tradional and novel diagnostic techniques to assess wound infection
Marieke Haalboom is a PhD student in Department of Research Methodology, Measurement and Data Analysis (OMD). Her supervisor is prof.dr. J.A.M. van der Palen from the Faculty of Behavioural, Management and Social sciences (BMS).
In healthcare, we frequently encounter patients who suffer from one or multiple wounds that show a delayed or even failed healing process. A major complication in these patients is the development of wound infection, which occurs when pathogens outcompete the immune system. The activity of both pathogens and the immune system leads to further damage to tissues and therefore delays the healing process even more. The ability to detect wound infection in an accurate and timely manner is essential to prevent further complications, such as spreading of the infection to the blood stream or the need for amputation. In addition, the ability to accurately rule out infection in wounds can prevent unnecessary use of antibiotic treatment. The aim of this Thesis was to assess traditional and novel diagnostic techniques that potentially fulfill this need for accurate and timely detection of wound infection.
Traditional diagnostic techniques mainly rely on the subjective interpretation of clinical information and microbiological culture results. In chapter 3 we describe a study in which we wanted to determine whether microbiological culture results would differ between two well-known wound sampling techniques; wound swab and wound biopsy. We sampled the wounds of 180 patients with both techniques and cultured all wound samples according to standard clinical procedures. We did not find meaningful differences between culture results from wound swabs and wound biopsies. Swabs were able to identify all potentially pathogenic microorganisms cultured from biopsies in 73% of all wounds, while agreement for specific organisms varied between 87-98%. Therefore, there seems to be no direct need for invasive wound biopsies for microbiological culture in clinical practice.
The assessment of wound infection, however, requires interpretation of (clinical information and) microbiological culture results. Therefore, in chapter 4, we raise the question whether the assessment of wound infection differs between the availability of culture results from wound swab versus wound biopsy, in addition to clinical information about the wound. Culture results from the 180 patients included in the study described in chapter 3 were supplemented with clinical information and provided to a panel of 6 experts who independently assessed each wound as infected or not, separately for swab and biopsy. Overall, the assessment of infection did not significantly differ between the availability of culture results from wound swab versus wound biopsy. However, there were substantial differences between the assessments provided by the experts; experts agreed about the assessment in 34% and 42% of all wounds for swab and biopsy respectively. This variability between experts can complicate accurate detection of infection.
To overcome the pitfalls of traditional methods, several novel diagnostic techniques for the detection of wound infection have been developed. In this Thesis, we were able to evaluate three promising novel techniques.
In chapters 5 and 6, we assess a novel diagnostic technique that measures the activity of proteolytic and antimicrobial enzymes that are released during the immune response against pathogens. In chapter 5, we demonstrate, in a study with 81 patients with a variety of wounds, that elevated activity of myeloperoxidase (MPO), lysozyme and human neutrophil elastase (HNE) seems to be a promising marker for infection that provides results within 30 minutes. The study demonstrated that diagnostic accuracy was somewhat improved when the enzyme measurements were combined in one result. Therefore, we combined the measurements of the three enzymes into one diagnostic tool which we assessed in chapter 6. Although the tool provided even faster results (<20 min), diagnostic accuracy was lower than expected based on the results in chapter 5. This could be explained by changes made to the chemistry in the enzyme assays or the use of a different reference standard to determine wound infection status. The reference standard in this study was based on (subjective) assessments based on clinical and microbiological information, while the reference standard in chapter 5 was solely based on microbiological information. In absence of a perfect reference standard, it might be possible that the high number of false positive results of the diagnostic tool in chapter 6 is rather a representation of the ability to detect wound infection at an early stage, before signs and symptoms are observable.
In chapter 7 we evaluate the use of wound pH measurement as an easy, inexpensive and fast diagnostic technique for wound infection. We demonstrated in a study with 120 patients that with increasing wound pH, there was an increase in the proportion of infected wounds as determined by experienced clinicians. The pH of infected wounds was, on average, 7.2 while in non-infected wounds pH was 6.5. Increases in the activity of MPO, lysozyme and HNE were also associated with increased wound pH. However, further research is needed to substantiate the relation between infection and wound pH and to exclude alternative explanations.
In chapter 8 we explore the possibility to use an electronic nose, Aetholab, to detect wound infection. Aetholab measures volatile organic compounds from (patient) samples, while its advanced software package can be used to differentiate between samples with and without a specific disease. We used Aetholab to analyze wound swabs from 77 patients and found fairly good diagnostic accuracy over a variety of reference standards based on clinical judgment and microbiological culture results. Aetholab provided the same classification of wound infection as the reference standards in 71-87% of all wounds. These promising first results encourage the initiation of a larger study to improve and validate the algorithms used in the electronic nose. Moreover, we would like to explore the possibility to explore the use of alternative outcomes in Aetholab measurements, such as the detection of microorganisms.
The novel diagnostic techniques assessed in this Thesis do have the potential to detect wound infection in a fast and easy manner. Each tool requires a wound swab, which is a relatively easy, inexpensive and non-invasive method, and all techniques provide results within 20 minutes. Moreover, these novel techniques could have the ability to reduce the variability in the assessment of wound infection through traditional techniques. However, it is difficult to determine the accuracy of novel diagnostic techniques when there is no measure to establish the true infection status of the wound. Nevertheless, we tried to determine the diagnostic accuracy of each novel technique with the best reference standard available. We aim to verify our results in larger studies in which we might also use alternative methods to determine diagnostic performance. In the meantime, we hope our results create awareness amongst clinicians about the imperfections that exist in diagnostic techniques they currently use. Although there remains uncertainty about the diagnostic accuracy of novel techniques, it might be valuable to use these techniques to support assessment of wound infection in clinical practice.