Voxel-based morphometry and arterial spin labeling fMRI reveal neuropathic and neuroplastic features of brain processing of itch in end-stage renal disease

Alexandru D. P. Papoiu, Nichole M. Emerson, Tejesh S. Patel, Robert A. Kraft, Rodrigo Valdes-Rodriguez, Leigh A. Nattkemper, Robert C. Coghill, Gil Yosipovitch


Pruritus of end-stage renal disease (ESRD) is a multifactorial symptom of complex etiology not yet fully understood. In this study we have investigated the cerebral perfusion patterns at rest in ESRD patients on hemodialysis, compared with those in healthy volunteers. We have also studied the brain responses evoked by experimental itch induction in ESRD, after stimulating the two distinct histamine and cowhage itch pathways, and compared them with the responses evoked in healthy volunteers. To identify potential structural alterations in ESRD patients compared with a group of age-matched healthy volunteers, we calculated the density of gray matter for the entire brain using a voxel-based morphometric analysis. Our results indicated that gray matter density was significantly reduced in ESRD patients in the frontal, parietal, temporal, and occipital cortices, as well as in the S1, precuneus, and insula, whereas the brain stem, hippocampus, amygdala, midcingulate cortex, and nucleus accumbens displayed an increased gray matter density. Functionally, we found a significantly higher brain perfusion at baseline associated with ESRD pruritus in the anterior cingulate, insula, claustrum, hippocampus, and nucleus accumbens. The brain responses evoked by cowhage itch, which are mediated by protease-activated receptors (PAR2), displayed significant differences compared with responses in healthy individuals and were correlated with perceived itch intensity in a dual, complex manner. The inverse correlations in particular suggested that a negative feedback mechanism modulated itch intensity, when elicited in a preexistent chronic itch background.

  • chronic itch
  • end-stage renal disease, histamine itch
  • cowhage itch
  • functional brain imaging (fMRI)
  • voxel-based-morphometry
  • cerebral perfusion changes

end-stage renal disease (ESRD) pruritus is a distressing symptom seriously affecting the quality of life of ∼60% patients diagnosed with chronic kidney disease undergoing hemodialysis (Pisoni et al. 2006). Its exact pathophysiology remains obscure, being considered multifactorial. Possible mechanisms include peripheral C-fiber neuropathy or an abnormal innervation, μ-opioid/κ-opioid imbalance, immune system dysfunction with a predominant TH1 proinflammatory response and skin barrier abnormalities (Patel et al. 2007). Ultimately, itch perception takes shape in the higher structures of the central nervous system, which can be affected in ESRD in multiple ways (Battaglia et al. 2005; Gambaro et al. 1987; Hsieh et al. 2009; Kalita et al. 2006; Krishnan 2006; Michaelis et al. 1996; Prohovnik et al. 2007; Savazzi et al. 2001; Tarhan et al. 2004; Tryc et al. 2011). Recent brain imaging studies have described the central processing of pruritus in healthy subjects as well as in chronic itch states, such as atopic dermatitis. Distinct cerebral areas were involved in processing itch in patients with atopic dermatitis compared with healthy subjects, revealing patterns associated with itch intensity and disease severity (Ishiuji et al. 2009; Schneider et al. 2008). Therefore, it is important to explore the brain processing of pruritus in patients with ESRD pruritus undergoing hemodialysis to better understand its central mechanism. Inducing acute itch in a preconditioned background, in patients habituated with a chronic itch experience, can provide insight into the functional status of networks known to process itch information. In addition, it may reveal adaptive mechanisms that could have been developed for coping with the unpleasant sensation of itch.

Two main neuronal transmission pathways have been described for itch: one mediated by histamine and the other by protease-activated receptors (PAR2) that can be exogenously stimulated by spicules of the tropical plant cowhage (Mucuna pruriens; Johanek et al. 2006; Namer et al. 2008). Upon skin contact, cowhage releases a cysteine protease, mucunain, that acts as a ligand for PAR2 and PAR4 receptors (Reddy et al. 2008), of which only PAR2 has been identified in the skin. These two itch modalities are conveyed via distinct peripheral and spinothalamic pathways, up to their thalamic (3rd neuron) stations (Davidson et al. 2007, 2009, 2012). Until recently, mostly histamine has been used as the experimental pruritogen for brain imaging of itch (Herde et al. 2007; Valet et al. 2008). However, because of a general lack of therapeutic efficacy of antihistamines in chronic itch, including ESRD pruritus, and the mediation of pruritus by PAR2 in conditions such as atopic eczema (Steinhoff et al. 2003), the cowhage-based model has been proposed as more relevant for chronic pruritus (Papoiu et al. 2011a). For these reasons, we employed both modalities of itch induction in our studies. In a recently published functional MRI (fMRI) study, we discovered common features, but also significant differences, in brain activation induced by the stimulation of these two itch pathways in healthy individuals (Papoiu et al. 2012). In the present study, we expanded our investigation to examine the networks, mechanisms, and patterns of brain activity evoked by these two itch modalities in ESRD patients on hemodialysis experiencing pruritus, and contrasted them with the patterns observed in healthy individuals. Differences in brain responses induced by the stimulation of these two itch pathways were also contrasted within the ESRD group.



Thirteen ESRD patients on hemodialysis with chronic itch (8 men, 5 women, ages 29–65 yr, average 50.9 ± 11.0 yr) and 15 healthy volunteers (8 women, 7 men, ages 20–42 yr, mean 31.0 ± 7.0 yr), all right-handed, who signed a written informed consent and complied with all study criteria, participated in this study. ESRD patients were on hemodialysis for an average of 51 mo and suffered with chronic itch of an intensity higher than 3 on a scale of 0 to 10. All procedures were approved by the Institutional Review Board of Wake Forest University School of Medicine and were conducted in accordance with the Declaration of Helsinki. Patients with antecedents of stroke, significant cerebral or systemic vascular pathology, gross structural cerebral alterations, abnormal ventricular space architecture, HIV, hepatitis B or C, neuropathy, depression, and significant cognitive deficits were excluded. A separate control group of 13 healthy volunteers, age and sex matched, was used for the voxel-based-morphometry (VBM) analysis (average age: 51.5 ± 12.7 yr).

Study Design

We employed a sequential arterial spin labeling (ASL) scan model consisting of two successive ASL scans of 390 s (6.5 min) performed immediately after itch induction, as described previously (Papoiu et al. 2012). Functional MRI data were acquired in baseline conditions (at rest) and immediately after itch stimulus administration. The stimuli used were histamine and cowhage, which were applied separately on the right (dominant) volar aspect of the forearm. Subsequent stimuli were applied to an area situated at least 5 cm away from the area used in a previous application, to avoid application on areas affected by alloknesis. When switching from one itch stimulus to another, a pause was taken to allow itch to completely subside (intensity ratings to return to 0).

Induction of Itch

Histamine itch was induced by iontophoresis to the volar aspect of the forearm. A 1% solution of histamine dissolved in 2% methylcellulose gel (Sigma, St. Louis, MO) was delivered using a current of 200 μA through a round iontophoresis electrode, 14 mm in diameter, for 30 s (PF 382b Perilont power device; Perimed, Järfälla, Sweden) as we previously reported (Papoiu et al. 2011a). After the itch sensation from histamine iontophoresis had completely subsided, another area at least 5 cm away, on the volar aspect of the forearm, was used for cowhage application. A total of 40–45 cowhage spicules were counted under a magnifying lens, picked up with a microtweezer, and applied within a 4-cm2 circular area on the skin. The spicules were gently rubbed for 30 s onto the subject's skin with a circular motion to facilitate contact; a cotton cloth was used to demarcate the area to prevent any stray spicules from stimulating surrounding skin.

Psychophysical Measurements

Subjects used a 100-mm visual analog scale (VAS) to rate itch intensity induced by the two stimuli at the end of each fMRI series.

fMRI Scanning Parameters

All images were acquired on a 1.5-T MRI scanner (GE Healthcare, Milwaukee, WI) using an eight-channel MR head coil array (Medrad, Warrendale, PA). The total imaging protocol consisted of localizer images acquired for graphical prescription, a high-resolution anatomic T1-weighted image, and six fMRI experiments (2 baseline, 2 histamine-induced itch, 2 cowhage itch). For the acquisition of structural high-resolution images, high-resolution T1-weighted anatomic images were used to identify regions of activation, classify tissues, normalize images to a standard space, and screen for abnormalities. T1-weighted structural scans were obtained using an inversion-prepared 3D spoiled gradient echo sequence with a matrix size of 256 × 192, field of view (FOV) of 24 × 18 cm, echo time (TE) of 1.9 ms, inversion preparation time of 600 ms, flip angle of 20°, slice thickness of 1.5 mm with no gap between slices, and 128 slices, giving an in-plane resolution of 0.94 mm.

T1 Map Image Acquisition of the Brain Tissue Signals

Accurate quantitative cerebral blood flow (CBF) maps require that the T1 of the tissue be measured at each voxel. T1 maps were calculated from data acquired from a separate inversion recovery (IR) echo-planar imaging (EPI) sequence. A global inversion C-shaped frequency offset corrected inversion (C-FOCI) pulse is used to minimize underestimation of T1 due to fresh spins perfusing into the imaging slice. Twelve inversion times (TIs) were acquired logarithmically from 10 ms to 6 s with a repetition time (TR) of 10 s by changing the order of slice acquisition. After a deliberate 6-s delay, the first volume was acquired without an inversion pulse to obtain an M0 image. Thirteen imaging volumes were then acquired in a total scan time of 2 min 10 s. All other imaging acquisition parameters (FOV, matrix size, TE, flip angle, slice thickness, slice location, etc.) are identical to the Q2TIPS-FAIR protocol as described in Acquisition of Cerebral Blood Perfusion Data. The T1 for each voxel was calculated by curve fitting the IR curve to a three-parameter decaying exponential model.

Acquisition of Cerebral Blood Perfusion Data

CBF was measured with quantitative imaging of perfusion using a single subtraction with thin-slice TI1 periodic saturation (QUIPSS II TIPS; also known as Q2TIPS) with flow-sensitive alternating inversion recovery (FAIR). The Q2TIPS-FAIR sequence is a multi-slice sequence that incorporates saturation pulses to minimize the uncertainty associated with tagged blood's transit time into the imaging slice. The saturation pulses in implementation of Q2TIPS are very selective suppression (VSS) of radiofrequency pulses, which are applied every 25 ms between 800 (TI1) and 1,200 ms (TI1s). The VSS pulses saturate a 2-cm slab of tissue with a 1-cm gap between the saturation slab and the first imaging slice. Our current implementation of the Q2TIPS-FAIR sequence uses a C-FOCI pulse (β = 1,361, μ = 6). A C-FOCI pulse is used instead of the standard adiabatic hyperbolic secant pulse to improve perfusion sensitivity by minimizing slice imperfections. The 11 oblique slices were prescribed parallel to the anterior commissure-posterior commissure line and were acquired sequentially, inferior to superior. This provided coverage from the vertex to the ventral aspect of the thalamus and prefrontal cortex. Other imaging parameters were as follows: TE = 28 ms, TI1 = 800 ms, TI1s = 1,200 ms, TI = 2,000 ms, TR = 3,000 ms, receiver bandwidth = 62.5 kHz, flip angle = 90°, FOV = 24 cm (frequency) × 18 cm (phase), 64 × 48 acquisition matrix (11 slices, 8-mm thickness, 0-mm slice gap), and anterior-posterior frequency encoding direction. A diffusion gradient with an equivalent b value of 5.25 mm2/s is added to suppress intra-arterial spins. The Q2TIPS-FAIR sequence was used to acquire 60 label-control pairs (slice-selective inversion/global inversion) in 6 min 30 s. These label-control pairs are pairwise subtracted and then averaged to generate a perfusion-weighted image. The first 30 s (5 label-control pairs) are used to establish steady state. During these 30 s, a single-shot EPI proton density (M0) image is acquired. This M0 image serves as an internal reference to scale the perfusion-weighted images appropriately to calculate absolute quantitative CBF maps according to the general kinetic model described by Buxton et al. (1998).

Statistical Analysis of Local CBF Signal Fluctuations

The functional image analysis package FSL [Functional Magnetic Resonance Imaging of the Brain (FMRIB) Software Library, Center for FMRIB, University of Oxford, Oxford, UK] was used for image processing and statistical analysis. The CBF data were movement corrected and spatially smoothed using a 5-mm 3D isotropic Gaussian kernel. Each CBF image was scaled by its mean global intensity (intensity normalization) to minimize variability due to global CBF changes. Next, each subject's CBF images were registered to their structural data using a 7-parameter linear 3D transformation and transformed into standard stereotaxic space (as defined by the Montreal Neurological Institute, MNI) using a 12-parameter linear 3D transformation. Standard general linear model-based analyses using fixed-effects models within subjects and mixed-effects models between subjects were performed to 1) identify significant effects evoked by histamine and cowhage itches in reference to a baseline (resting) state and to contrast them against each other, and 2) to test direct or inverse correlations of brain activation/deactivation with stimulus intensity. 3) The patterns of activation observed for both cowhage- and histamine-induced itch in ESRD patients on dialysis were compared with the results obtained in a group of healthy volunteers (n = 15) using mixed-effects (FLAME1+2) analyses. Analysis of brain activations induced by the same itch stimulus over time between ESRD and healthy subjects was performed using an unpaired t-test. Clusters of voxels exceeding a Z score >2.3 and P < 0.05 were considered statistically significant.

Contrast analysis of the processing of cowhage and histamine itch in ESRD vs. healthy volunteers was performed using an unpaired t-test. Linear regression analyses using VAS ratings of intensity as covariates of interest were employed to detect areas that might have a role in the modulation of itch perception. These areas were compared (overlaid) to mapped areas of activation or deactivation evoked by either itch modality.

Voxel-Based Morphometry

Structural data were analyzed with FSL-VBM, a voxel-based morphometry-style analysis (Ashburner and Friston 2000; Good et al. 2001) carried out with FSL tools (Smith et al. 2004). First, structural images were brain-extracted using BET (Smith 2002). Next, tissue-type segmentation was carried out using FAST4 (Zhang et al. 2001). The segmented gray-matter images were aligned to MNI152 standard space using the 12-parameter affine registration tool FLIRT (Jenkinson et al. 2001, 2002), followed by nonlinear registration using FNIRT (Andersson et al. 2007a, 2007b), which uses a b-spline representation of the registration warp field (Rueckert et al. 1999). The resulting images were averaged to create a study-specific template, to which the native gray matter images were then nonlinearly re-registered. The registered partial volume images were then modulated (to correct for local expansion or contraction) by dividing by the Jacobian of the warp field. The modulated segmented images were then smoothed with an isotropic Gaussian kernel with a sigma of 3 mm. Finally, a voxelwise general linear model (GLM) was applied by permutation-based nonparametric testing (randomise) using 5,000 permutations and corrected for multiple comparisons across space, with the statistical significance threshold set at P < 0.05.

Resting State Baseline Brain Perfusion Analysis

Two consecutive baseline series of 6-min 30-s duration each were acquired in immediate sequence for each participant (780 s = 13 min). Perfusion images were acquired by ASL. A total of 26 series acquired in ESRD patients were compared with 30 series acquired in healthy individuals. Using FSL software, perfusion maps were transformed into MNI152 standard space and analyzed by randomise inference method under a GLM design for an unpaired t-test. Statistical significance was set for P < 0.05. The model employed a threshold-free cluster-enhanced design (TFCE), with correction for multiple comparisons across space (the most conservative approach) using 5,000 permutations.


Resting State Brain Perfusion Analysis

Resting state brain perfusion analysis comparing baseline perfusion maps in ESRD patients and healthy subjects revealed that perfusion was significantly higher in ESRD in well-defined, discrete cerebral regions: anterior cingulate cortex (ACC), bilateral insula, and claustrum, areas that were previously described to be activated by acute itch. This is the first report that identifies a persistent activation at rest in areas known to be involved in itch processing, in a chronic pruritus condition. In addition, discrete clusters significantly activated in ESRD were found in a subcallosal gray matter region consistent with the location of nucleus accumbens, in the secondary somatosensory area (S2), and in superior temporal gyrus (Fig. 1, Table 1). Other discrete activations were mapped to hippocampus (enthorinal cortex), amygdala, and dentate gyrus.

Fig. 1.

Higher levels of brain perfusion were found at baseline in end-stage renal disease (ESRD) patients with chronic pruritus compared with healthy individuals in the insula, claustrum, anterior cingulate cortex (ACC), amygdala, enthorinal cortex, secondary somatosensory area (S2), and nucleus accumbens. STG, superior temporal gyrus. Arterial spin labeling functional MRI: P < 0.05. Montreal Neurological Institute (MNI) standard space coordinates (x, y, z).

View this table:
Table 1.

Brain areas with significantly higher cerebral perfusion at baseline in ESRD patients than in healthy subjects

A higher perfusion at baseline in healthy individuals was limited to cuneus, precuneus, occipital pole, visual cortex, and the head of the caudate nucleus. Notably, these areas did not overlap with regions where the dynamic physiological response to experimental cowhage itch was higher in healthy subjects than in ESRD patients (therefore, baseline perfusion differences could not account for these differences).

Functional Imaging of Brain Responses in ESRD Following the Stimulation of Two Distinct Itch Pathways

Cowhage itch.

In ESRD patients, cowhage itch induced a significant activation of rostral and pregenual ACC (Brodmann areas 24–25), middle frontal gyrus, caudate nucleus head, lateral globus pallidus, and putamen and presented a limited involvement of the somatosensory area S1, superior parietal lobule (SPL), precuneus, and insular cortex (Fig. 2). These findings are in notable contrast with healthy subjects, where the pattern was more extensive, bilateral activation that was scattered in the precuneus, posterior parietal cortex, supramarginal gyrus, angular gyrus, and posterior cingulate cortex (PCC). Importantly, there was a reduced activation of S1, precuneus, insular cortex, and PCC in ESRD, a feature that was not observed for histamine itch. These differences were significant in a direct contrast analysis by t-test (P < 0.05; Fig. 3).

Fig. 2.

A qualitative comparison of the significant responses evoked by cowhage itch in ESRD patients and healthy subjects. In ESRD patients (red-yellow), cowhage induced a significant activation of ACC (Brodmann areas 24 and 25), lateral globus pallidus, and putamen, and a limited involvement of the insular cortex, in contrast with healthy subjects (blue). The activation of precuneus appeared strikingly limited in ESRD. SMG, supramarginal gyrus; lat., lateral; occ., occipital. Z scores > 2.3; P < 0.05. MNI standard space coordinates (x, y, z).

Fig. 3.

Cortical activations evoked by protease-activated receptor 2 (PAR2) mediated cowhage itch in ESRD patients were significantly reduced in somatosensory area S1, superior parietal lobule (SPL), precuneus, cuneus, and posterior cingulate cortex (PCC) compared with healthy volunteers. Z score > 2.3; P < 0.05. MNI standard space coordinates (x, y, z).

Histamine itch.

Histamine itch in ESRD patients induced an extensive, contralateral activation in the insula, claustrum, ACC, hippocampal formations, and the primary and secondary somatosensory cortices. Although the patterns of brain activation evoked by histamine-induced itch in ESRD patients appeared to differ qualitatively from the pattern observed in healthy volunteers, a contrast analysis (unpaired t-test) did not reveal any significant differences in histamine itch processing between the healthy and ESRD groups. The pattern observed for histamine itch in ESRD patients resembled the pattern of cowhage-induced itch in healthy subjects, prominently involving the contralateral insula and claustrum.

Qualitative Comparison of Brain Activation by Cowhage and Histamine

A qualitative comparison of brain activation maps induced by cowhage and histamine in ESRD inspected by overlay appeared to suggest differences in brain processing. A contrast analysis of brain processing patterns of these two itch pathways within the ESRD group did not yield significant differences, although significant differences between histamine and cowhage itch processing patterns were found in healthy individuals (Papoiu et al. 2012).

Psychophysical Data on Perceived Itch Intensity and Their Correlation With Brain Activation

VAS retrospective ratings of itch intensity experienced during the MRI scanning sessions (taken immediately after the scan) showed that itch intensity ratings were higher for cowhage itch (5.42 ± 1.45, mean ± SD) than for histamine itch (3.09 ± 1.51; P < 0.01) but were not different from the ratings reported by healthy individuals (cowhage itch: 5.66 ± 2.06; histamine itch: 3.17 ± 1.14).

We identified significant correlations between the activation of cerebral areas involved in itch processing and perceived itch intensity (Tables 2 and 3). Linear regression analysis using VAS ratings as covariates of interest revealed that some of these areas coincided with the areas of activation, whereas other regions were distinct. The activations induced by cowhage itch appeared inversely correlated with perceived itch intensity in S1, SPL, S2, supramarginal gyrus, angular gyrus, precuneus, superior temporal gyrus, and ACC (Fig. 4), whereas in the insula*, claustrum*, and subcallosal gray matter*, brain responses were directly correlated with itch intensity. The areas marked with an asterisk were highly perfused at baseline in ESRD patients with chronic itch (Tables 1 and 4). A pattern of inverse correlations was also observed for the responses elicited by histamine itch in the hippocampus, midbrain, periaqueductal gray, inferior temporal gyrus, and the intracalcarine sulcus (Table 3); however, these areas were different compared with the regions linked with the intensity of PAR2-mediated itch.

View this table:
Table 2.

Brain clusters activated by cowhage itch in ESRD patients and their significant inverse correlations with itch intensity

View this table:
Table 3.

Histamine itch-evoked activations and their inverse correlation with VAS ratings of itch perception in ESRD

Fig. 4.

Cowhage itch evoked activations in S1, SPL, SMG, angular gyrus, precuneus, PCC, STG, depicted in red-yellow tones corresponding with their Z scores, overlapped to a significant extent with areas where responses were inversely correlated with itch intensity (blue) in the left cerebral hemisphere and coincided in S2 bilaterally. The overlap (purple) occurred predominantly in the left hemisphere, contralateral to the itch stimulation site, where pruritoceptive information (from the right forearm) is projected to the cortex. Z score > 2.3; P < 0.05. MNI standard space coordinates (x, y, z).

View this table:
Table 4.

Brain areas where responses evoked by cowhage itch were directly correlated with itch intensity

Gray Matter Changes in ESRD Patients

Using high-resolution structural images acquired in baseline conditions, a VBM analysis was performed to identify cerebral structures where the density of gray matter was significantly different in ESRD patients compared with that in a group of age- and sex-matched healthy subjects (Fig. 5, A and B). The density of gray matter was significantly and extensively decreased in ERSD patients in the prefrontal, temporal, parietal, and occipital cortices, S1, SPL, insula, and claustrum (Fig. 5A). The ESRD patient group also displayed areas of increased gray matter density in the brain stem, hippocampus, amygdala, thalamus, midcingulate ACC, and nucleus accumbens (Fig. 5B). These areas coincide anatomically with areas where an increased perfusion at baseline in the ESRD patients was found: in nucleus accumbens and the subcallosal gray matter, hippocampus, and amygdala, as well as certain subdivisions of the ACC (BA 23, 24, 25, and 32).

Fig. 5.

Brain areas showing significant changes in gray matter density in ESRD patients on hemodialysis, compared with changes in healthy subjects (P < 0.05) by FSL-VBM, a voxel-based morphometric analysis. A: areas of significantly decreased density of gray matter in ESRD were found in S1, SPL (bilaterally), the prefrontal, temporal, parietal, and occipital cortices, and also the subgenual and pregenual ACC, insula, and claustrum. B: areas displaying a significant increase in gray matter density in ESRD were mapped to the brain stem, hippocampus (Hc), amygdala, thalamus, ACC, and nucleus accumbens (NAc). IFG, inferior frontal gyrus; MFG, middle frontal gyrus.


On examination of the evoked cerebral responses induced by the stimulation of two distinct itch pathways in ESRD, only PAR2-mediated cowhage itch displayed significant differences in brain activation compared with responses in the healthy group, whereas histamine itch responses were not significantly different. Therefore, our findings point to an altered pattern of processing cowhage itch in ESRD. The response to cowhage itch evoked a reduced activation in S1, SPL, precuneus, insula, and ACC compared with healthy subjects. A structural basis of these differences could emerge from the significantly decreased gray matter density in these areas (Fig. 5A). This could represent signs of an underlying central neuropathy in ESRD, which affects cerebral structures involved in itch processing. Although the thinning of gray matter has been reported as a structural abnormality in ESRD patients on hemodialysis, in areas such as the caudate nucleus and the midbrain (Prohovnik et al. 2007), we hereby provide a systematic report where gray matter changes were mapped in the entire brain. Decreases in gray matter thickness have been observed also in chronic pain conditions, such as chronic back pain (Apkarian et al. 2004), complex regional pain syndrome (CRPS), spinal cord injury, and fibromyalgia (Henry et al. 2011). Interestingly, we have also found that the density of gray matter was increased in the brain stem, hippocampus, amygdala, and nucleus accumbens in ESRD patients. This change could emerge from an increased pruritoceptive input, since there is evidence that experience can change gray matter density in the brain. Teutsch et al. (2008) found that 8 consecutive days of repetitive noxious stimulation resulted in increases in gray matter in pain-related regions. In addition, a neuroimaging study in individuals with painful osteoarthritis of the hip before and after surgery showed that removing chronic pain through hip surgery led to a reversal in gray matter changes (Gwilym et al. 2010). In summary, both significant increases and decreases in gray matter density are noted to occur in different cerebral areas in ESRD patients experiencing chronic itch.

The significant brain activations in ESRD identified at baseline underscore the role of limbic system circuits: ACC, amygdala, and hippocampus (enthorinal cortex), as well as that of insula and claustrum, in itch processing in chronic disease and introduce nucleus accumbens as a new region of interest. The implication of nucleus accumbens, found to be overactive in ESRD patients at rest, is consistent with its role in depression and sleep disturbances, which have been known to affect patients undergoing hemodialysis, and may point, additionally, to a central dysfunction of the dopaminergic system in this disease. The link between nucleus accumbens and thalamocortical circuits that control sleep and wakefulness may add a new layer of understanding of the sleep loss, chronic suffering, and depressed mood characteristic of this condition. This hypothesis is consistent with our recent findings on the brain mechanism of itch relief induced by active (self) scratching (Papoiu et al. 2013), according to which the dopaminergic system of the midbrain (the ventral tegmentum, substantia nigra) together with nucleus accumbens are involved in the relief of itch induced by scratching. It is interesting to note that bilateral posterior insula and claustrum, recently described to play a distinct role in the processing of PAR2-mediated itch (Papoiu et al. 2012), were active at rest (highly perfused) in this chronic itch condition (Fig. 1).

Relationship Between Brain Activation and Perceived Itch Intensity

Itch intensity ratings in ESRD were not significantly different compared with those in healthy subjects for the two itch modalities; therefore, the ability to perceive itch does not appear affected in these patients. Of particular interest is that activation of certain discrete regions within somatosensory areas S1 and S2 was inversely correlated with itch intensity, which could be interpreted as a negative feedback into S1 and S2 exerted by other centers. Of note, these unique features of itch processing have not been observed in healthy individuals or in other chronic pruritus conditions we have studied, such as atopic dermatitis or psoriasis (Papoiu et al. 2011b).

We hypothesize that the reduced activation of S1 observed in ESRD in response to cowhage itch could reflect a tonic inhibition of PAR2-mediated itch. This could occur as a consequence of a preexisting PAR2 stimulation, considering that changes occurring in ESRD selectively impact the processing of cowhage itch. This may reflect a form of functional plasticity, developed in ESRD patients as an adaptation, in the context of reduced gray matter density. The thinning of gray matter has been proposed to act as a pressure factor inducing neocortical plasticity (Greenwood 2007). Plastic changes have been reported to occur in chronic pain states: a reduction of S1 receptive fields was documented in chronic back pain, and tonic inhibition compensating for persistent excitatory inputs was reported in CRPS (Henry et al. 2011). Therefore, it is likely that chronic pruritus is associated with similar neocortical changes. Previous studies suggested that neocortical maps of receptivity can be reduced as a consequence of constant afferent inputs (Barth 2002). Synaptic plasticity is manifested when inhibitory inputs to a cortical area are strengthened in response to greater excitatory drive in the same brain region. Evidence supports the notion that different cell types in the various layers of the neocortex are differentially activated by changes in sensory inputs and may play different roles in changing receptive field properties within primary sensory cortex (Bartley et al. 2008). It is consistent under this hypothesis that the limited activation of S1, SPL, precuneus, and insula evoked by cowhage itch in ESRD unmasked a preexisting adaptive inhibition at the neocortical level. Thus neocortical changes occurring in ESRD may differentially impact the processing of cowhage vs. histamine itch. The question is whether the gray matter changes are necessary and sufficient to induce these changes or whether they require in addition a persistent (pathological) overstimulation of PAR2. Interestingly, the overexpression of PAR2 receptors in the skin has been linked with ESRD pruritus (Kim et al. 2010).

Taken together, our structural and functional findings suggest that a neocortical functional reorganization occurs in several areas of the brain in ESRD, selectively affecting the processing of PAR2 mediated itch. These neocortical plasticity features are reflected in the unique functional activation patterns indicating an itch modulation mechanism not seen previously in healthy individuals or in other chronic itch conditions, such as atopic dermatitis or psoriasis.


This study was funded by National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant 5-R01-AR-055902.


No conflicts of interest, financial or otherwise, are declared by the author(s).


A.D.P.P., T.S.P., R.A.K., R.C.C., and G.Y. conception and design of research; A.D.P.P. and L.A.N. performed experiments; A.D.P.P., N.M.E., R.A.K., R.V.-R., L.A.N., and R.C.C. analyzed data; A.D.P.P., N.M.E., and G.Y. interpreted results of experiments; A.D.P.P. prepared figures; A.D.P.P. drafted manuscript; A.D.P.P. and G.Y. edited and revised manuscript; A.D.P.P., N.M.E., T.S.P., R.A.K., R.V.-R., L.A.N., R.C.C., and G.Y. approved final version of manuscript.


View Abstract