Diagnostic technologies for wound monitoring

Abstract

Chronic wound infections represent a worldwide problem, generating high morbidity and medical expense. Failure to control infections such as MRSA in the reparative process of a wound can cause disruption of normal anatomical structure and function, resulting in a chronic wound. Existing approaches to identifying infection largely involve surveying a range of physical parameters, and a limited use of non-invasive technologies. Evaluation is time consuming, and often results in inconsistencies in patient care.

This project researches three possible alternative methodologies/technologies for the monitoring of wounds, by measuring components of wound fluid. Two of the three technologies are designed to be used by physicians and patients, similarly to commercially available home blood glucose test kits, and are based on the measurement of three biomarkers: glucose, ethanol and H2O2 in PBS, and in serum as surrogate wound fluid. The first is a voltammetric technique known as dual pulse staircase voltametry (DPSV), which produces peaks characteristic of particular analytes at an electrode. The second is an amperometric biosensor array, based on screen printed three electrode assembies of carbon, rhodinised carbon (glucose biosensor only) and Ag/AgCl reference. The glucose biosensor uses glucose oxidase enzyme as the biorecognition agent, the H2O2 biosensor is a mediated system using horseradish peroxidase enzyme and dimethylferrocene mediator, and the ethanol biosensor is a bienzyme mediated system utilising alcohol oxidase enzyme horseradish peroxidase enzyme and coupled dimethylferrocene mediator. Wounds are known to produce characteristic odours, therefore the third technology studied is a single sensor odour analyser with advanced data analysis to detect five commonly occuring wound bacteria, S.aureus, K.pneumoniae, S.pyogenes, E.coli and P.aeruginosa in growth media and surrogate wound fluid. This technology would be used as a 'near patient' monitoring system and is based on machine olfaction similar to that of a commercial electronic nose, but uses a single metal oxide sensor in combination with principle components analysis.

DPSV scans of the individual analytes demonstrated distinctive peaks, exhibiting nonlinear relationships with concentration. A great deal of useful information was generated using this technique, however, limitations were discovered regarding repeatability and inter-analyte interference in mixtures. Limits of detection in surrogate wound fluid with the glucose biosensor, hydrogen peroxide biosensor, and ethanol biosensor were as follows: 169.5 µM glucose, 8.43 µM hydrogen peroxide, and 7.94 µM ethanol respectively (all at 99.7% confidence). Direct detection of ethanol from metabolically active S.aureus in surrogate wound fluid yielded a limit of detection of 1.23 x 108 CFU/ml at 99.7% confidence, and 19 µM in terms of ethanol specific response. The single sensor odour analyser demonstrated the ability to detect and discriminate between the three biomarkers, between five bacteria individually, and partial discrimination of paired bacteria (in broth and surrogate wound fluid). It was also found that S.aureus could be detected down to a cell density of 5x106CFU/ml in surrogate wound fluid, lower than that found for the biosensor concept.