The thesis comprises literature reviews which present arguments novel to the field and two discrete but related studies, which in combination make a contribution to the classification, assessment of risk and prevention of pressure sores. The first study, a randomised controlled trial involving 446 patients undergoing vascular, general and gynaecology surgery, the use of a dry visco-elastic polymer pad intra-operatively reduced the probability of pressure sore development by half. Pressure sore incidence was 11 % (22/205) for patients allocated to the dry polymer pad and 20% (43/211) for patients allocated to the standard operating table mattress. Both studies explored key prognostic factors using multi-variate methods. Analysis of data derived from the randomised controlled trial found four factors to be independently associated with post-operative pressure sore development including intra-operative hypotensive episodes, Day I Braden mobility scale and intraoperative mean core temperature. The second study, a prospective cohort study involving 101 patients identified non-blanching erythema, pre-operative albumin, weight loss preceding admission and intra-operative minimum diastolic blood pressure. Results are consistent with findings from the literature review which identified key factors in the prediction of pressure sore development (reduced mobility, nutrition, perfusion, age and skin condition). The second study also explored the clinical significance of erythema in defining and classifying the term 'pressure sore'. Using laser Doppler imaging it was determined that blanching and non-blanching erythema are characterised by high blood flow of differing intensity. Discriminant analysis identified three general patterns in skin blood flow, which enabled scan classification with good agreement between clinical and predicted classifications. The results confirm data derived from the prospective observations of skin suggesting that non-blanching erythema is not indicative of irreversible ischaemic damage and resolves in approximately two thirds of cases. The point at which non-blanching erythema becomes irreversible remains unknown