Objective
The problems of accurate in vivo thermometry during ultrasound exposures are well known. When metallic thermocouple probes are inserted into tissue the relative movement between the tissue and the thermocouple caused by the ultrasonic wave give rise to a viscous heating artefact which must be accounted for. In addition, the insertion of a needle into the tissue may reduce the cavitation threshold in that region, and lead to bubble mediated thermal effects. The aim of this study was to investigate these effects for different thermocouple types currently in common use.

Methods
Thin film (100 x 200 μm sensitive element (National Physical Laboratories, UK)), needle (320 μm diameter) and fine wire thermocouples (200 μm (made in house) and 260 μm (commercially available) diameter) were compared in fresh bovine liver ex vivo. The extent of the viscous heating artefact was determined by a subtraction technique, and by analysis of the heating and cooling curves. Passive cavitation detectors and B-mode ultrasound imaging were used to study the effects of cavitation on the heating produced.

Results
The presence of the thermocouple reduced the cavitation threshold in the tissue. In the presence of cavitation, an artifactually high temperature was recorded. This effect was least obvious for the needle thermocouples. Viscous heating artefact was seen for all thermocouple types with the exception of the thin film thermocouple and may account for more than 50% of the measured temperature rise. The 260 μm needle thermocouples appeared to exhibit the lowest viscous heating artefact.

Conclusions
Artifactually high temperatures may be measured in therapeutic ultrasound fields in tissue. These errors must be accounted for.