The lack of knowledge of blood perfusion distribution in a tumour and its surrounding tissues remains a major source of uncertainty of the induced temperature distribution during hyperthermia treatments. In addition, large blood vessels cool significant tissue volumes around them, making it very difficult to cover the whole tumour volume with therapeutic thermal exposure even with modern highly controllable hyperthermia systems. In this paper the earlier theoretical and experimental studies indicating the feasibility of inducing perfusion-independent thermal exposure using high-power, short ultrasound pulses have been continued. The purpose of this research was to investigate temperature elevation variations caused by tissue location, tissue interfaces, and large blood vessels through a series of in vivo experiments. Results show that high-temperature ultrasound hyperthermia does not suffer extensively from tissue heterogeneities, and even tissues within a few millimetres from an artery could be adequately exposed.