P 23 fMRI-mapping of somatosensory finger representation in complex regional pain syndrome

Objective Somatosensory deficits are often reported in patients with complex regional pain syndrome (CRPS). In particular reduced spatial tactile resolution is a typical clinical finding in CRPS patients. The detailed underlying mechanism for development and preservation of these deficits is not yet fully understood. However, a systematic review of studies on pain representation and cerebral representation showed a smaller representation of the hand in the primary somatosensory cortex (S1) contralateral to the affected upper limb ( Di Pietro et al., 2013 ). Associations between S1 hand representation size, pattern of pain and clinical deficits in patients with CRPS are suspected, but studies using high resolution fMRI are rare. Methods In a group of 13 patients diagnosed with CRPS type 1 (CSS: 13.1 ± 9.8; age 54 ± 10 years; handedness: 90 ± 11) and unilateral affection of the upper limb, the thumb (D1) and pinky (D5) representation in Brodmann area (BA) 3b of S1 was determined with our new high resolution fMRI protocol ( Pfannmoller et al., 2016 ). The surface distance between D1 and D5 was used as a measure for the size of the hand representation. In order to investigate the impact of the hand representation size on behavior measures, associations with pain intensity (VAS pain intensity: 4.0 ± 2.5), pinch grip performance (Roeder), and two-point-discrimination (TPD D1) as a measure for the spatial tactile resolution were analyzed. Results Overall, the affected and non-affected hand differed in motor function (Roeder: t (12) = 2.36; p  = .036) and tactile resolution (e.g. TPD: t (12) = 2.65; p  = .021). Patients with higher D1-D5 distance showed better TPD ( r  = −.62; p  = .024). We found no association between D1-D5 distance and patient’s pain intensity ( r  = −.39; n.s.). Our optimized evaluation revealed no difference between D1-D5 distances of the affected and non-affected side. Without quality optimization the non-affected hand (20.9 ± 4.3 mm) was larger than the affected hand (16.2 ± 3.5 mm) and this difference was highly significant ( t (10) = 4.85; p  = .001). However, the non-optimized distances showed no associations to the behavioral measures. Conclusion Our data on an association of D1-D5 S1-distance with spatial tactile resolution are in good agreement with other studies( Pleger et al., 2006 ). The only difference to the existing literature is the similarity of the representation size of non-affected and affected hand. Since our optimized evaluation decreases the statistical significance of the results we assume that this finding is due to our sample size and that the difference between the hand representations may be recovered if more patients are included.