prediction and monitoring the therapeutic response of chronic dermal wounds

Prediction and monitoringthe therapeutic response ofchronic dermal woundsKeith Moore, Roisin McCallion, Richard J. Searle, Michael C. Stacey, KeithG. Harding

Moore K, McCallion R, Searle RJ, Stacey MC, Harding KG. Prediction and monitoring the therapeutic response ofchronic dermal wounds. Int Wound J 2006;3:89—96.

ABSTRACTA significant proportion of chronic wounds fail to heal in response to treatment of underlying pathologiescombined with good wound care practice. Current prognostic tests to identify these wounds rely on the use ofalgorithms based on clinically measurable parameters such as wound dimensions and wound duration. Venousleg ulcers may be stratified into healing/non healing at 24 weeks of compression therapy and diabetic foot ulcertreatment outcome assessed using a 3-parameter algorithm. Accurate and reproducible measurement of woundarea is required for these algorithms to have clinical utility. Whilst a number of attempts have been made todevelop computerised wound-assessment techniques, wound tracing by clinicians combined with planimetryremains the standard methodology. Once treatment has been initiated, it is important to continuously monitorthe wound to assess efficacy of treatment. This can be achieved by measuring wound area change over the firstweeks of treatment to identify whether re-assessment of treatment strategy is required. A number of algorithmsfor assessing rate of wound area change have been evaluated to determine a surrogate endpoint for healing.Retrospective analysis of large patient groups indicates that approximately 75% correct prediction of healingoutcome can be achieved.

Key words: Analysis . Chronic wound . Diagnostic . Outcome . Prognosis

INTRODUCTIONFor the healthy individual, dermal healingis a structured process leading to rapid re-establishment of skin barrier function. It followsa defined temporal sequence of haemostasis,early and late inflammation, granulation tis-sue formation, extracellular matrix synthesis,re-modelling and scar formation (1). This gen-erates a healing trajectory that can be used topredict time to closure of wounds healing by

secondary intention after measurement ofinitial wound dimensions (2). However, manyfactors such as infection, comorbidities and age-ing may exert a negative influence on healing toinduce deviation from an optimal trajectory.This can lead to impaired healing resulting indelay or even prevention of wound closure.Chronic wounds with impaired healing includevenous leg ulcers, diabetic foot ulcers and pres-sure ulcers. They occur at a relatively high fre-quency estimated at >1% of an aged populationfor venous leg ulcer (VLU) (3), with 15% ofdiabetics likely to develop foot ulceration (4),and the prevalence of pressure ulcers can beup to 15% in an acute care setting (5).

Many chronic wounds respond to a combi-nation of treatment of underlying pathologies(compression therapy for VLU, pressure relieffor DFU and PU) and good wound care

Authors: K Moore, PhD, WoundSci, Usk, Monmouthshire,UK; R McCallion, PhD, Smith & Nephew Medical Ltd, Hull,UK; RJ Searle, PhD, Smith & Nephew Medical Ltd, Hull, UK; MCStacey, FRACS, Department of Surgery, University of WesternAustralia, Freemantle Hospital, Australia; KG Harding, FRCS,Wound Healing Research Unit, Department of Surgery, CardiffUniversity, Cardiff, UKAddress for correspondence: K Moore, PhD, WoundSci,Usk, Monmouthshire, UKE-mail: [email protected]

� 2006 The Authors. Journal compilation � 2006 Blackwell Publishing Ltd and Medicalhelplines.com Inc 89

Key Points

. for the healthy individual,dermal healing is a structuredprocess leading to rapidre-establishment of skin barrierfunction

. this follows a defined temporalsequence of hemostasis, earlyand late inflammation, granula-tion tissue formation, extracellu-lar matrix synthesis remodellingand scar formation

. this generates a healing trajec-tory that can be used to predicttime to closure of wounds heal-ing by secondary intention aftermeasurement of initial wounddimensions

. however, many factors such asinfection, comorbidities and age-ing may exert a negative influenceon healing to induce deviationfrom an optional trajectory

. this can lead to impaired heal-ing resulting in delay or evenprevention of wound closure

REVIEW&

practice by using wound contact dressingsthat generate a moist wound environment.However, a substantial minority remainsrefractory to treatment (6) and may requireadjunctive therapies to achieve healing. Anexpanding body of knowledge characterisingthe differences between healing and non heal-ing wounds has identified potential therapeu-tic targets and has led to the development of anumber of treatments that are designed tomodulate healing by interacting with keyaspects of biological events that regulate thehealing process. Examples include recombi-nant growth factors such as platelet-derivedgrowth factor (7), keratinocyte growth factor(8), granulocyte-monocyte colony-stimulatingfactor (9), tissue-engineered dermal replace-ments (10), wound dressings with potentialbioactivity (11,12) or synthetic protease inhibi-tors (13).

The increasing number of therapeutic tar-gets and treatments for chronic wounds gen-erates a need to make early and rationaltherapeutic choices by identifying thosewounds requiring advanced therapies and tomonitor treatment efficacy. The demand forvalidated objective wound-monitoring assess-ment systems at all levels of care was empha-sised by a recent publication entitled ‘Why

won’t this wound heal and what should I do

about it?’ (14). An ideal solution would be aminimally invasive diagnostic/prognosticmonitoring system analogous to those usedin other pathologies such as diabetes wherethe concentration of an analyte, glucose, isknown to be closely linked to disease progres-sion and response to treatment. However, thechronic wound may have a number of contri-buting factors causing multiple pathologicaldifferences compared with a healing wound.Our current understanding of the physio-logical events associated with healing does notallow development of a single parameter-based assay, and it is probable that a success-ful wound-monitoring system will utilisemultiparametric assessment possibly incor-porating as yet unidentified biomarkers.

The absence of such a system necessitatesreliance on existing clinical and laboratorymeasurements, and it is the objective of thisreview to focus on clinically measurablewound parameters that may be used for pre-dicting outcome and monitoring treatmentefficacy in patients with chronic wounds.

WOUND PROGNOSTICINDICATORSIf available, a validated prognostic indicatorwould be used to stratify wounds into thoserequiring advanced therapies and thosepotentially responsive to standard treatment,and would confer quality of life and costbenefits by accelerating healing and reducingusage of inappropriate treatment. Althoughmany parameters are known to be associatedwith a poor outcome for chronic wounds(Table 1), only initial wound area and woundduration have been incorporated into statistic-ally evaluated prognostic algorithms.

Venous leg ulcers of long duration (15,16)and with a larger area at initiation of treat-ment have a decreased chance of healing inresponse to compression therapy (17). Anulcer length greater than 10 cm (18), increasedwidth, length, length � width and area allcorrelate with failure to heal, but when theulcer area is >40 cm2 the correlation betweenarea and healing/non healing decreases (19).A simple model for VLU prognosis based onscoring 1 point for ulcers >5 cm2 and 1 pointfor >6 months’ duration was developed todiscriminate between those wounds that willheal after 24 weeks compression bandaging(95% of wounds with a score of 0) and thosewith a decreased chance of healing (37% ofwounds with a score of 2) (20). Followinganalysis of a larger patient cohort, this modelwas subsequently refined so that a woundwith an initial area <10 cm2 of <12 months’duration will have a 81% chance of healing by24 weeks of compression bandaging, whilstone, that is, area of >10 cm2 and >12 months’duration has only a 22% chance of healing

Table 1 Risk factors for delayed healing of venous leg ulcers

Deep vein incompetence (55)

Popliteal vein reflux (56)

Socio-economic status (57)

History of previous ulceration (16)

Highly exuding ulcers (16)

Increasing age (16)

Male gender (58)

High body mass index (59)

Low serum zinc (60)

Low albumin levels (61)

Anxiety and depression (62)

Long duration (16)

Large area (17)

Therapeutic response of chronic dermal wounds

� 2006 The Authors. Journal compilation � 2006 Blackwell Publishing Ltd and Medicalhelplines.com Inc90

Key Points

. the increasing number of thera-peutic targets and treatmentsfor chronic wounds generates aneed to make early and rationaltherapeutic choices by identify-ing those wounds requiringadvanced therapies and tomonitor treatment efficacy

. an ideal situation would be aminimally invasive diagnostic/prognostic monitoring systemanalogous to those used inother pathologies such as dia-betes where the concentrationof an analyte, glucose, isknown to be closely linked todisease progression andresponse to treatment

. the absence of such a systemnecessitates reliance on existingclinical and laboratory measure-ments and it is the objective ofthis review to focus on clinicallymeasurable wound parametersthat may be used for predictingoutcome and monitoring treat-ment efficacy in patients withchronic wounds

. a validated prognostic indicatorwould be used to stratifywounds into those requiringadvanced therapies and thosepotentially responsive to stan-dard treatment and would con-fer quality of life and costbenefits by accelerating healingand reducing usage of inap-propriate treatment

(21). It is important to recognise that thesedata were determined for VLU undergoingcompression therapy. The authors introducedthe important caveat that the prognostic indica-tors may not be valid for predicting outcomesof new therapies untested in this model.

The same relationship between healing, sizeand duration applies to diabetic neuropathicfoot ulcers in that ulcers healing within 20weeks of standard care are more likely to besmaller and of shorter duration. Incorporationof ulcer grade allows a 3-parameter model assuperficial diabetic wounds heal more effec-tively than deeper wounds and those withabscess and osteomyelitis (22). The model isbased on a scoring system in which a point isscored if the wound is older than 2 months,larger than 2 cm2, or has a grade �3 on a6-point scale. An increased score indicates agreater chance of non healing so that a score of3 indicates an 81% chance of non healingcompared to 35% for a score of 0 (23). Largediabetic ulcer size also correlates with a riskof subsequent limb amputation (24).

MONITORING RESPONSE TOTREATMENTPrognostic algorithms are intended to assessthe patient and wound at a single time pointbefore treatment as an aid to the developmentof a treatment strategy. Once treatment hasbeen initiated, a dynamic model is requiredto monitor change in wound status and toevaluate therapeutic efficacy with sufficientaccuracy that it can be used as a decision-making tool. Wound closure is the obviousendpoint of healing, and the logical and sim-ple measure of progress towards that end-point is to monitor response to treatment byassessing change in wound size over time. It isaccepted clinically that measurements taken atregular intervals give a good indication of awound’s healing progress and allow for atimely recognition of improvement or dete-rioration of the wound condition (25). Usingthis parameter allows monitoring algorithms

to be devised using historical data and then tobe evaluated in prospective studies.

Wound area change over 4 (26—28) or 3weeks (15) has been identified as the bestindicator for use as a surrogate endpoint tomonitor healing response and predict out-come. A complicating factor in evaluatinghealing is the consideration of which para-meter to use in calculating change in woundarea. In attempting to use absolute woundarea change after 4 weeks of treatment to pre-dict outcome at 24 weeks, the rate of healingor area healed per week did not differentiatebetween healing and non healing (19). Percentarea reduction has been recommended as thebest way of predicting healing rates (29),although this has been challenged on thebasis that it biases wound closure rate infavour of smaller wounds (30). Expression ofthe data as percentage change in area frombaseline to 4 weeks provides the best com-bination of positive and negative predictivevalues (68�2 and 74�7%, respectively) and thelargest area (0�75) under the ReceiverOperator Characteristis (ROC) curve (Box 1, 31).A retrospective study of 56 488 wounds (32)found that the log healing rate [(Log area attime 0 – Log area at time þ t weeks later)/t],and percentage change in wound area mea-sured over a 4-week period could also be usedas a surrogate marker of healing at 12 or 24weeks. The ROC value found was 0�72—0�80for change in wound area over 4 weeks.

In other studies, estimating change in areaat 3 weeks by comparison with initial mea-surement, the so-called baseline adjustedhealing rate, was found to be unsatisfactoryin predicting outcome (33). However, by takingthe mean of healing rates between each visit todetermine a mean-adjusted healing rate, it didprove possible at 3 weeks after starting treat-ment to predict eventual healing outcome. Bycomparison with absolute area change or per-centage area change, it has been suggestedthat use of the rate of change of wound peri-meter may provide a more accurate reflectionof healing rates when comparing ulcers of

BOX 1.The Receiver Operator Characteristic

The ROC curve is often used to evaluate diagnostic tests. It provides a measure of the ability of a test to properly detect an event

that actually occurs, called sensitivity or the true-positive fraction, relative to false detection in the absence of the event, called the

false-positive fraction (63). The closer the area under the ROC curve approaches a value of 1, the greater the utility of the test.



Key Points

. prognostic algorithms areintended to assess the patientand wound at a single timepoint before treatment as anadd to the development of atreatment strategy

. wound closure is the obviousendpoint of healing and thelogical and simple measure ofprogress towards that endpointis to monitor response to treat-ment by assessing change inwound size over time

. it is accepted clinically that meas-urements taken at regular inter-vals give a good indication of awound’s healing progress andallow for a timely recognition ofimprovement or deterioration ofthe wound condition

. it has been suggested that use ofthe rate of change of wound peri-meter may provide a more accu-rate reflection of healing rateswhen comparing ulcers of differ-ent size, as this parameter is inde-pendent of wound geometry

. clearly, further developmentwork is required before a reli-able and universally applicablewound monitoring method isavailable

different size, as this parameter is independentof wound geometry (34). Although woundperimeter is rarely calculated in clinicalpractice, it is readily available if the area ismeasured by wound tracing.

The utility of measuring early area changeon a single patient basis (as opposed to com-parison of patient groups) has recently beenquestioned. A study of 17 patients indicatedthat the ability of initial healing rate to predicthealing time for patients undergoing com-pression therapy was poor (35). Although anincrease in wound area over 4 weeks consis-tently correlated with non healing, it did notprove possible to relate rate of area decreaseand outcome with any accuracy. This mayhave been a consequence of the heterogeneityof the wounds included and the low healingrates (29%) achieved in this study comparedwith the higher rates (66%) in others where apositive correlation was demonstrated (32).For a prognostic test to be of value, it musthave utility when applied to a patient in aclinical setting. On the basis of Hill’s results(35), the value of this approach may berestricted to identifying those patients whorequire a re-assessment of their treatmentregime after 4 weeks of treatment. In contrast,a study of PU healing (36) concluded that ifstage IV ulcers and wounds with initial size<2 cm2 are assessed by initial area change,then use of a Gompertzian statistical modelprovides a relevant method to evaluate thera-peutic interventions. Clearly, further develo-pment work is required before a reliable anduniversally applicable wound monitoringmethod is available.

MEASUREMENT OF WOUND AREAAssessment of wound appearance has tradi-tionally been evaluated visually by cliniciansadministering patient care. Such assessment ishighly dependant on individual proficiency(37). The obvious problems of lack of objectiv-ity and requirement for skill and experienceare compounded if different clinicians makeassessments at different times on the samewound. There is thus a need for objectiveand documented information to monitorwound progress and to determine the effectof therapeutic interventions. Instrumentationdeveloped for this purpose needs to duplicatethe information derived by human assessmentand, if possible, derive additional information

using techniques such as colour or surfacetexture analysis.

Manual techniquesA crude assessment of wound area may bederived by measuring length and breadthwith a ruler. Chronic wounds are rarely sym-metrical, and this approach does not generatethe desired accuracy (38). The widely usedalternative method of tracing the wound mar-gin onto a transparent film is more accurateand simple to use. It requires skill in use withcomplex or circumferential wounds and issubjective in that the clinician has to decidewhere the actual wound margin is when tra-cing onto the film. The tracing forms a perma-nent record, and area can be calculated bycounting squares if a grid is placed over thetracing or by mechanical or computerised pla-nimetry. A hand-held device, Visitrak Digital(Smith & Nephew Medical, Hull, UK), hasrecently been introduced to reduce the timeinvolved in calculation of area from woundtracings and improve inter- and intraoperatoraccuracy so that area change over time can bemore accurately monitored and recorded (39).

Automated techniquesIdentification of the wound margin to allowautomated definition of the wound peripheryand calculation of the wound area remains amajor challenge. This can be achieved in aresearch setting using deformation of struc-tured light to define the wound surface.Laser systems have been demonstrated to befaster and of equivalent accuracy to tracing formeasurement of simulated wounds in vitro

(40). Use of a laser mounted on a motorisedX-Y table located above the wound surfacegives less than 1% error in measuring thevolume of such model wounds (41).

Two systems using structured light techni-ques are currently available to measurewound parameters directly in a clinical situa-tion. Patterns of laser-generated lines or dotsare projected onto the area around the wound.The measurement of area and volume instru-ment (MAVIS) system generates a calculatedvolume from the degree of observed distor-tion of parallel lines (42) but still requires theintervention of a clinician to define the woundmargin on a computer display. The require-ment for full automation is addressed by analternative system utilising a pattern of dots



Key Points

. assessment of wound appearancehas traditionally been evaluatedvisually by clinicians administeringpatient care and such assessmenthas been highly dependent of indi-vidual proficiency

. this is compounded if differentclinicians make assessments atdifferent times on the samewound

. instrumentation developed forthis purpose needs to duplicatethe information derived byhuman assessment and if possi-ble, derive additional informationusing techniques such as colouror surface texture techniques

. a crude assessment of woundarea may be derived by measur-ing length and breadth with aruler

. chronic wounds are rarely symme-trical and this approach does notgenerate the desired accuracy

. a hand held device, VisitrakDigital (Smith & NephewMedical) has recently been intro-duced to reduce the timeinvolved in calculation of areafrom wound tracings andimprove inter and intraoperatoraccuracy so that area changeover time can be more accu-rately monitored and recorded

. laser systems have been demon-strated to be faster and of equiva-lent accuracy to tracing formeasurement of simulatedwounds in vitro but these techni-ques require sophisticated expen-sive and cumbersomeinstrumentation at the patient’sside

. development of computerisedimage analysis for measurementof wound area brings with it thepossibility of simultaneous colouranalysis to impose objectivity toan assessment that has reliedheavily on clinical expertise

projected over the wound surface. After imagecapture, the distortion of each dot is assessed,and where this falls outside the predictedshape, a computerised decision is made sothat the dot falls on the wound margin toallow perimeter definition by interpolatingdots at the margin (43).

These techniques require sophisticated,expensive and relatively cumbersome instru-mentation at the patient’s side. With wideavailability of digital cameras, a more attrac-tive option is to capture a digital image of thewound for subsequent computerised analysis(44). In effect, this converts a 3D image into 2dimensions, and single plane photographs areinherently limited by this transformation. If awound extends around the curvature of theleg then a 2D image under-represents the truewound area. A further limitation is that the 2Dimage takes no account of the depth of thewound. Commercial products have beendeveloped for photographic image analysissuch as the VERGE Videometer system (45).A digital image of the wound is captured forcomputerised analysis, but the system suffersfrom the same limitation as the MAVIS systemin that the clinician has to delineate thewound margin.

Progress has been made towards com-puterised delineation of the wound margin.Using an adaptive spline technique (46), it ispossible to define the wound margin byperforming analysis of image features in theregion of the wound margin. Currently, this isa semi-automatic technique as a sequence ofinitial control points have to be defined on theimage at the wound margin. This techniquehas been incorporated into a system for colouranalysis of venous leg ulcers (47).

Colour analysis can be of value in assessinga wound’s progress towards healing.Depending upon severity and aetiology,wounds may to some extent be covered withblack eschar and epithelialisation will failuntil it is removed by debridement. As treat-ment succeeds and healing progresses, thewound colour changes from black to yellowto red and finally granular red. Less severewounds may appear a healthy red colour onpresentation. Wound appearance has beenwidely documented on a red/yellow/blacksystem (48) and although easy to use hasbeen criticised as difficult to quantify manu-ally (49). Development of computerised image

analysis for measurement of wound areabrings with it the possibility of simultaneouscolour analysis to impose objectivity to anassessment that has relied heavily on clinicalexpertise.

CONCLUSIONFollowing an initial diagnostic assessment,current treatment strategies follow a trial-and-error approach. For a venous leg ulcer,the first line approach is compression band-aging that will induce healing in a proportionof patients. Those non responsive to compres-sion therapy alone may require adjunctivetherapies such as tissue-engineered dermalequivalents that provide temporary woundcover or growth factors that may stimulatehealing (50). The remaining non responsivewounds may then be treated with aggressivesurgical debridement or venous surgery.There will remain a residual population ofwounds that are refractory to any treatment.These wounds will require management tomaintain patient quality of life. As a non heal-ing wound passes through each treatmentiteration, the total treatment costs and impacton patient quality of life increase. These costsmight be minimised if a prognostic indicatorwas available to identify at treatment initi-ation those wounds potentially non responsiveto compression therapy. Additional clinicalbenefit may be gained by the use of a mon-itoring system that would give an earlywarning of poor treatment response and anindication that a change to the treatmentregime was required to stimulate healing.

The biological complexity of the healingprocess and the interrelatedness of the manyprocesses involved in healing present manyopportunities for underlying pathologies tolead to impaired healing. It is unsurprisingtherefore that many risk factors for non heal-ing have been identified (Table 1). Patient andwound assessment form the basis for chronicwound management, and consideration ofrisk factors at initial sassessment can providean experienced clinician with an indication ofthe potential outcome of treatment. However,the only non subjective prognostic indicatorcurrently available for the less experiencedpractitioner is an algorithm based on woundarea and wound duration. Measurement ofarea by tracing is relatively simple and com-bined with wound duration can give an



Key Points

. following an initial diagnosticassessment, current treatmentstrategies follow a trial anderror approach

. as a non healing wound passesthrough each treatment itera-tion, the total treatment costsand impact on patient qualityof life increases

. clinical benefit may be gainedby the use of a monitoring sys-tem that would give and earlywarning of poor treatmentresponse and an indicationthat a change to the treatmentregime was required to stimu-late healing

. consideration of risk factors atinitial assessment can provideclinician with an indication ofthe potential outcome oftreatment

. however, the only non sub-jective prognostic indicatorcurrently available is an algo-rithm based on wound areaand wound duration

. monitoring wound area changeon a regular basis remains themain method for evaluation oftreatment response

. the possible combination ofbiomarkers characterizing thewound microenvironment withdynamic clinically measurablewound parameters may yieldan accurate

. wound assessment and moni-toring system

estimate of healing potential of venous ulcersin response to compression therapy with a rea-sonable level of confidence. Whilst not givingan absolute prediction of outcome, this algo-rithm may be used to identify those woundsthat require more rigorous evaluation ofresponse to treatment.

As well as depending on successful treat-ment of the underlying pathologies, woundmanagement outcome is dependent on exo-genous factors such as bacterial proliferationin wound tissue (51). This requires continuousmonitoring of the wound to ensure thatappropriate interventions are implemented.Whilst spreading infection is identified bythe clinical symptoms of odour, pain and cel-lulitis, a lower level of bacterial bioburdenresulting in critical colonisation or local infec-tion may cause inhibition of healing withoutthe associated symptoms of infection (52).Critical colonisation may manifest in impairedhealing and may be observed as a non initia-tion or a cessation of healing in response tocompression therapy. Whilst laboratory teststo quantify bacterial bioburden are availableon biopsy tissue, these are not routinely per-formed clinically on a prospective basis.

Monitoring wound area change on a regularbasis remains the main method for evaluationof treatment response. Whilst this can be per-formed on an ad hoc basis, it requires accurateand reproducible measurement at weeklyintervals to accurately monitor and documentwound progress. Additional to monitoringresponse, wound area change over time canbe used as a predictor of outcome with anapproximate 75% accuracy as shown by thearea under the ROC curve. It has been sug-gested that this level of discriminationbetween healing and non healing is sufficientfor the rate of wound area change over aninitial 4 weeks of treatment to be used as asurrogate endpoint for clinical trials, and toidentify patients unlikely to heal early in thecourse of treatment (32). The latter concept issupported by prospective data where nonresponding wounds were accurately identi-fied by increasing wound area (35). However,this study did not find a good correlationbetween rate of wound area decrease andsubsequent healing. These data strike a noteof caution because the clinical value in thismethod lies in its use as an early prospectivetest to evaluate the success of a particular

treatment regime. Validation in clinical usefor the prediction of time to healing willhave to await the outcome of further prospec-tive studies.

Whilst the measurement of clinically mea-surable wound parameters has value inpredicting and monitoring healing outcome,these are at best 80% accurate. A large bodyof data is available to characterise the molecu-lar environment of the chronic wound (53),but no analytes have been identified forincorporation into a test of wound prognosisand monitoring. The possible combinationof biomarkers characterising the woundmicroenvironment (54) with dynamic clini-cally measurable wound parameters mayyield an accurate wound assessment andmonitoring system.

REFERENCES1 Cherry GW, Hughes MA, Leaper DJ, Ferguson MWJ.

Wound healing. In: Morris PJ, Wood WC, editors.Oxford text book of surgery, 2nd edn. OxfordUniversity Press, Oxford, 2002:129—59.

2 Marks JL, Hughes E, Harding KG, Campbell H,Ribeiro CD. Prediction of healing time as an aidto the management of open granulating wounds.World J Surg 1983;7:641—5.

3 Margolis DJ, Bilker W, Santanna J, Baumgarten M.Venous leg ulcer: incidence and prevalence in theelderly. J Am Acad Dermatol 2002;46:381—6.

4 Reiber GE. The epidemiology of diabetic footproblems. Diabet Med 1998;13 Suppl:S6—S11.

5 Amlung SR, Miller WL, Bosley LM. The 1999national pressure ulcer prevalence survey: abenchmarking approach. Adv Skin Wound Care2001;14(6):297—301.

6 Franks PJ, Moffatt CJ, Connolly M, Bosanquet N,Oldroyd MI, Greenhalgh RM, McCollum CN.Factors associated with healing leg ulcerationwith high compression. Age Ageing1995;24:407—10.

7 Guzman-Gardearzabal E, Leyva-Bohorquez g Salas-Colin S, Paz-Janeiro JL, Alvarado-Ruiz R,Garcia-Salazar R. Treatment of chronic ulcers inthe lower extremities with topical becaplermingel .01%: a multicenter open-label study. Adv Ther2000;17:184—9.

8 Robson MC, Phillips TJ, Falanga V, Odenheimer DJ,Parish LC, Jensen JL, Steed DL. Randomized trialof topically applied repifermin (recombinanthuman keratinocyte growth factor-2) to acceleratewound healing in venous ulcers. Wound RepairRegen 2001;9:347—52.

9 Da Costa RM, Ribeiro Jesus FM, Aniceto C,Mendes M. Randomized, double-blind, placebo-controlled, dose-ranging study of granulocyte-macrophage colony stimulating factor in patients



with chronic venous leg ulcers. Wound RepairRegen 1999;7:17—25.

10 Marston WA, Hanft J, Norwood P, Pollak R. Theefficacy and safety of dermagraft in improvingthe healing of chronic diabetic foot ulcers: resultsof a prospective randomized trial. Diabetes Care2003;26:1701—5.

11 Cullen B, Smith R, McCulloch E, Silcock D,Morrison L. Mechanism of action ofPROMOGRAN, a protease modulating matrix,for the treatment of diabetic foot ulcers. WoundRepair Regen 2002; 10:16—25.

12 Colletta V, Dioguardi D, Di Lonardo A, Maggio G,Torasso F. A trial to assess the efficacy and toler-ability of Hyalofill-F in non-healing venous legulcers. J Wound Care 2003;12:357—60.

13 Fray MJ, Dickinson RP, Huggins JP, OcclestonNLA. Potent selective inhibitor of matrix metal-loproteinase-3 for the topical treatment ofchronic dermal ulcers. J Med Chem 2003; 46:3514—25.

14 Doughty DB. Why won’t this wound heal, and whatshould i do about it? Home Healthc Nurse2004;22:18—20.

15 Phillips TJ, Machado F, Trout R, Porter J, Olin J,Falanga V. Prognostic indicators in venous ulcers.J Am Acad Dermatol 2000;43:627—30.

16 Gohel MS, Taylor M, Earnshaw JJ, Heather BP,Poskitt KR, Whyman MR. Risk factors for delayedhealing and recurrence of chronic venous legulcers — an analysis of 1324 legs. Eur J VascEndovasc Surg 2005;29(1):74—7.

17 Skene AI, Smith JM, Dore CJ, Charlett A, Lewis JD.Venous leg ulcers: a prognostic index to predicttime to healing. BMJ 1992;305:1119—21.

18 Meaume S, Couilliet D, Vin F. Prognostic factors forvenous ulcer healing in a non-selected populationof ambulatory patients. J Wound Care 2005;14(1):31—4.

19 Kantor J, Margolis DJ. Efficacy and prognostic valueof simple wound measurements. Arch Dermatol1998;134:1571—4.

20 Margolis DJ, Berlin JA, Strom BL. Which venous legulcers will heal with limb compression bandages?Am J Med 2000;109:15—9.

21 Margolis DJ, Allen-Taylor L, Hoffstad O, Berlin JA.The accuracy of venous leg ulcer prognosticmodels in a wound care system. Wound RepairRegen 2004;12:163—8.

22 Apelqvist J, Castenfors J, Larsson J, Stenstrom A,Agardh CD. Wound classification is more impor-tant than site of ulceration in the outcome ofdiabetic foot ulcers. Diabet Med 1989;6:526—30.

23 Margolis DJ, Kantor J, Santanna J, Strom BL, Berlin JA.Risk factors for delayed healing of neuropathicdiabetic foot ulcers: a pooled analysis. ArchDermatol 2000;136:1531—5.

24 Oyibo SO, Jude EB, Tarawneh I, Nguyen HC,Armstrong DG, Harkless LB, Boulton AJ. Theeffects of ulcer size and site, patient’s age, sexand type and duration of diabetes on the outcomeof diabetic foot ulcers. Diabet Med 2001;18:133—8.

25 Charles H. Wound assessment: measuring the areaof a leg ulcer. Br J Nurs 1998;7:765—72.

26 Cukjati D, Robnik-Sikonja M, Rebersek S, Kononenko I,Miklavcic D. Prognostic factors in the prediction ofchronic wound healing by electrical stimulation.Med Biol Eng Comput 2001;39:542—50.

27 Sheehan P, Jones P, Caselli A, Giurini JM, Veves A.Percent change in wound area of diabetic footulcers over a 4-week period is a robust predictorof complete healing in a 12-week prospective trial.Diabetes Care 2003;26:1879—82.

28 Margolis DJ, Gelfand JM, Hoffstad O, Berlin JA.Surrogate end points for the treatment of diabeticneuropathic foot ulcers. Diabetes Care2003;26:1696—700.

29 Flanagan M. Wound measurement: can it help us tomonitor progression to healing? J Wound Care2003;12:189—94.

30 Gilman T. Wound outcomes: the utility of surfacemeasures. Int J Low Extrem Wounds 2004;3: 125—32.

31 Kantor J, Margolis DJ. A multicentre study of per-centage change in venous leg ulcer area as a prog-nostic index of healing at 24 weeks. Br J Dermatol2000;142:960—4.

32 Gelfand JM, Hoffstad O, Margolis DJ. Surrogateendpoints for the treatment of venous leg ulcers.J Invest Dermatol 2002;119:1420—5.

33 Tallman P, Muscare E, Carson P, Eaglstein WH,Falanga V. Initial rate of healing predicts completehealing of venous ulcers. Arch Dermatol1997;133:1231—4.

34 Gorin DR, Cordts PR, LaMorte WW, Manzoian JO.The influence of wound geometry on the mea-surement of wound healing rates in clinical trials.J Vasc Surg 1996;23:524—8.

35 Hill DP, Poore S, Wilson J, Robson MC, Cherry GW.Initial healing rates of venous ulcers: are theyuseful as predictors of healing? Am J Surg2004;188(1A Suppl):22—5.

36 Wallenstein S, H.Brem. Statistical analysis ofwound-healing rates for pressure ulcers. Am JSurg., 2004;188(1A Suppl):73—8.

37 Dhawan AP, Sicsu A. Segmentation of skin imagesusing colour texture information of surface pig-mentation. Comput Med Imaging Graph1992;16:163—77.

38 Majeske C. Reliability of wound surface area mea-surements. Phys Ther 1992;72(2):138—41.

39 Moore K. Using wound area measurement to pre-dict and monitor response to treatment of chronicwounds. J Wound Care 2005;14(5):229—32.

40 Smith RB, Rogers B, Tolstykh GP, Walsh NE, Davis MG,Bunegin L, Williams RL. Three-dimensional laserimaging system for measuring wound geometry.Lasers Surg Med 1998;23:87—93.

41 Patete PV, Bulgrin JP, Shabani MM, Smith DJ. Anon-invasive, three-dimensional, diagnostic laserimaging system for accurate wound analysis.Physiol Meas 1996;17:71—9.

42 Plassmann P, Jones TD. MAVIS: a non-invasiveinstrument to measure area and volume ofwounds. Measurement of area and volume instru-ment system. Med Eng Phys 1998;20: 332—8.

43 Krouskop TA, Baker R, Wilson MS. A noncontactwound measurement system. J Rehabil Res Dev2002;39:337—45.



44 Taylor RJ. Mouseyes revisited: upgrading a com-puter program that aids wound measurement.J Wound Care 2002;11:213—6.

45 Williams C. The Verge Videometer wound measure-ment package. Br J Nurs 2000;9:237—9.

46 Hoppe A, Wertheim D, Melhuish J, Harding KG,Williams RJ. A spline based method for assess-ment of wound images. In: Advances in medicalsignal and information processing. ConferencePublication no. 475. Bristol, UK: Institute ofElectrical Engineers, 2000, 206—11.

47 Oduncu H, Hoppe A, Clark M, Williams RJ,Harding KG. Analysis of skin wound images usingdigital color image processing: a preliminary com-munication. Int J Low Extrem Wounds 2004;3:151—6.

48 Goldman RJ, Salcido R. More than one way to mea-sure a wound: an overview of tools and techni-ques. Adv Skin Wound Care 2002;15:236—45.

49 Maklebust J. Pressure ulcer assessment. Clin GeriatrMed 1997;13:455—81.

50 Roberts C, Mansbridge J. The scientific basis anddifferentiating features of dermagraft. Can J PlastSurg 2002;10 Suppl A:6A—13A.

51 Bowler PG, Duerden BI, Armstrong DG. Woundmicrobiology and associated approaches to woundmanagement. Clin Microbiol Rev 2001;14:244—69.

52 Edwards R, Harding KG. Bacteria and wound heal-ing. Curr Opin Infect Dis 2004;17:91—6.

53 Drinkwater SL, Smith A, Burnand KG. What canwound fluids tell us about the venous ulcer microen-vironment? Int J Low Extrem Wounds 2002;1:184—90.

54 Pleat JM, Hebestreit HF, Dwek RA. Identifying pro-tein production in wound healing: current techni-ques. J Wound Care 2002;11:63—6.

55 Bradbury AW, Stonebridge PA, Callam MJ, Ruckley CV,Allan PL. Foot volume try and duplex ultrasono-graphy after saphenous and subfascial perforat-ing vein ligation for recurrent venous ulceration.Br J Surg 1993;80:845—8.

56 Barwell J, Ghauri ASK, Taylor M, Deacon J, Wakely C,Poskitt KR, Whyman MR. Risk factors for healingand recurrence of chronic venous leg ulcers.Phlebology 2000;15:49—52.

57 Franks PJ, Bosanquet N, Connolly M, Oldroyd MI,Moffatt CJ, Greenhalgh RM, McCollum CN.Venous ulcer healing: effect of socioeconomic fac-tors in London. J Epidemiol Community Health1995;49:385—8.

58 Taylor RJ, Taylor AD, Smyth JV. Using an artificialneural network to predict healing times and riskfactors for venous leg ulcers. J Wound Care2002;11:101—5.

59 Platt AJ, Mohan D, Baguley P. The effect of bodymass index and wound irrigation on outcomeafter bilateral breast reduction. Ann Plast Surg2003;51:552—5.

60 Zorrilla P, Salido JA, Lopez-Alonso A, Silva A.Serum zinc as a prognostic tool for wound healingin hip hemiarthroplasty. Clin Orthop 2004;420:304—8.

61 James TJ, Hughes MA, Cherry GW, Taylor RP.Simple Biochemical markers to assess chronicwounds. Wound Repair Regen 2000;8:264—9.

62 Cole-King A, Harding KG. Psychological factorsand delayed healing in chronic wounds.Psychosom Med 2001;63:216—20.

63 Hanley JA, McNeil BJ. The meaning and use of thearea under a receiver operating characteristic(ROC) curve. Radiology 1982;143:29—36.



prediction and monitoring the therapeutic response of chronic dermal wounds

Documents