Reproducibility and normal Variation of cephalic warm and cold detecti
on thresholds were investigated in three healthy subject groups. The f
ace, the mastoid process, and the hands were studied. No significant i
ntra-observer test-retest difference (n =20) was found. Good reliabili
ty (intra-class correlation coefficient [ICC]>0.4) was found for 13 of
14 measurements. A small significant inter-observer difference (n =20
) was found for cold thresholds. Good reliability (lCC>0.4) was observ
ed for both cold and warm thresholds in most of the test locations (6
of 8). In general, the largest variability was found in the mastoid an
d frontal lateral regions. Thermal thresholds varied with investigatio
n site in 56 controls (ANOVA, p<0.0005). No significant gender differe
nces were found for cephalic warm and cold thresholds. Most cold thres
holds (4 of 5) but also some warm thresholds (2 of 5) increased with a
ge at the cephalic sites. Our results reveal the frontal medial, the m
axillar medial, and lateral regions as the most reliable cephalic test
locations. The mastoid region may also be useful for investigating th
e upper cervical small-fiber function.