Since the number of organisms isolated from a medical device is crucial in assessing the likelihood of device-associated infection, we examined whether incubation of catheters in trypsin before sonication can increase the yield of superficially colonised vascular catheters in vitro and those removed from patients. Polyurethane and silicone catheters were individually colonised in vitro with individual clinical isolates including Staphylococcus aureus and Escherichia coli. Equal numbers of 1 cm segments of colonised catheters were then individually incubated either in a trypsin-containing solution or a control solution without trypsin. Each solution containing the segment was then sonicated and cultured quantitatively. In the clinical arm, indwelling catheters removed from patients were also cut into 1 cm segments that were equally suspended in the trypsin-containing or control solution and then sonicated and cultured quantitatively. Trypsin-based sonication enhanced the detection of S. aureus on colonised polyurethane and silicone catheters in vitro by 14- and 30-fold, respectively (P = 0.03 and P = 0.04), and the detection of E. coli on colonised polyurethane and silicone catheters by 3- and 6-fold, respectively (P = 0.04 and P = 0.05). Compared with sonication alone, trypsin followed by sonication resulted in 10% increase in the detectability of significant colonisation of indwelling catheters removed from patients and 11% increase in the mean colony counts of colonising organisms (P = 0.04). Exposure of catheters to trypsin before sonication improves the sensitivity of sonication and enhances the accuracy of assessing significant catheter colonisation.