c consult author(s) regarding copyright matters licenseeprints.qut.edu.au/54360/2/54360.pdf ·...
Post on 22-Aug-2020
1 Views
Preview:
TRANSCRIPT
This may be the author’s version of a work that was submitted/acceptedfor publication in the following source:
Dargaville, Tim, Farrugia, Brooke, Broadbent, James, Pace, Stephanie,Upton, Zee, & Voelcker, Nico(2013)Sensors and imaging for wound healing: A review.Biosensors and Bioelectronics, 41, pp. 30-42.
This file was downloaded from: https://eprints.qut.edu.au/54360/
c© Consult author(s) regarding copyright matters
This work is covered by copyright. Unless the document is being made available under aCreative Commons Licence, you must assume that re-use is limited to personal use andthat permission from the copyright owner must be obtained for all other uses. If the docu-ment is available under a Creative Commons License (or other specified license) then referto the Licence for details of permitted re-use. It is a condition of access that users recog-nise and abide by the legal requirements associated with these rights. If you believe thatthis work infringes copyright please provide details by email to qut.copyright@qut.edu.au
License: Creative Commons: Attribution-Noncommercial-No DerivativeWorks 2.5
Notice: Please note that this document may not be the Version of Record(i.e. published version) of the work. Author manuscript versions (as Sub-mitted for peer review or as Accepted for publication after peer review) canbe identified by an absence of publisher branding and/or typeset appear-ance. If there is any doubt, please refer to the published source.
https://doi.org/10.1016/j.bios.2012.09.029
Sensors and Imaging for Wound Healing: A Review
Tim R. Dargaville1*, Brooke L. Farrugia1, James A. Broadbent1, Stephanie Pace2,
Zee Upton1 and Nicolas H. Voelcker2
1. Tissue Repair and Regeneration Program, Queensland University of Technology,
Institute of Health and Biomedical Innovation, 60 Musk Avenue, Kelvin Grove,
Queensland, Australia 4059
2. Mawson Institute, University of South Australia, GPO Box 2471, Adelaide, South
Australia, Australia 5001
* corresponding author: t.dargaville@qut.edu.aau
Keywords: wound healing, sensors, diagnostic, theranostic, chronic wound,
infection, imaging, point‐of‐care, dressing, smart bandage
Abstract (150250 words)
Wound healing involves a complex series of biochemical events and has
traditionally been managed with 'low tech' dressings and bandages. The concept
that diagnostic and theranostic sensors can complement wound management is
rapidly growing in popularity as there is tremendous potential to apply this
technology to both acute and chronic wounds. Benefits in sensing the wound
environment include reduction of hospitalization time, prevention of
amputations and better understanding of the processes which impair healing.
This review discusses the state‐of‐the‐art in detection of markers associated
with wound healing and infection, utilizing devices imbedded within dressings
or as point‐of‐care techniques to allow for continual or rapid wound assessment
and monitoring. Approaches include using biological or chemical sensors of
wound exudates and volatiles to directly or indirectly detect bacteria, monitor
pH, temperature, oxygen and enzymes. Spectroscopic and imaging techniques
are also reviewed as advanced wound monitoring techniques. The review
concludes with a discussion of the limitations of and future directions for this
field.
Table of Contents
1. Introduction
2. Potential markers
3. Sensors for detection of infection
3.1 Sensors targeting bacterial biochemistry
3.2 Porous silicon sensors
3.3 Odor Sensors
3.4 Temperature as an indicator for infection
4. Sensors for detection of pH changes
4.1 Importance of pH
4.2 pH sensors
5. Moisture
6. Spectroscopy
7. Wound imaging
8. Conclusions and Future Directions
1. Introduction
In the last century, there have only been a handful of technical advances that
have contributed to changes in the discipline of wound management. One of the
most important was in the 1960s when it was found that keeping a wound moist
accelerates the healing processes. This was reported in the pivotal study by
Winter in 1962 (Winter 1962) and later became accepted practice and indeed a
key design parameter in the development of dressings (Wu et al. 1995). The
other crucial developments have been the management of infections in wounds
through the use of anti‐microbial agents, most notably, silver and iodine (Mertz
et al. 1999; Wright et al. 2002), the use of compression pressure therapy (for
chronic venous leg ulcers) (Wong et al. 2012), skin grafts (Rizzi et al. 2010) and
hyperbaric oxygen therapy (breathing 100% oxygen at elevated pressure)
(Malda et al. 2007). Each of these breakthroughs has resulted in new commercial
products. For instance, there are now numerous different hydration‐controlling
dressings available (Queen et al. 1987; Queen et al. 2004); while for infection
control dressings impregnated with silver have become ubiquitous in wound
management (Wright et al. 2002) (Figure 1). Despite these advancements and
the wide range of dressings available, wound management is still extremely
challenging due to its subjectivity, the complexity of the wound healing process
itself, and patient variability.
There is now growing evidence to suggest that we are currently on the
verge of the next significant advance in wound care where sensors will be used
as diagnostic tools in wound healing to revolutionize wound care practice. The
main types of wounds which will benefit most from sensor technology are
chronic ulcers, and to a lesser extent infected acute wounds and large full‐
thickness burns. Chronic ulcers can be especially difficult to treat, highly
susceptible to infection and can cause long‐term suffering for the patient. Sensor
innovation in the management of these wounds has the potential to impact
clinical practice, patient outcomes and economic policy.
Chronic ulcers are a substantial cause of morbidity and societal burden
worldwide. Foot ulcers, for instance, occur in approximately 25% of diabetics
and are the leading cause of non‐traumatic amputation in developed countries
(Turns 2011). In the United States, chronic wounds as a whole are estimated to
affect approximately 1–2% of the population during their lifetime (Gottrup
2004), translating to an estimated morbidity of 6.5 million sufferers at a cost of
US$25 billion per annum (Crovetti et al. 2004; Singer and Clark 1999). Equally
concerning is the rate at which chronic wound incidence is increasing due to
lifestyle changes and the aging population. Together these phenomena pose a
substantial threat to the future of health provision and the management of
national economies (Sen et al. 2009). Wound management technologies and
strategies are underdeveloped and need to evolve to meet this present and
increasing challenge. Accordingly, reflecting on current practice may identify
avenues for innovation and improvement.
Currently, when a patient presents with a wound there are a series of
standard steps that will be followed by the clinician. Initially, their past medical
history, including cardiovascular profile and peripheral vascular disease, and the
wound history would be established together with a physical examination of the
patient and of their wound (Turns 2011). This may be followed by an array of
tests including physical examination and imaging of the wound. In a best‐case
scenario once the wound aetiology and the current status of the wound have
been established, a management plan will be put in place. This can involve
swabbing for detection of infection, preparation of the wound (cleaning and
possibly debridement), dressing of the wound and coverage with bandages.
The time taken from initial presentation of a wound to the
commencement of a wound management plan may be lengthy and require
multiple appointments with a clinician, as any laboratory tests ordered can take
hours or days to complete. Once treatment has commenced the patient would
then be re‐evaluated at subsequent appointments which can be days to months
apart. These cumulative delays and follow‐up appointments stall the
administration of appropriate treatment and lead to increased cost. This is
critical in terms of chronic wound management, where it has been shown that
the longer the delay in treatment, the more difficult a wound is to heal (Margolis
et al. 2000; Moffatt et al. 2009). As such, the use of rapid, specific and
quantitative assessments, that can be completed during a standard medical
consultation, would be better suited to wound management.
Point‐of‐care (POC) technologies are designed to provide rapid medical
assessments at, or near, the site of patient care. These technologies are well
suited to wound management due to their designed ease‐of‐use, convenience
and rapid turnaround. In terms of wounds, POC assessment has the potential to
ensure that effective management, based on relevant biochemistry, is
administered without delay, rather than after the fact. This type of assessment is
further suited to wound management due to the abundance of assayable
biochemistry found within the wound environment such as proteolysis and
reduction‐oxidation events. Tests could be modelled from the modern pregnancy
test whereby a simple readout is diagnostic of a particular wound parameter.
Indeed, this has been the approach of Systagenix in the development of a POC
device for evaluating elevated protease activity through their Woundcheck™
Protease Status diagnostic (Serena et al. 2011) (Figure 1). While POC
technology, such as that developed by Systagenix, is well‐suited to wound
management, technological innovations are paving the way for the development
of next‐generation wound assessment tools.
An area of growing research which could revolutionize wound care is in
‘smart dressings’. Similar to the emerging field of “wearable” or “epidermal”
electronics (Kim et al. 2011), smart dressings use biochemical cues to generate
readable output which has diagnostic or theranostic value. The difference
between these two terms is that a diagnostic will give an accurate, non‐
subjective decision‐making output (e.g. the wound is or is not infected), whereas
a theranostic may help guide treatment (e.g. the wound has a certain level of
matrix metalloproteinases) (Harding 2007). Diagnostics would be aimed at the
deskilled care‐giver, whereas theranostics would be useful for experienced
wound specialists and researchers. ‘Smart dressings’ have the potential to
expedite the diagnosis of wounds through the use of biosensors either
incorporated into or near wound dressings, similar to POC dip‐stick‐type tests, to
generate a result in minutes. When applied as continuous or semi‐continuous
monitoring devices, smart dressing readouts may also be used to resolve
predictive trends. The ability to not just monitor wound healing but also to
predict it will help ensure potentially problematic wounds receive the
appropriate attention at the earliest possible stage. This information may be
used to help guide wound practitioners to adopt a more or less aggressive
management strategy, at the right time, to achieve healing.
As ‘smart dressings’ and POC diagnostics/theranostics are on the verge of
becoming a commercial reality, it is an opportune time to review the literature in
order to examine the wide range of approaches being used by multiple
disciplines to apply sensor and imaging methods to the wound environment.
While there has been extensive research into sensors for biomedical
applications, this review will be restricted to technologies pertaining to wound
healing.
2. Potential markers
The complexity of the wound healing process means there ought to be an
abundance of possible diagnostic and theranostic targets. A list of markers under
investigation for use in wound healing was recently published by Harding and an
expert working group in “Diagnostics and wounds – a consensus document”
(Harding 2007) and is reproduced in Table 1 below.
Table 1. List of potential markers for wound healing (Harding 2007).
Bacterial load/specific microbial species/biofilms Cytokine release in response to specific microbial antigens DNA – e.g. gene polymorphisms to indicate susceptibility to disease, poor
healing or infection Enzymes and their substrates – e.g. matrix metalloproteinases and
extracellular matrix Exposed bone Growth factors and hormones – e.g. platelet‐derived growth factor
(PDGF), sex steroids (androgens/oestrogens), thyroid hormones Immunohistochemical markers – e.g. integrins, chemokine receptors and
transforming growth factor beta II receptors to monitor healing status Inflammatory mediators – e.g. cytokines and interleukins to monitor
healing status and guide use of anti‐inflammatory treatments Nitric oxide Nutritional factors – e.g. zinc, glutamine, vitamins pH of wound fluid Reactive oxygen species Temperature Transepidermal water loss from periwound skin
What is noticeable about this table is that it relatively general and highlights the
importance of continued research into better understanding the specific markers
involved in wound healing. This is not to say that the search for markers of
chronic wound healing has been stagnant. Many of the biochemical components
found in Table 1 are underpinned by numerous clinical investigations involving
the analysis of wound tissues and exudates (Broadbent et al. 2010). These
investigations detail the analysis of over 150 proteins and metabolites and,
despite limitations in sample number or exclusion criteria (Liu et al. 2011), are
the foundation of our current understanding of chronic wound biochemistry.
Noteworthy examples include the identification of over‐abundant proteases and
their therapeutic and prognostic potential (Cao et al. 2011; Liu et al. 2009), and
identification of growth factor dysregulation (Cooper et al. 1994) and the
potential to restore healing through their inactivation (Streit et al. 2006).
Regardless of these successes, wound investigations are yet to result in the
identification of specific and quantitative targets for use in clinical practice (note
that the Systagenix elevated protease activity diagnostic is not currently used in
clinical practice). Concomitantly, chronic wound biochemistry is generally still
regarded as controversial (Liu et al. 2011).
Recent literature has provided new insights into mechanisms taking place
in the chronic wound environment. Work by Fernandez et al. (2012) has
demonstrated a correlation between uric acid concentration and wound severity
in clinically uninfected wounds. The authors further showed the concurrent
depletion of uric acid precursors and demonstrated the ex vivo activity of the
responsible enzyme, xanthine oxidoreductase, in wound fluid. (Hoffmann et al.
2011) investigated the association of the protease inhibitor, α1‐
antichymotrypsin, with wound healing and was able to show a functional
relationship between its inactivation and wound healing. Both of these
investigations outline mechanisms for inflammatory mediation and offer
potential targets for the development of innovative sensing devices. In addition
to these autogenic molecular targets, complications such as infection have been
sufficiently examined in other medical contexts to be the focus of sensor
development for wound management.
3. Sensors for detection of infection
One of the single most common complications preventing wounds from healing
is infection. Staphylococci and Streptococci are the two prevalent opportunistic
pathogenic organisms found in community‐acquired superficial wounds and are
also common to many chronic wounds (Davies et al. 2001) known to harbour
diverse microbial populations (Dowd et al. 2008). There are many obvious signs
of advanced infection including redness, heat, swelling, purulent exudate, smell,
pain, systemic illness, “foamy” granulation tissue, contact bleeding, tissue
breakdown and epithelial bridging (Grey and Harding 2006) but the challenge is
to detect infection before it reaches this stage.
The current method for characterization of infection is usually taking a
swab from the wound site. The swab is then analysed in a microbiology
laboratory for bacterial growth (e.g. semi‐quantitative culture) and graded on a
scale ranging from “scanty” through to “heavy”. There is frequent prejudice
towards motile and fast growing organisms while fastidious organisms, such as
anaerobes, may be under‐represented (Healy and Freedman 2006). Swabs can
often only detect superficial pathogenic organisms and fail to indicate what is
happening deeper in the wound. Other analyses might include gram stain and
quantitative culture. Most wound swabs are prone to false positives as most will
yield bacterial growth which is not always due to infection. Treatments may
include antiseptic topical treatment with silver compounds or iodine or systemic
treatment with antibiotics. While designed to inhibit growth of micro‐organisms,
the compound‐loaded dressings can also lead to impairment of functions
important to wound healing. For instance, the silver‐loaded dressing Acticoat has
been shown to cause impairment of skin cell proliferation and function in
addition to its desired microbicidal activity (Poon and Burd 2004).
There is a clear need for methods to enable early detection of infection to
guide the use of antibiotic treatments. Moreover, indication of when treatment is
not required would be valuable in reducing costs and overuse of antibiotics and
antibacterial substances (e.g. silver) leading to resistant strains and delayed
healing, respectively (Burd et al. 2007). The large number of markers for
infection has led to many diverse approaches by researchers to use sensors
either incorporated into dressings or as POC tools.
3.1 Sensors targeting bacterial biochemistry
Bacteria secrete a variety of tell‐tale biochemical by‐products. For
instance, pyocyanin ‐ the blue‐green pigment secreted by most Pseudomonas
(Ps.) aeruginosa strains, has been used by Sharp et al. (2010) to indirectly detect
bacteria. Using a carbon fibre tow consisting of several thousand carbon
filaments with diameters of 10 m thermally sandwiched into a lamination
pouch, they were able to observe the oxidation and reduction events of
pyocyanin using square wave voltammetry. This allowed them to reproducibly
quantify its presence in buffered solutions over a concentration range of 1 – 100
M. Furthermore, they were able to monitor pyocyanin in broth culture of Ps.
aeruginosa using the carbon fibre electrodes and observed good correlation with
a standard chloroform‐acid photometric method for quantification. Advantages
of this method include a small voltage scan range which limits possible
interference from endogenous compounds, small/cheap electrode materials
which can be incorporated into existing dressings or bandages and the
possibility for early detection of colonising Ps. aeruginosa. The obvious
disadvantages are that it requires a power source and an electrochemical
detector.
Urate or uric acid is another metabolite found within wounds which may
be used as a diagnostic marker for colonization of S. aureus and Ps. aeruginosa. It
is believed these bacteria rapidly metabolize uric acid via uricase synthesis,
therefore a rapid decrease in uric acid concentration within wound fluid may be
a generic indicator for bacterial colonization. In a similar approach to the
pyocyanin sensor above, Sharp et al. (2008) used a carbon fibre sensor laminate
and square wave voltammetry to detect the oxidation of urate at +0.23 V as a
sharp peak free from interference from other metabolites. Successful
measurements in blood and blister fluid were also reported although problems
were encountered with loss of signal over time due to electrode fouling. This was
partially solved by using a cellulose acetate barrier around the electrodes to
filter out large proteins and fats thought to be responsible for the fouling. While
generally applicable to acute wounds, this diagnostic test may suffer in its
application to chronic wounds. Work by Fernandez et al. (2012) recently showed
that chronic venous leg ulcers, with no clinical sign of infection, actually had
elevated uric acid concentration. Furthermore, its concentration correlated
positively with wound severity. The increase in uric acid in severe chronic
wounds and the decrease in infected wounds clearly indicates the biochemical
complexity of this approach.
The toxins produced by bacteria found in wounds may be used to trigger
a sensor output indicating early infection. Inspired by the mechanisms in which
toxins phospholipase A2 and ‐hemolysin from S. aureus and P. aeruginosa
permeabilize or hydrolyze cell membranes, Zhou et al. reported synthetic
Trojan‐like phospholipid vesicles, which are disguised as eukaryotic cells but
release a fluorescent dye cargo when acted on by these toxins (Figure 2) (Zhou
et al. 2010; Zhou et al. 2011). They achieved this by synthesizing phospholipid‐
based giant (ca. 50 m) unilamellar vesicles containing the dye
carboxyfluorescein which were stabilized by UV crosslinking of a
photopolymerizable fatty acid and immobilized onto surface‐modified
polypropylene fabric. Proof‐of‐principle experiments incubating the fabric‐
bound vesicles with P. aeruginosa and S. aureus demonstrated emission of green
fluorescent light on irradiation with low intensity UV light while the control
incubated with E. coli, which does not produce the lysing toxins, did not produce
significant fluorescence. This same approach was used to release the
antimicrobial sodium azide from the vesicles (Zhou et al. 2010), further
increasing their use as a responsive dressing. This work according to the authors
is continuing and is now investigating sensitivity to a variety of strains of P.
aeruginosa and S. aureus (Figure 2) (Zhou et al. 2011).
Neutrophils are a cell type other than the bacteria themselves that are a
marker for infection. In normal wound healing, neutrophilic granulocytes are
rapidly recruited to the site of injury where they release enzymes to purge the
wound of necrotic tissue. Once the wound has been cleansed, the neutrophil
activity should cease. Infection also stimulates neutrophils and while they may
help fight off infection there is evidence that they prevent wound closure
through the degradation of important growth factors and ECM proteins (Yager et
al. 1997). Two of the enzymes excreted by neutrophils, namely, human
neutrophil elastase (HNE) and cathepsin G (CatG), have recently been reported
as early stage warning markers for wound dressing‐based diagnostics by
Hasmann et al. (2011). They used an approach similar to an earlier report by
Edwards et al. (2005) which exploited the enzymatic nature of HNE and CatG
towards peptide substrates to produce a sensor made from a chromophore
linked by a short peptide sequence susceptible to cleavage by each of the
enzymes. The chromophore/peptide combinations used were N‐
methoxysuccinyl‐Ala‐Ala‐Pro‐Val‐p‐nitroanilide (MeOSuc‐AAPV‐pNA) and N‐
Succinyl‐Ala‐Ala‐Pro‐Phe‐p‐nitroanilide (Suc‐AAPF‐pNA) which are hydrolyzed
by HNE and CatG, respectively. These were immobilized onto several typical
wound dressing materials: silica gel, collagen, collagen/hyaluronic acid (with
and without thiol groups), polyamide and polyethylene terephthalate. These
dressings were incubated with fluid from either infected or non‐infected wounds
and monitored by UV absorbance for detection of the cleaved p‐nitroanilide. In
all cases the fluid from infected wounds led to greater absorbance measurements
which can be correlated back to increased HNE or CatG (silica gel only)
concentrations for the respective peptide sequences. It was argued that while
this approach requires long incubation times (12+ hours) the fact that the
substrates were incorporated into the dressings constantly in contact with the
wound, these times were not prohibitive. Much shorter incubation times of 2‐5
minutes have been reported by Edwards et al. (2008) using cellulose‐AP‐suc‐Ala‐
Ala‐Pro‐Ala‐pNA substrate. A dip‐stick approach with the same types of
substrates would be much more acceptable if the incubation time were within
the time of a patient consultation.
The group of Rimmer has recently reported an elegant method for
binding either gram positive or gram negative bacteria to a polymer which
changes conformation upon binding and has potential as a bacterial sensor.
Highly branched poly(N‐isopropylacrylamide) (HB‐PNIPAM) was modified with
the peptidic antibiotics polymyxin or vancomycin which bind to the cell
membranes of gram negative (e.g. Ps. aeruginosa) or gram positive (e.g. S.
aureus) bacteria, respectively (Sarker et al. 2011; Shepherd et al. 2010). When
incubated with their bacterial targets, the HB‐PNIPAM‐polymyxin and HB‐
PNIPAM‐vancomycin conjugates changed from soluble open coil structures to
agglomerated globule structures which were used to deplete bacteria from an
infected 3D tissue engineered skin model (Shepherd et al. 2011). This same
approach may be used to selectively detect bacteria by incorporation of a
fluorescent probe.
3.2 Porous silicon
Porous silicon (pSi) has become a popular substrate for optical biosensor
manufacture and may have potential application in wound sensors. A number of
specific characteristics underscore the popularity of pSi including: cost‐
effectiveness; ease of use in fabrication via anodization of semiconductor grade
silicon; flexibility in fabricating different architectures including multilayers of
alternating low and high porosity (through adjustment of etching parameters)
(Jane A. et al. 2009); a well understood chemistry for biofunctionalization
(Stewart and Buriak 2000); and, tuneable optical properties. The nanoscale
pores and crystalline features of the pSi lead to quantum confinement effects
causing visible spectrum luminescence (Canham 1990). Furthermore, the large
surface area of pSi nanostructures is well suited to the detection of changes in
refractive index, leading to the prospect of gas sensing or biosensing (Jane A. et
al. 2009; Letant and Sailor 2000; Worsfold et al. 2006). In fact, the optical
reflectance properties of the thin pSi film result in the appearance of Fabry‐Pérot
interference fringes. Changes in the effective refractive index of a thin pSi film
cause a shift in the interference pattern and affect the position of the photonic
resonance peak in multilayered pSi resonators. Hence, capture of biomolecules
within the film alters the optical properties of the film, in some cases even visible
to the unaided eye, and afford the transduction of the binding event (Amato and
Rosenbauer 1997; Bisi et al. 2000; Chazalviel and Ozanam 1997; Fauchet 1998;
Ouyang et al. 2005; Sailor and Link 2005; Vincent 1994).
To demonstrate pSi as a substrate for bacteria common to wounds the
groups of Miller and Fauchet coated a photoluminescent pSi microcavity with a
peptidomimetic receptor for diphosphoryl lipid A, a component of
lipopolysaccharide found in the outer cellular membrane of gram negative
bacteria. When exposed to E. coli, Salmonella and Ps. Aeruginosa (all gram
negative), a red shift in the photoluminescence spectra was readily observed
spectroscopically, whereas the photoluminescence was not affected when
exposed to gram positive bacteria B. subtilis and L. acidiophilus (Figure 3a)
(Chan et al. 2001; Fauchet 1998; Ouyang et al. 2005).
Biosensors adapted to cell and tissue culture situations have substantial
potential to not only detect the presence of certain cells but also changes in cell
behavior, which is highly relevant to wound healing (Chan et al. 2001; Li et al.
2011; Massad‐Ivanir et al. 2011; Massad‐Ivanir et al. 2010). Schwartz et al.
(2006) have monitored the health of prokaryotic and eukaryotic cell cultures by
means of light scattering from the surface of a pSi resonator where the
introduction of a scattering centre, such as a bacterium or a mammalian cell
resulted in oblique angle detection of a pSi resonance. This concept was used to
monitor physiological changes occurring in primary rat hepatocytes that affected
their viability (Schwartz et al. 2006) and to monitor the proliferation of
Pseudomonas bacteria and their lysis upon viral infection in real time (Alvarez S.
D. et al. 2007).
Recent studies have also demonstrated the detection of proteases using
pSi resonators. Matrix metalloproteinases (MMP) are zinc‐dependent
endopeptidases with increased activity in chronic wound fluid (Rayment et al.
2008). Gao et al. sensed MMP‐2 at concentrations as low as 0.1 ng/mL by spin‐
coating a protective gelatin film on top of a pSi resonator (Gao et al. 2008). Upon
contact with MMP‐2, the film was digested and the resulting peptides were able
to enter the pores and induced color changes visible to the unaided eye due to
red shifts in the position of the photonic resonance (Figure 3b). Likewise,
Orosco et al. (2006) obtained red shifts in resonance peaks upon digestion of
spin‐coated corn protein (zein) films over a pSi resonator in the presence of
pepsin. In contrast, Kilian et al. (2007) exploited a blue shift upon release of
peptide fragments from angiotensin immobilized into a pSi resonator in the
presence of the endoprotease subtilisin. Similarly, Martin et al. (2010) developed
a pSi microcavity to detect MMP‐8. Instead of exploiting the enzymatic activity of
MMP, the researchers attached anti‐MMP antibodies on the walls of pSi and were
able to detect the presence of MMP‐8 at the level of 100 ng/mL through immune‐
capture into the pores.
In summary, pSi based structures capable of detecting bacteria and
enzymes via shifts in their photonic properties lend themselves to the
development of stand‐alone optical sensors which could be incorporated into
dressings, such as hydrogels (Jane A. et al. 2009) to provide real‐time data via
colorimetric readouts.
3.3 Odor Sensors
Artificial olfaction systems, commonly referred to as electronic noses, have long
been used to detect volatile chemicals produced by bacteria (Allardyce et al.
2006; Pavlou et al. 2002; Wiggins et al. 1985). Many of the by‐products of
bacterial metabolism are airborne (spores, volatile organic compounds and
mycotoxins) and have characteristic odors which can give clues to the identity of
the bacteria, for instance specific anaerobic bacteria give rise to unpleasant
malodor (Bowler et al. 1999) The advantage odor sensors have over dressing‐
bound detection systems is that they can be used in situations where the
dressing is under layers of compression bandages which, to be effective, need to
be kept in place for long periods without being disturbed.
Tandem gas chromatography–mass spectrometry (GC‐MS) is the
traditional method for analysis of volatile species but the instrumentation is
bulky and very costly. The common alternative is to use solid‐state gas sensors
which rely on a change in conductivity, capacitance, work function, mass, optical
and reaction energy which can vary depending on the gas‐solid interaction
(Korotcenkov 2007). The common types of gas sensors used as odor sensors in
medical applications have recently been reviewed by Persaud (2005) and vary in
specificity, portability and cost. While gas sensors have attracted intense interest
in disease and biosecurity applications, only a few reports have addressed their
use in wound healing. To be successful, it has been suggested that a database of
wound volatiles is required, together with technology miniaturisation and cost
reduction (Thomas et al. 2010).
The volatile organic compound profile emitted by chronic wound lesions
can be extremely complex. Polydimethylsiloxane skin‐sampling patches fed into
GC‐MS revealed over 300 resolved peaks and many unresolved peaks from
venous ulcer samples (Thomas et al. 2010). While GC‐MS analysis generates a
wealth of data, the cost and bulk of the instrumentation makes it prohibitive for
most wound clinics. The challenge is to develop sensors which offer similar
information but at real‐time, lower cost and better portability. The use of
chemiresistors, such as conducting polymer arrays made from substituted
pyrroles, constitutes a relatively cheap approach to gas analysis but suffers from
sensitivity problems, sensor response time and interference by humidity (Bailey
et al. 2008). Despite these disadvantages, a clinical trial by Parry et al. (1995)
involving 21 patients using an Odormapper instrument housing 20
semiconducting polymer elements was able to instantly detect haemolytic
streptococci in infected wounds.
Metal oxide sensors made from a large selection of metal types are an
attractive low‐cost alternative which are based on the principle of temperature
dependent resistance modulation of thin semi‐conducting metal oxide films by
the adsorption and desorption of gases (Wang et al. 2010). Tin and indium
oxides in an array format have been used to analyse wound swabs later
identified using standard microbiological assays to be infected with E. coli, Ps.
aeruginosa and S. aureus (Setkus et al. 2006). Processing of the electrical output
from the array format generated two‐dimensional multi‐layered plots that were
matched to the type of bacteria by simple visual inspection. However, this study
was carried out in a closed laboratory system with synthetic atmosphere to
reduce unpredictable sensor response; reproduction of this environment is
clearly a hurdle to translating this technique to the clinic.
The problems associated with using metal oxide sensors in the presence
of background gases was addressed in a study by Feng et al. (2011). They used
wavelet analysis of fourteen metal oxide sensors and one electrochemical sensor,
all commercially sourced, to extract wound information in mice from the
background of the natural and rich odor of the animals. This algorithmic
approach to extracting fine distinctions between channels does not overcome the
inherent poor discriminating power of the narrowly tuned metal oxide sensors.
In future it may be possible to develop biomimetic sensors inspired by odorant
receptors in insects which have superior independent sampling of broader
regions of odorant space for better discrimination of volatiles compared with
currently available sensors (Berna et al. 2009).
3.4 Temperature as an indicator for infection
Calor (heat) is an established marker of infection in wounds and may be used as
an early predictor of chronicity before any obvious changes to the appearance of
the wound are observed (Nakagami et al. 2010). It is relatively easy and cheap to
measure temperature of skin and wounds using hand‐held infrared
thermometers. In a study of at‐risk patients it was shown that infrared
thermometers have the sensitivity to detect early signs of ulceration when used
to measure temperature gradients between feet where differences of >4°F were
used as the trigger point (Armstrong et al. 2007). While temperature is a very
useful parameter to measure when monitoring wound healing there has been
limited research into incorporating temperature sensors into wound dressings.
One approach reported is to use wireless temperature sensors based on the
resistivity of carbon nanotubes coupled to a transponder to predict bedsores and
inflammation (Matzeu et al. 2011). These sensors were tested on healthy
volunteers but no reports have been made where they were used on patients.
Wireless technology may be useful in a clinic but ideally the wound
dressing would incorporate some sort of stand‐alone indicator of temperature,
for example colorimetric sensors for measuring temperature. Color changing
fibers with sensitivity of +/‐ 0.5 °C developed specifically for detecting infection
have been patented by the group of Cranston but no research papers exist on this
work (Mestrovic 2009). Many other types of low‐cost colorimetric temperature
sensors have been developed which could also be applied to wound management
such as stimulus‐responsive polymers attached to pSi in order to generate
temperature and also pH sensors (Sciacca et al. 2011; Segal et al. 2007; Wu and
Sailor 2009). For example, Vasani et al. (2011) employed surface‐initiated atom
transfer radical polymerization to graft a thermo‐responsive polymer, poly(N‐
isopropylacrylamide) (PNIPAM), from pSi films producing a stimulus‐responsive
inorganic‐organic composite material. The optical properties of this material
changed significantly and reversibly when changing the temperature around the
lower critical solution temperature (LCST) of the PNIPAM component.
Microfabrication of single crystal silicon nanomembrane diodes has been
used by Kim et al. (2012) to develop electronic temperature sensors with
resolution of 0.2 °C. An additional feature of the sensors is that they can also
deliver heat to the tissue through a gold micro‐heater to promote healing. Their
sensors are intended for use in sutures but could equally be incorporated into
dressings.
4. Sensors for detection of pH changes
4.1 Importance of pH
All biochemical processes in the body, including wound healing, are influenced
by pH. Normally the pH of healthy skin is slightly acidic, in the range of pH 4‐6,
but when it is damaged this acidic milieu is disturbed as the body’s internal pH of
7.4 is exposed. In circumstances where the wound is acute, the pH follows a
relatively simple pathway through an acidic inflammation stage followed by a
more basic granulation step before re‐establishing in the pH 4‐6 range during re‐
epithelization (Figure 4a). The pH of chronic wounds is substantially more
complex. The somewhat simplified plot of wound pH in chronic wounds in
Figure 4b (Schneider et al. 2007) suggests the pH oscillates between pH 7‐8 as
the wound fails to heal, however, there is evidence that the real situation is far
more complex than this. Shi et al. (2011) found in a survey of the literature that
pH values measured using electrodes designed for wounds or pH paper ranged
from 5.4 up to 9.0 and that an increase in pH could be attributed to bacterial
colonization. Srinivasan and Madhadevan (2010) have shown that a bacterial
bioburden may not always necessarily result in an increase in pH as it is
dependent on the primary energy source, at least for E coli. Moreover there is
anecdotal evidence that the composition (biofilm or planktonic), type of bacteria,
anaerobic or aerobic conditions may influence pH so great caution should be
exercised when using pH alone to diagnose wound status.
Despite the conflicting information on wound pH, its importance to the
biochemical processes integral to wound healing means that pH is a marker of
both diagnostic and theranostic interest. Consequently, several research groups
have developed dressings which incorporate pH sensitive materials which are
reviewed in the following section.
4.2 pH sensors
One of the simplest and most cost effective methods for determining pH is by
using indicator dyes, which absorb different wavelengths of visible light
depending on the pH. The challenges when using dyes incorporated into wound
dressings is they must not leach from the dressing and they also have to be
sensitive to the pH range encountered in wounds. Trupp et al. (2010) and Mohr
et al. (2008) used the approach of synthesizing a series of hydroxyl‐substituted
azobenzene derivatives as indicator dyes for optically monitoring pH between 6‐
10 (Trupp et al. 2010) or 3‐12 (Mohr et al. 2008). They tailored the sensitivity of
the dyes by the substituent para or ortho to the hydroxyl group. The dyes were
then immobilized on cellulose films using their pendent vinylsulfonyl groups. To
protect the cellulose from mechanical and swelling forces and to prevent
shrivelling, it was laminated onto polyethylene terephthalate films. These
laminates exhibited good reversibility and could be patterned into arrays,
meaning a pH map of a wound may be possible. This technology has been used to
assemble prototypes which turn from yellow to purple between pH values of 6.5
and 8.5 (www.technewsdaily.com/1454‐expressive‐bandage‐displays‐
infectious‐spread‐.html)
Sridhar and Takahata (2009) have developed a microfabricated wireless
pH monitor intended to be imbedded into a wound dressing to continuously
track pH. The device uses a pair of wire coils, which sandwich a pH‐sensitive
hydrogel. As the hydrogel swells and deswells with pH, the distance separating
the two planar coils changes which results in a change of inductance (Figure 4c).
They observed a linear response in terms of inductance change with coil
separation distance and changes in frequency over the pH range 2‐7 when using
poly(vinyl alcohol)‐poly(acrylic acid) as the pH‐sensitive hydrogel between the
coils. The authors acknowledge that changes in moisture content may also
change the reading and suggest using a pH‐insensitive hydrogel in tandem to
compensate for this. Interestingly, they also used a network‐spectrum analyser
with antenna to allow for wireless measurement of the coil gap, and hence pH.
This approach may be valuable in cases where the smart dressing is covered by a
compression bandage, provided of course that the bandage does not exert any
force on the sensor.
A different approach to measuring pH of wound fluid has been reported
by Phair et al. (2011) who exploited the presence of substantial amounts of
endogenous uric acid in wound exudate to indirectly measure pH based on the
oxidation potential change in uric acid with pH. Uric acid is the final breakdown
product of purine catabolism and is present at concentrations >100 µmol L‐1 in
wound fluid (James et al. 2003). Their device features a laminated screen‐printed
planar electrode design with a cellulose filter paper wick for bringing exudate to
the sensor. Importantly, their device contains no expensive components and is
therefore disposable. The sensor was shown to be responsive to a wide range of
test solutions ranging in pH from 3.7 to 8.0. When tested against blood from a
healthy volunteer the measured pH matched the expected pH of 7.4, although
signal quality diminished over time as the uric acid was depleted because of
oxidation and fouling of the electrode. This type of sensor would be unsuitable
for incorporation into a dressing for continuous monitoring but may be useful as
a POC dip‐stick theranostic.
5. Moisture
Wound moisture levels are known to be critical to healing (Winter 1962; Wu et
al. 1995) – too much moisture can result in maceration while too little can lead
to the wound drying out (Banks et al. 1997). This key finding by Winter in the
1960s has led to a proliferation of dressing materials designed specifically to
control the moisture content including films, hydrocolloids, foams, alginates and
hydrogels (Queen et al. 1987; Queen et al. 2004). McColl et al. (2007) have
developed an electrical impedance sensor which allows for real‐time monitoring
of moisture in an array format so a map of the wound can be obtained. Their
sensor takes advantage of the ionic nature of wound fluid and quantifies fluid
volume using alternating current of varying frequency and measures the
impedance across Ag/AgCl electrodes insulated with a silicone elastomer
inserted into a range of commercially available dressings. By inserting eight
pairs of independent electrodes into the dressings, maps of hydration could be
obtained. The system was tested in vitro with a simulated wound bed made from
a Teflon mold with a liquid flow channel and syringe pump‐controlled injectors
to control delivery of an aqueous NaCl/CaCl2 solution at four points below a
wound dressing. The dressings were weighed periodically to obtain a
relationship between change in impedance and loss of moisture. Impedance
greater than 200 indicated dehydration of the dressing and drying of the
contact surface. Of the dressings examined, the polyurethane foams lost water
rapidly while the polyurethane films retained moisture, presumably by occlusion
which is not surprising as the film dressing is designed for lightly exuding
wounds.
6. Spectroscopy
Vibrational spectroscopy is a technique that may be useful in the monitoring of
wounds and although it is not strictly a sensor technique it is worth reviewing
the state‐of‐the‐art with respect to wounds as it may inspire sensor‐type
approaches in future. Hand‐held portable spectrometers are now available
making it feasible for a clinic to have one in their inventory (Erickson and
Godavarty 2009). Monitoring of deoxy‐ and oxy‐hemoglobin has been achieved
using near infrared (NIR) spectroscopy initially in diabetic rat wound models
(Weingarten et al. 2006; Weingarten et al. 2008) and recently in small human
trials (Weingarten et al. 2010). NIR spectroscopy has been used to monitor
oxygen concentration of hemoglobin in a number of other medical applications
(Erickson and Godavarty 2009) based on the ability to distinguish between
deoxy‐ and oxy‐hemoglobin by virtue of their different absorbances. This may
be exploited for use in monitoring wounds where it may be used as a reflection
of impaired blood flow. Indeed, NIR spectroscopy has been used to distinguish
between diabetic and non‐diabetic wounds in rats (Weingarten et al. 2006). In a
human study subsurface oxy‐hemoglobin concentration was measured using a
non‐invasive NIR fibre‐optic probe and found that hemoglobin concentration
remained elevated in non‐healing wounds. This was attributed to greater blood
vessel ingrowth when compared with normally healing wounds which
experience degradation and reabsorption of vessels in the late stage of healing
when angiogenesis ceases (Weingarten et al. 2010). Significantly, the NIR
technique offered early warning of non‐healing wounds where some wounds
which appeared to be healing based on traditional metrics were identified as
having elevated haemoglobin by NIR.
7. Wound imaging
Imaging of whole wounds has the potential to offer the clinician much greater
information than single‐target diagnostics. In fact, one of the most basic and yet
powerful tools in any wound clinic is visual observation of the site.
Unfortunately, this approach is only useful if the observer has seen enough
wounds, followed their progression and can confidently use this knowledge to
inform their management program. Appreciably, this accumulation of experience
may take many years. Imaging and the dissemination of the images, therefore,
has the ability to enable clinicians with a powerful, informative and non‐invasive
means to examine and quantify wound parameters. It further provides the
potential to generate extensive databases of wound images that will be valuable
for research, diagnostic and educational purposes.
Wound area is a key parameter used to monitoring healing and is often
measured by tracing of the wound using transparent film or simply measuring
with a ruler. Both these methods are subjective, time‐consuming and marginally
invasive. Commonly used alternatives, including ultrasound and digital
photography, are useful methods for recording an image of the wound and allow
for post‐visitation analysis but suffer from color calibration for photography and
identification of wound edges which can be difficult with complex wound
topography. Papazoglou et al. (2010) have developed an advanced method for
determining wound area using an algorithm that can detect and recognize a
wound from a digital photograph. They used a standard low‐cost point‐and‐
shoot camera equipped with a polarizing filter to reduce light reflection.
Consistency in lighting was achieved by photographing in the dark and relying
on only the flash of the camera to provide a light source. The images were
recorded as standard JPEG mode files and imported into a MATLAB program
which identified a wound area and a non‐wound area using grey scale images
derived from the red, blue and green channels, as well as other color spaces with
combinations of threshold and pixel‐based color comparing segmentation
methods (Figure 5a). They note that when there is clear color contrast it is easy
to accurately and reproducibly measure the wound area. However, when the
wound bed is inflamed, unclean, or if there is poor color contrast then
repeatability is reduced.
The use of digital photography need not be constrained to recording of
visible light. The ubiquity of cheap digital cameras may be the key to wound
imaging when used in tandem with sensor films and ratiometic imaging. Meier et
al. (2011) used a digital camera fitted with a 405 nm LED excitation ring to
capture a 2‐dimensional image of pH and partial pressure of oxygen (pO2) within
wounds. This work is based on luminescence life‐time imaging sensors
developed previously by the same group based on polystyrene‐co‐acrylonitrile
particles loaded with palladium(II)‐porphyrin complexes, which are sensors for
oxygen (Schreml et al. 2011b), or FITC as a pH‐dependent luminophore (Schreml
et al. 2011a). The mechanism for oxygen sensing is based on the oxygen‐
dependent quenching of the luminescence lifetime for the immobilized
porphyrin. When a thin polyurethane film containing particles loaded with the
pO2 and pH sensors and a reference diphenylanthracene dye were placed on
acute and chronic wounds, an image map showing pO2 an pH was obtained using
a digital camera with the 405 nm LED excitation (Figure 5b). This method works
by separating the red, green and blue (RGB) channels of the camera and
recording the output from different probes onto each channel. The emission of
the Pt (II) probe, which is dependent on pO2, and the luminescence of the FITC,
which is a probe for pH, are recorded on the red and green channels,
respectively, then referenced to diphenylanthracene dye in the blue channel. In
the oxygen and pH maps (Figure 5b) of the chronic wound, the upper left corner
is an area of low oxygen concentration and high pH, possibly indicative of
inflammation which is absent in the acute wound.
The ability to measure pO2 quickly and easily may make this technique
useful in other POC diagnostics, such as an alternative to transcutaneous
oxometry measurements (TCOM). This technique is routinely used on patients
undergoing hyperbaric oxygen therapy for treatment of chronic wounds to
confirm that oxygen is present in blood and plasma as a screening tool and
predictor of therapy outcomes (Grolman et al. 2001). TCOM uses a heated patch
on the skin to enhance oxygen diffusion to the surface where it is then measured.
The sensor developed by Meier et al. may offer a smaller, cheaper alternative to
traditional TCOM instruments, although these authors do not discuss this. In
their paper, Meier et al. (2011) do, however, suggest their technology may also
be used to probe temperature and to record videos of wounds.
8. Conclusions and Future Directions
Since the invention of the blood oxygen sensor in the 1960s by Clarke and Lyons
(1962), biosensors have become a ubiquitous part of the research landscape.
Two of the most successful and recognizable technologies which have gone on to
be commercially available are the blood glucose monitor for diabetics and the
hCG hormone pregnancy test, but there are other sensor technologies now
beginning to permeate the home diagnostics market targeted at variety of health
conditions (Lee 2008). With the technology for measuring certain metabolites,
proteins, gases and temperature now mature it is inevitable that these will make
the transition into wound sensors as either disposable dip‐stick type devices or
as continuous monitoring tools in dressings.
The need for sensors for wound healing stems from the complexity of not
just the dynamic wound healing biochemistry but also the management of
wounds from a clinical perspective. There is strong anecdotal evidence that an
important challenge in wound management is in the diversity of experience of
the clinicians themselves, many of whom do not specialize in wound
management. Sensors specially designed to detect key wound parameters have
the potential to add objectivity to the practice of wound management. For
example, there are indications that dressings are regularly being disturbed
unnecessarily by premature renewal because the dressing looks full of exudate
whereas the wound bed is still clean. A wound dressing with a built in moisture
or infection sensor may aid in this decision‐making.
The number of and sophistication of sensors is ever increasing but what is
apparent through this review of the literature on sensors specifically focussed on
wound healing, is that there aren't actually very many identified targets being
exploited, at least in the open literature, despite the long list of potential markers
in Table 1. To date, the only wound markers targeted for sensor applications
have been: i) infection (including odor and temperature), ii) pH, iii) oxygen iv)
uric acid, v) hemoglobin and iv) broad spectrum protease activity (Systagenix's
Woundcheck™ product). Beyond this, reliable and specific targets have yet to be
identified. Searches for biomarkers in wound fluid using proteomic approaches
have been successful in finding differing distributions of specific proteins
between acute healing and chronic non‐healing wounds (Eming et al. 2010),
which may be useful for future sensor development. However, it has been noted
that a single protein approach may be unrealistic due to the complex balance
between the proteins and their respective inhibitors (Wyffels et al. 2010).
Many of the sensors in this review (summarized in Table 2), although
intended for use on wounds, are still in the early preclinical stages of
development. The challenges of using sensors in real wound environments
should not be underestimated since hundreds of different proteins are present
at a range of concentrations (Eming et al. 2010) across pH levels as low as 5.4
and as high as 9.0 (Shi et al. 2011) within highly heterogeneous wounds.
Continuous monitoring in‐dressing sensors must be immune to protein fouling
and cross‐talk if they are to operate effectively. It may be that disposable dip‐
stick POC sensors are a more realistic option for many of the sensors discussed
in this review. It is also important that many of the sensor outputs discussed
should not be used in isolation. An example is pH, where an acidic or basic
wound environment could be normal or abnormal depending on the progression
of healing or state of infection.
The two types of sensors – diagnostic and theranostic, offer different
information to the wound manager. Diagnostic outputs would be intended to
inform what action is needed, for example, that the wound is infected and
requires anti‐microbial treatment. Theranostic output, conversely, for example,
pH or presence of particular proteins, may be much more valuable in the study of
wound healing especially if linked to databases (similar to the Proteomics
Discovery Pipeline). The pooled data could then be linked to healing outcomes to
further inform the practice of wound management.
Based on some of the leading edge research found in the literature
reviewed here and some artistic licence we have drawn what we think the
wound dressing of the future may conceivably look like (Figure 6). This dressing
of the future incorporates a map of the wound using some of the imaging
technology discussed as well as indicators for infection and pH and the time
since the last dressing change. This is a view of how a dressing may look in 5‐10
years but it’s conceivable that some of the current leading‐edge sensor
technology, such as stimuli responsive hologram images, wearable sensors and
biocompatible photonics, such as silk waveguides will play a part in future
concepts (Horgan et al. 2006; Kim et al. 2011; Parker et al. 2009).
Perhaps the most important outcome of this research is the impact it will
have on patients. While it’s likely the clinicians will largely decide which sensor
technologies become accepted, it should be remembered that the technology
should not be designed simply to make their job easier but to actually improve
patient outcomes. Any sensor technology used in isolation will also be sure to fail
and a holistic approach needs to be taken including patient nutrition and mental
well‐being.
Acknowledgments
Thanks to Michelle Gibb and Dr Gregory Schultz for helpful advice and to Mike
Plum for help with the figures. The funding for this project was provided by the
Wound Management Innovation Cooperative Research Centre (WMICRC) and
the Tissue Repair and Regeneration Program at QUT.
Conflicts
The authors have no conflict of interest.
Figure Captions Figure 1: Current commercial products for wound management. Clockwise from top – Acticoat silver dressing (Acticoat (wound.smith‐nephew.com), Mepilex moisture control dressing (http://www.molnlycke.com), Woundcheck™ Protease Status diagnostic (http://www.systagenix.com/our‐products/lets‐test) Figure 2: (A) Polypropylene fabric impregnated with vesicles containing carboxyfluoroescein under low UV intensity light in the presence of different bacteria. (B) The mechanism for bursting of the vesicles by bacteria releasing the fluorescent dye or an antimicrobial agent. (i) vesicle prior to rupture, (ii) toxins from bacteria lyse vesicle wall, (iii) contents of vesicle released. Adapted from Zhou et al. (2010) and Zhou et al. (2011). Figure 3: a. Photoluminescence spectra (675 to 900 nm) of a porous silicon microcavity biosensor in the presence and absence of bacterial cell lysates. Blue spectra: sensor alone following derivatization with TWTCP and glycine methyl ester. Red spectra: sensor photoluminescence following incubation with cell lysates from Gram‐(+) bacteria (B. subtilis, left) or Gram‐(‐) bacteria (E. coli, right). The green spectra in each case are the difference between spectra obtained in the presence and absence of bacterial cell lysates. Reproduced, with permission, from Chan et al. (2001). b. Detection of MMP‐2 on a partially dried chip. (i) An optical image of the chip. (ii) Spectra of the reflected light taken from the spots that have been loaded with (from bottom to top) blank control, 0.1, 1, 10, 100, and 1000 ng/mL MMP‐2 samples. Spectra are offset along the y axis for clarity. Reproduced with permission, from Gao et al. (2008). Figure 4: a. pH of acute wounds, b. pH of chronic wounds. Images taken from Schneider et al. (2007). c. pH sensitive gel between inductance coils for continuous in situ pH monitoring (Sridhar and Takahata (2009)). Figure 5: a. animal and human wounds traced manually (black line) and using image analysis (green, white, yellow lines) developed by Papazoglou et al. (2010). Scale bar=1cm b. Left: A digital camera fitted with a 405nm‐LED ring light and an emission filter for photographing wounds covered in a sensor film. Right: visible light pictures, pO2 and pH maps comparing acute and chronic wounds. Image adapted from Meier et al. (2011). Figure 6: The future of wound dressings? Our impression of a hypothetical wound dressing incorporating a color map of the wound, pH readout, infection feedback and indicator displaying if the dressing needs changing.
References http://www.systagenix.com/our‐products/lets‐test (accessed 5 Sept 2012) www.technewsdaily.com/1454‐expressive‐bandage‐displays‐infectious‐spread‐.html (accessed 21 Nov 2011). Allardyce, R.A., Langford, V.S., Hill, A.L., Murdoch, D.R., 2006. J. Microbiol. Methods 65(2), 361‐365. Alvarez S. D., Schwartz M. P., Migliori B., Rang C.U., Chao L., J., S.M., 2007. phys. stat. sol. a 204(5), 1439‐1443. Amato G, Rosenbauer M. In Optoelectronic Properties of Semiconductors and Superlattices, Amato, G., Delerue, C., Bardeleben, H.J. (eds). Gordon and Breach: Amsterdam, 1997; 5, 3‐52 Armstrong, D.G., Holtz‐Neiderer, K., Wendel, C., Mohler, M.J., Kimbriel, H.R., Lavery, L.A., 2007. Am. J. Med. 120(12), 1042‐1046. Bailey, A.L.P.S., Pisanelli, A.M., Persaud, K.C., 2008. Sens. Actuators. B. Chem. 131(1), 5‐9. Banks, V., Bale, S., Harding, K., Harding, E.F., 1997. J. Wound Care 6(6), 266‐269. Berna, A.Z., Anderson, A.R., Trowell, S.C., 2009. PLoS One 4(7), e6406. Bisi, O., Ossicini, S., Pavesi, L., 2000. Surf. Sci. Rep. 38(1‐3), 1‐126. Bowler, P.G., Davies, B.J., Jones, S.A., 1999. J. Wound Care 8(5), 216‐218. Broadbent, J., Walsh, T., Upton, Z., 2010. Proteomics Clin. Appl. 4(2), 204‐2014. Burd, A., Kwok, C.H., Hung, S.C., Chan, H.S., Gu, H., Lam, W.K., Huang, L., 2007. Wound Repair Regen. 15(1), 94‐104. Canham, L.T., 1990. Appl. Phys. Lett. 57(10), 1046‐1048. Cao, Y., Croll, T.I., Rizzi, S.C., Shooter, G.K., Edwards, H., Finlayson, K., Upton, Z., Dargaville, T.R., 2011. J. Biomed. Mater. Res. A 96 A(4), 663‐672. Chan, S., Horner, S.R., Fauchet, P.M., Miller, B.L., 2001. J. Am. Chem. Soc. 123(47), 11797‐11798. Chazalviel J.N., Ozanam, F. In Optoelectronic Properties of Semiconductor and Superlattices Amato, G., Delerue, C., Bardeleben, H.J. (eds). Gordon and Breach: Amsterdam, 1997; 5, 53‐71. Clark Jr, L.C., Lyons, C., 1962. Ann. N.Y. Acad. Sci. 102, 29‐45. Cooper, D.M., Yu, E.Z., Hennessey, P., Ko, F., Robson, M.C., 1994. Ann. Surg. 219(6), 688‐691. Crovetti, G., Martinelli, G., Issi, M., Barone, M., Guizzardi, M., Campanati, B., Moroni, M., Carabelli, A., 2004. Transfus Apher Sci. 30(2), 145‐151. Davies, C.E., Wilson, M.J., Hill, K.E., Stephens, P., Hill, C.M., Harding, K.G., Thomas, D.W., 2001. Wound Repair Regen. 9(5), 332‐340. Dowd, S.E., Wolcott, R.D., Sun, Y., McKeehan, T., Smith, E., Rhoads, D., 2008. PLoS One 3(10), 3326. Edwards, J.V., Caston‐Pierre, S., Bopp, A.F., Goynes, W., 2005. J. Pept. Res. 66, 160‐168. Edwards, J.V., Caston‐Pierre, S., Howley, P., Condon, B., Arnold, J., 2008. Sens. Lett. 6(4), 518‐523. Eming, S.A., Koch, M., Krieger, A., Brachvogel, B., Kreft, S., Bruckner‐Tuderman, L., Krieg, T., Shannon, J.D., Fox, J.W., 2010. J. Proteome Res. 9(9), 4758‐4766. Erickson, S.J., Godavarty, A., 2009. Med. Eng. Phys. 31(5), 495‐509.
Fauchet, P.M., 1998. Light Emission in Silicon: From Physics to Devices 49, 205‐ 252. Feng, J., Tian, F., Yan, J., He, Q., Shen, Y., Pan, L., 2011. Sens. Actuators. B. Chem. 157(2), 395‐400. Fernandez, M.L., Upton, Z., Edwards, H., Finlayson, K., Shooter, G.K., 2012. Int. Wound J. 9(2), 139‐149. Gao, L.Z., Mbonu, N., Cao, L.L., Gao, D., 2008. Anal. Chem. 80(5), 1468‐1473. Gottrup, F., 2004. Am. J. Surg. 187, 38‐43. Grey, J.E., Harding, K.G., 2006. ABC of Wound Healing. Grolman, R., Wilkerson, D., Taylor, J., Allinson, P., Zatina, M., 2001. Am. Surg. 67(11), 1072‐1079. Harding, K., 2007. Diagnostics and Wounds. A Consensus Document. Hasmann, A., Gewessler, U., Hulla, E., Schneider, K.P., Binder, B., Francesko, A., Tzanov, T., Schintler, M., Van der Palen, J., Guebitz, G.M., Wehrschuetz‐Sigl, E., 2011. Exp. Dermatol. 20(6), 508‐513. Healy, B., Freedman, A., 2006. Br. Med. J. 332(7545), 838‐841. Hoffmann, D.C., Textoris, C., Oehme, F., Klaassen, T., Goppelt, A., Romer, A., Fugmann, B., Davidson, J.M., Werner, S., Krieg, T., Eming, S.A., 2011. J. Biol. Chem. 286(33), 28889‐28901. Horgan, A.M., Marshall, A.J., Kew, S.J., Dean, K.E.S., Creasey, C.D., Kabilan, S., 2006. Biosens. Bioelectron. 21(9), 1838‐1845. James, T.J., Hughes, M.A., Cherry, G.W., Taylor, R.P., 2003. Wound Repair Regen. 11(3), 172‐176. Jane A., Dronov R, Hodges A, N.H, V., 2009. Trends Biotechnol. 27(4), 230‐239. Kilian, K.A., Bocking, T., Gaus, K., King‐Lacroix, J., Gal, M., Gooding, J.J., 2007. Chem. Commun.(19), 1936‐1938. Kim, D.H., Lu, N., Ma, R., Kim, Y.S., Kim, R.H., Wang, S., Wu, J., Won, S.M., Tao, H., Islam, A., Yu, K.J., Kim, T.I., Chowdhury, R., Ying, M., Xu, L., Li, M., Chung, H.J., Keum, H., McCormick, M., Liu, P., Zhang, Y.W., Omenetto, F.G., Huang, Y., Coleman, T., Rogers, J.A., 2011. Science 333(6044), 838‐843. Kim, D.H., Wang, S., Keum, H., Ghaffari, R., Kim, Y.S., Tao, H., Panilaitis, B., Li, M.,
Kang, Z., Omenetto, F., Huang, Y., Rogers, J.A., 2012. Small in press DOI: 10.1002/smll.201200933 Korotcenkov, G., 2007. Mater. Sci. Eng. B. Solid State Mater. Adv. Technol. 139(1), 1‐23. Lee, T.M.H., 2008. Sensors 8(9), 5535‐5559. Letant, S., Sailor, M., 2000. Adv. Mater. 12(5), 355‐359. Li, S., Huang, J.F., Cai, L.T., 2011. Nanotechnology 22(42), 425502. Liu, Y., Min, D., Bolton, T., Nube, V., Twigg, S.M., Yue, D.K., McLennan, S.V., 2009. Diabetes Care 32(1), 117‐119. Liu, Y.C., Margolis, D.J., Isseroff, R.R., 2011. J. Invest. Dermatol. 131(4), 818‐827. Malda, J., Klein, T., Upton, Z., 2007. Tissue Eng. 13(9), 2153‐2162. Margolis, D., Berlin, J., Strom, B., 2000. Am. J. Med. 109(1), 15‐19. Martin, M., Taleb Bendiab, C., Massif, L., Palestino, G., Agarwal, V., Cuisinier, F., Gergely, C., 2010. Phys. Status Solidi C 8(6), 1888‐1892. Massad‐Ivanir, N., Shtenberg, G., Tzur, A., Krepker, M.A., Segal, E., 2011. Anal. Chem. 83(9), 3282‐3289. Massad‐Ivanir, N., Shtenberg, G., Zeidman, T., Segal, E., 2010. Adv. Funct. Mater. 20(14), 2269‐2277.
Matzeu, G., Losacco, M., Parducci, E., Pucci, A., Dini, V., Romanelli, M., Di Francesco, F., 2011. Skin temperature monitoring by a wireless sensor. 37th Annual Conference of the IEEE Industrial Electronics Society, IECON 2011 pp. 3533‐3535, Melbourne. McColl, D., Cartlidge, B., Connolly, P., 2007. Int. J. Surg. 5(5), 316‐322. Meier, R.J., Schreml, S., Wang, X.D., Landthaler, M., Babilas, P., Wolfbeis, O.S., 2011. Angew. Chem. Int. Ed. 50(46), 10893‐10896. Mertz, P.M., Oliveira‐Gandia, M.F., Davis, S.C., 1999. Dermatol. Surg. 25(2), 89‐93. Mestrovic, M., patent WO 2009/023937. Moffatt, C., Doherty, D., Smithdale, R., Franks, P., 2009. Br. J. Dermatol. 162, 51‐ 58. Mohr, G.J., Muller, H., Bussemer, B., Stark, A., Carofiglio, T., Trupp, S., Heuermann, R., Henkel, T., Escudero, D., Gonzalez, L., 2008. Anal. Bioanal. Chem. 392(7‐8), 1411‐1418. Nakagami, G., Sanada, H., Iizaka, S., Kadono, T., Higashino, T., Koyanagi, H., Haga, N., 2010. J. Wound Care 19(11), 465‐472. Orosco, M.M., Pacholski, C., Miskelly, G.M., Sailor, M.J., 2006. Adv. Mater. 18(11), 1393‐1396. Ouyang, H., Christophersen, M., Viard, R., Miller, B.L., Fauchet, P.M., 2005. Adv. Funct. Mater. 15(11), 1851‐1859. Papazoglou, E.S., Zubkov, L., Mao, X., Neidrauer, M., Rannou, N., Weingarten, M.S., 2010. Wound Repair Regen. 18(4), 349‐358. Parker, S.T., Domachuk, P., Amsden, J., Bressner, J., Lewis, J.A., Kaplan, D.L., Omenetto, F.C., 2009. Adv. Mater. 21(23), 2411‐2415. Parry, A.D., Chadwick, P.R., Simon, D., Oppenheim, B., McCollum, C.N., 1995. J. Wound Care 4(9), 404‐406. Pavlou, A., Turner, A.P.F., Magan, N., 2002. Lett. Appl. Microbiol. 35(5), 366‐369. Persaud, K.C., 2005. Int. J. Low. Extrem. Wounds 4(1), 50‐56. Phair, J., Newton, L., McCormac, C., Cardosi, M.F., Leslie, R., Davis, J., 2011. Analyst 136(22), 4692‐4695. Poon, V.K.M., Burd, A., 2004. Burns 30(2), 140‐147. Queen, D., Gaylor, J.D.S., Evans, J.H., Courtney, J.M., Reid, W.H., 1987. Biomaterials 8(5), 367‐371. Queen, D., Orsted, H., Sanada, H., Sussman, G., 2004. Int. Wound J. 1(1), 59‐77. Rayment, E.A., Dargaville, T.R., Shooter, G.K., George, G.A., Upton, Z., 2008. Biomaterials 29(12), 1785‐1795. Rizzi, S.C., Upton, Z., Bott, K., Dargaville, T.R., 2010. Expert Rev. Med. Devices 7(1), 143‐154. Sailor, M.J., Link, J.R., 2005. Chem. Commun. 11, 1375‐1383. Sarker, P., Shepherd, J., Swindells, K., Douglas, I., MacNeil, S., Swanson, L., Rimmer, S., 2011. Biomacromolecules 12(1), 1‐5. Schneider, L.A., Korber, A., Grabbe, S., Dissemond, J., 2007. Arch. Dermatol. Res. 298(9), 413‐420. Schreml, S., Meier, R.J., Wolfbeis, O.S., Landthaler, M., Szeimies, R.M., Babilas, P., 2011a. Proc. Natl. Acad. Sci. USA 108(6), 2432‐2437. Schreml, S., Meier, R.J., Wolfbeis, O.S., Maisch, T., Szeimies, R.M., Landthaler, M., Regensburger, J., Santarelli, F., Klimant, I., Babilas, P., 2011b. Exp. Dermatol. 20(7), 550‐554.
Schwartz, M.P., Derfus, A.M., Alvarez, S.D., Bhatia, S.N., Sailor, M.J., 2006. Langmuir 22(16), 7084‐7090. Sciacca, B., Secret, E., Pace, S., Gonzalez, P., Geobaldo, F., Quignarda, F., Cunin, F., 2011. J. Mater. Chem. 21, 2294‐2302. Segal, E., Perelman, L.A., Cunin, F., Di Renzo, F., Devoisselle, J.M., Li, Y.Y., Sailor, M.J., 2007. Adv. Funct. Mater. 17(7), 1153‐1162. Sen, C., Gordillo, G., Roy, S., Kirsner, R., Lambert, L., Hunt, T., Gottrup, F., Gurtner, G., Longaker, M., 2009. Wound Repair Regen. 17, 763‐771. Serena, T., Cullen, B., Bayliff, S., Gibson, M., DeMarco, D., Galbraith, J., Le, N., Mancinelli, M., Sabo, M., Samies, J., 2011. http://www.systagenix.com/cms/uploads/NG42‐ 11_Serena_et_al_Protease_Activity_Levels_Associated_with_Healing_Status _of_Chronic_Wounds_Wounds_UK_2011_004.pdf (accessed 21 Nov 2011) Setkus, A., Galdikas, A.J., Kancleris, Z.A., Olekas, A., Senuliene, D., Strazdiene, V., Rimdeika, R., Bagdonas, R., 2006. Sens. Actuators. B. Chem. 115(1), 412‐ 420. Sharp, D., Forsythe, S., Davis, J., 2008. J. Biochem. 144(1), 87‐93. Sharp, D., Gladstone, P., Smith, R.B., Forsythe, S., Davis, J., 2010. Bioelectrochemistry 77(2), 114‐119. Shepherd, J., Sarker, P., Rimmer, S., Swanson, L., MacNeil, S., Douglas, I., 2011. Biomaterials 32(1), 258‐267. Shepherd, J., Sarker, P., Swindells, K., Douglas, I., MacNeil, S., Swanson, L., Rimmer, S., 2010. J. Am. Chem. Soc. 132(6), 1736‐1737. Shi, L., Ramsay, S., Ermis, R., Carson, D., 2011. J. Wound Ostomy. Continence Nurs. 38(5), 514‐521. Singer, A., Clark, R., 1999. N. Eng. J. Med. 341, 738‐746. Sridhar, V., Takahata, K., 2009. Sens. Actuators A. Phys. 155(1), 58‐65. Srinivasan, K., Mahadevan, R., 2010. BMC Biotechnol. 10(2), 1‐10. Stewart, M., Buriak, J., 2000. Adv. Mater. 12(2), 859‐869. Streit, M., Beleznay, Z., Braathen, L.R., 2006. Int. Wound J. 3(3), 171‐179. Thomas, A.N., Riazanskaia, S., Cheung, W., Xu, Y., Goodacre, R., Thomas, C.L.P., Baguneid, M.S., Bayat, A., 2010. Wound Repair Regen. 18(4), 391‐400. Trupp, S., Alberti, M., Carofiglio, T., Lubian, E., Lehmann, H., Heuermann, R., Yacoub‐George, E., Bock, K., Mohr, G.J., 2010. Sens. Actuators. B. Chem. 150(1), 206‐210. Turns, M., 2011. Br. J. Nurs. 20(15 suppl.), s19‐25. Vasani, R.B., McInnes, S.J.P., Cole, M.A., Jani, A.M.M., Ellis, A.V., Voelcker, N.H., 2011. Langmuir 27(12), 7843‐7853. Vincent, G., 1994. Appl. Phys. Lett. 64(18), 2367‐2369. Wang, C., Yin, L., Zhang, L., Xiang, D., Gao, R., 2010. Sensors 10(3), 2088‐2106. Weingarten, M.S., Neidrauer, M., Mateo, A., Mao, X., McDaniel, J.E., Jenkins, L., Bouraee, S., Zubkov, L., Pourrezaei, K., Papazoglou, E.S., 2010. Wound Repair Regen. 18(2), 180‐185. Weingarten, M.S., Papazoglou, E., Zubkov, L., Zhu, L., Vorona, G., Walchack, A., 2006. Wound Repair Regen. 14(3), 364‐370. Weingarten, M.S., Papazoglou, E.S., Zubkov, L., Zhu, L., Neidrauer, M., Savir, G., Peace, K., Newby, J.G., Pourrezaei, K., 2008. Wound Repair Regen. 16(2), 234‐242. Wiggins, R.J., Wilks, M., Tabaqchali, S., 1985. J. Clin. Pathol. 38(8), 933‐936.
Winter, G.D., 1962. Nature 193(4812), 293‐294. Wong, I.K.Y., Andriessen, A., Charles, H.E., Thompson, D., Lee, D.T.F., So, W.K.W., Abel, M., 2012. J. Eur. Acad. Dermatol. Venereol. 26(1), 102‐110. Worsfold, O., Voelcker, N.H., Nishiya, T., 2006. Langmuir 22(16), 7078‐7083. Wright, J.B., Lam, K., Buret, A.G., Olson, M.E., Burrell, R.E., 2002. Wound Repair Regen. 10(3), 141‐151. Wu, J., Sailor, M.J., 2009. Adv. Funct. Mater. 19(5), 733‐741. Wu, P., Fisher, A.C., Foo, P.P., Queen, D., Gaylor, J.D.S., 1995. Biomaterials 16(3), 171‐175. Wyffels, J.T., Fries, K.M., Randall, J.S., Ha, D.S., Lodwig, C.A., Brogan, M.S., Shero, M., Edsberg, L.E., 2010. Int. Wound J. 7(4), 236‐248. Yager, D.R., Chen, S.M., Ward, S.I., Olutoye, O.O., Diegelmann, R.F., Cohen, I.K., 1997. Wound Repair Regen. 5(1), 23‐32. Zhou, J., Loftus, A.L., Mulley, G., Jenkins, A.T.A., 2010. J. Am. Chem. Soc. 132(18), 6566‐6570. Zhou, J., Tun, T.N., Hong, S.h., Mercer‐Chalmers, J.D., Laabei, M., Young, A.E.R., Jenkins, A.T.A., 2011. Biosens. Bioelectron. 30(1), 67‐72.
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Table 2. Summary of approaches used to monitor the wound environment Category Marker Approach Test Condition Reference
infection pyocyanin square wave voltammetry broth culture Sharp et al. (2010)
uric acid square wave voltammetry blood and blister fulid
Sharp et al. (2008)
phospholipase A2
and -hemolysin
phospholipid vesicles incubation with P. aeruginosa and S. aureus
Zhou et al. (2010), Zhou et al. 2011
neutrophil elastase and cathepsin G
chromophore/enzyme substrates
wound fluid Hasmann et al. (2011)
Edwards et al. (2005)
Ps. Aeruginosa, S. aureus
changes in conformation of polymer (PNIPAM-antibiotic) conjugates
infected 3D tissue engineered skin model
Sarker et al. (2011), Shepherd et al. (2010)
E. coli, Salmonella and Ps. Aeruginosa
photoluminescent pSi microcavity functionalized with receptor for diphosphoryl lipid A
incubation with E. coli, Salmonella and Ps. Aeruginosa
Chan et al. (2001), Fauchet (1998) Ouyang et al. (2005)
-haemolytic streptococci
odor sensors based on conducting polymer arrays
clinical trial of infected wounds
Parry et al. (1995)
E. coli, Ps. aeruginosa and S. aureus
odor sensors based on semi-conducting metal oxide films
infected wound swabs
Setkus et al. (2006)
temperature for bedsores/ inflammation
wireless carbon nanotube temperature sensors
healthy volunteers
Matzeu et al. (2011)
temperature / infection
color changing fibers n.a. Mestrovic (2009)
Infection / Healing Markers
pH (infection, healing marker)
indicator dyes n.a. Trupp et al. (2010), Mohr et al. (2008)
pH (infection, healing marker)
inductance change from pH sensitive hydrogel with wireless transponder
buffer solutions Sridhar and Takahata (2009)
Healing Markers
pH printed Electrodes measuring oxidation potential change in uric acid with pH
wound exudate Phair et al. (2011)
pH and partial luminescence life-time chronic wound Meier et al.
pressure of oxygen (pO2)
imaging sensors with digital photography
(2011), Schreml et al. 2011a), Schreml et al. (2011b)
pO2 transcutaneous oxometry measurements
hyperbaric oxygen patients
Grolman et al. (2001)
MMP-2, MMP-8, zein
Color change based on Protein and anti-body films on pSi resonators
Incubation with enzymes
Gao et al. (2008), Orosco et al. (2006) , Kilian et al. (2007), Martin et al. (2010)
moisture impedance across Ag/AgCl electrodes
simulated wound bed
McColl et al. (2007)
deoxy- and oxy-hemoglobin / blood flow
near infrared (NIR) spectroscopy
diabetic rat wound models, human trials
Weingarten et al. (2006) Weingarten et al. (2008) Weingarten et al. (2010)
wound area digital photograph algorithm
wound photographs
Papazoglou et al. (2010)
top related