Hand and finger skin temperatures in convective and contact cold exposure

The present study aimed at investigating the spatial variability of skin temperature (Tsk) measured at various points on the hand during convective and cold contact exposure. A group of 8 subjects participated in a study of convective cooling of the hand (60 min) and 20 subjects to contact cooling of the finger pad (5 min). Experiments were carried out in a small climatic chamber into which the hand was inserted. For convective cold exposure,Tsk was measured at seven points on the palmar surface of the fingers of the left hand, one on the palmar surface and one on the dorsal surface of the hand. The air temperature inside the mini-chamber was 0, 4, 10 and 16°C. With the contact cold exposure, the subjects touched at constant pressures an aluminium cube cooled to temperatures of −7, 0 and 7°C in the same mini-chamber. ContactTsk was measured on the finger pad of the index finger of the left hand. TheTsk of the proximal phalanx of the index finger (on both palm and back sides), and of the middle phalanx of the little finger was also measured. The variation ofTsk between the proximal and the distal phalanx of the index finger was between 1.5 to 10°C during the convective cold exposure to an air temperature of 0°C. Considerable gradients persisted between the hand and fingers (from 2 to 17°C at 0°C air temperature) and between the phalanges of the finger (from 0.5 to 11.4°C at 0°C air temperature). The onset of cold induced vasodilatation (CIVD) on different fingers varied from about 5 to 15 min and it did not always appear in every finger. For contact cold exposure, whenTsk on the contact skin cooled down to nearly 0°C, the temperature at the area close to the contact skin could still be 30°C. Some cases of CIVD were observed in the contact skin area, but not on other measuring points of the same finger. These results indicated that local thermal stimuli were the main determinents of CIVD. Representative hand skin temperature may require five or more measuring points. Our results strongly emphasised a need to consider the large spatial and individual variations in the prediction and modelling of extremity cooling.