On the interpretation of temporal differences of BOLD fMRI responses to nociceptive stimulation.

On the interpretation of temporal differences of BOLD fMRI responses to nociceptive stimulation To the Editor: We have read with interest the paper by Becerra and colleagues in the January 2004 issue of the Journal of Neurophysiology (Becerra et al. 2004). In this article the authors use functional magnetic resonance imaging (fMRI) to investigate the temporal aspects of peripheral and CNS processing of nociceptive inputs. By applying noxious thermal stimuli to the dorsum of the hand and foot, they present two main findings. First, the hemodynamic response evoked in " primary sensory regions " by foot stimulation occurs with an average delay of 3.6 s compared with the response evoked when a matched stimulus is delivered to the hand of the same subject. Second, the response in " regions involved in emotions " does not show any temporal delay between hand and foot stimulation and precedes the response in somatosensory regions. We would like to comment briefly regarding these findings and their possible interpretation. Concerning the response observed in sensory regions, the authors affirm that a temporal delay of 3.6 s is consistent with the greater peripheral conduction distance after foot stimulation. Although one would expect to observe differences in the transit times for nociceptive information to reach the brain after hand and foot stimulation, the delay should be of the order of hundreds of milliseconds. This is based on the following considerations. The temperature of stimulation used by Becerra and colleagues (46°C applied to the skin for 29 s using a contact thermode) corresponds to the median thermal threshold of type II A␦ mechano-heat nociceptors (AMH-II units, thermal threshold 46 – 47°C, conduction velocity about 15 m/s), and is well above the median thermal threshold of C mechano-heat nociceptors (CMH units, thermal threshold of about 42°C, conduction velocity about 1 m/s) (Treede et al. 1995, 1998). Even if we assume that the somatosensory input provided by their stimulus was exclusively mediated by CMH nociceptors and corresponding unmyelinated primary sensory fibers, this would still not explain the large delay reported. In fact, considering a 30-cm difference in peripheral nerve conduction distance between arm and leg, a CMH-mediated somatosen-sory input delivered to the foot would arrive at the spinal cord about 300 ms after a similar hand input—and temporal delays in this range have been reproduced in several experiments (e.g., Opsommer et al. 1999; Towell et al. 1996; Tran et al. …