Cellular versus acellular matrix devices in treatment of diabetic foot ulcers: study protocol for a comparative efficacy randomized controlled trial.

0.2 MB·English

by  Lev-Tov, Hadar

Lev-Tovetal.Trials2013,14:8
http://www.trialsjournal.com/content/14/1/8 TRIALS
STUDY PROTOCOL Open Access
Cellular versus acellular matrix devices in
treatment of diabetic foot ulcers: study protocol
for a comparative efficacy randomized controlled
trial
Hadar Lev-Tov1,2, Chin-Shang Li3, Sara Dahle1,2* and Roslyn Rivkah Isseroff1,2*
Abstract
Background: Diabetic foot ulcers (DFUs) represent a significant source of morbidity and anenormous financial
burden. Standard care for DFUs involves systemic glucose control, ensuring adequate perfusion, debridement of
nonviable tissue, off-loading, control of infection, local wound care and patienteducation, all administered by a
multidisciplinary team.Unfortunately, even with the beststandard ofcare (SOC) available, only24% or 30% of DFUs
will heal atweeks 12 or 20,respectively.
The extracellular matrix(ECM) in DFUs is abnormal and itsimpairment has been proposed as a key target for new
therapeutic devices. These devices intendto replace theaberrant ECM by implanting a matrix, either devoid ofcells
or enhanced with fibroblasts, keratinocytes or both as well as various growth factors. These new bioengineered skin
substitutes are proposed to encourage angiogenesis and in-growthof new tissue, and to utilize livingcells to
generate cytokines neededfor wound repair.
Todate, the efficacy of bioengineered ECM containing live cellular elements for improving healing above that of a
SOC control group has not been compared withthe efficacy of anECM devoid of cellsrelative to the same SOC.
Our hypothesis is thatthere is nodifference inthe improved healing effected byeitherof thesetwo product types
relative to SOC.
Methods/Design:To testthis hypothesis we propose a randomized, single-blind, clinical trial withthree arms: SOC,
W W
SOC plus Dermagraft (bioengineered ECM containing livingfibroblasts) and SOC plus Oasis (ECM devoid of
living cells) in patients with nonhealing DFUs. The primary outcome is thepercentage of subjects that achieved
complete wound closure by week 12.
Discussion: Ifour hypothesisis correct, then immensecost savings could be realized by using the
orders-of-magnitude less expensive acellular ECM device without compromising patient health outcomes.The
article describes theprotocol proposed to testour hypothesis.
Trial registration: ClinicalTrials.gov: NCT01450943. Registered:7October2011
Keywords: Diabetic foot ulcer, Chronicwounds,Nonhealingwounds,Oasis,Dermagraft, Wound matrix
*Correspondence:[email protected];[email protected]
1VeteransAffairsMedicalCenter,NorthernCaliforniaHealthcareSystem,
10535HospitalWay,Mather,CA95655,USA
2DepartmentofDermatology,UniversityofCaliforniaDavis,3301CStreet,
Sacramento,CA95816,USA
Fulllistofauthorinformationisavailableattheendofthearticle
©2013Lev-Tovetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Lev-Tovetal.Trials2013,14:8 Page2of8
http://www.trialsjournal.com/content/14/1/8
Background Keratinocytes in diabetic wounds exhibit impaired mi-
Diabetes affects nearly one-third of the adult population gration,whichmaybemediatedbyc-Mycandβ-catenin.
in the United States [1]. For individuals with diabetes, A drastic decrease in available inorganic phosphate leads
the lifetime probability of developing a diabetic foot to decreased levels ofATPinDFUs–adevastating blow
ulcer (DFU) is estimated at 10 to 25% [2]. DFUs repre- to all healing-related processes. Immune response is
sent a significant source of morbidity and an enormous impaired, with reduced inflammatory cell recruitment at
financial burden [3,4]. Diabetes is the leading cause of the initial phase and later a skewed immune response
nontraumaticlower-extremityamputationsintheUnited towards macrophage and B-cell infiltrates. That phase is
States [1] and a DFU is often the initial insult leading to also characterized by increased TNFα and IL-1β expres-
these amputations [5]. The pathophysiology of DFUs is sion, which are known MMP stimulators. Leukocyte
thought to result from the combined comorbidities of function and intracellular killing are impaired in DFUs,
neuropathy, vascular deficits, impaired immunity, infec- renderingtheseulcersexceptionallysusceptibletosuper-
tion and trauma, all occurring in no particular order and ficial infections and resistant biofilms. Neuropathy
overlapping to produce a vicious cycle [6]. The standard leads to decreased levels of neuropeptides that nor-
of care (SOC) for DFUs involves systemic glucose con- mally contribute to healing. In addition, neuropathy
trol,ensuringadequateextremityperfusion,debridement reduces capillary blood flow. Impaired gap junction
ofnonviabletissue,off-loading,controlofinfection,local function has recently immerged as an additional
wound care and patient education, all administered by a pathological mechanism leading to impaired wound
multidisciplinary team [7-10]. Unfortunately, even with healing in DFUs.
the best SOC available, only 24% or 30% of DFUs will As described above, the ECM is abnormal in DFUs
heal atweeks12or20,respectively [11]. and its impairment has been proposed as a key target
Classic wound repair has been described as a rela- for newtherapeutic devices[19].Thesedevices intend to
tively linear process (albeit with significant complexity replace the aberrant ECM by implanting a matrix, either
and overlap), progressing through three general phases: devoid of cells or enhanced with fibroblasts, keratino-
inflammatory, including platelet aggregation, release of cytes or both as well as various growth factors. These
proinflammatory cytokines and recruitment of neutro- new bioengineered skin substitutes are proposed to en-
phils and macrophages; proliferative, including fibro- courage angiogenesis and in-growth of new tissue, and
blast proliferation, angiogenesis and formation of to utilize living cells to generate cytokines needed for
W
granulation tissue and keratinocyte proliferation and wound repair. Dermagraft (Shire Regenerative Medi-
re-epithelialization; and remodeling, the longest phase, cine, Inc. La Jolla, California, United States ) is one ex-
including myofibroblast transformation and slow re- ample of bioengineered matrix supplemented with
structuring of the healed wound to increase its fibroblasts (cellular matrix (CM)). Other “new-gener-
strength [6,12]. ation” ECM replacement tissues, devoid of cellular com-
In nonhealing DFUs the normal healing process is ponents, have been created for wound repair, and the
W
stalled. The exact mechanisms involved in this impair- Oasis (Healthpoint, Ltd Fort Worth, Texas, United
ment are not fully understood but an increasing number States)acellularmatrix(ACM)isanexampleofthiscat-
ofpathwaysaresuggestedwithvaryinglevelsofevidence egory. Interestingly, in industry-supported randomized
and are summarized herein as reviewed in a number of controlled trials, the reported rates of wound closure at
recent excellent articles [13-18]. Hyperglycemia is week 12 are approximately 50% for both devices [20,21].
thought to adversely affect healing by increased levels of A success rate of 50% represents about 20% added bene-
advanced glycation end products that inhibit normal fit to the use of such devices over the SOC. This min-
extracellular matrix (ECM) deposition and upregulate imal benefit may be disproportionate to the expenditure
activity of matrix metalloproteinases (MMPs). Prolonged these devices accumulate (for example, up to $1,800 per
hyperglycemia may also offset the delicate balance be- application, and up to eight applications as the recom-
tween reactive oxygen species and various antioxidants mended regimen).
by depletion of nicotinamide adenine dinucleotide Theefficacyof bioengineeredECMcontaininglive cel-
phosphate on one hand and increasing reactive oxygen lular elements (CM) for improving healing above that of
species production on the other. DFUs have increased a SOC control group has not to date been compared
activity of MMPs coupled with decreased activity with the efficacy of an ECM devoid of cells (ACM) rela-
of MMP inhibitors (tissue inhibitors of metalloprotei- tive to the same SOC. Our hypothesis is that there is no
nases). Fibroblasts in DFUs are often senescent, ex- difference in the improved healing effected by either of
hibit decreased proliferation and are less responsive these two product types relative to the SOC. If this hy-
to growth factors when compared with fibroblasts pothesis is correct, then immense cost savings could
from age-matched diabetic controls without ulceration. be realized by using the orders-of-magnitude less
Lev-Tovetal.Trials2013,14:8 Page3of8
http://www.trialsjournal.com/content/14/1/8
expensive acellular ECM device without compromising staff at participating Veterans Affairs Northern Califor-
patient health outcomes. To test this hypothesis we pro- niaHealthcareSystem woundcenters.
posed a randomized, single-blind, clinical trial with three
W
arms: SOC, SOC plus Dermagraft (bioengineered ECM Inclusioncriteria
W
containing living fibroblasts) and SOC plus Oasis (ECM
devoidoflivingcells)inpatientswithnonhealingDFUs. (cid:1) InstitutionalReviewBoard(IRB)InformedConsent
Formissigned anddated prior toanystudy-related
Methods/Design activities.
Participants (cid:1) Areaofthestudy ulcerafterdebridementisbetween
The study protocol is approved by both the Veterans 1and 25cm2atweek0/visit3.
Affairs’ Institutional Research and Development Com- (cid:1) Subjectisbetween18 and85 yearsofage.
mittee and their Institutional Review Board (IRB). Study (cid:1) Subject's highest Ankle Brachial Pressure Index
participants are veterans eligible for Veterans Affairs’ (ABPI)/Ankle Arm Index (AAI) ≥ 0.80 and < 1.4
medicalbenefits.Patientsare recruited from allclinicsat (Highest ABPI/AAI value from three
the Veterans Affairs’ Northern California Healthcare measurements within last 6 months shall apply) or
System via provider referral and IRB-approved flyers. toe–arm index ≥ 0.6.
The study is supported by a Veterans Affairs’ MERIT (cid:1) Patient hasoneormorediabeticulcersonthetarget
award (project ID SURG-369-10S). During the “run-in” footwith onlyoneulcer selectedasthestudy
phase (seeTable 1) all participants are screened for eligi- (target) ulcer.Thetargetulcermustbeatleast4cm
bility based on the inclusion and exclusion criteria. All fromanontarget ulcerand, intheinvestigator’s
study-related procedures (except diagnostic tests such as opinion, beunlikelyto coalescewithanotherulcer
radiography or blood testing) are performed by study within12 weeksofrandomization.
Table1Studyprocedureschedule
Procedure Visit1 Visit2 Visit3(week0), Visits4to10(weeks Visits11to14 Visit15(week Visits16to19
(week−2) (week−1) randomization 1to7),treatment (weeks8to11), 12),study (weeks13to28),
(variable) treatment endpoint follow-up
Informedconsent X
Inclusion/exclusion X X X
criteria
Physicalexamination X X
Vitalsigns X X X X X X X
Ankle-brachialindex X
Fungalinfection X
cultureand
malignancy
evaluation
Bloodsampledrawn X X
forlaboratories
SF-36v2™ X X
questionnaire
Ulcerassessment X X X X X X X
Debridement X X X X X X X
Ulcerphotography X X X X X X X
andarea
measurement
Randomization X
DermagraftWTx+ X X
SOC
OasisWTx+SOC X X
SOCandoff-loading X X X X X X X
Concomitant X X X X X X X
medicationsand
adverseevents
DermagraftW(AdvancedBioHealing,Inc.LaJolla,California,UnitedStates),OasisW(Healthpoint,Ltd,FortWorth,Texas,UnitedStates).SF-36v2™,ShortForm-36,
version2;SOC,standardofcare;Tx,treatment.
Lev-Tovetal.Trials2013,14:8 Page4of8
http://www.trialsjournal.com/content/14/1/8
(cid:1) Subject'sstudyulcer isfullthickness anddoesnot DermagraftW (Advanced BioHealing, Inc. La Jolla,
W
extend tobone, muscle,ortendon. California, United States), Oasis (Healthpoint, Ltd Fort
(cid:1) Subject'sstudyulcer hasbeenpresent atleast4 Worth,Texas,UnitedStates).
weekspriortothe initialscreening(visit 1)or6
weeksatrandomization(visit3). Participantconsent
(cid:1) Subjecthasbeen diagnosed with type1ortype2 Detailed informed consent is obtained from all partici-
diabetes andhemoglobinA1c<10%. pantsandaHIPAAHealthInsurancePortabilityandAc-
(cid:1) Study ulcerhasnoclinicalfeature ofinfection (two countability Act release form is signed. Prior to any
signsofinflammationand elevatedbacterialloadof study-related intervention, study staff provide each pro-
thewound). spective subject with adetailedexplanation ofthe nature
(cid:1) Female subjectsofchildbearingage potential havea and purpose of the trial. Once study staff are assured
negative pregnancy testand arelactatingforthe that the prospective subject understands all study proce-
duration ofthestudy. dures and potential implications, an IRB-approved
(cid:1) Subjectunderstandstherequirementsofthisstudyand informed consent form is signed. The signed copy
iswillingtocomplywithallthestudyrequirements. becomes part of the subject’s record. An additional copy
of the informed consent is provided to the subject for
Exclusioncriteria futurereference.
(cid:1) Subjectisdiagnosedwithcancerand isundergoing Criteriaforwithdrawal
treatment with immunosuppressiveor Subjects are informed that they may withdraw from the
chemotherapeutic agents,radiotherapyorsystemic trial for any reason at any time and that the investigator
corticosteroids<30daysbeforeenrollment. may decide to withdraw them from the trial for safety-
(cid:1) SubjectisdiagnosedwithHIV/AIDS. related issues. Study staff explain that failure to comply
(cid:1) Subjectisdiagnosedwithanybleedingdisorders. with the study’s procedure may result in withdrawal
(cid:1) Subjectisdiagnosedwithanyconnective tissue from the trial. In the case of such an event, the reason
diseases. for withdrawal is documented in the subject’s file and
(cid:1) Forfemalesubjects,thesubjectispregnantorlactating. the subject is referred to the wound clinic for further
(cid:1) Subjecthasahistoryofillicitdrugusewithin 1year follow-up.
ofenrollment.
(cid:1) Inthe pastyear,the subjectexperienced episodesof Participantcompliance
drinking morethan fivealcoholic beveragesin<2 Overall participant compliance with study-related proce-
hoursand/ordrinking alcohol became aproblemin dures is assessed at each study visitby the staff. Specific-
theirinterpersonal relationships,work,driving and/ ally, detailed questioning regarding compliance with off-
ortheirbehavioringeneral. loading and any tampering with study-related dressings
(cid:1) Subjecthasanyactive infectedwoundsor is performed at each study visit and any deviations are
osteomyelitis(confirmed bybonebiopsy,magnetic noted inthesubject’sfileforlater analysis.
resonanceimagingorbonescan).
(cid:1) Subjecthad Doppler examinationwithin thelast365 Summarizeddetailsofproductscompared(interventions)
W
days,and itdemonstrated reflux >0.5 seconds. The CM of this study, Dermagraft , is a cryopreserved
(cid:1) Subjectisdiagnosedwithactive Charcot as human fibroblast-derived dermal substitute composed of
describedbySaunder’sclassificationsystem. viablenewbornforeskin fibroblasts,seededonto abioab-
(cid:1) Subjectmanifestssignsofpoor nutritionalstatus sorbable polyglactin mesh. The product is supplied fro-
and/oralbumin level<2.9g/dl. zen (−75°C) in a clear bag containing a 2 inch×3 inch
(cid:1) SubjectreceivedDermagraftWand/orOasisWinthe matrix[22].Theproductisstoredaccordingtotheman-
last60 days. ufacturer’s instructions and records of quality control,
(cid:1) Study ulcersize<1.0cm2or >25cm2. including constant temperature monitoring, are kept by
(cid:1) Subjecthasanyporcineallergyorcowproductallergy. study staff. DermagraftW is applied according to the
(cid:1) Subject’srecent (last30 days)chemistrytest’sserum manufacturer’sinstructions[22].
W
creatinine is twotimesabovethe upper limitof The ACM of this study, OASIS , is a bioabsorbable
normaland/orliver enzymes (AST,ALT)are three ECM derived from porcine small intestinal submucosa.
timesabovethe upper limitofnormal. The product is processed using proprietary methods to
(cid:1) Between week−2/visit1and week0/visit 3 retain its original structure while stored at room
(randomization) thestudy ulcerdecreasedinsizeby temperature [23]. The product is stored and applied
>40%,orincreasedinsizeby>50%. accordingtothemanufacturer’sinstructions[23].
Lev-Tovetal.Trials2013,14:8 Page5of8
http://www.trialsjournal.com/content/14/1/8
W
TheSOCdressingiscomprisedofIodosorb gel(Smith withnodrainage,orcallusformationwithunderlyingulcer
W
& Nephew Largo, Florida, United States), Adaptic ontwoconsecutivevisits1weekapart.
(Johnson&JohnsonGargrave,Skipton,UnitedKingdom) Another secondary outcome is the rate of wound heal-
and gauze. This dressing also serves as the secondary ing to achieve complete closure based on weekly wound
dressingforthetwointerventionarms.Ifasubjecthasan areameasurements.
allergy to iodine products, then Bacitracin antibiotic oint- Third is the incidence of ulcer recurrence at week 20.
mentisusedasareplacementforIodosorb. Recurrence will be defined as an ulcer occurring at the
We will provide all subjects with a removable walking same locationasthehealedstudyulcer.
boot,suchasacamwalkerordiabeticconformerBledsoe Fourth is the association of wound healing (overall and
boot. The clinician will make further accommodation/ given a particular treatment) with the following wound
adjustment to offload plantar ulcers with excavation or characteristics: peri-ulcer erythema, induration, tender-
aperature using plastazote, foam or felt padding. Those ness, pain, local warmth, size, depth, undermining, ulcer
patientsthatuseexclusively manualormotorizedwheel- location, ulcer age at presentation, ulcer precipitating
chairs will be identified to differentiate patients that are eventandtotalwoundassessmentscoreatpresentation.
fully ambulatory from those with limited ambulatory Another outcome is the association of wound healing
capacity. While total contact casting is considered the (overall and given a particular treatment) with the fol-
gold standard for offloading plantar ulcers by many aca- lowing patient characteristics on presentation: age, sex,
demicians, in our experience total contact casting is not body mass index, smoking history, family history of dia-
in standard use owing to time constraints with cast ap- betes, complete blood count results, chemistry panel
plication and accessibility to materials. Indeed, a recent results, erythrocyte sedimentation rate and C-reactive
study found that fewer than 2% of diabetic foot specia- protein serum concentration and hemoglobin A1C
lists utilize this form of offloading [24]. Our study there- serumconcentration.
fore models the standard care for all arms after realistic Sixth is the association of a particular treatment with
clinicalpractice. the incidence of the following during the study period:
All products used in this trial are approved by the cellulitis, osteomyelitis, acute Charcot disease and over-
Food and Drug Administration for use in treatment of alladverseeventsrate.
DFUs. Another outcome is the association of a particular
treatment with a change in quality of life assessments
Hypothesis(objectives) (quality-adjustedlife-years).
There is no significant difference in healing at weeks 12 The final secondary outcome is cost-effectiveness of a
and week 20 or in the rate of healing when treating particulartreatmentcomparedwith SOC.
DFUswith SOCascompared with CMorACM.
Samplesize
Alternativehypothesis Based on previous trials [20,21] it is estimated that ap-
There is a significant difference in healing at weeks 12 proximately 50% of the subjects in the intervention arms
and week 20 or in the rate of healing when treating will achieve complete wound closure at week 12 (pri-
DFUswith SOCascompared with CMorACM. mary outcome) and approximately 25% of the SOC arm
will achieve complete wound closure by week 12. The
Primaryoutcome sample size was estimated based on 80% power and a
The primary outcome is the percentage of subjects that 0.05significanceleveltodetecta25%differencebetween
achieved complete wound closure by week 12. Complete the intervention and control arms (Stplan 4.5 statistical
healing is defined as full re-epithelialization of the ulcer software, Department of Biomathematics, University of
with no drainage, or callus formation with underlying Texas MDAnderson Cancer Center, Houston,TX, USA,
ulcer on two consecutive visits 1 week apart. Callus 2010). Based on the method of normal approximation to
(hyperkeratosis)formationwillbedebridedatthewound the arcsin transformation of the binomial distribution
site to determine whether complete closure has been and, to be conservative, a two-sided test, enrollment of
obtained. The wound is deemed healed if there is no 57subjectsineacharmisrequired.
underlying open lesion after debridement (complete
epithelialization). Randomization
Sequencegeneration
Secondaryoutcomes Since healinghasbeenshowntodependonageandrace
The first secondary outcome is the percentage of subjects [25,26], randomization was done in small blocks (six
that achieved complete wound closure by week 20. subjects per block) in order to match subjects according
Completehealingwillbedefinedasfullre-epithelialization to age groups (< 50 years old, 50 through 59 years old,
Lev-Tovetal.Trials2013,14:8 Page6of8
http://www.trialsjournal.com/content/14/1/8
W
60 through 69 years old and 70 through 85 years old) We will evaluate the cost-effectiveness of Oasis com-
W
and race (black and nonblack). We used a computer- pared with Dermagraft and the SOC by measuring
generated algorithm to randomly assign treatments to quality-adjusted life-years as the measure of effective-
subjects within a block and to randomly determine the ness, based on resultsobtained from Short Form-36 (SF-
block sequence. Subjects presenting to the study will be 36v2™) questionnaires. Multiple regression will be used
entered into their appropriate block in a strict chrono- to evaluate continuous dependent variable outcomes, in-
logical order. cluding change in SF-36v2™ physical and mental compo-
nent summaries scores between baseline and the end of
Allocationconcealmentandimplementation thestudy.
The randomization table was generated prior to study Rates of healing among the groups will be analyzed
initiation by the study’s biostatistician. Allocation is con- using a log-rank test to compare the time of healing
cealed bytheNorthernCaliforniaVeterans AffairsInves- within 20weeks.
tigational Drug Service, an independent third party, until Secondary outcomes such as complete healing at 20
themomentofrandomization. weeks and rate of ulcer recurrence at 20 weeks will be
analyzed using chi-square tests or Fisher’s exact tests
Blinding when appropriate. In addition, demographics, smoking
Owing to the nature of the interventions studied, history, and other characteristics mentioned in Second-
complete blinding is not realistic (for example, Derma- ary outcomes will be summarized in a table. For com-
graftW requires storage at −75 ± 10°C and a specific ma- parison of nominal categorical secondary outcome
nipulation immediately prior to application). An variables,wewilluse the chi-square testorFisher’sexact
independent observer was therefore assigned to assess test when appropriate. For comparison of ordinal cat-
the primary and secondary outcomes. Clinical images egorical variables, we will use the Wilcoxon rank-sum
will be captured at each visit before and after debride- test (for two independent group comparisons) and the
ment and will be de-identified and uploaded to an exter- Kruskal–Wallis test (for three independent group
W
nal server (Silhouette Central Aranz Medical, comparisons).
Christchurch, New Zealand). The observer will log on to
the server independently and determine whether the Datamanagement
wound has closed or not. In addition, the observer will Data will be managed by the study’s biostatistician (C-SL)
determine whether the margins used for area measure- at the University of California Davis Clinical and Transla-
mentareaccurate. tionalScienceCenter.Datawillbeautomaticallyimported
from digital subject records into pre-designed spread
W
Statisticalmethods sheetsusingExcel andanalyzedusingStatisticalAnalysis
All data will be analyzed based on intention to treat. System,version9.3[28].
The primary outcome, wound closure by week 12, will
be analyzed using the chi-square test or Fisher’s exact Planandtrialdesign
test when appropriate to compare the percentages of This is a randomized, controlled, single-blind, trial com-
subjects with complete closure in each group. We will paringtheefficacyofACMwiththatofCMintreatment
also perform exploratory analysis with logistic regres- of DFUs. A total of 171 subjects will be enrolled and
sion, using complete wound closure at 12 weeks as the randomly assigned to one of three treatment groups:
dependent variable and the independent variables would ACM, CM and SOC. Subjects are followed for a total of
include the characteristics described as the fourth to 30 weeks in three major phases: run-in phase, treatment
sixth items in Secondary outcomes above. We will run phase and follow-up phase. The study’s procedures are
preliminary analysis on selected covariates to determine detailedbelow.
whether therearesignificantdifferences.
Cost-effectiveness analysis will be performed to com- Run-inphase
pare the treatment groups DermagraftW and OasisW and During this 2-week period (weeks −2 through 0) pro-
the standard care group using the analysis of variance F spective subjects are seen on a weekly basis and rigor-
test or Kruskal–Wallis test according to the data distri- ously evaluated for eligibility. Once informed consent is
butionpattern.Ifsignificant,wewillfurtheranalyzewith secured, detailed health-related history is solicited and a
pair-wise comparisons using Tukey’s test or Dunn’s test. thorough physical examination is performed. Compre-
A multiple-comparison post-hoc test for Kruskal–Wallis hensive lower-extremity ulceration evaluation is done
analysis will be performed with the Statistical Analysis (including imaging and area measurements) and compli-
System (Cary, NC, USA) macro developed by Elliott and ance with an off-loading device is assessed. The follow-
Hynan [27]. ing laboratory studies are ordered to exclude
Lev-Tovetal.Trials2013,14:8 Page7of8
http://www.trialsjournal.com/content/14/1/8
comorbidities and establish a baseline: vital signs, body from the confirmatory visit and in week 12 or in week
mass index, pregnancy test, Ankle-Brachial Index, quan- 12 only, whichever comes first, and will return to the
titative bacterial cultures, fungal cultures, tissue path- schedule above thereafter. In the event of failure to heal
ology, complete blood count, comprehensive metabolic after week 12, subjects will be followed at the wound
panels (including liver enzymes and albumin levels), clinic based on clinical need but will return for all
erythrocyte sedimentation rate, C-reactive protein level, follow-upphasemandatedvisits.
hemoglobin A1C level and lower-extremity X-rays. Once
eligibility is established, subjects are randomized (as
described above) to one of three treatment arms. SF-36 Ethicalconsiderations
questionnairesarefilledoutbythesubjects. The DOLCE trial is approved by the Veterans Affairs
Northern California Healthcare System IRB and the Re-
Treatmentphase search andDevelopment Committee.Theprimaryinves-
This phase is divided into a variable treatment period tigators will ensure that the study is conducted in full
(weeks 0 through 7) and a SOC treatment period (weeks compliance with the protocol and IRB regulations as
8 through 11). Subjects will be evaluated on a weekly well as international standards on human subjects’ re-
basis. search. Theprimaryinvestigators will ensure compliance
During the variable treatment period, subjects in each with institutional regulations as well as local and na-
arm will receive its corresponding primary dressing tional law. Compliance will be administered by a dedi-
treatment(thatis,ACM,CMorSOC).Duringeachvisit, cated, certified Clinical Research Coordinator. A data
vital signs will be measured, a comprehensive lower- monitoring committee has been set up to review safety
extremity assessment will be performed (including pho- data. All adverse events will be reported to the IRB as
tography and ulcer area measurement), health status and stipulated bytheIRBinitsprotocol-approvalletter.
medication changes will be recorded and adverse events
will be assessed. Compliance with the off-loading device
willalsobeassessed. Discussion
During the SOC treatment phase the primary dressing The DOLCE trial is a randomized, single-blind, com-
applied to all subjects will be identical and will comprise parative efficacy trial to assess the difference between
the SOC dressing. The rest of the visit will be identical ACM, CM and SOC in the treatment of DFUs. As the
tothe variabletreatmentperiod asdescribed above. healthcare-related industry and science are marching to-
On week 12, a study endpoint visit will be conducted. wards development of increasingly complex and expen-
During this visit, vital signs will be measured, a compre- sive devices, it is crucial to regularly and rigorously
hensive lower-extremity assessment will be performed reassess the efficacy of various available interventions.
(including photography and ulcer area measurement), Healthcare providers are constantly presented with new
health status and medication changes will be recorded devices and need solid evidence to support clinical deci-
and adverse events will be assessed. Compliance with sion-making.However,manystudies designed to provide
the off-loading device will also be assessed. In addition, such evidence are industry supported, and thus subject
the SF-36 questionnaire is filled out for the second time to concerns regarding inherent bias – if not with the de-
by the subjects and repeat complete blood count and sign and execution of the trial, then certainly with the
comprehensive metabolic panels are ordered. SOC treat- dissemination of the results, should those results not
mentwillbeprovided toallsubjects. favor the product whose manufacturer sponsored the
trial ([29]). In addition, in this study, subanalysis may
Follow-upphase allow for a better understanding of the indications for
Thisphaseiscomprised offour monthlyvisits(weeks13 preferring one device over the other. Practicing within
through 28). During each visit, vital signs are measured, financial constraints is especially important when treat-
a comprehensive lower-extremity assessment is per- ing a prevalent condition such as DFUs. Providing the
formed (including photography and ulcer area measure- wound carespecialistwith the data to support the use of
ment), health status and medication changes are less expensive treatments without compromising patient
recorded and adverse events are assessed. Compliance outcomes could potentially save the healthcare system
with the off-loading device is also assessed. During the significant resources. The DOLCE trial is an attempt to
follow-up phase the primary dressing applied to all sub- provide wound care specialists with relevant evidence to
jects is identical and will comprise the SOC dressing. an overwhelming clinical problem in order to support
SOCtreatmentisprovided toallsubjects. meaningful clinical decision-making. In addition, we
In the event of early wound closure (before week 12), hope to set an example for the importance of independ-
the subject will return for follow-up visits every 4 weeks ent,investigator-initiatedcomparativeefficacy research.
Lev-Tovetal.Trials2013,14:8 Page8of8
http://www.trialsjournal.com/content/14/1/8
Trialstatus 10. LepantaloM,ApelqvistJ,SetacciC,RiccoJB,deDonatoG,BeckerF,Robert-
Activelyrecruiting. EbadiH,CaoP,EcksteinHH,DeRangoP,DiehmN,SchmidliJ,TeraaM,
MollFL,DickF,DaviesAH:ChapterV:Diabeticfoot.EurJVascEndovasc
Surg2011,42(Suppl2):S60–S74.
Abbreviations
11. MargolisDJ,KantorJ,BerlinJA:Healingofdiabeticneuropathicfoot
ACM:acellularmatrix;CM:cellularmatrix;DFU:diabeticfootulcer;
ulcersreceivingstandardtreatment.Ameta-analysis.DiabetesCare1999,
ECM:extracellularmatrix;IL:interleukin;IRB:InstitutionalReviewBoard;
22:692–695.
MMP:matrixmetalloproteinase;SF-36:ShortForm-36;SOC:standardofcare;
12. SingerAJ,ClarkRA:Cutaneouswoundhealing.NEnglJMed1999,
TNF:tumornecrosisfactor.
341:738–746.
13. GarySibbaldR,WooKY:Thebiologyofchronicfootulcersinpersons
Competinginterests
withdiabetes.DiabetesMetabResRev2008,24(Suppl1):S25–S30.
Theauthorsdeclarethattheyhavenocompetinginterests.
14. LootsMA,LammeEN,ZeegelaarJ,MekkesJR,BosJD,MiddelkoopE:
Differencesincellularinfiltrateandextracellularmatrixofchronic
Authors’contributions
diabeticandvenousulcersversusacutewounds.JInvestDermatol1998,
HL-Tcontributedtotheprotocoldesign,wroteIRBamendments,andwrote 111:850–857.
thefirstdraftofthemanuscript.C-SLcontributedtothestatisticaldesign
15. MedinaA,ScottPG,GhaharyA,TredgetEE:Pathophysiologyofchronic
andeditedthemanuscript.SDcontributedtotheoverallstudydesign, nonhealingwounds.JBurnCareRehabil2005,26:306–319.
wrotethefirstdraftoftheprotocol,editedthemanuscript,andwrotethe
16. Mendoza-NaranjoA,CormieP,SerranoAE,WangCM,ThrasivoulouC,
MERITgrant.RIconceivedthestudyidea,supervisedthestudydesignand
SutcliffeJE,GilmartinDJ,TsuiJ,SerenaTE,PhillipsAR,BeckerDL:
protocoldevelopment,editedthemanuscript,editedIRBdocuments,and
OverexpressionofthegapjunctionproteinCx43asfoundindiabetic
wrotetheMERITgrant.Allauthorsreadandapprovedthefinalmanuscript. footulcerscanretardfibroblastmigration.CellBiolInt2012,36:661–667.
17. NeutD,Tijdens-CreusenEJ,BulstraSK,vanderMeiHC,BusscherHJ:
Acknowledgements Biofilmsinchronicdiabeticfootulcers–astudyof2cases.ActaOrthop
TheauthorsacknowledgetheseminalcontributionsofDrHuongLLeinher 2011,82:383–385.
roledevelopingthisprojectwithheroverarchinggoalofprovidingthebest
18. BajpaiS,ShuklaVK,TripathiK,SrikrishnaS,SinghRK:Targetingconnexin43
caretoherpatients.Sadly,heruntimelydeathpreventedherfrom
indiabeticwoundhealing:futureperspectives.JPostgradMed2009,
completingthistask.ThistrialissupportedbyaVeteranAffairs’MERITaward 55:143–149.
(ProjectIDSURG-369-10S).StatisticalservicesaresupportedbytheNational
19. PanuncialmanJ,FalangaV:Thescienceofwoundbedpreparation.Surg
CenterforAdvancingTranslationalSciences,NationalInstitutesofHealth, ClinNorthAm2009,89:611–626.
throughgrant#UL1TR000002.
20. GentzkowGD,IwasakiSD,HershonKS,MengelM,PrendergastJJ,RicottaJJ,
SteedDP,LipkinS:Useofdermagraft,aculturedhumandermis,totreat
Authordetails diabeticfootulcers.DiabetesCare1996,19:350–354.
1VeteransAffairsMedicalCenter,NorthernCaliforniaHealthcareSystem,
21. NiezgodaJA,VanGilsCC,FrykbergRG,HoddeJP:Randomizedclinicaltrial
10535HospitalWay,Mather,CA95655,USA.2DepartmentofDermatology,
comparingOASISWoundMatrixtoRegranexGelfordiabeticulcers.Adv
UniversityofCaliforniaDavis,3301CStreet,Sacramento,CA95816,USA. SkinWoundCare2005,18(5Pt1):258–266.
3DepartmentofPublicHealthSciences,DivisionofBiostatistics,Universityof
22. ForNon-healingDiabeticFootUlcers.[http://www.dermagraft.com/about/
CaliforniaDavis,MS1CRoom145,Davis,CA95616,USA.
overview]
23. HealthpointOffersTwoOasisWMatrixProducts.[http://www.
Received:22July2012Accepted:18December2012
oasiswoundmatrix.com/]
Published:9January2013
24. WuSC,JensenJL,WeberAK,RobinsonDE,ArmstrongDG:Useofpressure
offloadingdevicesindiabeticfootulcers:dowepracticewhatwe
References preach?DiabetesCare2008,31:2118–2119.
1. CenterforDiseaseControlandPrevention:NationalDiabetesFactSheet:
25. RanjitaMisraLL,DavidV,AshleyM,KhannaSR,SenCK:PredictorsofDiabetic
NationalEstimatesandGeneralInformationonDiabetesandPrediabetesin
WoundHealingbyRacial/EthnicCategories[poster].ColumbusOH:TheOhio
theUnitedStates,2011.Atlanta,GA:DepartmentofHealthandHuman
StateUniversityComprehensiveWoundCenter;2012[http://misra.tamu.
Services;2011.
edu/chronic_healing.html]
2. MargolisDJ,HoffstadO,NafashJ,LeonardCE,FreemanCP,HennessyS,
26. GoodsonWH,GoodsonWH,HuntTK:Woundhealingandaging.JInvest
WiebeDJ:Location,location,location:geographicclusteringoflower- Dermatol1979,73:88–91.
extremityamputationamongMedicarebeneficiarieswithdiabetes. 27. ElliottAC,HynanLS:ASASWmacroimplementationofamultiple
DiabetesCare2011,34:2363–2367. comparisonposthoctestforaKruskal–Wallisanalysis.ComputMethods
3. BoultonAJ,VileikyteL,Ragnarson-TennvallG,ApelqvistJ:Theglobal ProgramsBiomed2011,102:75–80.
burdenofdiabeticfootdisease.Lancet2005,366:1719–1724.
28. StatisticalAnalysisSystem.[www.sas.com]
4. SenCK,GordilloGM,RoyS,KirsnerR,LambertL,HuntTK,GottrupF,
29. LundhA,KrogsbollLT,GotzschePC:Sponsors'participationinconduct
GurtnerGC,LongakerMT:Humanskinwounds:amajorandsnowballing
andreportingofindustrytrials:adescriptivestudy.Trials2012,13:146.
threattopublichealthandtheeconomy.WoundRepairRegen2009,
17:763–771.
doi:10.1186/1745-6215-14-8
5. ApelqvistJ,BakkerK,vanHoutumWH,SchaperNC:Practicalguidelineson
Citethisarticleas:Lev-Tovetal.:Cellularversusacellularmatrixdevices
themanagementandpreventionofthediabeticfoot:baseduponthe
intreatmentofdiabeticfootulcers:studyprotocolforacomparative
InternationalConsensusontheDiabeticFoot(2007)Preparedbythe efficacyrandomizedcontrolledtrial.Trials201314:8.
InternationalWorkingGroupontheDiabeticFoot.DiabetesMetabResRev
2008,24(Suppl1):S181–S187.
6. FalangaV:Woundhealinganditsimpairmentinthediabeticfoot.Lancet
2005,366:1736–1743.
7. BremH,SheehanP,BoultonAJ:Protocolfortreatmentofdiabeticfoot
ulcers.AmJSurg2004,187:1S–10S.
8. GameFL,HinchliffeRJ,ApelqvistJ,ArmstrongDG,BakkerK,HartemannA,
LöndahlM,PricePE,JeffcoateWJ,InternationalWorkingGrouponDiabetic
Foot:Specificguidelinesonwoundandwound-bedmanagement2011.
DiabetesMetabResRev2012,28(Suppl1):232–233.
9. BakkerK,ApelqvistJ,SchaperNC:Practicalguidelinesonthe
managementandpreventionofthediabeticfoot2011.DiabetesMetab
ResRev2012,28(Suppl1):225–231.