Predicting delayed pressure ulcer healing using thermography: a prospective cohort study.

ABSTRACT

OBJECTIVE: To investigate whether thermography can be used to detect latent inflammation in pressure ulcers and predict pressure ulcer prognosis in a clinical setting. METHOD: For this cohort study, we recruited 35 patients with stage II-IV pressure ulcers on the torso, who underwent thermographic assessment on discovery of their pressure ulcer. The patients were followed up for at least 3 weeks. Thermography was performed immediately after dressing removal. Pressure ulcers were classified into two groups depending on whether or not the wound site temperature was lower or higher than the periwound skin the low temperature group and the high temperature group respectively. A generalised estimation equation was used to estimate the relative risk of delayed healing of pressure ulcers, comparing wounds with high temperatures and low temperatures. RESULTS: Of the 35 patients, 21 had 'low temperature' wounds and 14 had 'high temperature' wounds at baseline. Two patients in the high temperature group presented with overt infection, and were excluded from further analysis. Twenty-two pressure ulcers were considered to heal 'normally' (that is, the wound area reduced by 30% or more within 3 weeks) and 16 did not heal. The baseline DESIGN score (a measure of gross wound status) did not differ in any subscales between the high and low temperature groups. The relative risk for delayed healing in high temperature cases was 2.25 (95% confidence intervals; 1.13-4.47, p=0.021). Sensitivity was 0.56, specificity was 0.82, positive predictive value was 0.75, and negative predictive value was 0.67. CONCLUSION: Our results indicate that using thermography to classify pressure ulcers according to temperature could be a useful predictor of healing at 3 weeks, even though wound appearances may not differ at the point of thermographical assessment. The higher temperature in the wound site, when compared with periwound skin, may imply the presence of critical colonisation, or other factors which disturb the wound healing.