ناتوانی در ادراک بیماری برای بیهوشی : مطالعه نقشه برداری ضایعه علامت مبتنی بر وکسل

Abstract Brain-damaged patients affected by hemianaesthesia (i.e., the loss of tactile sensibility on the contralesional side of the body) may deny their deficits (i.e., anosognosia for tactile deficits) even reporting tactile experience when stimuli are delivered on the impaired side. So far, descriptive analysis on small samples of patients reported that the insular cortex, the internal/external capsule, the basal ganglia and the periventricular white matter would subserve anosognosia for hemianaesthesia. Here, we aimed at examining in depth the anatomo-functional nature of anosognosia for hemianaesthesia by means of a voxelwise statistical analysis. We compared two groups of left hemiplegic patients due to right brain damages differing only for the presence/absence of anosognosia for left hemianaesthesia. Our findings showed a lesional cluster confined mainly to the anterior part of the putamen. According to the current anatomical evidence on the neural basis of sensory expectancies, we suggested that anosognosia for hemianaesthesia might be explained as a failure to detect the mismatch between expected and actual tactile stimulation.

. Introduction Anosognosia (from the Greek nosos disease and gnosis knowledge; an-/a-is a negative prefix) is the lack of awareness for neurological/neuropsychological deficits following focal brain lesions. Such a denial behavior has been reported selectively for motor (e.g., hemiplegia), sensory (cortical blindness, hemianopia, hemianaesthesia), and cognitive deficits (see Prigatano, 2010 for a review), and it has been taken as evidence of modality-specific disorders of consciousness. Indeed, when different symptoms are simultaneously present due to a brain damage, patients may be unaware of one of them but aware of another revealing that the monitoring of different aspects of behavior is underpinned by discrete brain mechanisms (see Berti, Làdavas, & Della Corte, 1996 for details on this point). Within the sensory domain, anosognosia for hemianaesthesia (hereinafter AHA) is diagnosed when patients are persuaded that they are still able to perceive contralesional tactile stimuli despite the fact that, during the standard neurological examination with eye closed they never report of being touched on the affected side (Bottini et al., 2009, Marcel, 2004, Marcel et al., 2004, Vallar et al., 2003a and Vallar et al., 2003b). AHA patients may also report an actual tactile sensation when they see a stimulus delivered to their anesthetic body parts (Pia et al., 2013a and Romano et al., 2014). Such a subjective report seems to reflect a real subjective experience of touch rather than a mere verbal confabulation and/or a bias to simply report what is seen, because AHA patients may show normal physiological reactions (i.e., skin conductance response to incoming stimuli delivered to their anesthetic body part; Romano et al., 2014). Interestingly, tactile sensations arise when the physical counterpart is absent (Pia et al., 2013a and Romano et al., 2014). To the best of our knowledge, only two studies have directly examined AHA (Marcel et al., 2004 and Spinazzola et al., 2008). Both of them demonstrated that when hemianaesthesia (hereinafter HA) co-occur with hemiplegia (i.e., the complete paralysis of the contralesional side of the body; hereinafter HP), AHA can be dissociated from unawareness of HP (hereinafter AHP). In other words, patients can deny their contralesional somatosensory deficits but not their contralesional motor deficits and vice versa. Additionally, Spinazzola and coworkers (Spinazzola et al., 2008) analyzed the individual lesional pattern of four patients affected by AHA reporting that lesions to the insular cortex and to the basal ganglia were crucially associated to AHA (see also (Romano et al., 2014) for similar findings). It was suggested (Spinazzola et al., 2008) that brain damage would impair the ability to distinguish between an internal representation of the sensation and the actual perception of the physical stimulus. The false belief of being still able to perceive tactile stimuli would arise from the intact brain activity within spared areas of the somatosensory system. In the present paper, we aimed at obtaining a clearer anatomical picture of AHA in order to better understand the nature of the unawareness behavior. As first, we compared the lesional patterns of groups of right brain damaged patients differing only for the presence/absence of AHA. Secondly, on the bases of the anatomical pattern we draw inferences about the functional meaning of the damaged areas.

. Results Fig. 1A and B show the lesion overlapping of AHA_HP and HA_HP groups, respectively, whereas Table 2A and B report their quantitative estimate. The AHA_HP group showed a lesional pattern mainly involving the insula, the caudate nucleus, the posterior limb of internal capsule, the putamen, the superior corona radiata, the superior fronto-occipital fasciculus and the superior longitudinal fasciculus. The HA_HP group displayed a lesional pattern mainly involving the insula, the superior corona radiata and the superior longitudinal fasciculus. Overlays of regional lesion plots of the AHA_HP (A) and HA_HP (B) groups. The ... Fig. 1. Overlays of regional lesion plots of the AHA_HP (A) and HA_HP (B) groups. The frequency is represented trough a color scale ranging from black (lesion in one patient) to red (lesion in eleven patients). Subtraction of regional lesion plots (C). Regions damaged more frequently in the AHA_HP group respect to HA_HP group are displayed in warm colors, from dark red to white. MNI coordinates of each transverse section are reported. Figure options Table 2. Quantitative estimate of the regional lesion plots of the AHA_HP (A) and the HA_HP groups (B). Region overlaps in three patients or more in at least one of the two groups are reported. For each region, the percentage of lesioned voxels (Region %), the number of patients (Patients #) and MNI coordinates are reported, Quantitative estimate of the subtraction of regional lesion plots (C). Only brain regions that were damaged the 50% or more frequently in the AHA_HP group respect to HA_HP group are reported. For each region, the number of lesioned voxels (Voxels %), the percentage of lesioned voxels (Region %), the percentage of patients (Patients %) and MNI coordinates (MNI) are reported. Grey/White matter (A) AHA_HP (B) HA_HP Region % Patients # MNI Region % Patients # MNI X Y Z X Y Z Amygdala 67 5 31 0 −12 8 2 36 3 −25 Angular 93 5 42 −50 22 77 5 42 −50 22 Anterior_cor_rad 79 6 23 20 5 33 4 25 22 9 Anterior_limb_int_cap 92 7 22 7 13 71 4 21 −5 18 Calcarine 0 0 0 0 0 18 3 28 −69 6 Caudate 54 8 21 4 21 44 5 20 −12 21 Fornix 21 3 30 −16 −10 0 0 0 0 0 Frontal_Inf_Ope 95 6 39 17 5 97 6 48 12 0 Frontal_Inf_Orb 85 4 46 19 −12 77 5 52 19 −9 Frontal_Inf_Tri 89 5 46 21 0 87 5 47 19 3 Frontal_Mid 82 4 49 43 11 40 2 41 41 3 Frontal_Sup 80 3 24 48 23 14 2 30 3 47 Heschl 100 7 42 −16 6 100 6 41 −21 5 Hippocampus 21 4 36 −21 −6 3 2 35 −19 −8 Insula 100 9 31 −17 16 99 8 31 −16 20 Occipital_Inf 3 2 49 −74 −3 8 3 33 −81 0 Occipital_Mid 77 4 49 −67 25 80 3 33 −80 0 Occipital_Sup 22 3 34 −71 41 0 0 0 0 0 Olfactory 43 3 26 9 −12 0 0 0 0 0 Pallidum 100 7 28 −4 −3 56 3 28 −3 −5 Parietal_Inf 92 4 52 −48 38 31 3 37 −50 38 Postcentral 83 4 53 −6 25 68 5 60 −16 19 Posterior_corona_rad 30 7 27 −24 20 39 7 29 −24 23 Posterior_limb_int_cap 86 8 26 −11 18 55 6 26 −12 16 Posterior_tha 29 3 37 −52 −3 62 3 27 −72 5 Precentral 85 5 37 3 29 70 3 58 10 16 Putamen 100 9 24 6 4 84 6 36 1 −2 Retrolenticular_part 66 5 26 −21 2 42 3 37 −34 5 Rolandic_Oper 100 7 38 −4 13 98 7 42 −2 10 Sagittal_stratum 57 5 43 −28 −10 27 2 39 −18 −9 Superior_corona_rad 88 9 29 −18 20 77 8 27 −18 21 Superior_fronto-occi_fas 95 8 21 7 20 100 4 17 −6 19 Superior_long_fas 100 8 30 −4 19 100 8 31 −16 22 SupraMarginal 91 5 47 −42 22 76 6 46 −35 29 Temporal_Inf 46 4 56 −16 −18 30 3 53 −6 −27 Temporal_Gyurus_Mid 92 5 48 −8 −18 88 5 46 0 −16 Temporal_Pole_Mid 78 4 47 3 −17 75 5 52 15 −22 Temporal_Pole_Sup 89 6 44 4 −17 87 6 54 9 0 Temporal_Gyrus_Sup 94 7 41 −14 3 95 7 43 −9 −5 Thalamus 55 6 22 −19 13 20 3 22 −18 6 Uncinate_fasc 92 6 33 2 −14 0 0 0 0 0 Grey/White matter (C) AHA_HP minus HA_HP Region % Patients % MNI X Y Z Anterior_limb_int_cap 86 55 22 7 13 Pallidum 100 64 22 6 1 Putamen 98 73 24 12 0 Uncinate_fasc 92 55 33 2 −14 Table options Fig. 1C shows the lesion plot subtraction between the two groups, whereas Table 2C displays its quantitative estimate. The subtraction analysis showed that the anterior limb of the internal capsule, the pallidum, the putamen and the uncinate fasciculus were damaged at least the 50% more frequently in the AHA_HP group respect to HA_HP group. It is worth noticing that the reverse subtraction showed that no regions were injured at least 50% more in the HA_HP, respect to AHA_HP, groups. The Voxel-by-voxel test comparing the two groups with binomial data (Fig. 2A and Table 3A) revealed a cluster mainly involving the anterior putamen, with a minor involvement of the uncinate fasciculus (and a very small involvement of the pallidum, part of the lentiform nucleus together with the putamen). In the group of five pure AHA patients, the lesional pattern mainly involved the insula, the postcentral gyrus, the rolandic operculum, the superior longitudinal fasciculus, the supramarginal gyrus and the putamen (Fig. 2B and Table 3B). (A) Brain regions significantly associated to AHA (AHA_HP group vs HA_HP group). ... Fig. 2. (A) Brain regions significantly associated to AHA (AHA_HP group vs HA_HP group). All voxels which survived to the binomial test are displayed. The color scale represents Z-Libermesiter scores. (B) Overlays of regional lesion plots of the AHA group. The frequency is represented trough a color scale ranging from black (lesion in one patient) to red (lesion in five patients). Figure options Table 3. (A) Quantitative estimate of the brain structures significantly associated to AHA (AHA_HP group vs HA_HP group). For each brain structure, the number of clustering voxels, z score, and MNI coordinates of the center of mass are reported. Quantitative estimate of the regional lesion plots of the AHA group (B). Region overlaps in at least three patients or more are reported. For each region, the percentage of lesioned voxels (Region %), the number of patients (Patients #) and MNI coordinates are reported. Grey/White matter (A) AHA_HP vs HA_HP Region % z score MNI X Y Z Putamen 17 3.58 24 12 0 Pallidum 1 3.2 22 6 1 Uncinate_fasciculus 8 2.83 33 2 −14 Grey/White matter (B) AHA Region % Patients # MNI X Y Z Angular 55 2 40 −57 22 Anterior_limb_int_cap 42 2 13 11 −5 Caudate 19 2 13 11 −9 Frontal_Inf_Ope 85 2 62 14 4 Frontal_Mid 37 2 32 2 36 Heschl 99 2 43 −17 5 Insula 92 3 37 −28 22 Occipital_Mid 20 2 34 −62 33 Pallidum 47 2 14 10 −4 Parietal_Inf 67 2 44 −45 38 Postcentral 71 3 51 −21 31 Posterior_cor_rad 35 2 30 −50 19 Posterior_limb_int_cap 43 2 24 −22 10 Precentral 37 3 40 −14 36 Putamen 91 2 16 14 −10 Rectus 4 2 19 15 −11 Retrolenticular_part 48 2 25 −24 10 Rolandic_Oper 97 3 42 −34 22 Superior_corona_rad 53 3 30 −13 24 Superior_long_fas 96 3 32 −19 24 SupraMarginal 71 3 46 −36 23 Temporal_Mid 35 2 41 −57 21 Temporal_Pole_Sup 11 2 56 5 −3 Temporal_Sup 76 3 51 −38 22 Thalamus 18 2 23 −23 11