Eur J Appl Physiol 2004 (May); 91 (5-6): 516-524
Huygen FJ, Niehof S, Klein J, Zijlstra FJ
Department of Anesthesiology,
Pain Treatment Centre,
PO Box 2040, 3000 CA, Rotterdam, The Netherlands
The use of thermography in the diagnosis and evaluation of complex regional pain syndrome type 1 (CRPS1) is based on the presence of temperature asymmetries between the involved area of the extremity and the corresponding area of the uninvolved extremity. The interpretation of thermographic images is, however, subjective and not validated for routine use. The objective of the present study was to develop a sensitive, specific and reproducible arithmetical model as the result of computer-assisted infrared thermography in patients with early stage CRPS1 in one hand. Eighteen patients with CRPS1 on one hand and 13 healthy volunteers were included in the study. The severity of the disease was determined by means of pain questionnaires [visual analogue scale (VAS) pain and McGill Pain Questionnaire], measurements of mobility (active range of motion) and oedema volume.
Asymmetry between the involved and the uninvolved extremities was calculated by means of the asymmetry factor, the ratio and the average temperature differences. The discrimination power of the three methods was determined by the receiver-operating curve (ROC). The regression between the determined temperature distributions of both extremities was plotted. Subsequently the correlation of the data was calculated. In normal healthy individuals the asymmetry factor was 0.91 (0.01) (SD), whereas in CRPS1 patients this factor was 0.45 (0.07) (SD). The performance of the arithmetic model based on the ROC curve was excellent. The area under the curve was 0.97, the P value was <0.001, the sensitivity 92% and specificity 94%. Furthermore, the temperature asymmetry factor was correlated with the duration of the disease and VAS pain. In conclusion, in resting condition, videothermography is a reliable additive diagnostic tool of early stage CRPS1. This objective tool could be used for monitoring purposes during experimental therapeutic intervention.