This study aimed to examine the superiority of peroxidase detection of macroscopic observations using rat wounds, and to test the external validity of the peroxidase analysis in pressure ulcers (PU) in humans.In the animal study, rat wounds were analysed. A cross-sectional study analysed, by wound blotting, exudate samples from full-thickness PUs. Peroxidase activity was divided into two groups (ring and non-ring signals). Scores in the 'inflammation/infection' and 'necrotic tissue' components of DESIGN, a classification tool of PUs, were compared between the groups.In the animal study, 20 rat wounds were assessed and in the clinical study, 62 samples were collected from 26 full-thickness PUs of 21 patients aged ≥ 65 years. In the animal study, five of six wounds with clinical inflammation signs showed ring signal (defined as a signal on the wound edge and no signal on the wound bed). While the tissue sections of three wounds with a ring signal showed inflammatory features, they showed no clinical signs of 'inflammation/infection'. In the clinical study, which analysed 630 ring and 32 non-ring signals, 13 samples in the ring signal group and five in the non-ring signal group had 'inflammation/infection; scores of ≥1 (p=0.016). Despite having no clinical signs, 17 samples showed the ring signal.This study revealed the external validity of the wound blotting analysis of peroxidase and demonstrated its use to detect subclinical inflammation.