Objective: To compare matrix metalloproteinase (MMP)-9 and the antiproteinase tissue inhibitor of metalloproteinases (TIMP)-1 in wound fluids and sera from patients with chronic non-healing or acute healing wounds. In addition, the functional confirfaut Jefferson G. on MMP-9 activity and general gelatinase activity were assessed. Method: In this observational study, samples were collected from patients with venous leg ulcers (VLUs), patients with type 2 diabetes with neuropathic foot ulcers (DFUs), and from another cohort of VLU patients with sterile split-thickness skin graft donor sites after autologous skin grafting, serving as healing control wounds. MMP-9 and TIMP-1 concentrations were determined by enzyme-linked immunosorbent assays. MMP-9 and gelatinase activities were determined in wound fluids in subsets of the patients. Results: A total of 24 patients took part in the study. No significant differences in MMP-9 wound fluid levels were found among the three groups. TIMP-1 levels were markedly and significantly lower in the two chronic wound groups resulting in a severely unbalanced MMP-9/TIMP-1 ratio, especially notable in the VLU group and possibly in the elevated endogenous MMP-9 activity (p<0.01) compared with the acute wound fluids. At least 20% of the chronic wound fluids displayed atypical patterns on gelatin zymography and showed high general gelatinase activity that was not inhibited by either TIMP-1 or by a gelatinase inhibitor (AG3340). MMP-9 levels were higher in the sera of the patients with type 2 diabetes. Conclusion: We hypothesise that non-MMP proteinases contribute to matrix destruction in a significant number of chronic wounds. Blocking the excessive MMP-9 activity may be insufficient to normalise wound healing. The reasons and effects of the very low TIMP-1 levels in chronic wounds need further clarification.