Nociceptive local field potentials recorded from the human insula are not specific for nociception

Liberati, Giulia [UCL] Klöcker, Anne [UCL] Maia da Cunha Oliveira Safronova, Marta [UCL] Ferrao Santos, Susana [UCL] Mouraux, André [UCL]

Introduction: The insula, and especially its posterior portion, is generally believed to play a specific role in nociception, as its injury can alter the perception of pain, and because electrical stimulation and epileptic seizures in this region may generate pain­related experiences (Garcia­Larrea 2012). Moreover, depth recordings in humans have shown that nociceptive stimuli elicit robust local field potentials (LFPs) in the insula, often regarded as pain­specific (Frot et al. 2014). Nevertheless, these observations are not sufficient to justify the conclusion that the posterior insula is specifically involved in the perception of pain, as this region is also involved in the processing of a wide range of non­nociceptive sensory inputs, and contributes to a large number of cognitive, affective and homeostatic functions (Cauda et al. 2012). Taking advantage of the high spatial resolution of direct intracerebral recordings performed in humans, we assessed whether the insula exhibits nociceptive­specific responses. Based on the results of previous EEG and functional magnetic resonance imaging (fMRI) experiments (Mouraux et al. 2009, 2011), and on findings showing multisensory responses in the insula (zu Eulenburg et al. 2013), we hypothesized that LFPs recorded in this region would not reflect nociceptive­specific activity, but instead, multimodal activity. Methods: Five patients (two females, mean age: 29) suffering from focal epilepsy were investigated using depth electrodes implanted at different locations, comprising the anterior and posterior insula, for a total of 70 insular sites. The experiment consisted of two sessions of four blocks, one session per side of stimulation (right and left side of the body). In each block, patients received stimuli belonging to one of four sensory modalities: thermonociceptive, vibrotactile, auditory, and visual. Each block comprised 40 stimuli, each lasting 40 ms. Nociceptive stimuli consisted of pulses of heat generated by a CO2 laser applied to the hand dorsum. Non­nociceptive somatosensory stimuli were delivered via a recoil­type vibrotactile transducer applied to the index fingertip. Visual stimuli were delivered by means of a light­emitting diode placed on the hand dorsum. Auditory stimuli consisted of loud, lateralized 800­ Hz sounds delivered through earphones. The order of the blocks was randomized across patients. Participants were asked to press a button as soon as they felt a stimulus, and to rate its intensity on a numerical scale ranging from 0 to 10. Results: All patients described the sensation elicited by the nociceptive stimuli as painful and pricking. However, the average ratings for stimulus intensity did not differ significantly across modalities (repeated­measures ANOVA, p=0.7). Nociceptive laser stimulation elicited LFPs at the same electrode contacts as non­nociceptive vibrotactile, auditory, and visual stimulation. All four types of stimuli elicited consistent LFPs in the posterior and anterior insula, appearing as large biphasic waves (Fig. 1). A blind source separation procedure based on a probabilistic independent component analysis (PICA) showed that the insular LFPs elicited by nociceptive stimulation can be entirely explained by multimodal neural activity also contributing to the LFPs elicited by non­nociceptive tactile, auditory and visual stimulation (Fig. 2). Conclusions: In contrast with current assumptions, our results indicate that insular LFPs elicited by transient nociceptive stimuli reflect multimodal cortical activities unspecific for pain. These responses could reflect mechanisms of attentional re­orientation towards salient stimuli, including, but not limited to, painful stimuli. This issue is particularly relevant, as these brain responses are commonly used to draw strong conclusions on how pain is represented in the brain (Bushnell et al. 2006; Tracey et al. 2007, 2008; Frot et al. 2013).