Skin alterations of intact skin and risk factors associated with pressure ulcer development in surgical patients: A cohort study

Background: The pathology literature suggests three types of pressure ulcer with six possible mechanisms leading to tissue breakdown. A limitation of current evidence is the difficulty in replicating the clinical situation and in determining the point at which a tissue assault becomes irreversible and results in tissue breakdown. In particular clinical observations of alteration in darkly pigmented skin, blanching erythema, non-blanching erythema and non-blanching erythema with other skin changes including induration, oedema, pain, warmth or discolouration have not been assessed in relation to subsequent skin/tissue loss and their path ophysiological and aetiological importance is not fully understood. Objectives: To assess the validity of clinical signs of erythema as predictors of pressure ulcer development and identify variables which independently are predictive of >= Grade 2 pressure ulcer development. Design: Prospective cohort study. Participants: 109 general, vascular and orthopaedic hospital patients, aged over 55 years with an expected length of stay of >= 5 days were recruited. Of these 97 were pressure ulcer free at baseline and/or had complete follow-up including 59 women and 38 men with a median age of 75 years (range 55-95). Setting: Single centre large acute UK NHS hospital. Methods: To identify clinical signs of erythema predictive of skin loss, the odds of pressure ulcer development were examined using logistic regression. To identify variables independently predictive of >= Grade 2 pressure ulcer development logistic regression modeling was undertaken. Results: There was significantly increased odds of pressure ulcer development associated with non-blanching erythema (7.98, p = 0.002) and non-blanching erythema with other skin changes (9.17, p = 0.035). Logistic regression modeling identified non-blanching erythema, pre-operative albumin, weight loss, and intra-operative minimum diastolic blood pressure, as independent predictors of Grade >= 2 pressure ulcer development. Conclusions: Non-blanching erythema with or without other skin changes is distinct from normal skin/blanching erythema and is associated with subsequent pressure ulcer development. (c) 2006 Elsevier Ltd. All rights reserved.