The hypothesis that capillary blood sampling is made easier by warming
the heel was examined in a randomised, controlled trial of healthy ne
wborn infants. Sampling was performed using an automated lancet with o
r without prior warming. The time taken to collect a standard volume o
f blood, the number of repeat procedures needed, and the infants' beha
vioural responses were measured. Eighty one procedures were studied in
57 infants. Warming produced a median rise in heel skin temperature o
f 4.7 degrees C. However, there were no significant differences betwee
n the warmed and unwarmed groups in any of the outcome measures.Heel s
kin temperature is not an important factor in capillary blood sampling
. Attention should be directed towards improving sampling devices and
technique.