Margolis DJ, Kantor J, Santanna J, Strom BL, Berlin JA. Risk Factors for Delayed Healing of Neuropathic Diabetic Foot UlcersA Pooled Analysis. Arch Dermatol. 2000;136(12):1531-1535. doi:10.1001/archderm.136.12.1531
DamianoAbeniMD, MPHMichaelBigbyMDPaoloPasquiniMD, MPHMoysesSzkloMD, MPH, DrPHHywelWilliamsPhD, FRCP
To estimate the effect of various risk factors on the probability thatneuropathic diabetic foot ulcers will heal within 20 weeks of care.
Design and Setting
A pooled or meta-analysis of individual patient data from the standardcare arms of 5 randomized clinical trials was conducted. We analyzed 586 subjectswith diabetes mellitus who had a neuropathic diabetic foot ulcer. All patientsreceived good wound care, debridement, and "off-loading" of the wound.
Main Outcome Measure
Multivariable logistic regression was used to calculate the magnitudeof the association of each risk factor with patients having healed wounds.
Logistic regression odds ratios (ORs; 95% confidence intervals [95%CIs]) revealed that those patients with a diabetic neuropathic foot ulcerthat healed within 20 weeks using standard care were more likely to have asmaller wound (OR = 0.67; 95% CI, 0.55-0.81), a wound that existed for a shorterperiod (OR = 0.73; 95% CI, 0.61-0.87), and be nonwhite (OR = 0.64; 95% CI,0.43-0.96) compared with patients whose wounds did not heal within 20 weeks.The patient's age (OR = 0.99; 95% CI, 0.89-1.01), serum level of glycosylatedhemoglobin at the start of the study (OR = 1.03; 95% CI, 0.97-1.10), and sex(OR = 1.02; 95% CI, 0.69-1.50) were unassociated with the probability of woundhealing. Substantial heterogeneity was not found among the studies.
A standard care regimen for diabetic neuropathic foot ulcers is mostlikely to be effective for patients who have wounds that are small and ofbrief duration. This information should help dermatologists decide initiallywhether to use standard care, to try a new treatment, or to refer the patientto a specialty center.
FOOT ULCERS and lower extremity amputations are among the most seriouscomplications associated with diabetes melitus (DM). Lower extremity amputationis more common in persons with DM than in those without DM and affects morethan 80,000 persons with DM each year in the United States.1- 4A foot ulcer precedes approximately 85% of lower extremity amputations inpatients with DM.3,5 Foot ulcersare present on 2.7% of all patients with DM who are hospitalized.1 Most diabetic foot ulcers are due, at least in part,to loss of cutaneous sensation (peripheral insensate neuropathy), althoughsome of these patients will also have poor arterial flow to their limb.2,5,6 Due to the peripheralneuropathy, patients with DM are unable to perceive sensations, making themvulnerable to trauma.
The Food and Drug Administration (FDA) recently approved a new agentfor specifically treating patients with a neuropathic diabetic foot ulcerand adequate lower limb arterial blood flow.7- 9Another agent will, in the near future, be approved by the FDA for this sameindication.10 In other words, by virtue ofrecent advances in the treatment of diabetic foot ulcers, a subset of thisailment, diabetic foot ulcers primarily due to neuropathy can now be treatedwith new medical therapies. Even with these advances their care usually includeswound debridement, appropriate wound dressings, and "off-loading" the areaof the foot that has ulcerated.11- 13Off-loading refers to methods that protect the ulcer from the repetitive traumaof activities of daily living. Off-loading methods include molded shoes, orthotics,crutches, wheelchairs, contact casts, and other devices. Off-loading helpsprotect the wound so that it can heal. Yet, little has been published on theefficacy of standard care for neuropathic diabetic foot ulcers. Furthermore,to our knowledge, prognostic factors related to the successful treatment ofa neuropathic diabetic foot ulcer with standard care have not been well studied.Most of the previous studies that have examined risk factors associated withpoor healing of diabetic foot ulcers enrolled neuropathic, arterial insufficient,and combination-type patients.3,4,14- 23
Previously, we conducted a group level meta-analysis to determine thepercentage of patients with a neuropathic ulcer that could be expected toheal within 12 or 20 weeks of care.11 As isoften the case, the information available on a group level was insufficientto evaluate risk factors. In the pooled analysis reported here, we use patientlevel data from the standard care arms of randomized clinical trials thatinvestigated the efficacy of new therapies for the treatment of neuropathicdiabetic foot ulcers. Our primary objective was to pool the control arms ofseveral randomized clinical trials to estimate the association of variouspotential risk factors with the probability that a neuropathic diabetic footulcer will heal within 20 weeks of care.
We previously reported the results of a meta-analysis on the probabilitythat a patient with a diabetic foot ulcer will heal.11While conducting this study, we realized that approximately 95% of the totalnumber of patients that we identified for the group level meta-analysis werepart of a few recently conducted randomized clinical trials. We contactedthe industry sponsors of these trials (OrthoMcNeil, Raritan, NJ, and AdvancedTissue Sciences, La Jolla, Calif), both of which agreed to provide us withdata and specified variables from the control-arm patients.7- 10All of these studies were randomized phase 2 or phase 3 clinical trials thathad been reviewed by the FDA; all were of high quality. These studies usedsimilar criteria to diagnosis neuropathic diabetic foot ulcers, which werethat the limbs needed to have adequate arterial perfusion (ie, a transcutaneousoxygen level of >30 mm Hg, or an ankle-brachial index >0.80), and documentedfoot neuropathy. These patients also received similar care, which includedoff-loading, debridement, and a moist wound dressing. For these clinical trials,wounds were debrided to remove callous and necrotic tissue. Wounds were oftendebrided several times during the study. Good wound care included frequentwound cleansing, usually with a saline solution, a moist wound dressing, suchas saline-moistened gauze, and the avoidance of topical antiseptics on thewounds. In addition, to ensure that wounds, which were primarily on the forefootor midfoot, were properly off-loaded several techniques were used. These includedmolded shoes, half-shoes, crutches, wheelchairs, and bed rest. No sponsoringcompany was involved, directly or indirectly, in the design or execution ofthis meta-analysis or in decisions related to the writing or submission ofthis manuscript.
Risk factors were selected for evaluation if they had been identifiedas such in previous publications, and if information about the risk factorwas available in all 5 data sets.3,4,14- 23Therefore, as with meta-analyses in general, not all potential interestingrisk factors could be evaluated. The patient risk factors we evaluated wereas follows: age at the start of the study, sex, ethnicity (white vs all others),serum level of glycosylated hemoglobin at the start of the study, the areaof the wound (in square centimeters) at the start of the study, and the patient'sreport of duration (in weeks) of the wound at the start of the study. Datafor all 6 of these risk factors were incomplete in all 5 studies. Data onwound duration were unavailable for 102 patients (17%), and data on serumlevels of glycosylated hemoglobin were unavailable for 25 patients (4%). Theprimary outcome for our analysis was a healed wound within 20 weeks of care.Additional analyses included a healed wound within 12 weeks of care and timeto healing.
All analyses used an intention-to-treat approach that included all patientsassigned to the standard care arm of each of the 5 studies. Patients who wererandomly assigned to the experimental therapy arms were excluded from ouranalysis even if they actually received standard care.
To calculate the effect of a given risk factor, pooled unadjusted (single-variable)odds ratios (ORs) and 95% confidence intervals (95% CIs) were estimated usingordinary logistic regression models. Since data on all the risk factors wereunavailable for all patients, ORs and 95% CIs are reported separately forall of the patients who had data for each risk factor, and for the patientswho had data for all of the risk factors.
The outcomes of interest were modeled using 2 different time end points:patients who healed within 20 weeks of the start of care, and patients whohealed within 12 weeks of the start of care. The distributions of both woundarea and wound duration were highly skewed. To reduce the influence of largevalues, the natural logarithms of these variables were used in all models.Each model also included a set of indicator variables representing the studyin which each patient had participated. These indicator variables adjustedfor variability in the probability of healing across studies that is unexplainedby the other risk factors.
Fully adjusted pooled ORs with 95% CIs are also reported. Full adjustmentmeans that the OR for the risk factor of interest is adjusted to considerall of the other risk factors. Since data were unavailable for all of thepatients for all risk factors, fully adjusted ORs are reported only for thosepatients who had complete data for all risk factors. The fully adjusted ORswere also estimated using an imputation algorithm for the missing wound durationdata (NORM24).
In the ordinary, or fixed-effects regression approach, the indicatorvariables allow inferences about the variability in healing rates across thestudies included in the analysis. We also constructed random-effects models,which assume that there is a population of studies and that the studies includedin our analysis are a random sample from this population.25This alternative approach can have an effect on the estimated variance ofthe logistic regression coefficients. The data set was also reanalyzed usinga proportional hazards model. The results did not change using the random-effectsor proportional hazards models. Therefore, only the ordinary logistic regression(fixed-effects) results are reported.
The heterogeneity of the effect of each variable was assessed usinginteraction terms between predictors and between each predictor and the studyindicator variables. Testing study indicator by risk factor interactions showswhether the association between the risk factor and the probability of healingvaries across studies. These interactions were evaluated using the fixed-effectsand random-effects models.25- 27
To estimate the impact of publication bias, both the Begg test and theEgger publication bias plot were evaluated for all risk factors.28Neither test gave statistically significant evidence of a publication bias.All analyses were done using SAS software, version 6.12 (SAS Institute Inc,Cary, NC), or Stata, version 5.0, with Stata Technical Bulletin updates (CollegeStation, Tex).
Aggregating data from the 5 clinical trials, the patients' mean agewas 58 years (age range, 25-93 years). The mean wound size was 1.61 cm2 (range, 0.11-111.22 cm2); 94% of the wounds were smallerthan 10 cm2. The average serum level of glycosylated hemoglobinwas 7.5% (range, 3.8%-20.5%). The average wound duration prior to enrollmentin the study was 30 weeks (range, 2-1444 weeks). The fraction of men was 73.2%,and 77.8% were white. The fraction of patients who completely healed within12 weeks of care was 23.9% (95% CI, 20.5%-27.3%); the fraction who healedwithin 20 weeks was 32.8% (95% CI, 29.0%-36.6%).
Our unadjusted analysis showed that patients with a neuropathic diabeticfoot ulcer were more likely to heal within 20 weeks if their wounds were smaller(<2 cm2) or had existed for a shorter period (<6 months)before they began the clinical trial, or if the patients were nonwhite (Table 1). Similar findings were seen forhealing within 12 weeks (Table 1).The patient's age, sex, serum glycosylated hemoglobin level at the start ofthe clinical trial, and the clinical trial in which the patient participatedwere unassociated with the probability of wound healing.
The adjusted, risk factor associations that we found for wound healingwithin 12 and 20 weeks were similar for patients with complete data recordsand for patients who had incomplete data, with the exception of ethnicity(Table 1). In our analysis ofethnicity, the actual OR for the analysis with imputation was similar to theunadjusted value, 0.69 and 0.64, respectively. However, the P value was not significant when the imputation model was used (P = .09). This, in part, was due to an increased varianceproduced by the imputation method.
Twenty-week healing rates stratified for 2 risk factor variables, woundsize and wound duration, are listed in Table 2. This table clearly shows the association of these 2 variablesindependently. The bottom third of Table2 also shows 2 of the 9 possible stratified combinations of woundsize and duration on the percentage of patients healed within 20 weeks ofcare. These combinations are the shorter duration, smaller wound vs the longerduration, larger wound.
Heterogeneity can exist when differences in outcomes are not accountedfor by sampling variation. In the analysis of these data, the clinical trialin which the patient was enrolled may be a source of 2 types of heterogeneity.The first type is variability in the probability of healing from clinicaltrial to clinical trial that is not accounted for by other patient characteristics.To address this type of heterogeneity, fixed-effects and random-effects Qstatistics were used. Neither of these approaches led to any statisticallysignificant differences in the model results, and when adjusted for patientcharacteristics, there was no statistically significant variation in healingrates across studies.
The second type of heterogeneity occurs when the association betweenpatient characteristics and healing varies from clinical trial to clinicaltrial. Using clinical trial × risk factor interaction terms in the logisticregression models tests this type of heterogeneity. Again, fixed-effects andrandom-effects models failed to detect any significant heterogeneity, suggestingthat the relationship between patient risk factors and healing was constantacross studies.
To our knowledge, this pooled analysis of patient level data is thefirst analysis of risk factors associated with the healing of neuropathicdiabetic foot ulcers to be done using a large sample. We identified 3 predictorsof wound healing within 12 or 20 weeks of standard care: the initial woundarea, the duration of the wound at the start of treatment, and the patient'sethnicity. Standard care was defined to include debridement, good wound care,and off-loading of the ulcer. Equally important, we found that 3 other factorswere not associated with wound healing within 12 or 20 weeks: the patient'sage, sex, and serum level of glycosylated hemoglobin measured at the startof care.
Previous assessments of diabetic foot ulcer healing, or lack of healingas represented by a lower extremity amputation, have often concluded thatlower limb perfusion is the key risk factor that leads to a poor outcome.3,4,14- 23These studies had inconsistent reports on the associations of sex, age, ethnicity,wound size, and wound duration with the probability of wound healing.3,4,14- 23These reports analyzed diabetic foot ulcer of mixed origins.
In our study, we only analyzed patients who were defined objectivelyas having arterial perfusion of their lower limbs that was adequate for woundhealing, and so we did not consider the role of limb arterial perfusion. Weused a definition of adequate limb perfusion that was developed in prior studies.6,17 We believe that the results of ourstudy and previous studies may differ because the healing associations thatwe found may have been modified in prior studies by poor limb perfusion. Inother words, it is possible that sex, ethnicity, wound size, and wound durationare only associated with wound healing in patients who already have adequatearterial perfusion of their lower limbs and have what can be described aspure neuropathic wounds. Previous investigations may have studied a mix ofpatients with neuropathic wounds and those with wounds due to arterial insufficiency.Therefore, by design, our results are relevant specifically to patients withDM who have neuropathic foot ulcers and adequate perfusion of their lowerlimbs. This is the group of patients with DM who had foot ulcers and mostlikely to respond to nonsurgical care and the group of patients who are thetarget of the new FDA-approved topical therapies.
There was a remarkable lack of significant heterogeneity in risk factorassociations among the studies we analyzed. Therefore, combining all of thestudy results to obtain estimates of the association between each risk factorand the probability of wound healing is appropriate. The lack of substantialheterogeneity probably relates to the fact that the same sponsor designedmany of these studies. Also, all were FDA-reviewed clinical trials. But allof these studies were also multicenter clinical trials. In total, this meta-analysisincludes observations made by more than 50 different health care providers(eg, general surgeons, vascular surgeons, dermatologists, internists, primarycare physicians, nurses, nurse practitioners, podiatrists, physical therapists).It is therefore likely that this meta-analysis is generalizable to the entirecommunity of health care providers with experience in caring for patientswith DM who have neuropathic foot ulcers who use standard care that consistsof wound debridement, good wound care, and off-loading of the wound.
Our study has several limitations. We could only evaluate risk factorsthat were common to all of the clinical trials we included. For example, wedid not evaluate associations between wound healing and a patient's weight,cigarette smoking, type and duration of DM, type of glucose level control,method for off-loading the ulcer, type of wound dressing used on the ulcer,patient compliance, and the influence of comorbid illness. Furthermore, theremay have been other large, randomized clinical trials that used a standardcare arm that were not published and were not identified when we contactedexperts and, therefore, are excluded from our analysis. Since studies thatshow a positive treatment effect is more likely to be published, publicationbias can be an important source of bias when combining studies to estimatethe effect of a treatment. However, in our study we evaluated the associationof risk factors on wound healing. Treatment benefit, not these associations,was the primary end point of the studies. It is, therefore, doubtful thatan association or lack of an association of these risk factors with woundhealing would have influenced an author's decision to publish. In addition,both the Begg test and the Egger bias plot were applied, and neither was consistentwith substantial publication bias.28,29While the effect that this type of publication bias might have had on ourresults cannot be predicted, we doubt that the effect could have been substantial.
Neuropathic diabetic foot ulcers are often treated in the ambulatorycare setting.3,5,17,30In fact, the standard-care regimen described in this article can be providedto patients entirely as outpatients. Our meta-analysis of patient level datashowed that the use of a standard-care regimen for neuropathic diabetic footulcers is most likely to be successful for patients who have more recent (<6months) and smaller wounds (<2 cm2) (Table 2, bottom third). There also seems to be an influence of ethnicity,in that nonwhite patients do better than white patients. However, there isno biological basis for this finding. It is likely that this finding reflectsan overall unmeasured bias in the enrollment of patients into these randomizedclinical trials (eg, only nonwhites who were perceived to be compliant bythe local study investigators were enrolled in these clinical trials). Asnew treatments for neuropathic diabetic foot ulcers become available, dermatologistsand other medical providers should find the risk factor information from thisstudy useful to help them decide whether to use a new treatment in associationwith standard care, refer to a specialist, or start with standard care alone.Furthermore, when researchers plan future clinical trials for studying nonhealingneuropathic wounds in patients with DM, it is prudent for the investigatorsto understand these risk factors so that patients who are most likely to failstandard care can be enrolled in the study.
A cooperative effort of the Clinical Epidemiology Unit of the IstitutoDermopatico dell'Immacolata–Istituto di Recovero e Cura a CarattereScientifico (IDI-IRCCS) and the Archives of Dermatology
Accepted for publication August 8, 2000.
This study was funded in part by a grant from Curative Health Servicesto the Trustees of the University of Pennsylvania (Dr Margolis) and by grantAG 00715 from the National Institutes of Health, Bethesda, Md (Dr Margolis).
Sandy Masiak ably provided secretarial support.
Corresponding author: David J. Margolis, MD, MSCE, 815 Blockley Hall,423 Guardian Dr, Department of Epidemiology and Biostatistics, Universityof Pennsylvania School of Medicine, Philadelphia, PA 19104-6021 (e-mail: email@example.com).
Margolis et al use the creative approach of using individual patientdata obtained from the control arms of 5 industry-sponsored clinical trialsto investigate factors that are associated with healing of neuropathic diabeticfoot ulcers. Factors associated with healing within 20 weeks included durationless than 6 months, small size, and ethnicity. Age, diabetic control, andsex were not associated with outcome.
Please see Trisha Greenhalgh's "How to Read a Paper: Papers That SummariseOther Papers (Systematic Reviews and Meta-analyses)" in the British Medical Journal (1997;31:672-675 [Also available at: http://www.bmj.com/cgi/content/full/315/7109/672. Accessibility verified:October 17, 2000]) for a brief overview of meta-analysis and a series in the British Medical Journal by Matthias Egger for a more in-depthdiscussion of the pros and cons of meta-analysis including the detection ofpublication bias [Available at: http://www.bmj.com/cgi/content/full/315/7109/672. Accessibility verified: October 17, 2000]. Logistic regression isa method used to pool data from several different studies. Please see thereview article "Should We Adjust for Covariates in Nonlinear Regression Analysesof Randomized Trials?" by W. W. Hauck et al that was published in Controlled Clinical Trials (1998;19:249-256) for a discussion of theneed to adjust for covariates in meta-analyses.—MichaelBigby, MD, Boston, Mass