Article Text
Abstract
Introduction Ulcerative colitis (UC) is a colonic inflammatory disorder of unconfirmed aetiology. Clinical assessment involves invasive endoscopic examination with a small yet significant procedural risk, on which therapeutic decisions are made. Non-invasive biomarkers may be better tolerated and reduce procedural costs and risks. Matrix metalloproteinases (MMP) are enzymes involved in tissue remodelling; MMP are elevated in mucosa and urine of children with active UC. We measured urinary MMP activity in adult patients with UC, matched controls investigated for functional symptoms and normal healthy volunteers to evaluate as a potential biomarker of disease activity.
Methods Ethical approval and informed consent were obtained. Patients with UC and age-sex matched controls, identified during outpatient assessment, were prospectively recruited and flexible sigmoidoscopy (FS) performed. Endoscopic (Sutherland) and histological (Gomes) appearance in patients with UC was graded. A group of healthy volunteers were recruited at a local University. Urine samples, snap frozen at collection in liquid nitrogen, were thawed, centrifuged, and assayed using commercially obtained fluorescein-labelled gelatinase activity kits. MMP activity was corrected for creatinine concentration. Results were expressed as median ± IQR. Statistical tests included Kruskal–Wallis analysis and Spearman's correlation.
Results 80 active and 16 quiescent UC patients, 77 age-sex matched controls and 22 normal healthy volunteers were compared. Urinary MMP activity in active compared with quiescent UC (p=0.185) and in each compared with age-sex matched controls (p=0.237, p=0.525 respectively) was not significantly different. Exclusion of patients taking 5-aminosalicylates and corticosteroids did not alter significance. There was no correlation between urinary MMP activity and UC disease activity measured endoscopically (r=0.09, p=0.425) or histologically (r=0.178, p=0.127). Urinary MMP activity in healthy volunteers was significantly lower than patients with active UC (p<0.0001), quiescent UC (p<0.002), and controls (p<0.0001).
Conclusion Contrary to previously published work, our findings suggest that urinary MMP activity, measured using fluorescein-labelled gelatinase assay, does not discriminate between quiescent and active UC and does not correlate with UC disease activity. Significant differences noted between healthy volunteers and patients with UC were unexpected, but may reflect difference in group demographics. MMP gelatinase assays are therefore a poor non-invasive biomarker of disease activity in UC.
Competing interests None declared.