The Detection of Wound Infection by Ion Mobility Chemical Analysis

biosensors Article The Detection of Wound Infection by Ion Mobility Chemical Analysis Emma Daulton 1 , Alfian Wicaksono 1 , Janak Bechar 2 , James A. Covington 1, * Joseph Hardwicke 2,3 1 2 3 * and School of Engineering, University of Warwick, Coventry CV4 7AL, UK; (E.D.); (A.W.) Warwick Medical School, University of Warwick, Medical School Building, Coventry CV4 7HL, UK; (J.B.); (J.H.) Department of Plastic Surgery, University Hospitals of Coventry and Warwickshire NHS Trust, Clifford Bridge Road, Coventry, CV2 2DX, UK Correspondence: Received: 6 February 2020; Accepted: 26 February 2020; Published: 29 February 2020



 Abstract: Surgical site infection represents a large burden of care in the National Health Service. Current methods for diagnosis include a subjective clinical assessment and wound swab culture that may take several days to return a result. Both techniques are potentially unreliable and result in delays in using targeted antibiotics. Volatile organic compounds (VOCs) are produced by micro-organisms such as those present in an infected wound. This study describes the use of a device to differentiate VOCs produced by an infected wound vs. colonised wound. Malodourous wound dressings were collected from patients, these were a mix of post-operative wounds and vascular leg ulcers. Wound microbiology swabs were taken and antibiotics commenced as clinically appropriate. A control group of soiled, but not malodorous wound dressings were collected from patients who had a split skin graft (SSG) donor site. The analyser used was a G.A.S. GC-IMS. The results from the samples had a sensitivity of 100% and a specificity of 88%, with a positive predictive value of 90%. An area under the curve (AUC) of 91% demonstrates an excellent ability to discriminate those with an infected wound from those without. VOC detection using GC-IMS has the potential to serve as a diagnostic tool for the differentiation of infected and non-infected wounds and facilitate the treatment of wound infections that is cost effective, non-invasive, acceptable to patients, portable, and reliable. Keywords: wound infection; gas analysis; diagnosis; VOC; GC-IMS 1. Introduction The skin is the largest organ in the human body and provides a critically important barrier to the external environment, allowing homeostasis and protecting against infection [1]. The skin is also important in temperature control and allows for the sensing of the external environment. Volatile organic compounds (VOCs) originate from skin structures (such as eccrine, apocrine, and sebaceous glands) as well as from skin commensal organisms [2,3]. VOCs are thought to be by-products of normal metabolic pathways [4]. Their production is usually dependent on species, strain, growth phase, pH, nutrients, and co-existing environmental conditions [5]. VOCs represent a heterogenous cohort of chemicals and include ketones, alcohols, esters, and sulphur compounds amongst others [6]. The five most common VOCs from human skin are 6-methyl-5-hepten-2-one, nonanal, decanal, geranylacetine and (E)-2- nonenal [7]. VOCs are potential biomarkers of occult disease. To date, these VOC patterns have been identified in patients with melanoma using electronic noses [8]. Biosensors 2020, 10, 19; doi:10.3390/bios10030019 www.mdpi.com/journal/biosensors Biosensors 2020, 10, 19 2 of 9 Chronic wounds are a significant financial burden to the National Health Service, with expenditure exceeding £1 billion per year [9]. A total of 50 million surgical procedures are performed in the United States, with surgical site infection incidence being as high as 5% with a high mortality association [10]. Open wounds resulting from trauma or surgery are of particular importance to clinicians because the loss of the primary barrier of skin makes them much more susceptible to infection. At present, the standard of care for large open wounds is to assess them primarily by their clinical appearance in addition to the examination of the host response to a possible infection. Wounds may also have an offensive odour resulting from either specific bacterial colonisation or from tissue necrosis. If there is a clinical suspicion of wound and/or surrounding skin infection (redness, swelling, heat, pain), empirical antibiotics therapy is commenced. However, there is no single objective point of care (POC) method of distinguishing between an infected malodorous wound and a non-infected colonized wound, and there are no clinical signs or immediate POC tests available to give a causative microorganism. For infected wounds, the current standard of care is empirical treatment with antibiotics in addition to specialist wound care. Antimicrobial therapy is commenced immediately based on a “best guess” approach to the common causative organism. The wound is swabbed, or tissue is taken for microscopy and culture, which typically takes between 48–72 h. This helps determine bacterial sensitivity or resistance to the empirical treatment. However, swab results may be misleading, as clinical microbiology laboratories use culture methods that select for planktonic bacteria or are not always suitable for anaerobic species. A wound culture also might not capture bacteria protected within a biofilm, so the result can be inconclusive [11–13]. The disadvantage with this current methodology is two-fold. First, patients with a truly infected wound may receive incorrect antibiotic therapy, resulting in the worsening of infection and increased hospital stay. Second, empirical antibiotics, either due to over-prescription or incorrect provision for colonised wounds, remain a major risk factor for antimicrobial resistance (AMR). In over prescription, the rise of easily transmissible genetic elements encoding resistance to last line antimicrobials raises the real possibility of a post-antibiotic era. This demonstrates a pressing need for a novel, fast and reliable POC test to guide correct antibiotic prescription in primary and secondary care. A potential solution to this is to detect and monitor the odours emanating from wounds and the dressings in contact with the wound. These gas phase biomarkers can be detected with a range of different analytical instrumentation. To this end, researchers have previously shown that high-end analytical instrumentation, such as GC-MS (gas chromatograph mass spectrometers—the gold standard for this purpose) is able to detect and identify infections [14]. However, such instruments are large, expensive, bulky, and require highly trained staff to use. For this reason, a number of researchers have used electronic noses—instruments that attempt to mimic the biological olfactory system—as a means of detecting these biomarkers [15]. Though they show considerable promise, they are unable to detect individual chemical components in a complex mixture. Furthermore, many of these instruments are formed from arrays of metal-oxide gas sensors that drift (...truncated)