UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Improving footwear to prevent ulcer recurrence in diabetes: Analysis of
adherence and pressure reduction
Waaijman, R.
Publication date
2013
Link to publication
Citation for published version (APA):
Waaijman, R. (2013). Improving footwear to prevent ulcer recurrence in diabetes: Analysis of
adherence and pressure reduction.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
Chapter 7
In preparation
Roelof Waaijman
Mirjam de Haart
Mark L.J. Arts
Daan Wever
J.W.E. Verlouw
Frans Nollet
Sicco A. Bus
PROGNOSTIC FACTORS OF PLANTAR FOOT ULCER
RECUR-RENCE IN NEUROPATHIC DIABETIC PATIENTS
ABSTRACT
Background: Plantar foot ulcer recurrence in patients with diabetes mellitus and
pe-ripheral neuropathy is common and its risk factors are not well understood. A compre-hensive analysis of biomechanical, behavioural, and disease-related factors was con-ducted to assess prognostic determinants of plantar foot ulcer recurrence.
Methods: 171 neuropathic diabetic patients with prior plantar foot ulceration and
prescription therapeutic footwear were followed for 18 months or until ulceration. Demographic data, disease parameters, barefoot and in-shoe plantar peak pressures, footwear adherence, and daily ambulatory activity were entered in multivariate logistic regression models of ulcer recurrence: one model for all recurrences and one for recur-rences at the prior ulcer location with a non-traumatic cause, to isolate those ulcers suggestive of a pressure-related aetiology (model 2).
Findings: Significant determinants for all 71 recurrences were: minor lesion at study
entry (OR 5.24, 95%CI[2.83-9.68]), day-to-day variation in stride count (0.92[0.88-0.95]), and past cumulative ulcer months (1.04[1.01-1.06]), predicting 68% of ulcers correctly. Significant determinants for the 41 recurrences at the prior ulcer location were: minor lesion at study entry (9.12[4.14-20.09]), in-shoe peak pressure <200kPa + adherence >80% (0.40[0.18-0.90]), barefoot peak pressure (1.10[1.00-1.21]), day-to-day variation in stride count (0.92[0.87-0.97]), and past cumulative ulcer months (1.02[1.00-1.05]), predicting 76% of ulcers correctly.
Interpretation: Prediction of plantar foot ulcer recurrence was moderately
success-ful. Minor lesion was the most dominant predictor, while sustained adequate offload-ing was protective against pressure-related recurrence. These findoffload-ings suggest that to reduce ulcer recurrence risk, minor lesions should be adequately managed, barefoot walking prevented, and adequate offloading and adherence guaranteed.
7
INTRODUCTION
In patients with diabetes mellitus, foot ulcers are a serious risk for infection and am-putation1, 2. The prevention of foot ulcers is paramount to avoid these devastating
out-comes. Several studies have identified risk factors for diabetic foot ulceration, which include, among others, peripheral neuropathy, peripheral arterial disease, and foot de-formity3-8. The strongest predictors of ulceration are peripheral neuropathy and a
his-tory of ulceration, which illustrates that ulcers often recur, up to 40% annually9.
Elevated dynamic barefoot plantar pressure during walking is, in the presence of neu-ropathy, an important predictor of diabetic foot ulceration10-12. One biomechanical study
found the most optimal barefoot peak pressure cut-off level, 700kPa, to be 70% sensi-tive and 65% specific for ulceration13, while another study found a barefoot peak
pres-sure of 875kPa to be 64% sensitive and 46% specific11. This suggests that barefoot
plan-tar peak pressure, although important, is a moderate predictor. Adding in-shoe planplan-tar peak pressure may improve prediction since patients, in particular those with healed foot ulcers, often have prescription therapeutic footwear with the goal to prevent ulcer recurrence. Controversy still exists on the effectiveness of this footwear14-16, but none of
these studies measured in-shoe plantar pressures. Therefore the role of this parameter in ulcer development is so far unknown.
Factors that may further contribute to ulceration are ambulatory weight-bearing activ-ity and adherence to wearing adequately offloading footwear. More steps taken pre-sumes a higher cumulative load on the foot during the day, which may increase the risk of ulceration, although to date a clear association between activity and ulceration has not been found17, 18. Adherence to wearing prescription footwear is a behavioural
pa-rameter that also influences the amount of biomechanical stress that the foot experi-ences since we know that barefoot peak pressures are generally much higher than in-shoe peak pressures. Combining pressure, activity, and adherence data best represents the cumulative load on the foot and may prove valuable in understanding ulcer recur-rence in diabetes19.
To the best of our knowledge, a prospective study that integrates disease-related, bio-mechanical, and behavioural factors into a model of plantar foot ulcer recurrence does not exist. Such a model may support clinicians to decide on potentially preventative measures. The aim of this prospective study was to assess the prognostic value of dis-ease-related, biomechanical, and behavioural factors in plantar foot ulcer recurrence in diabetic patients who are at high risk for foot ulceration.
METHODS
SubjectsOne hundred seventy-one patients with diagnosed diabetes mellitus, peripheral neu-ropathy, a recently healed plantar foot ulcer (<18 months prior to study entry) and new prescription footwear were included in the study. These were patients who participated in a randomized controlled trial on therapeutic footwear effectiveness (Dutch trial
reg-ister, NTR1091) and who were consecutively recruited from 10 participating centres. Loss of protective sensation due to peripheral neuropathy was confirmed present by 10-g Semmes Weinstein monofilament and vibration perception threshold testing5. The
exclusion criteria were an active plantar foot ulcer, bilateral amputation proximal to the tarso-metatarsal level, severe illness that would make 18-months survival unlikely, and the inability to walk unaided. Written informed consent was obtained from each patient prior to inclusion of the study, which was approved by the local Research Ethics Com-mittee of each participating centre.
Outcome
The outcome in this study was plantar foot ulcer recurrence in 18 months. A plantar foot ulcer was defined as a full-thickness lesion of the skin, i.e. a wound penetrating through the dermis at the plantar side of the foot, without reference to time present20, 21. Photographs of the plantar foot, taken at each study visit or in-between visits when
an ulcer occurred, were assessed for outcome by 3 independent diabetic foot experts who were blinded to patient identity and characteristics. When agreement on outcome was not present, 2 additional foot experts assessed the photographs to obtain a clear (4 against 1) majority on outcome.
Two outcomes of plantar foot ulcer recurrence were considered: all ulcer recurrences, and only those recurrences with a suggested pressure-related aetiology (model 2), be-cause of our specific interest in biomechanical determinants. A pressure-related aetiol-ogy was suggested for those plantar foot ulcers of which patients did not report trauma as cause (e.g. bumping the foot, stepping into a nail) and which developed at the loca-tion where the prior ulcer healed22.
Procedures
Patients were prospectively evaluated every 3 months until a plantar foot ulcer oc-curred or until 18 months follow-up, whichever came first. At study entry, demographic and disease-related data were collected, a foot examination was performed, and new custom-made therapeutic footwear was delivered to the patient. Footwear consisted of custom-made insoles either in custom-made shoes or in extra-depth shoes. Photo-graphs of the foot taken at each study visit using a standardized protocol were assessed for presence of foot deformities and minor lesions, which were defined as the presence of abundant callus, haemorrhage, or a blister on the plantar aspect of the foot. Two observers independently scored foot deformity and minor lesions and reached consen-sus on outcome. Furthermore, at each visit, the frequency of treatment by a foot care provider was assessed.
Barefoot dynamic plantar foot pressures were measured at study entry using an EMED-X system (Novel, Munich, Germany) with sensors arranged in a spatial resolution of 4 sen-sors per cm2, sampling at 50 Hz. Data was collected using a two step gait approach to the
platform23. At each study visit, in-shoe dynamic plantar foot pressures were measured
in a maximum two pair of prescription footwear per patient, including the newly deliv-ered footwear. In-shoe plantar pressures were measured at 50Hz sampling frequency using the Pedar-X system (Novel, Munich, Germany). This system includes wide-sized
7
pressure measurement insoles with a sensor resolution of approximately one sensor per cm2. Patients walked forth and back along a 10-meter long walkway assuring that a
minimum 12 midgait steps per foot were collected24. Each Pedar insole was calibrated
each 3 months using a calibration device and protocol from the manufacturer.
Use of prescription footwear was objectively assessed once during the study over a 7-day period using a shoe-worn monitor (@monitor, Academic Medical Center, Amster-dam) that accurately and reliably distinguished periods that shoes were on and off25.
Over the same 7-day period, daily stride count was measured using an ankle-worn step activity monitor (StepWatchTM, Orthocare Innovations LLC, Oklahoma, United States).
Patients were asked to keep a daily log of the time periods that they were away from home.
Data analysis
In the barefoot and in-shoe plantar pressure distribution pictures, regions of interest were masked, including the plantar foot region distal to the heel (named ‘plantar foot’), the prior ulcer location, and the new ulcer location in those patients who ulcerated. Mean peak pressure and pressure-time integral over the steps taken were calculated for each mask using Novel multimask software (version 20.3.32). Pressure data was averaged over two consecutive study visits to reflect as adequate as possible the loading condition over the 3-month period in-between visits.
To calculate adherence to wearing prescription footwear, the @monitor and StepWatch data were synchronized and analysed using Matlab R2011a software (MathWorks, Na-tick, MA). Adherence was calculated as the percentage of foot strides over the measu-rement period that the patient wore their prescribed footwear. Using the daily log data, adherence and stride count were calculated for the periods that patients were at home and away from home.
Data on plantar foot pressure, adherence, and stride count were combined to define two new parameters that represented the actual load on the foot in daily life: weighted
pressure (WP) and cumulative plantar tissue stress (CPTS): WP = inshoe PP * adherence + barefoot PP * (1 - adherence) [kPa]
CPTS = (inshoe PTI * adherence + barefoot PTI * (1 - adherence)) * strides [MPa.s/day]
where PP = peak pressure; PTI = pressure time integral and strides = daily stride count
Independent factors
Multiple patient and disease-related factors were included in the analysis: age, gender, diabetes type and duration, body-mass index, HbA1c, vibration perception threshold, cumulative duration of past foot ulcers, history of amputation, presence of peripheral arterial disease20, smoking, alcohol consumption (>2 units/day), living alone, being
employed, highest education level, and time between healing of the prior ulcer and study entry. Foot deformity was also included and was classified as “absent”, “mild” (i.e.
pes planus, pes cavus, hallux valgus or limitus, hammer toes, and lesser toe amputa-tion), “moderate” (i.e. hallux rigidus, hallux or ray amputation, prominent metatarsal heads, and claw toes), or “severe” (i.e. Charcot deformity, (fore)foot amputation and pes equines). For minor lesions, presence of a minor lesion at study entry was included, as well as the ratio of the first two visits (entry and first follow-up) that a minor lesion was present (“minor lesion index”, [0, 0.5, 1]).
Biomechanical factors included in the analysis were barefoot plantar peak pressure at the plantar foot, the prior ulcer location, and the new ulcer location, and in-shoe peak plantar pressure at the plantar foot, prior ulcer location, and the new ulcer location (for each pair of shoes included in the study). Additionally, patients were categorized into two groups based on an in-shoe peak pressure cut-off level of 200 kPa at the plantar foot and the prior ulcer location, as this was indicated to be potentially discriminative for ulcers remaining healed26.
Behavioural factors included in the analysis were: average daily stride count, day-to-day variation in stride count (calculated as the standard deviation in daily stride count over the 7-day measurement period)27, footwear adherence, adherence at home, adherence
away from home, and the average number of visits per month to a foot care provider. Additionally, patients were categorized into two groups based on the combined pres-ence, or abspres-ence, of an in-shoe peak pressure <200kPa and adherence ≥80%. This pa-rameter, together with weighted pressure and cumulative plantar tissue stress were included in the analysis.
Statistical analysis
Descriptive statistics were performed using SPSS Statistics 19 (IBM Corporation, NY, USA). To combine patient-related data (one patient) and foot related data (two feet) in the analysis, we selected the foot with the new ulcer for those patients who re-ulcer-ated and the foot with the prior ulcer for those patients who survived ulceration in 18 months.
Including all independent factors, univariate and multivariate multi-level logistic re-gression models were developed of all plantar foot ulcer recurrences (model 1) and all recurrences with pressure-related aetiology (model 2) using MLWIN software, version 2.23 (Institute of Education, University of London, London, UK)28. Foot ulcer recurrence
was nested at four levels – participating centre (fourth level), patient (third), footwear (second) and follow-up period (first) – and a random intercept at patient level or foot-wear level was calculated, whichever fitted the model best. A significance level of P < .10 was used. Significant factors from the univariate analysis were entered in a multivariate multi-level logistic regression model of plantar foot ulcer recurrence (with backward selection, significance at P < .10). To assess the fit of the model, the percentage of cor-rectly classified ulcer recurrences was calculated.
Pearson correlation coefficients were calculated between selected factors, to explore associations between these factors.
7
RESULTS
Baseline patient characteristics and descriptive statistics for all factors in the study are shown in table 1. Seventy-one patients developed a plantar foot ulcer, in a mean±SD 28±22 weeks. Forty-one of the 71 ulcers developed at the prior ulcer location and had a suggested pressure-related aetiology. These 41 ulcers developed in a mean±SD 25±19 weeks. In 23 of the 71 patients with ulcer recurrence, a minor lesion directly preceded the ulcer, meaning that a minor lesion was present at the last follow-up visit before ul-ceration. Of all 135 minor lesions that were identified during the study, 17% developed into a foot ulcer.
Results of the univariate and multivariate multi-level logistic regression analysis for all 71 recurrences (model 1) are shown in table 2. The univariate analysis showed that having a severe foot deformity, a higher minor lesion index, a minor lesion at entry, in-creased weighted pressure, inin-creased barefoot peak pressure, and a longer cumulative duration of past foot ulceration increased the odds of developing ulcer recurrence. A combination of in-shoe peak pressure <200kPa with adherence >80%, and more day-to-day variation in stride count decreased the odds of developing ulcer recurrence. In multivariate analysis, a minor lesion at entry, cumulative duration of past foot ulcera-tion in months, and day-to-day variaulcera-tion in stride count remained independently signif-icant prognostic factors. With this model, 68% of the ulcer recurrences were correctly classified.
The results for the 41 ‘pressure-related’ recurrences (model 2) are shown in table 2. The univariate analysis showed that having a higher minor lesions index, a minor le-sion at entry, increased in-shoe peak pressures at the (prior) ulcer location, increased weighted pressure, increased barefoot peak pressure, and a longer cumulative dura-tion of past foot ulceradura-tion increased the odds of developing a recurrent ulcer. A combi-nation of in-shoe peak pressure <200kPa with adherence >80%, longer time between prior ulcer healing and study entry, more day-to-day variation in stride count, and a longer diabetes duration decreased the odds for a recurrent ulcer. In the multivariate analysis, a minor lesion at entry, combination of in-shoe peak pressure <200kPa with adherence >80%, barefoot peak pressure, day-to-day variation in stride count, and cu-mulative months of past foot ulceration remained independent significant prognostic factors. With this model, 76% of the ulcer recurrences were correctly classified. The correlation coefficient between minor lesion index and the time between prior ul-ceration and study entry was r = -0.23 (P < 0.01); and between minor lesion index and the cumulative number of months patients had a foot ulcer in the past was r = 0.18 (P < 0.05).
Table 1. Base line and out come dat a expr essed as mean (SD), per cent ag e (%), or as fr equenc y (N) for all inc luded patients, for the patients wit h and wit hout ulc er recurr enc e in t he analysis of al l 71 plant ar ulc ers, and f or t he patients wit h and wit hout ulc er r ecurr enc e in t he analysis of t he 41 non-tr aumatic plant ar ulc ers t hat dev eloped at t he pr evious ulc er loc ation. Variable All patients Anal
ysis of all ulc
er r ecurr enc es (N=71) Anal
ysis of all non-tr
aumatic
ulc
ers at the prior ulc
er location (N=41) No ulc er Ulc er No ulc er Ulc er Subjects (N) 171 100 71 130 41 Age ( years) 63.3 (10.1) 63.6 (9.4) 62.8 (11.2) 63.4 (9.8) 62.9 (11.3) Male gender (%) 82.5 80 85.9 80.8 87.8 Diabet es type 2 (%) 71.3 71.0 71.8 70.8 73.2 Diabet es dur ation ( years) 17.3 (13.5) 17.7 (13.8) 16.7 (13.2) 18.2 (14.2) 14.3 (10.5) BMI (k g/m 2) 30.7 (5.7) 30.7 (5.3) 30.6 (6.2) 30.8 (5.7) 30.3 (5.7) HbA1c (%) 7.6 (1.4) 7.5 (1.3) 7.7 (1.6) 7.6 (1.4) 7.6 (1.5) Vibr ation P er ception thr eshold (V) 45.4 (10.2) 46.2 (9.9) 44.3 (10.6) 45.6 (10.3) 44.8 (10.0) Cumulati ve dur ation of past f
oot ulcers (months)
14.1 (19.0) 10.7 (12.5) 20.3 (26.2) 13.3 (17.9) 16.9 (22.5) Smok er or hist ory of smoking (%) 66.7 66.0 67.6 63.8 75.6 Mor
e than 2 units alcohol intak
e per da y (%) 11.7 13.0 9.9 11.5 12.2 Li ving alone (%) 26.9 22.0 33.8 26.2 29.3 Emplo yed (%) 21.6 20.0 23.9 20.8 24.4 Education le vel (Lo w / Medium / Hig h %) 56 / 18 / 26 56 / 18 / 26 56 / 18 / 25 57 / 17 / 26 54 / 22 / 24 Peripher al art erial disease, gr ade II (%) a 35.2 38.5 30.4 36.5 30.8
Visits per month t
o a f oot car e pr ovider (N) 1.3 (0.8) 1.2 (0.7) 1.4 (0.8) 1.2 (0.8) 1.5 (8.3)
Minor lesion at entry (%)
29.2
18.0
45.1
17.7
65.9
Minor lesion inde
x (0-1) 0.27 (0.38) 0.17 (0.31) 0.41 (0.43) 0.17 (0.32) 0.57 (0.41) Hist ory of amputation (%) 29.8 28.0 32.4 31.5 24.4 Foot def ormity (%) Absent 9.9 13.0 5.6 11.5 4.9
7
Variable
All patients
Anal
ysis of all ulc
er r ecurr enc es (N=71) Anal
ysis of all non-tr
aumatic
ulc
ers at the prior ulc
er location (N=41) No ulc er Ulc er No ulc er Ulc er Mild 36.8 38.0 35.2 36.2 39.0 Moder at e 41.5 42.0 40.8 40.8 43.9 Se ver e 11.7 7.0 18.3 11.5 12.2 Months healed fr om ulcer ation bef or e stud y entry 4.99 (5.50) 5.34 (5.71) 4.45 (5.19) 5.45 (5.67) 3.54 (4.73) Dail y stride count (N) 3359 (1749) 3437 (1990) 3238 (1287) 3404 (1857) 3209 (1331) Variation in dail y stride count (N) 1194 (713) 1276 (793) 1068 (549) 1234 (747) 1062 (578) Adher ence (%) 72.9 (24.3) 72.7 (24.1) 73.1 (24.7) 73.1 (23.7) 72.2 (26.2) Adher ence at home (%) 61.7 (32.3) 63.2 (32.3) 59.3 (32.6) 64.4 (31.2) 53.7 (34.7) Adher ence a w ay fr om home (%) 87.6 (26.7) 84.9 (30.8) 91.7 (18.3) 86.1 (28.8) 91.9 (19.1) In-shoe peak pr essur e @plantar f oot (kP a) 254 (79) 249 (77) 261 (83) In-shoe peak pr essur e @plantar f oot >200kP a at entry (%) 76.6 76.0 77.5 Bar ef oot peak pr essur e @plantar f oot (kP a) 979 (293) 935 (307) 1042 (260) W eig ht ed pr essur e @plantar f oot (kP a) 445 (198) 429 (188) 472 (212) Cum. tissue str ess @plantar f oot (MP a.s/da y) 675 (475) 652 (436) 715 (538) In-shoe peak pr essur e @plantar f oot <200kP a and adher ence >80% at entry (%) 9.7 10.2 9.0 In-shoe peak pr essur e @(prior)ulcer (kP a) 187 (88) 178 (82) 212 (99) In-shoe peak pr essur e @(prior)ulcer >200kP a at entry (%) 40.1 36.2 51.2 Bar ef oot peak pr essur e @(prior)ulcer (kP a) 738 (393) 699 (393) 849 (375) W eig ht ed pr essur e @(prior)ulcer (kP a) 329 (192) 314 (188) 372 (200) Cum. tissue str ess @(prior)ulcer (MP a.s/da y) 377 (283) 361 (279) 423 (292) In-shoe peak pr essur e @(prior)ulcer <200kP a and adher ence >80% at entry (%) 24.8 27.3 17.9 Pr esenc e of peripher al art erial disease (gr ade I=No, gr ade II =Y es) w as asse ssed using t he PEDIS c lassific ation 20
Table 2. Re sults f or t he univ ariat e and multiv ariat e multi-lev el logistic r egr ession analysis of f act ors r elat ed t o all plant ar f oot ulc er r ecurr enc e (N=71) and non-tr aumatic ulc er r ecurr enc es at t he pr evious ulc er loc ation (N=41). Anal
ysis of all ulc
er r
ecurr
enc
es (N=71)
Anal
ysis of all non-tr
aumatic ulc
ers at the prior ulc
er location (N=41) Uni variat e model Multi variat e model Uni variat e model Multi variat e model Variable ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue Age ( years) 0.99 [0.96-1.02] .678 1.00 [0.96-1.03] .779 Gender (M/F) 0.66 [0.3-1.47] .310 0.56 [0.22-1.44] .230 Diabet es type (1/2) 1.03 [0.52-2.01] .935 1.12 [0.52-2.44] .769 Diabet es dur ation ( years) 1.00 [0.97-1.02] .705 0.97 [0.95-1.00] .041* BMI (k g/m 2) 1.00 [0.95-1.05] .977 0.98 [0.93-1.05] .627 HbA1c (%) 1.14 [0.91-1.41] .255 1.00 [0.77-1.29] .985 Vibr ation P er ception thr eshold (V) 0.98 [0.95-1.01] .267 0.99 [0.96-1.03] .662 Cumulati ve dur ation of past f oot ulcers (months) 1.04 [1.01-1.06] .006* 1.04 [1.02-1.06] .001* 1.04 [1.01-1.07] .012* 1.02 [1.00-1.05] .063* Smoking (No / Y es or smoking hist ory) 1.08 [0.56-2.08] .810 1.69 [0.79-3.62] .178 Alcohol intak e >2 units/da y (No/Y es) 0.75 [0.29-1.93] .547 1.06 [0.36-3.15] .918 Li
ving alone (No/Y
es) 1.79 [0.89-3.57] .102 1.14 [0.52-2.53] .745 Emplo yment (No/Y es) 1.26 [0.6-2.63] .545 1.24 [0.53-2.91] .616 Education le vel Low ref er ence ref er ence Medium 0.99 [0.43-2.26] .983 1.36 [0.53-3.51] .528 Hig h 0.95 [0.46-1.97] .898 1.00 [0.43-2.31] .993 PAD gr ade (I / II)a 0.78 [0.53-1.15] .206 0.79 [0.37-1.68] .540 Monthl y visits t o f oot car e pr ovider (N) 1.35 [0.87-2.12] .185 1.44 [0.85-2.45] .179
Minor lesion at entry (No/Y
es) 3.80 [1.88-7.65] .000* 5.24 [2.83-9.68] .000* 9.75 [3.81-24.91] .000* 9.12 [4.14-20.09] .000*
Minor lesion inde
x (0-1)
5.60 [2.39-13.13]
.000*
15.27 [5.03- 46.39]
7
Anal
ysis of all ulc
er r
ecurr
enc
es (N=71)
Anal
ysis of all non-tr
aumatic ulc
ers at the prior ulc
er location (N=41) Uni variat e model Multi variat e model Uni variat e model Multi variat e model Variable ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue Hist
ory of amputation (No/Y
es) 1.21 [0.62-2.37] .567 0.69 [0.32-1.51] .355 Se verity of def ormity Absent ref er ence ref er ence Mild 2.24 [0.73-6.87] .158 2.69 [0.73-9.9] .136 Moder at e 2.33 [0.77-7.05] .135 2.66 [0.73-9.66] .136 Se ver e 6.10 [1.57-23.69] .009* 2.69 [0.56-12.91] .217
Months healed bef
or e stud y entry 0.97 [0.92-1.03] .334 0.94 [0.88-1.01] .081* Dail
y stride count (per 100 strides)
0.99 [0.97-1.01]
.360
0.99 [0.97-1.01]
.404
Variation in dail
y stride count (per 100
strides) 0.95 [0.90-0.99] .023* 0.92 [0.88-0.95] .000* 0.96 [0.91-1.01] .096* 0.92 [0.87-0.97] .003* Adher ence (%) 1.00 [0.99-1.01] .989 1.00 [0.98-1.01] .823 Adher ence abo ve 80% 0.98 [0.51-1.9] .958 1.06 [0.5-2.23] .883 Adher ence at home (%) 1.00 [0.98-1.01] .495 0.99 [0.98-1.00] .146 Adher ence a w ay fr om home (%) 1.01 [1.00-1.03] .114 1.01 [0.99-1.03] .216 In-shoe peak pr essur e @plantar f oot (per 100 kP a) 1.21 [0.95-1.53] .120 In-shoe peak pr essur e @plantar f oot >200kP a 1.24 [0.85-1.82] .261 Bar ef oot peak pr essur e @plantar f oot (per 100kP a) 1.18 [1.09-1.27] .000* W eig ht ed pr essur e @plantar f oot (per 100 kP a) 1.22 [1.08-1.38] .001* Cum. tissue str ess @plantar f oot (per MP a.s/ da y) 1.00 [1.00-1.00] .453
Anal
ysis of all ulc
er r
ecurr
enc
es (N=71)
Anal
ysis of all non-tr
aumatic ulc
ers at the prior ulc
er location (N=41) Uni variat e model Multi variat e model Uni variat e model Multi variat e model Variable ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue ß [95% CI] P v alue In-shoe peak pr essur e @plantar f oot <200kP a and adher ence >80% 0.47 [0.26-0.85] .012* In-shoe peak pr essur e @(prior)ulcer (per 100 kP a) 1.38 [1.05-1.81] .023* In-shoe peak pr essur e @(prior)ulcer >200kP a 1.43 [0.89-2.32] .142 Bar ef oot peak pr essur e @(prior)ulcer (per 100 kP a) 1.12 [1.05-1.21] .001* 1.10 [1.00-1.21] .053* W eig ht ed pr essur
e @(prior) ulcer (per 100
kP a) 1.22 [1.06-1.40] .005* Cum. tissue str ess @(prior)ulcer (per MP a.s/ da y) 1.00 [1.00-1.00] .162 In-shoe pr essur e @(prior)ulcer <200kP a and adher ence >80% 0.5 [0.28-0.89] .019* 0.40 [0.18-0.90] .027* Dat a ar e mean (SD) value s per fact or , un-st andar diz ed linear regr ession coeffi cients (i.e. odds ratio’s) wit h 95% confidenc e int er vals, and P-v alue s. * = signific ant association (P < 0.10); r ef er enc e = r ef er enc e c at eg or y
7
DISCUSSION
Taking into account a wide range of biomechanical and behavioural factors that are suggested to be important in causing plantar foot ulcer recurrence in diabetes, 68% of all recurrences and 76% of recurrences most supposed to be pressure-related were correctly classified. This is moderate given the large number of risk factors included and the fact that by chance alone, 50% of ulcers are correctly classified. Factors that could be expected to play a role in plantar foot ulcer recurrence proved to be the strongest de-terminants. The presence of a minor lesion (i.e. haemorrhage, abundant callus or blis-ter) at study entry caused the largest increase in ulcer recurrence risk (odds ratio’s > 5). History of patient burden, expressed as the cumulative number of months that patients had a foot ulcer, could also be expected as significant factor. And for biomechanics and patient behaviour, the analyses showed that both peak pressure while walking barefoot (predictor) and continuous offloading in worn prescription footwear (protector) were associated with ulcer recurrence.
The strong association between minor lesions and foot ulceration has been long recog-nized in the aetiological pathways of foot ulceration in diabetes21. A minor lesion
direct-ly preceded 32% of recurrences and 17% of minor lesions resulted in a foot ulcer. Both the minor lesion index and minor lesion at entry were strong univariate predictors of recurrence, but only lesions at entry remained independently significant in multivariate analysis. A shorter period between healing of the prior foot ulcer and entry in the study was positively correlated with minor lesion index. This suggests that the previous ulcer, although fully reepithelialised, had not yet fully recovered to intact skin making it more vulnerable for injury. Furthermore, minor lesion index was also significantly correlated with the cumulative number of months patients had a foot ulcer in the past. This may explain why a history of ulceration is one of the strongest predictors of new ulceration in diabetes3, 4.
For the first time, prospectively an association is found between pressures measured in-side prescription footwear and ulcer recurrence in neuropathic diabetic patients. This accounted for the analysis of ‘pressure-related’ ulcer recurrence (model 2), not for the analysis of all recurrences (model 1), for which none biomechanical parameter was a prognostic factor in the multivariate analysis. This difference in outcome between the two models is understandable because for model 2 only recurrences were taken into account with a suggested pressure-related cause, not being the result of direct trauma acting on the foot. Although we only had data on ulcer cause from patient reports, we suggested that recurrence at the prior ulcer location would best represent a non-trau-matic pressure-related cause. While both in-shoe peak pressure alone (OR 1.4) and low in-shoe peak pressure combined with high adherence (OR 0.50) were significant fac-tors in the univariate analysis of model 2, only the latter remained a prognostic factor in multivariate analysis, thereby confirming earlier results from Chantelau et al.29. This
stresses the importance of continuous offloading to lower risk of ulcer recurrence. High barefoot peak pressure also increased the risk of recurrence, which is in accordance with several previous studies5, 10, 11. Predictive power was, however, small, both
univari-ately and multivariunivari-ately, compared to these previous analyses. This may be because we tested a selected group of high risk patients (all with a prior foot ulcer and peripheral neuropathy) which may weaken associations compared to studies that assess patients
across a wider spectrum of complications. In any case, identification of a common bare-foot peak pressure threshold for ulcer recurrence remains an elusive goal.
Footwear adherence and ambulatory activity level were not independently predictive of ulcer recurrence. In combination with low in-shoe peak pressures, high adherence was, however, protective of ulcer recurrence. Weighted pressure also combined plantar peak pressures with adherence. This showed to be a significant factor in univariate, but not in multivariate analysis, maybe because it does not use specific cut-off levels for low pressure and high adherence like the first parameter does. No association was found with ulcer recurrence when combining pressure-time integral with adherence and daily stride count as cumulative plantar tissue stress parameter, despite that we defined cu-mulative stress in a more sophisticated way than others did18. Maybe this is because
the role of ambulatory activity level in ulcer risk is not straightforward17, 27. Further
sophistication of the model, incorporating the measurement of all footwear that the patient wears (also non-prescription footwear), and, when possible, shear pressure as a parameter, may be needed to better predict plantar foot ulcer recurrence in high-risk diabetic patients.
Understanding which factors determine ulcer recurrence in this high-risk group of pa-tients is important given the high prevalence of recurrence after an episode of ulceration. However, factors known to be the strongest predictors of ulceration, i.e. prior ulceration and presence of peripheral neuropathy, were conditions all patients already had, and could therefore not be included as dichotomous parameter in the regression analyses. Furthermore, all patients had prescription custom-made footwear, which prevented as-sessing the role of having prescription footwear or not on ulcer recurrence. These facts may explain the moderate success in correctly classifying ulcer recurrence with both models in this study. Measuring disease-related, biomechanical, and behavioural fac-tors in a diabetic patient group ranging in complication level from non-neuropathic, to neuropathic without ulcer history, to neuropathic with ulcer history and without fixed types of footwear may better predict ulceration. Nevertheless, the results showed that changeable as well as non-changeable factors were associated with ulcer recurrence, and clinical practice should focus on managing the changeable factors.
For clinical practise the findings in this study implicate that to lower the risk of plantar foot ulcer recurrence in diabetes, the primary focus should be on frequently screening patients’ feet for presence of minor lesions as ‘warning signs’ of ulceration. It can be questioned whether current recommendations for foot screening suffice; more frequent screening may be required21. If minor lesions are present, they should be managed in
a timely fashion to prevent further complications. The findings on minor lesions also suggest that patients may benefit from continued offloading with a healing device after re-epithelialisation of the ulcer and before transferring to preventative footwear to al-low the skin to regenerate and regain strength. Regarding biomechanics and patient behaviour, the predictive effect of barefoot peak pressures and the protective effect of adequate footwear offloading and adherence stresses the importance for patients to avoid walking barefoot and to wear adequately offloaded footwear at all times.
7
CONCLUSION
In conclusion, the study showed that prediction of plantar foot ulcer recurrence in dia-betes was moderately successful from disease-related, biomechanical and behavioural factors. The presence of minor lesions most strongly increases the odds to develop ulcer recurrence. High barefoot peak pressure increases the risk, while continuous effective offloading in worn prescription footwear reduces the risk for ulcer recurrences with a suggested pressure-related aetiology. Therefore, to prevent ulcer recurrence in diabe-tes, the focus should be on managing the changeable prognostic factors, which means reducing risk and improving early recognition of minor lesions, urging patients not to walk barefoot, guaranteeing adequate offloading in prescription footwear, and impro-ving adherence to prescription footwear.
Acknowledgements
The Academic Medical Centre in Amsterdam collaborated with 9 other multidiscipli-nary diabetic foot centres and 9 orthopaedic footwear companies in the Netherlands in this study. The authors like to acknowledge the contribution of Renske Keukenkamp in collecting data for the study and the following persons in patient recruitment and footwear prescription: T.E. Busch-Westbroek, P.J.A. Mooren, I. Ruijs, H. van Wessel, J.P.J. Bakker, C. van den Eijnde, H. Wessendorf, R. Dahmen, B. Koomen, J.G. van Baal, R. Has-pels, J. Harlaar, V. de Groot, J. Pulles, W. Polomski, R. Lever, G. du Mont, H.G.A. Hacking, J. de Bruin, H. Berendsen, W. Custers. This study was supported by project grants from the Dutch Diabetes Research Foundation (project 2007.00.067), the Dutch Foundation for the Development of Orthopaedic Footwear Technology (OFOM), and the Dutch Organi-zation for Health Research and Development (ZonMw, project 14350054).
REFERENCES
1. Lavery LA, Armstrong DG, Wunderlich RP, Mohler MJ, Wendel CS, Lipsky BA. Risk factors for foot infections in individuals with diabetes. Diabetes Care 2006; 29: 1288-1293.
2. Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA 2005; 293: 217-228.
3. Abbott CA, Carrington AL, Ashe H, Bath S, Every LC, Griffiths J, Hann AW, Hussein A, Jackson N, Johnson KE, Ryder CH, Torkington R, Van Ross ER, Whalley AM, Widdows P, Williamson S, Boulton AJ. The North-West Diabetes Foot Care Study: incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort. Diabet Med 2002; 19: 377-384.
4. Boyko EJ, Ahroni JH, Stensel V, Forsberg RC, Davignon DR, Smith DG. A prospective study of risk factors for diabetic foot ulcer. The Seattle Diabetic Foot Study. Diabetes Care 1999; 22: 1036-1042.
5. Pham H, Armstrong DG, Harvey C, Harkless LB, Giurini JM, Veves A. Screening techniques to identify people at high risk for diabetic foot ulceration: a prospective multicenter trial. Diabetes Care 2000; 23: 606-611.
6. Apelqvist J, Larsson J, Agardh CD. Long-term prognosis for diabetic patients with foot ulcers. J Intern Med 1993; 233: 485-491.
7. Dubsky M, Jirkovska A, Bem R, Fejfarova V, Skibova J, Schaper NC, Lipsky BA. Risk factors for recurrence of diabetic foot ulcers: prospective follow-up analysis of a Eurodiale subgroup. Int Wound J 2012; doi: 10.1111/j.1742-481X.2012.01022.x.
8. Monteiro-Soares M, Boyko E, Ribeiro J, Ribeiro I, Dinis-Ribeiro M. Predictive factors for dia-betic foot ulceration: a systematic review. Diabetes Metab Res Rev 2012; 28:574-600. 9. Pound N, Chipchase S, Treece K, Game F, Jeffcoate W. Ulcer-free survival following
manage-ment of foot ulcers in diabetes. Diabet Med 2005; 22: 1306-1309.
10. Frykberg RG, Lavery LA, Pham H, Harvey C, Harkless L, Veves A. Role of neuropathy and high foot pressures in diabetic foot ulceration. Diabetes Care 1998; 21: 1714-1719.
11. Lavery LA, Armstrong DG, Wunderlich RP, Tredwell J, Boulton AJ. Predictive value of foot pressure assessment as part of a population-based diabetes disease management program. Diabetes Care 2003; 26: 1069-1073.
12. Veves A, Murray HJ, Young MJ, Boulton AJ. The risk of foot ulceration in diabetic patients with high foot pressure: a prospective study. Diabetologia 1992; 35: 660-663.
13. Armstrong DG, Peters EJ, Athanasiou KA, Lavery LA. Is there a critical level of plantar foot pressure to identify patients at risk for neuropathic foot ulceration? J Foot Ankle Surg 1998; 37: 303-307.
14. Reiber GE, Smith DG, Wallace C, Sullivan K, Hayes S, Vath C, Maciejewski ML, Yu O, Heagerty PJ, Lemaster J. Effect of therapeutic footwear on foot reulceration in patients with diabetes: a randomized controlled trial. JAMA 2002; 287: 2552-2558.
15. Uccioli L, Faglia E, Monticone G, Favales F, Durola L, Aldeghi A, Quarantiello A, Calia P, Men-zinger G. Manufactured shoes in the prevention of diabetic foot ulcers. Diabetes Care 1995; 18: 1376-1378.
16. Bus SA, Valk GD, van Deursen RW, Armstrong DG, Caravaggi C, Hlavacek P, Bakker K, Cava-nagh PR. The effectiveness of footwear and offloading interventions to prevent and heal foot
7
ulcers and reduce plantar pressure in diabetes: a systematic review. Diabetes Metab Res Rev 2008; 24 Suppl 1: S162-S180.
17. Lemaster JW, Reiber GE, Smith DG, Heagerty PJ, Wallace C. Daily weight-bearing activity does not increase the risk of diabetic foot ulcers. Med Sci Sports Exerc 2003; 35: 1093-1099. 18. Maluf KS, Mueller MJ. Novel Award 2002. Comparison of physical activity and cumulative
plantar tissue stress among subjects with and without diabetes mellitus and a history of recurrent plantar ulcers. Clin Biomech 2003; 18: 567-575.
19. Bus SA. Priorities in offloading the diabetic foot. Diabetes Metab Res Rev 2012; 28 Suppl 1: 54-59.
20. Schaper NC. Diabetic foot ulcer classification system for research purposes: a progress re-port on criteria for including patients in research studies. Diabetes Metab Res Rev 2004; 20 Suppl 1: S90-S95.
21. Bakker K, Apelqvist J, Schaper NC. Practical guidelines on the management and prevention of the diabetic foot 2011. Diabetes Metab Res Rev 2012; 28 Suppl 1: 225-231.
22. Boulton AJ, Hardisty CA, Betts RP, Franks CI, Worth RC, Ward JD, Duckworth T. Dynamic foot pressure and other studies as diagnostic and management aids in diabetic neuropathy. Dia-betes Care 1983; 6: 26-33.
23. Bus SA, de Lange A. A comparison of the 1-step, 2-step, and 3-step protocols for obtaining barefoot plantar pressure data in the diabetic neuropathic foot. Clin Biomech 2005; 20: 892-899.
24. Arts ML, Bus SA. Twelve steps per foot are recommended for valid and reliable in-shoe plan-tar pressure data in neuropathic diabetic patients wearing custom made footwear. Clin Bio-mech 2011; 26: 880-884.
25. Bus SA, Waaijman R, Nollet F. New monitoring technology to objectively assess adherence to prescribed footwear and assistive devices during ambulatory activity. Arch Phys Med Reha-bil 2012; 93: 2075-2079.
26. Owings TM, Apelqvist J, Stenstrom A, Becker M, Bus SA, Kalpen A, Ulbrecht JS, Cavanagh PR. Plantar pressures in diabetic patients with foot ulcers which have remained healed. Diabet Med 2009; 26: 1141-1146.
27. Armstrong DG, Lavery LA, Holtz-Neiderer K, Mohler MJ, Wendel CS, Nixon BP, Boulton AJ. Variability in activity may precede diabetic foot ulceration. Diabetes Care 2004; 27: 1980-1984.
28. Twisk JWR. Applied Multilevel Analysis. A Practical Guide. New York: Cambridge University Press, 2006.
29. Chantelau E, Haage P. An audit of cushioned diabetic footwear: relation to patient compli-ance. Diabet Med 1994; 11: 114-116.
