MMP-2 concentrations in stroke according to etiology: adjusting for enzyme degradation in stored deep-frozen serum and other methodological pitfalls

Matrix metalloproteinases (MMP) have a prominent role in the pathophysiology of stroke. We investigated potential differences in MMP-2 concentrations with respect to acute stroke etiology. For another MMP family member, MMP-9, significant degradation over time has been found even when stored at -80 °C, so we measured temporal degradation of MMP-2 and adjusted for this and other factors potentially affecting our results. For 264 patients with acute stroke at baseline and a control cohort of 120 subjects, MMP-2 concentrations were measured using commercially available enzyme-linked immunosorbent assay (ELISA) kits. For each stroke patient, stroke etiology was categorized as cardioembolic, large vessel or small vessel ischemic stroke, or primary hemorrhage. Stroke patients had significantly lower MMP-2 concentrations than controls (mean ± standard deviation: 175.6 ± 65.6 ng/mL versus 212.0 ± 54.8 ng/mL, p<0.001). However, sample degradation (average sample storage time: 240.0 ± 113.7 days) was considerable, amounting to approximately 15% per year. The full extent of differences in MMP-2 concentrations between stroke of different subtypes only became evident when results were adjusted for enzyme degradation during storage and other methodological pitfalls. Before adjustment, the only significant difference between etiologies was that the cardioembolic stroke group had a significantly higher concentration of MMP-2 than the hemorrhage group. After adjustment for time to analysis and ELISA plate clustering, patients with cardioembolic stroke had significantly higher MMP-2 concentrations in comparison to all other stroke subtypes.