of respiratory sensations. Alternatively, an attention-distracting context has been shown to reduce the awareness of breathlessness, which might be an effective intervention in patients with some conditions (eg, COPD).
Cortical Representation of Dyspnea
Despite a growing understanding of the possible pathways leading to breathlessness, relatively little is known about higher brain centers in humans that process this sensation. In particular, the brain areas associated with the perception of the experience have not been well-explored. This is in part attributable to a lack of adequate animal models properly simulating human dyspnea perception and, furthermore, is due to an absence of high-resolution imaging techniques, which allow a nonin-vasive study of human brain activity.
Introduction to Findings Without Imaging Techniques
Early research” on experimental animals by means of evoked potentials, which were recorded with cortical surface electrodes after electrical or mechanical stimulation of different respiratory affer-ents, has demonstrated that afferents from airways and respiratory muscles project to the cerebral cortex in cats and monkeys. Prominent activations have been found in the somatosensory cortex, in the motor cortex and in the mesocortex. Results like these have suggested a role for higher cerebral involvement in respiratory sensations besides the pontomedullary respiratory oscillator. Canadian Health&Care Mall is the Internet supplier of drugs.
Activation of the somatosensory cortex in adult humans by means of RREPs was first shown by Davenport and coworkers. These RREPs were recorded after short inspiratory occlusions by means of scalp surface electrodes placed over the somatosensory region of the cortex measuring electroence-phalic activity below the surface, which is similar to the evoked potential techniques in other sensory systems using EEGs. Short states of breathlessness predominantly evoked early RREPs, namely, the primary positive voltage peak (P1), the primary negative voltage peak (N1), the secondary positive voltage peak (P2), and the secondary negative voltage peak (N2), which occur within about 100 ms (P1 and N1) or 200 ms (P2 and N2), after stimulus onset.” P1, P2, N1, and N2 are peaks in the EEG signal due to dipoles occurring when a cerebral column is depolarized by the arrival of activity from respiratory afferents activated by the inspiratory occlusion. Thus, the obtained components represent the arrival and first processing of respiratory-related afferent sensory information in the somatosensory cortex. A subsequent study demonstrated a positive correlation between the amplitude of P1 and the perceived magnitude of the respiratory load. Other studies conducted by means of percutaneous electrical or transcranial magnetic stimulation have provided further evidence for fast-conducting afferent and efferent connections between the cerebral cortex and respiratory muscles.
Another study by Davenport and coworkers using the RREP methodology is important with regard to the reported blunted perception of breathlessness in subgroups of asthmatic patients, specifically those with a history of near-fatal attacks., In a group of asthmatic children with a history of life-threatening asthma, they found an absence of the P1 component after respiratory occlusion (ie, the dyspneic sensory signal was not activating the somatosensory cortex), These results suggest a deficit in the neural processing of information related to breathlessness, which in turn might be an important mechanism underlying blunted symptom perception. Further evidence for this assumption has arisen from a study by Webster and Colrain, which was performed by means of RREPs. They demonstrated reduced late tertiary positive voltage peak components (P3) in asthmatic adults when compared to healthy control subjects following midinspiratory occlusion. Most interestingly, these reductions in the P3 amplitude were also obtained after a short auditory stimulus, whereas early P1 components showed comparable amplitudes in both groups. The data suggest the presence of an asthma-specific deficit in the later cortical processing of respiratory load information. A rather speculative interpretation of the findings would be that of a general deficit in the cortical processing of perceptual information in specific groups of asthma patients. However, more research is clearly needed to explore these first results further.
Imaging Techniques and Dyspnea
The development and increasing availability of high-resolution imaging techniques have provided a great tool for the noninvasive study of brain structures in the conscious human. Specifically, positron emission tomography (PET) scanning and functional MRI (fMRI) have been used in a vast variety of scientific contexts. Both methods measure regional cerebral blood flow, which is increased locally during neural activity. While PET scanning requires the IV application of a nuclear tracer (eg, H2O), fMRI employs the blood oxygen level dependence effect to contrast areas of different cerebral blood flow. However, compared to other sensory systems, the proportion of and interest in imaging studies examining sensations of breathlessness is markedly reduced.
The first studies in this field have focused primarily on aspects of volitional breathing or compensation of induced breathlessness, while the perception of dyspnea has not been systematically examined. Volitional breathing in the PET or fMRI scanner has been achieved by voluntary targeted breathing, either with or without added respiratory loads, and has been compared with unloaded spontaneous breathing or passive mechanical ventilation. The predominant activation has been obtained bilaterally in the primary motor cortex (M1), in the right premotor area (PMA), in the supplementary motor area (SMA), in the cerebellum, and in the thalamus. Some studies have reported several further activations in regions of the pontomedullary respiratory oscillator, in sensorimotor areas, in anterior cingulate structures, in the prefrontal cor-tex,’ and in the parietal cortex. These findings have demonstrated that higher motor cortex areas are involved in volitional breathing and in the compensation of breathlessness. McKay and coworkers concluded that the voluntary control of breathing is similar to other voluntary movements, and requires the activation of an integrated network of cortical and subcortical areas.
Further work has studied the effects of increased inspiratory CO2 on brain activity, which, as described above, is one known source leading to dyspnea. But again, a systematic examination of the perceived sensation of breathlessness has not been provided. However, besides activity in cerebellar, frontal, and occipital regions, hypercapnia was predominantly associated with activations in the limbic system (eg, the hypothalamus, hippocampus, cingulate cortex, and insula). No activations were seen in motor cortex areas that have been shown to be associated with volitional breathing. These results further suggest a participation of brain areas above the pon-tomedullary respiratory oscillator in the processing of sensory information related to breathlessness.
At present, only four imaging studies have been published in six reports, with three reports explicitly examining the perception of dyspnea, and different aspects of the fourth study having been reported in three different reports. Only one of these studies employed fMRI, which offers a higher resolution compared to PET scanning.
In the studies by Liotti et al, Brannan et al, and Parsons et al, breathlessness was induced in nine healthy volunteers by the inspiration of increased CO2 (8%) using either a facemask or a mouthpiece, Data from these subjects were contrasted with those of subjects having several other conditions, of which the episodes occurring without dyspnea that were due to facemask breathing of increased O2 (91%) and inspiration of room air (O2, 21%) are the most relevant for the present review. Evans and cowork-ers induced breathlessness in six healthy mechanically ventilated participants by restraining the tidal volume below spontaneous levels in combination with constantly elevating arterial Pco2 levels by manipulating inspired Pco2. This methodology was also used in the study by Banzett and coworkers, which included eight healthy volunteers. Both studies compared dyspneic conditions with episodes of higher tidal volume combined with normal arterial Pco2, which relieved breathlessness. Thus, all three studies stimulated chemoreceptors or induced increases in the respiratory motor drive leading to sensations that were verbally expressed as air hunger, urge to breath, and like breathholding. In contrast, Peiffer and coworkers applied external resistive loads during inspiration and expiration in eight healthy participants. These loads were introduced to a breathing circuit to induce moderate-to-severe breathlessness and were removed during unrestricted control conditions while the volunteers breathed at spontaneous levels. An additional load condition including the inhalation of menthol in order to reduce the dyspneic sensation revealed no prominent effects. The resistive-load technique predominantly stimulates respiratory mechanorecep-tors, which, as already described, can lead to breathlessness due to increases in perceived respiratory effort and work.
While subjects in all studies were lying in the supine position in the scanner, the different interventions were contrasted with episodes that occurred without breathlessness (ie, isocapnia with normal tidal volume or load-free breathing, respectively). Furthermore, all studies continuously monitored respiratory responses in airflow, volume, end-tidal Pco2, and mouth pressure to control the effectiveness of experimental stimulation. The perceived degree of breathlessness was assessed directly after interventions with appropriate rating scales and was compared to episodes with unrestricted breathing. Some studies further obtained verbal descriptions of the quality of the perceived sensation. Although the induction of breathlessness and the assessment of respiratory responses in the spatially limited, magnetically sensitive scanner environment are complicated procedures, all four studies overcame these restraints with appropriate designs. However, the small number of included participants is a shortcoming across all of these studies.
Despite the use of different intervention techniques, common activations in several brain areas have been demonstrated in at least three of the four studies. Predominant neural activity has been found in the insula, in insular agranular extensions (eg, operculum and frontal cortex areas), the anterior cingulate cortex (ACC), the posterior cingulate cortex (PCC), the cerebellum, the thalamus, and the amygdala. Further activations have been observed in the M1, the PMA, the SMA, and somatosensory areas, which were all involved during voluntary breathing. Additionally, neural activity has been observed in the pons, the putamen, the hypothalamus, the hippocampus, and the frontoparietal network.
Among all structures, the anterior insula, a multifunctional sensorimotor integration area, showed strong activations in all four studies, predominantly in the right hemisphere. This fact leads to the assumption that the insula is the crucial component in a larger cortical network underlying the perception of breathlessness. Like the ACC and the amygdala, the insula is part of the limbic network, and has dense connections to other limbic, somatosensory, and motor structures. Studies in rats have further demonstrated that the insula receives afferents from respiratory chemoreceptors, from mechanoreceptors, and from projections from the medulla, which underlines its important role in the integration of respiratory sensations. Moreover, data have shown altered activity or response timing patterns within several cortical areas, including the insula, cerebellum, ACC, and hippocampus in patients experiencing obstructive sleep apnea syndrome when compared to control subjects. These patterns were obtained during Valsalva maneuvers, which are associated with breathlessness due to prolonged expiratory effort against a high load. These findings further suggest the presence of neural deficits in the processing of respiratory sensations in patients with pulmonary diseases, specifically in insular, cingulate, and cerebellar areas.
A Cortical Scheme of Dyspnea Perception
Summarizing the work conducted on the different input mechanisms, transmitting pathways, and processing brain areas involved in sensations of breathlessness, a preliminary scheme for the cortical substrates underlying the perception of dyspnea can be derived (Fig 1). For a detailed review of the afferent and efferent pathways between peripheral receptors and the brain, the reader is referred to two excellent reviews.
Two major pathways have been suggested to process respiratory sensations to the cortex. The first pathway arises predominantly from respiratory muscle afferents, is relayed in the brainstem medulla, and projects to the ventroposterior thalamus area, from where thalamocortical projections ascend to the primary and secondary somatosensory cortex. In accordance with other interoceptive sensations, these structures might process the sensory or intensity aspects of dyspnea. The second pathway includes mainly vagal afferents from the lungs and airways, which are relayed in the brainstem medulla. Brainstem projections ascend to the amygdala and medial dorsal areas of the thalamus, and further to the insula and cingu-late cortex. This predominantly limbic pathway might further include the hippocampus, operculum, putamen, and other prefrontal areas, and might be more associated with the affective components of the experienced breathlessness. Both pathways include final projections to the higher motor cortex (ie, M1, PMA, and SMA), from where efferent motor commands project to the brainstem and/or respiratory muscles. The cerebellum might receive afferents from the pon-tomedullary respiratory oscillator in the brainstem or from the higher motor cortex, as a similar cerebellar activation has been shown during volitional breathing. Alternatively, the cerebellum might also be involved in affective functions of breathlessness, an idea that has been suggested for other primary sensations.> However, a clear differentiation between specific sensory and affective functions of brain areas associated with dyspnea still has to be explored.
Figure 1. Cortical areas involved in the perception of dyspnea. This preliminary scheme is based on the first results of neuroimaging studies and includes two major pathways signaling afferent information from peripheral receptors to the cortex, which have been derived from information from earlier psychophysical studies. The first pathway (black line) arises mainly from respiratory muscle afferents, and the second pathway (dashed line) includes mainly vagal afferents from the lungs and airways. Embedded brain areas represent only their proximate localization. AMYG = amygdala; CB = cerebellum; MDT = medial dorsal thalamus; MO = medulla oblongata; PFC = prefrontal cortex; PPC = posterior parietal cortex; S1 = primary somatosensory cortex; S2 = secondary somatosensory cortex; VPT = ventroposterior thalamus.