Custom-made footwear in diabetes: Offloading, usability and ulcer recurrence Arts, Marc

Transcription

1 UvA-DARE (Digital Academic Repository) Custom-made footwear in diabetes: Offloading, usability and ulcer recurrence Arts, Marc Link to publication Citation for published version (APA): Arts, M. L. J. (2013). Custom-made footwear in diabetes: Offloading, usability and ulcer recurrence 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: 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. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 14 Jan 2019

2 UITNODIGING MLJ Arts 2013 CUSTOM-MADE FOOTWEAR IN DIABETES Offloading, usability and ulcer recurrence Marc LJ Arts Voor het bijwonen van de openbare verdediging van het proefschrift CUSTOM-MADE FOOTWEAR IN DIABETES Offloading, usability and ulcer recurrence Door Marc LJ Arts Op woensdag 11 december 2013 om 11:00 In de Aula der Universiteit Lutherse kerk Singel 411 hoek Spui 1012 XM Amsterdam Custom-made footwear in diabetes Receptie na afloop PARANIMFEN Marieke van Venrooij-Arts Robin Coolen MLJ Arts Ceramplein BS Amsterdam

3 Custom-made footwear in diabetes Offloading, usability and ulcer recurrence Marc Leonardus Johannus Arts

4 Arts, Marc LJ Custom-made footwear in diabetes: Offloading, usability and ulcer recurrence Doctoral thesis, University of Amsterdam ISBN/EAN: Copyright 2013 Marc Leonardus Johannes Arts, Amsterdam, The Netherlands All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any form or by any means, without prior permission of the author, or, when applicable, of the publishers of the scientific papers Cover design: JJ Vaessen / MLJ Arts Photography (cover): LMH Vaessen Layout: MLJ Arts / RLH Coolen Printed by Gildeprint Drukkerijen, Enschede

5 Custom-made footwear in diabetes Offloading, usability and ulcer recurrence Academisch proefschrift ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Aula der Universiteit op woensdag 11 December 2013, te 11:00 uur door Marc Leonardus Johannes Arts geboren te Berghem

6 Promotiecommissie Promotor prof. dr. F. Nollet Co-promotores dr. S.A. Bus dr. M. de Haart Overige leden prof. dr. J.B.L. Hoekstra prof. dr. D.A. Legemate prof. dr. ir. J. Harlaar prof. dr. K. Postema dr. H.H.C.M. Savelberg dr. E.J.G. Peters Faculteit der Geneeskunde

7 The DIAbetic Foot Orthopedic Shoe (DIAFOS) trial was financially supported by: The Dutch Foundation for the Development of Orthopedic Footwear Technology (OFOM) The Dutch Diabetes Research Foundation (Diabetes Fonds) The Dutch Organization for Health Research and Development (ZonMW) Department of Rehabilitation, Academic Medical Center Amsterdam Printing of this thesis was financially supported by: Van der Burgh medical supplies BV Buchrnhornen Novel GmbH Roessingh Revalidatie Techniek OIM Orthopedie Livit Orthopedie Ben at Work music production Academic Medical Center George In der Maur orthopedische schoentechniek Anna Fonds Penders Voetzorg

8

9 CONTENTS Chapter 1 General introduction 9 Chapter 2 Twelve steps per foot are recommended for valid and reliable in-shoe plantar pressure data in neuropathic diabetic patients wearing custom-made footwear 27 Chapter 3 Offloading effect of therapeutic footwear in patients with diabetic neuropathy at high risk for plantar foot ulceration 43 Chapter 4 Pressure-reduction and preservation in custom-made footwear of patients with diabetes and a history of plantar ulceration 59 Chapter 5 Data-driven directions for effective custom-made footwear provision in diabetic patients 77 Chapter 6 Perceived usability and use of custom-made footwear in diabetic patients who are at high risk for foot ulceration 97 Chapter 7 Effect of custom-made footwear on foot ulcer recurrence in diabetes: a multicenter randomized trial 115 Chapter 8 General discussion 135 Summary / Samenvatting 153 Dankwoord 161 List of publications 169 Curriculum Vitae / Portfolio 173

10

11 Chapter 1 General introduction 9

12 Chapter 1 THE DIABETIC FOOT Epidemiology and complications Diabetes is one of the most common chronic diseases. In 2012, approximately 370 million people worldwide suffered from diabetes, representing approximately 8.3% of the world s adult population 1. Because of the strong increase in the percentage of elderly people and people with obesity, this number is expected to rise above 552 million in the next 20 years 2. In the Netherlands, people had diabetes in 2011, a number that is expected to increase to 1.1 million in The economic burden of diabetes is substantial. Over 814 million euro was spent on diabetes-related health care in the Netherlands in Costs will further increase markedly in future years 3. A prevalent and debilitating complication of diabetes is foot ulceration, as shown in Figure 1.1a. The lifetime risk for ulceration in diabetic patients is 15-25% 4. Diabetic foot ulcers are a major cause of morbidity and are the most frequent reason for hospitalization. They significantly increase the risk for infection and amputation 5-8. Globally, every 30 seconds a limb is lost due to diabetic foot disease 9. Eighty-five percent of these amputations are preceded by a foot ulcer 10, 11. Once a patient has had a foot ulcer, the risk for recurrence is substantially increased (relative risk 1.6 to 5.3) Prevention and appropriate management of foot ulcers are therefore of paramount importance to decrease both the patient and economic burden of diabetic foot disease Etiology of diabetic foot ulcers Diabetic foot ulcers result from the simultaneous action of multiple contributing causes 15. Often reported underlying pathologic conditions such as vascular dysfunction and peripheral nerve damage contribute to the development foot ulcers 16. Peripheral vascular disease is present in about 50% of diabetic patients with a foot ulcer 17. In combination with minor trauma, peripheral vascular disease leads to tissue breakdown and impairs healing of foot ulcers because blood perfusion of the foot is affected 18. The key factor in diabetic foot ulceration is peripheral neuropathy, which is present in approximately 30% - 50% of all diabetic patients 19, 20. As a result of neuropathy, patients have loss of sensory, motor, and/ or autonomic nerve function. Sensory nerve dysfunction impairs protective sensation in the foot, and hampers the patient s ability to perceive increased mechanical load, friction, pain or heat. Skin damage may occur and remain unnoticed 21. Loss of motor nerve function can result in muscle atrophy in the foot muscles and subsequent changes in the shape of the foot 22, 23. Finally, a dry skin and hyperkeratosis, which are common pre-signs for diabetic foot ulcers, and also edema, leading to a poor fit of shoes, may be a consequence of autonomic nerve damage 16, 18. Overall, peripheral neuropathy greatly increases the risk for foot ulceration in diabetic patients (relative risk 2.2)

13 General introduction 1 a b Figure 1.1 (a) A plantar foot ulcer located at the plantar site of the second metatarsal head of the left foot. (b) A peak plantar pressure (barefoot) plot in which the high (pink colored) pressure peak corresponds with the ulcer location seen in the photograph in figure 1a. PLANTAR PRESSURE AND FOOT ULCERATION In the presence of neuropathy, elevated plantar pressure is considered one of the most important risk factors of foot ulceration in diabetic patients Figure 1.1b shows excessively high peak pressure at the second metatarsal head in a patient walking barefoot. The location of the high-pressure peak corresponds with the location of the ulcer shown in Figure 1.1a. Repetitive exposure to high plantar pressure under bony prominences during unprotected ambulatory activity may lead to local skin breakdown 29. The association between elevated plantar pressure during barefoot walking and ulcer development has been confirmed by several authors, but there is no consensus on the exact pressure threshold for tissue breakdown This may be explained by the fact that although the likelihood of ulceration increases with increased plantar pressure, factors such as severity of neuropathy, level of foot deformity and limited joint mobility contribute to ulceration, which complicates ulcer prediction on an individual level. Furthermore, barefoot pressures do not represent the true mechanical load exposed to the foot, as patients often wear offloading footwear. The addition of in-shoe plantar pressures may therefore better indicate the actual load on the foot in these patients. Although a pressure level that reflects an individual peak pressure threshold for ulcer development has proven elusive to determine to date, a level <200kPa was found to be indicative for ulcer-free survival in diabetic patients who have had a previous foot ulcer 32. Until a better individual threshold is determined, achieving a in-shoe peak pressure level below 200kPa may be an effective target in footwear prescription. 11

14 Chapter 1 Elevated plantar pressures are strongly associated with the presence of foot deformity and a history of amputation Hammer toes or claw toes are among the most prevalent deformities in patients with diabetes (32% to 46%) 37, 38. Bus and co-workers showed that claw and hammer toe deformity is associated with a distal displacement and subsequent thinning of the sub-metatarsal head fat pads 23, which subsequently leads to increased plantar pressure under the metatarsal heads increasing the risk for ulceration 39. Other prevalent types of foot deformity in patients with diabetes are hallux valgus (33%) and limited joint mobility of the first metatarso-phalangeal joint (23% to 35%), both leading to local bony prominences and associated with increased plantar peak pressure under the hallux and the first metatarsal head 33, 37, 40, 41. Although Charcot foot deformity is not very prevalent, it is considered to be the most severe structural deformity of the diabetic foot 33, 38. A Charcot foot is a progressive condition characterized by pathological fractures, dislocation of joints, and often debilitating deformity 42, leading to increased plantar foot pressures mainly in the midfoot area 43. Finally, plantar pressure is markedly increased in patients with partial foot amputations, possibly due to biomechanical compensation or high prevalence of deformity and limited joint mobility in these patients 44. Thus, diabetic patients with neuropathy, foot deformity and/or partial foot amputation who show elevated plantar pressures are at high risk for the development of plantar foot ulcers. This statement is supported by the fact that inappropriate footwear (cramped toe box, inadequate offloading), has been reported to be an important factor in the development of foot ulcers in diabetic patients, as various authors have identified footwear as the root cause of 21 to 76% of ulcers and/or amputations Thus, these patients require adequate footwear that distributes these pressures across the foot to reduce pressure at at-risk foot regions. Offloading with custom-made therapeutic footwear Custom-made footwear Custom-made therapeutic footwear is recommended and often prescribed to high-risk diabetic patients to prevent ulcer development 21. For this, the footwear should reduce and redistribute mechanical pressure 49. To relieve mechanical pressure (also called offloading ) at high-pressure locations, custom insoles aim to redistribute the load from high-pressure locations to adjacent foot regions 50. In the Netherlands, two types of custom-made footwear are generally prescribed. These are semi-customized shoes (i.e. custom-made insoles worn in extra-depth off-the-shelf shoes) and fully customized shoes (i.e. custom-made insoles worn in custom-made shoes). In several studies, custom-made total-contact (i.e. accommodating) insoles have been found to 12

15 General introduction 1 be effective in the reduction of peak plantar pressure (range 16% - 32%), compared to control conditions Additional relief can be achieved by further modifying and individualizing the insole, e.g. with adding a metatarsal pad or bar (16% to 39%) 51, Substantial offloading effects have also been reported for the use of a rocker or roller configuration of the shoe outsole However, despite the positive effect on foot pressure of these custom interventions, large individual differences in offloading effect are common and many uncertainties exist on the effective application of footwear design elements in individual patients 52, 57, 63, 68. This concerns, for example, the optimal pivot point location and toe spring angle 67, 69. A more structured and evidence-based approach to footwear prescription and objective evaluation of the offloading success would probably reduce this variability in offloading effect and will assist prescribers to design properly offloaded footwear for diabetic patients 52, 57, 68. Effectiveness Despite widespread prescription of custom-made therapeutic footwear, evidence for its effectiveness to prevent ulcer recurrence is still meagre and conflicting results are present in the literature 70. In a prospective cohort study of Busch and Chantelau 71, and a randomized controlled trial of Uccioli et al. 72, reduced ulceration rates were shown in patients who wore custom-made therapeutic footwear compared to patients who wore their own (standard) footwear. Reiber et al., however, showed no significant benefit in ulcer recurrence prevention of custom-made footwear compared to off-the-shelf footwear in a large randomized controlled trial 73. The differences in these outcomes are difficult to explain, but may be because of a wide variety of methods and materials used in footwear prescription. None of the studies indicated whether their intervention footwear was effective in reducing pressure compared to the control footwear, nor was any accurate estimate of the patient s adherence to their prescribed footwear provided. In view of this, three systematic reviews confirm that there is still limited evidence to support the use of custom-made footwear in the prevention of foot ulcers and pointed out the need for well-designed prospective trials in which adequate offloading of footwear is assured 50, 70, 74. Plantar pressure assessment To design adequate offloading footwear that can prevent ulcer occurrence, high-pressure regions need to be identified, and pressure in these regions needs to be effectively reduced. Based on clinical evaluation, however, identification of such foot regions seems difficult 75. In clinical practice in the Netherlands, foot pressure prints on carbon paper sheets are commonly used to locate plantar foot regions with excessively high pressures. Although with this method pressure areas can be localized, the major disadvantage is that the load is not quantified. In the last 2 decades, more advanced equipment has been developed to assess dynamic pressures while walking barefoot and in shoes. With such equipment, high-pressure 13

16 Chapter 1 spots on the plantar site of the foot can be accurately identified and quantified. Additionally, offloading efficacy of custom-made therapeutic shoes can be evaluated. Although these systems are used more frequently nowadays in research and clinical practice, specific guidelines on how to obtain valid and reliable data are not widely available. Bus et al. 76 reported that at least 3 walking trials with a two-step approach to a pressure platform is most reliable for obtaining barefoot plantar pressures in neuropathic diabetic patients. Such recommendations are not available for in-shoe pressure measurements in these patients. Kernozek et al. 77 showed that a minimum 8 steps per foot were required for reliable in-shoe pressures data, but they tested healthy subjects walking in standard footwear on a treadmill at controlled speeds. A guideline on the required number of steps for reliable and valid in-shoe pressure data in neuropathic diabetic patients with foot deformity wearing custom-made footwear and walking over ground is needed to avoid collecting too many steps that may increase patient burden or too few steps that may compromise data quality. Offloading improvement Custom-made footwear design and evaluation of its effectiveness is generally based on a trial-and-error approach, meaning that ineffective footwear is primarily indicated when an ulcer recurs. Due to the lack of a structured conceptual approach in footwear prescription and evaluation as described above 52, 57, 68, offloading success is mostly dependent on the clinical experience and skills of the prescribing physician and shoe technician. The result of this trial-and-error approach is that offloading efficacy varies considerably between patients 57, 66, 78. Although methods for a more quantitative approach to footwear design are available, such as by using 3D foot scanning techniques, computer assisted design and manufacturing systems and foot pressure systems, these technologies are still not frequently used in clinical practice (situation in 2008) 49. Offloading success may be improved and intersubject variation may be reduced by the use of these technologies, which may potentially reduce the risk for foot ulceration 79. Moreover, the use of in-shoe plantar pressure assessment has been proven a valuable tool to evaluate the offloading efficacy of custom-made footwear and to guide modifications to the footwear that can reduce pressure at high-risk regions 49, 50. Bus et al. 80 showed this approach to be successful by improving offloading compared to baseline with on average 30% in a relatively small group of diabetic patients wearing custom-made footwear. Confirmation of these promising data is needed in a larger more homogeneous group of at-risk diabetic patients. Use and usability Besides that footwear should adequately offload at-risk foot regions to effectively contribute to ulcer prevention, custom-made footwear needs to be worn. Chantelau and col- 14

17 General introduction 1 leagues showed that, when protective footwear is worn for at least 60% of daytime, the risk for foot ulcer recurrence can be significantly reduced in diabetic patients 81. It is therefore worrying that several studies showed that therapeutic footwear is not often worn in this patient population: only 22 to 36% of patients use their footwear frequently The risk for ulcer development is therefore markedly elevated in these patients. Several authors reported the influence of perceived usability aspects of footwear on its actual use, including weight, appearance and comfort of the shoes Also the perceived benefit of using therapeutic footwear is an important factor in a patient s decision to wear the shoes 89. How diabetic patients with a high risk of ulceration, for whom wearing their prescription footwear is most important, perceive the usability of their footwear is not known, nor is the influence of this perception on footwear use. Insight in the perceived usability will contribute to the understanding and eventually improvement of patient s behaviour in wearing therapeutic footwear. The Diabetic Foot OrthopAedic Shoe (DIAFOS) trial In an ideal approach to footwear prescription and evaluation, footwear is objectively evaluated, for example by using in-shoe pressure analysis, and, if needed, modified to improve offloading 49. Furthermore, in such an approach the footwear should be monitored over time as used materials may be subject to wear and tear or foot structure may change over time 90. The effectiveness of such an approach in the prevention of foot ulceration has not been assessed before. The Department of Rehabilitation Medicine of the Academic Medical Center in Amsterdam collaborated with nine other hospital-based multidisciplinary diabetic foot clinics and nine orthopedic footwear companies across the Netherlands in a multicenter randomized controlled trial on custom-made footwear effectiveness called the DIAFOS trial (Diabetic Foot Orthopedic Shoe trial). The data on which the current thesis is based mainly originate from this trial. In the DIAFOS trial, neuropathic diabetic patients with a recently healed plantar foot ulcer were randomly assigned to an intervention group that had their custom-made footwear evaluated using in-shoe plantar pressure measurement and, if needed, modified based on this evaluation with the goal to improve offloading, or a control group that had their custom-made footwear evaluated according to usual practice in witch in-shoe plantar pressure measurements are not used. All patients in the trial were followed for 18 months or until foot ulcer recurrence. The primary outcome was the percentage of patients with a recurrent plantar foot ulcer in 18 months. Secondary outcomes of the trial that have been the subject of separate studies described in this thesis were (a) the offloading efficacy of the delivered footwear, (b) the pressure relieving effect and preservation of different types of footwear modification, and (c) the use and perceived usability of custommade footwear. 15

18 Chapter 1 Aims OF THIS THESIS The general aim of this thesis was to expand the knowledge on the effectiveness of custommade footwear to prevent plantar foot ulcer recurrence in diabetic patients, by applying objective measurements of offloading. Specific aims were: 1. To determine the minimum number of collected steps that is required to obtain representative in-shoe plantar pressure data in diabetic patients with neuropathy. 2. To assess the offloading efficacy of custom-made therapeutic footwear in high-risk diabetic patients. 3. To assess the value of using in-shoe plantar pressure assessment for evaluating, improving and preserving the offloading properties of newly prescribed custom-made footwear. 4. To assess the effect of footwear modifications on in-shoe plantar pressure in custom-made footwear for high-risk diabetic patients. 5. To evaluate the perceived usability of custom-made therapeutic footwear among high-risk diabetic patients and to assess how it is related to footwear use. 6. To assess whether offloading-improved custom-made footwear reduces ulcer recurrence in diabetic patients with neuropathy. Outline of the thesis Chapter two describes a study in which the number of footsteps required to obtain valid and reliable in-shoe plantar pressure data in diabetic patients with neuropathy wearing custom-made footwear is determined. In-shoe plantar pressure analyses were used in chapter 3 to evaluate to what extent offloading is adequate in the custom-made footwear worn by diabetic patients with neuropathy and a history of plantar foot ulceration. The contributing role of different types of foot deformity to footwear efficacy was specified in this study. Chapter 4 presents a study which aimed (1) to assess the value of using in-shoe plantar pressure analysis for improving the offloading properties of custom-made footwear and (2) to 16

19 General introduction 1 evaluate whether improved offloading results can be maintained over a 1-year period when compared with a control group. Where chapter 4 focused on if offloading capacity can be improved, chapter 5 evaluated how this can be done. In this chapter the offloading efficacy of the many footwear modifications performed in the DIAFOS trial were studied. These data were summarized in an offloading effect-matrix to assist footwear prescribers in the design and modification of adequately offloading footwear for high-risk diabetic patients. How diabetic patients with neuropathy, foot deformity and previous foot ulcers perceive the usability of their footwear and whether there are parameters associated with footwear use was shown in chapter 6. Chapter 7 describes the results of the DIAFOS randomized controlled trial in which we examined the effectiveness of offloading-improved custom-made footwear on plantar foot ulcer recurrence in comparison with usual care (i.e. non-improved custom-made footwear). In addition, we evaluated whether adherence to wearing custom-made footwear influenced ulcer recurrence. Finally, in chapter 8, the main findings of the studies in this thesis are discussed, the used methodology in the different studies is critically reviewed and implications for clinical practice and future research are presented. 17

20 Chapter 1 REFERENCES Whiting DR, Guariguata L, Weil C, Shaw J. IDF diabetes atlas: global estimates of the prevalence of diabetes for 2011 and Diabetes Res Clin Pract 2011;94(3): Bakker K, Schaper NC. The development of global consensus guidelines on the management and prevention of the diabetic foot Diabetes Metab Res Rev 2012;28 Suppl 1: Baan CA, van Baal PH, Jacobs-van der Bruggen MA et al. [Diabetes mellitus in the Netherlands: estimate of the current disease burden and prognosis for 2025]. Ned Tijdschr Geneeskd 2009;153(22): Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers in patients with diabetes. JAMA 2005;293(2): Apelqvist J, Ragnarson-Tennvall G, Larsson J, Persson U. Long-term costs for foot ulcers in diabetic patients in a multidisciplinary setting. Foot Ankle Int 1995;16(7): Boulton AJ, Kirsner RS, Vileikyte L. Clinical practice. Neuropathic diabetic foot ulcers. N Engl J Med 2004;351(1): Matricali GA, Dereymaeker G, Muls E, Flour M, Mathieu C. Economic aspects of diabetic foot care in a multidisciplinary setting: a review. Diabetes Metab Res Rev 2007;23(5): Reiber GE, Vileikyte L, Boyko EJ et al. Causal pathways for incident lower-extremity ulcers in patients with diabetes from two settings. Diabetes Care 1999;22(1): Boulton AJ, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J. The global burden of diabetic foot disease. Lancet 2005;366(9498): Pecoraro RE, Reiber GE, Burgess EM. Pathways to diabetic limb amputation. Basis for prevention. Diabetes Care 1990;13(5): Reiber GE. The epidemiology of diabetic foot problems. Diabet Med 1996;13 Suppl 1:S Abbott CA, Carrington AL, Ashe H et al. 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(5): 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(7):

21 General introduction Boyko EJ, Ahroni JH, Cohen V, Nelson KM, Heagerty PJ. Prediction of diabetic foot ulcer occurrence using commonly available clinical information: the Seattle Diabetic Foot Study. Diabetes Care 2006;29(6): Armstrong DG, Lavery LA. Diabetic foot ulcers: prevention, diagnosis and classification. Am Fam Physician 1998;57(6): Bowering CK. Diabetic foot ulcers. Pathophysiology, assessment, and therapy. Can Fam Physician 2001;47: Prompers L, Huijberts M, Apelqvist J et al. High prevalence of ischaemia, infection and serious comorbidity in patients with diabetic foot disease in Europe. Baseline results from the Eurodiale study. Diabetologia 2007;50(1): Boulton AJ. The pathogenesis of diabetic foot problems: an overview. Diabet Med 1996;13 Suppl 1:S12-S Adler AI, Boyko EJ, Ahroni JH, Stensel V, Forsberg RC, Smith DG. Risk factors for diabetic peripheral sensory neuropathy. Results of the Seattle Prospective Diabetic Foot Study. Diabetes Care 1997;20(7): Harris M, Eastman R, Cowie C. Symptoms of sensory neuropathy in adults with NIDDM in the U.S. population. Diabetes Care 1993;16(11): Apelqvist J, Bakker K, van Houtum WH, Schaper NC. Practical guidelines on the management and prevention of the diabetic foot: based upon the International Consensus on the Diabetic Foot (2007) Prepared by the International Working Group on the Diabetic Foot. Diabetes Metab Res Rev 2008;24 Suppl 1:S181-S Bus SA, Yang QX, Wang JH, Smith MB, Wunderlich R, Cavanagh PR. Intrinsic muscle atrophy and toe deformity in the diabetic neuropathic foot: a magnetic resonance imaging study. Diabetes Care 2002;25(8): Bus SA, Maas M, Cavanagh PR, Michels RP, Levi M. Plantar fat-pad displacement in neuropathic diabetic patients with toe deformity: a magnetic resonance imaging study. Diabetes Care 2004;27(10): 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(7): 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(7): Caselli A, Pham H, Giurini JM, Armstrong DG, Veves A. The forefoot-to-rearfoot plantar pressure ratio is increased in severe diabetic neuropathy and can predict foot ulceration. Diabetes Care 2002;25(6):

22 Chapter 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(10): 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(5): 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(4): Boulton AJ, Hardisty CA, Betts RP et al. Dynamic foot pressure and other studies as diagnostic and management aids in diabetic neuropathy. Diabetes Care 1983;6(1): 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(4): Owings TM, Apelqvist J, Stenstrom A et al. Plantar pressures in diabetic patients with foot ulcers which have remained healed. Diabet Med 2009;26(11): Ahroni JH, Boyko EJ, Forsberg RC. Clinical correlates of plantar pressure among diabetic veterans. Diabetes Care 1999;22(6): Lavery LA, Armstrong DG, Vela SA, Quebedeaux TL, Fleischli JG. Practical criteria for screening patients at high risk for diabetic foot ulceration. Arch Intern Med 1998;158(2): Armstrong DG, Lavery LA. Plantar pressures are higher in diabetic patients following partial foot amputation. Ostomy Wound Manage 1998;44(3): Cowley MS, Boyko EJ, Shofer JB, Ahroni JH, Ledoux WR. Foot ulcer risk and location in relation to prospective clinical assessment of foot shape and mobility among persons with diabetes. Diabetes Res Clin Pract 2008;82(2): Holewski JJ, Moss KM, Stess RM, Graf PM, Grunfeld C. Prevalence of foot pathology and lower extremity complications in a diabetic outpatient clinic. J Rehabil Res Dev 1989;26(3): Smith DG, Barnes BC, Sands AK, Boyko EJ, Ahroni JH. Prevalence of radiographic foot abnormalities in patients with diabetes. Foot Ankle Int 1997;18(6): Bus SA, Maas M, de LA, Michels RP, Levi M. Elevated plantar pressures in neuropathic diabetic patients with claw/hammer toe deformity. J Biomech 2005;38(9):

23 General introduction Ledoux WR, Shofer JB, Smith DG et al. Relationship between foot type, foot deformity, and ulcer occurrence in the high-risk diabetic foot. J Rehabil Res Dev 2005;42(5): Lindsay JR, Kennedy L, Atkinson AB et al. Reduced prevalence of limited joint mobility in type 1 diabetes in a U.K. clinic population over a 20-year period. Diabetes Care 2005;28(3): Armstrong DG, Todd WF, Lavery LA, Harkless LB, Bushman TR. The natural history of acute Charcot's arthropathy in a diabetic foot specialty clinic. Diabet Med 1997;14(5): Armstrong DG, Lavery LA. Elevated peak plantar pressures in patients who have Charcot arthropathy. J Bone Joint Surg Am 1998;80(3): Armstrong DG, Lavery LA. Plantar pressures are higher in diabetic patients following partial foot amputation. Ostomy Wound Manage 1998;44(3):30-2, 34, Apelqvist J, Larsson J, Agardh CD. The influence of external precipitating factors and peripheral neuropathy on the development and outcome of diabetic foot ulcers. J Diabet Complications 1990;4(1): Edmonds ME, Blundell MP, Morris ME, Thomas EM, Cotton LT, Watkins PJ. Improved survival of the diabetic foot: the role of a specialized foot clinic. Q J Med 1986;60(232): Macfarlane RM, Jeffcoate WJ. Factors contributing to the presentation of diabetic foot ulcers. Diabet Med 1997;14(10): Reiber GE. Who is at risk of limb loss and what to do about it? J Rehabil Res Dev 1994;31(4): Cavanagh PR. Therapeutic footwear for people with diabetes. Diabetes Metab Res Rev 2004;20 Suppl 1:S51-S Bus SA, Valk GD, van Deursen RW et al. The effectiveness of footwear and offloading interventions to prevent and heal foot ulcers and reduce plantar pressure in diabetes: a systematic review. Diabetes Metab Res Rev 2008;24 Suppl 1:S162-S Bus SA, Ulbrecht JS, Cavanagh PR. Pressure relief and load redistribution by custommade insoles in diabetic patients with neuropathy and foot deformity. Clin Biomech (Bristol, Avon) 2004;19(6): Lord M, Hosein R. Pressure redistribution by molded inserts in diabetic footwear: a pilot study. J Rehabil Res Dev 1994;31(3):

24 Chapter Lobmann R, Kayser R, Kasten G et al. Effects of preventative footwear on foot pressure as determined by pedobarography in diabetic patients: a prospective study. Diabet Med 2001;18(4): Postema K, Burm PE, Zande ME, Limbeek J. Primary metatarsalgia: the influence of a custom moulded insole and a rockerbar on plantar pressure. Prosthet Orthot Int 1998;22(1): Lott DJ, Hastings MK, Commean PK, Smith KE, Mueller MJ. Effect of footwear and orthotic devices on stress reduction and soft tissue strain of the neuropathic foot. Clin Biomech (Bristol, Avon ) 2007;22(3): Owings TM, Woerner JL, Frampton JD, Cavanagh PR, Botek G. Custom therapeutic insoles based on both foot shape and plantar pressure measurement provide enhanced pressure relief. Diabetes Care 2008;31(5): Raspovic A, Newcombe L, Lloyd J, Dalton E. Effect of customized insoles on vertical plantar pressures in sites of previous neuropathic ulceration in the diabetic foot. Foot 2000;10(3): Tsung BY, Zhang M, Mak AF, Wong MW. Effectiveness of insoles on plantar pressure redistribution. J Rehabil Res Dev 2004;41(6A): Viswanathan V, Madhavan S, Gnanasundaram S et al. Effectiveness of different types of footwear insoles for the diabetic neuropathic foot: a follow-up study. Diabetes Care 2004;27(2): Mueller MJ, Lott DJ, Hastings MK, Commean PK, Smith KE, Pilgram TK. Efficacy and mechanism of orthotic devices to unload metatarsal heads in people with diabetes and a history of plantar ulcers. Phys Ther 2006;86(6): Lott DJ, Hastings MK, Commean PK, Smith KE, Mueller MJ. Effect of footwear and orthotic devices on stress reduction and soft tissue strain of the neuropathic foot. Clin Biomech (Bristol, Avon) 2007;22(3): Guldemond NA, Leffers P, Schaper NC et al. The effects of insole configurations on forefoot plantar pressure and walking convenience in diabetic patients with neuropathic feet. Clin Biomech (Bristol, Avon) 2007;22(1): Hastings MK, Mueller MJ, Pilgram TK, Lott DJ, Commean PK, Johnson JE. Effect of metatarsal pad placement on plantar pressure in people with diabetes mellitus and peripheral neuropathy. Foot Ankle Int 2007;28(1): Hsi WL, Chai HM, Lai JS. Evaluation of rocker sole by pressure-time curves in insensate forefoot during gait. Am J Phys Med Rehabil 2004;83(7):

25 65. Schaff PS, Cavanagh PR. Shoes for the insensitive foot: the effect of a "rocker bottom" shoe modification on plantar pressure distribution. Foot Ankle 1990;11(3): General introduction Mueller MJ, Strube MJ, Allen BT. Therapeutic footwear can reduce plantar pressures in patients with diabetes and transmetatarsal amputation. Diabetes Care 1997;20(4): Praet SF, Louwerens JW. The influence of shoe design on plantar pressures in neuropathic feet. Diabetes Care 2003;26(2): Kato H, Takada T, Kawamura T, Hotta N, Torii S. The reduction and redistribution of plantar pressures using foot orthoses in diabetic patients. Diabetes Res Clin Pract 1996;31(1-3): van SC, Ulbrecht JS, Becker MB, Cavanagh PR. Design criteria for rigid rocker shoes. Foot Ankle Int 2000;21(10): Maciejewski ML, Reiber GE, Smith DG, Wallace C, Hayes S, Boyko EJ. Effectiveness of diabetic therapeutic footwear in preventing reulceration. Diabetes Care 2004;27(7): Busch K, Chantelau E. Effectiveness of a new brand of stock 'diabetic' shoes to protect against diabetic foot ulcer relapse. A prospective cohort study. Diabet Med 2003;20(8): Uccioli L, Faglia E, Monticone G et al. Manufactured shoes in the prevention of diabetic foot ulcers. Diabetes Care 1995;18(10): Reiber GE, Smith DG, Wallace C et al. Effect of therapeutic footwear on foot reulceration in patients with diabetes: a randomized controlled trial. JAMA 2002;287(19): Healy A, Naemi R, Chockalingam N. The effectiveness of footwear as an intervention to prevent or to reduce biomechanical risk factors associated with diabetic foot ulceration: A systematic review. J Diabetes Complications Guldemond NA, Leffers P, Nieman FH, Sanders AP, Schaper NC, Walenkamp GH. Testing the proficiency to distinguish locations with elevated plantar pressure within and between professional groups of foot therapists. BMC Musculoskelet Disord 2006;7: Bus SA, de LA. 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 (Bristol, Avon ) 2005;20(9): Kernozek TW, LaMott EE, Dancisak MJ. Reliability of an in-shoe pressure measurement system during treadmill walking. Foot Ankle Int 1996;17(4):

26 Chapter Guldemond NA, Leffers P, Schaper NC, Sanders AP, Nieman FH, Walenkamp GH. Comparison of foot orthoses made by podiatrists, pedorthists and orthotists regarding plantar pressure reduction in The Netherlands. BMC Musculoskelet Disord 2005;6: Bus SA. Priorities in offloading the diabetic foot. Diabetes Metab Res Rev 2012;28 Suppl 1: Bus SA, Haspels R, Busch-Westbroek TE. Evaluation and optimization of therapeutic footwear for neuropathic diabetic foot patients using in-shoe plantar pressure analysis. Diabetes Care 2011;34(7): Chantelau E, Haage P. An audit of cushioned diabetic footwear: relation to patient compliance. Diabet Med 1994;11(1): Jannink MJ, IJzerman MJ, Groothuis-Oudshoorn K, Stewart RE, Groothoff JW, Lankhorst GJ. Use of orthopedic shoes in patients with degenerative disorders of the foot. Arch Phys Med Rehabil 2005;86(4): van Netten JJ, Jannink MJ, Hijmans JM, Geertzen JH, Postema K. Use and usability of custom-made orthopedic shoes. J Rehabil Res Dev 2010;47(1): Breuer U. Diabetic patient's compliance with bespoke footwear after healing of neuropathic foot ulcers. Diabete Metab 1994;20(4): van Netten JJ, Dijkstra PU, Geertzen JH, Postema K. What influences a patient's decision to use custom-made orthopaedic shoes? BMC Musculoskelet Disord 2012;13(1): Knowles EA, Boulton AJ. Do people with diabetes wear their prescribed footwear? Diabet Med 1996;13(12): Williams AE, Nester CJ. Patient perceptions of stock footwear design features. Prosthet Orthot Int 2006;30(1): Ward AB. Footwear and orthoses for diabetic patients. Diabet Med 1993;10(6): Macfarlane DJ, Jensen JL. Factors in diabetic footwear compliance. J Am Podiatr Med Assoc 2003;93(6): Foto JG, Birke JA. Evaluation of multidensity orthotic materials used in footwear for patients with diabetes. Foot Ankle Int 1998;19(12):

27 General introduction 1 25

28 26

29 Chapter 2 Twelve steps per foot are recommended for valid and reliable in-shoe plantar pressure data in neuropathic diabetic patients wearing custom made footwear Clinical Biomechanics (Bristol, Avon) Oct; 26(8): Reprinted with permission from Elsevier MLJ Arts SA Bus 27

30 Chapter 2 Abstract Background Dynamic in-shoe plantar pressure assessment is used both in research and clinical practice to evaluate therapeutic footwear interventions in neuropathic diabetic patients. The aim was to determine the required number of footsteps for reliable and valid in-shoe plantar pressure data in these patients. Methods In 30 neuropathic diabetic patients wearing custom made therapeutic footwear, in-shoe plantar pressures were measured for a minimum of 20 midgait walking steps per foot. For each incremental number of steps and for each of six anatomical regions per foot, peak pressure, pressure-time integral, contact area, contact time, and force-time integral were calculated. Reliability was assessed by calculating intraclass correlation coefficients. Validity was assessed by calculating the coefficient of variation between each n-step protocol and the 20-step reference protocol based on limits of agreement analysis. Data was considered reliable with intraclass correlation coefficients >0.90 and valid with coefficients of variation <10%. Findings Three steps per foot were required to obtain reliable data for each foot region and parameter. Depending on the parameter, between 7 and 17 steps per foot were required to obtain valid data. With the exception of deviant outcomes in three forefoot regions for force-time integral, overall 12 steps per foot were required for valid data. Interpretation For neuropathic diabetic patients wearing custom-made therapeutic footwear, 12 midgait steps per foot are required to obtain valid and reliable in-shoe plantar pressure data. This provides directions for the use of in-shoe plantar pressure analysis in research and clinical practice in this patient group. 28

31 Number of required steps for representative in-shoe pressure data Introduction In-shoe dynamic plantar foot pressure assessment is used more and more in research and clinical practice to evaluate the efficacy of therapeutic footwear prescriptions in diabetic patients with peripheral neuropathy. Elevated plantar foot pressure is a causative factor of foot ulceration in these patients 1, 2, and foot ulceration is an important precursor of infection and amputation 3. To reduce the risk of ulceration, therapeutic footwear is commonly prescribed. This footwear primarily acts to redistribute plantar pressures on the foot and relieve pressure in regions at risk for ulceration 4, 5. To adequately assess the efficacy of therapeutic footwear interventions, one has to rely on representative and reliable in-shoe plantar pressure data. 2 To obtain such a representative (i.e. valid) and reliable estimate of the true in-shoe plantar pressures, data from multiple footsteps are generally collected, often in multiple walking trials, However, the precise number of collected steps per foot is often not reported and may vary considerably, from as few as three steps 6 to as many as steps 7. Too few collected footsteps may compromise data quality. Too many collected footsteps may fatigue patients, in particular with repeated measurements or multiple conditions tested and will increase data collection and analysis time. Guidelines for the required number of footsteps for valid and reliable in-shoe plantar pressure data in neuropathic diabetic patients currently do not exist. The only study we found on this topic assessed healthy subjects and found that eight steps per foot were required to obtain reliable in-shoe plantar pressure data 8. However, this study assessed only data reliability, not data validity, assessments were at the group level, and subjects were tested in standard footwear on a treadmill at controlled speeds. This limits the extrapolation of these findings to individual diabetic patients of which many have abnormalities in foot structure, gait, or balance 9-11, and who are generally tested in a clinical setting wearing therapeutic footwear and walking overground at comfortable speeds. The above considerations suggest that specific guidelines for the required number of footsteps for valid and reliable in-shoe pressure data in neuropathic diabetic patients are needed to help direct clinical practice and research toward proper use of in-shoe plantar pressure analysis. Therefore, the aim of this study was to determine the number of footsteps required to obtain valid and reliable in-shoe plantar pressure data in neuropathic diabetic patients wearing custom-made footwear. For this analysis, we focused on the most commonly reported pressure parameters in diabetic footwear studies. 29

32 Chapter 2 Methods Subjects Thirty neuropathic diabetic patients (22 males, 8 females, mean (SD) age 58.5 (10.9) years, mean body mass index (BMI) 31.5 (7.5)) who were at risk for plantar foot ulceration participated in the study. Mean time since diabetes onset was 25.2 (16.9) years. Twelve patients had diabetes type 1, 18 patients had diabetes type 2. All patients had peripheral neuropathy, indicated by a loss of protective sensation in the foot through the inability to sense the 10g Semmes-Weinstein monofilament at three plantar foot sites tested (hallux, first metatarsal head, and fifth metatarsal head) 12. Each patient presented with one or more foot deformities which were assessed clinically. These deformities included hammer/claw toes, pes cavus, prominent metatarsal heads, limited joint mobility at the first metatarsal-phalangeal joint, and hallux valgus. Subjects had to be able to walk unaided for a distance of at least 100m. Patients with non-diabetic causes of neurological deficit or lower extremity amputation were excluded. Written informed consent was obtained prior to the start of the study. All procedures were approved by the medical ethics committee of the University of Amsterdam Medical Centre. Instrumentation In-shoe dynamic plantar pressure was measured using the Novel Pedar-X system (Novel GmbH, Munich, Germany). The system comprises flexible 2mm thick insoles with a matrix of 99 capacitance-based sensors each sampling at 50Hz, which were placed in the shoes between the sock and insert. The measurement insoles are attached by leads to a data logger worn by the subject on a waist belt. Data was transmitted through a wireless Bluetooth connection to a laptop computer on which the data was stored. In-shoe plantar pressure was measured within a range of 20 to 600 kpa. Six different pairs of wide Pedar insoles were available to accommodate each foot size. Before data collection, each pair of insoles was calibrated according to the manufacturer s specifications. Footwear Patients wore newly-prescribed therapeutic footwear, which included fully customized footwear (orthopedic footwear), or custom molded inserts worn in extra-depth shoes (semiorthopedic footwear). Procedures Before pressure assessment, a zero-calibration was performed by unloading each measurement insole while the patient wore shoes. In-shoe plantar pressure was assessed while walking in multiple trials along a 12m walkway in a laboratory setting. Prior to the collection 30

33 Number of required steps for representative in-shoe pressure data of data, subjects performed two practice walking trials. Subjects walked at a self-selected comfortable speed, which was controlled during subsequent trials (± 5% variation allowed). Walking speed was measured using a photocell system. It was assured that in-shoe pressures from a minimum of 20 midgait steps per foot were collected. 2 Data analysis Software from Novel was used to analyze the in-shoe pressure data. The first and last footstep of each walking trial was removed automatically by the software to avoid acceleration and deceleration effects on the data. Additionally, footsteps showing sensor errors or major deviances in the ground reaction force curves were removed manually. Masks were used to divide the foot into six anatomical regions per foot: heel, midfoot, 1 st metatarsal, 2 nd to 5 th (lesser) metatarsals, hallux and 2 nd to 5 th (lesser) toes 13. For each incremental number of footsteps and for each foot region, mean values for the 30 subjects for peak pressure, pressure-time integral, contact area, contact time, and force-time integral were calculated. These five parameters were chosen because they are the most commonly reported parameters in pressure studies on diabetic footwear. Peak pressure and pressure-time integral are clinically most relevant, whereas force-time integral is biomechanically relevant in showing the mechanism of action in load (re-)distribution of footwear interventions 14. The first 20 clean midgait steps per foot were selected for data analysis. The condition with 20 footsteps was used as reference step protocol for statistical analysis. Each condition with an incremental number of footsteps, starting with two and finishing with 19, was defined as n-step protocol. The left and right feet were assessed separately. Statistical analysis Statistical analysis was carried out using SPSS statistical software (Version 18.0). To determine the number of footsteps required for reliable in-shoe pressure data, Intraclass Correlation Coefficients (ICC) were calculated per pressure parameter and foot region, for each incremental number of footsteps between two and 19 in comparison with the reference step protocol. An ICC > 0.90 was considered indicative of excellent reliability 15. To determine the number of footsteps required for valid in-shoe pressure data, limits of agreement analysis was performed in order to take into account inter-individual differences in gait variability and pressure recordings 16. First, for each pressure parameter and foot region, mean differences between values of each n-step protocol and the reference step protocol were calculated per subject. Ninety-five percent limits of agreement of these mean differences were then calculated per n-step protocol (Formula 1). Subsequently, a coefficient of variation was calculated between the 95% limits of agreement interval and the mean value for the 20-step reference protocol (Formula 2). Data was considered valid when this coefficient of variation was smaller than 10%. Overall, in-shoe plantar pressure data was considered 31

34 Chapter 2 Formula 1 Formula 2 Formula 1 and 2: LoA = limits of agreement; CoV = coefficient of variation; n = number of incremental footsteps in the calculation; µ = mean value per n footsteps; σ = standard deviation per n footsteps. valid and reliable in this study when for all five pressure parameters and for all foot regions the criteria for valid and reliable data were reached. The presence of a step or trial effect in the data (values increasing or decreasing with an incremental number of footsteps) was tested for each foot region and for each parameter. For this purpose, average values of the 30 subjects for each incremental number of footsteps were plotted. From these plots, linear regression coefficients were calculated and tested for statistical significance (P < 0.05). Additionally, ANOVA repeated measures and post-hoc analysis was used to compare the averaged values between each of four blocks of five subsequent footsteps (P < 0.05). Results Between two and five walking trials on the 12m walkway were required to obtain a minimum number of 20 left and right midgait footsteps in each subject. Six footsteps of six different patients showing sensor errors in the pressure recordings were removed manually. The mean (SD) walking speed of the subjects was 1.1 (0.2) m/s. Mean values and standard deviations for each parameter per foot region based on the 20-step reference protocol are shown in Table 2.1. In the analysis of a step or trial effect, none of the regression coefficients was statistically significant. Additionally, no significant differences were found between the average values for the four blocks of five footsteps, with the exception of contact time measured in the lesser metatarsal region. This means that there was no step or trial effect present in the data. Three steps per foot were required to obtain excellent reliability scores (ICC > 0.90) in each foot region for all pressure parameters (Table 2.2). For assessment of data validity, the coefficients of variation gradually decreased when adding more footsteps to the calculation (Figure 2.1). The number of footsteps required to reach coefficients of variation below 10% per foot region and parameter are shown in Table 2.3. For peak pressure, pressure-time integral, contact area, contact time, and force-time integral, the required number of steps per foot 32

35 Number of required steps for representative in-shoe pressure data Table 2.1. Mean (SD) values per foot region for each of the five parameters based on the 20-step reference protocol. Peak pressure (kpa) Pressuretime integral (kpa s) Contact area (cm 2 ) Contact time (ms) Force-time integral (N s) Left foot Heel 213 (75) 73.8 (31.4) 42.2 (5.7) 652 (116) (74.3) Midfoot 143 (74) 65.7 (26.8) 53.0 (9.4) 720 (70) (63.0) Metatarsal (102) 69.2 (29.3) 17.9 (3.0) 641 (109) 61.3 (30.5) Metatarsals (82) 70.4 (22.4) 40.3 (5.4) 690 (81) (52.5) Hallux 188 (84) 44.6 (7.10) 11.3 (2.6) 541 (77) 26.4 (16.5) Toes (86) 55.2 (19.8) 19.2 (7.1) 652 (84) 32.8 (16.9) 2 Right foot Heel 210 (77) 68.5 (21.4) 42.1 (5.8) 653 (80) (55.3) Midfoot 140 (65) 64.2 (29.2) 54.3 (9.2) 710 (63) (67.4) Metatarsal (70) 64.0 (23.0) 17.2 (2.7) 627 (88) 58.4 (24.1) Metatarsals (60) 65.8 (24.6) 40.8 (6.3) 681 (9) (75.5) Hallux 177 (69) 39.0 (17.5) 10.3 (2.2) 518 (41) 21.8 (11.8) Toes (79) 47.5 (19.2) 19.3 (4.6) 624 (97) 27.7 (13.8) for all foot regions together was 12, 11, 9, 7, and 17, respectively. More footsteps were generally required in the right foot compared to the left foot and in the forefoot compared to the rearfoot. In the right forefoot, a deviant number of 15, 16, and 17 footsteps were required to obtain valid force-time integral data in the lesser toes, hallux, and first metatarsal regions, respectively. The data for these foot regions showed outliers in three out of thirty patients. No specific disease characteristics were present in these patients that may explain these deviant outcomes. The decline of the coefficients of variation in these regions with increasing number of footsteps reached a plateau between 10 and 16 footsteps for force-time integral that was more significant than for the other parameters (Figure 2.1E). As a result, minor changes in the coefficient of variation may give larger changes in the number of required footsteps, which may explain the higher number of footsteps required for force-time integral in these three regions. With the exception of force-time integral values in these three forefoot regions of the right foot, 12 steps per foot were required to obtain valid and reliable data for all pressure parameters in all regions of both feet. 33

36 Chapter 2 Table 2.2. Intraclass correlation coefficients per foot region for each pressure parameter shown for the 2-step and 3-step protocols. Left foot Right foot Peak pressure Heel Midfoot Metatarsal 1 Metatarsals 2-5 Hallux Toes 2-5 Heel Midfoot Metatarsal 1 Metatarsals 2-5 Hallux Toes step protocol step protocol Pressure-time integral 2-step protocol step protocol Contact area 2-step protocol step protocol Contact time 2-step protocol step protocol Force-time integral 2-step protocol step protocol

37 Number of required steps for representative in-shoe pressure data Table 2.3. Number of steps required to obtain coefficients of variation <10% per foot region and parameter. Underlined numbers represent the highest number of required steps per parameter. Left foot Right foot Hallux Toes 2-5 Metatarsals 2-5 Heel Midfoot Metatarsal 1 Hallux Toes 2-5 Metatasals 2-5 Heel Midfoot Metatarsal 1 Peak pressure Pressure-time integral Contact area Contact time Force-time integral

38 Chapter 2 Figure 2.1. Coefficients of variation shown as a function of the number of footsteps. Data are presented as an average (+ 2 SDs) of all 12 foot regions. Curves are shown for (A) peak pressure, (B) pressure-time integral, (C) contact area, (D) contact time, and (E) force-time integral. The dotted vertical lines define the number of steps where the coefficients of variation were smaller than 0.10 (10%) in all 12 regions. 36

39 Number of required steps for representative in-shoe pressure data Discussion The aim of this study was to determine the required number of footsteps for valid and reliable in-shoe plantar pressure data in neuropathic diabetic patients wearing custom-made therapeutic footwear. For this analysis, we chose to select the five most commonly reported parameters in diabetic footwear studies. The results showed that only three steps per foot were required in these patients to obtain highly reliable data. To obtain valid in-shoe pressure data, between 7 and 17 steps per foot were required, depending on the parameter of interest. With the exception of some deviant outcomes for force-time integral, this study showed that 12 steps per foot were required to obtain both valid and reliable in-shoe pressure data in all parameters. Based on these results, we recommend that 12 midgait steps per foot are collected when in-shoe plantar pressure measurements are performed to evaluate custom-made therapeutic footwear in neuropathic diabetic patients who are at risk for ulceration. 2 Within only three steps per foot, ICCs reached levels above 0.90 indicating excellent reliability of the in-shoe plantar pressure data. This is not surprising considering the consistency people tend to show from step-to-step in level walking, resulting in only small differences in pressure and time parameters between subsequent steps. However, ICC calculations have the disadvantage that they are influenced by inter-subject variability, which was large in this study as the standard deviations of the mean outcomes in Table 2.1 show. Furthermore, ICC scores provide information at group level, not at the individual patient level. We aimed to provide recommendations on the use of in-shoe plantar pressure analysis in individual patients, as this seems more relevant for clinical practice. Therefore, a limits of agreement analysis was applied. This analysis showed that for contact time and contact area fewer footsteps were required to obtain valid data than for the pressure parameters, likely because they are not influenced by individual sensor values, giving more consistent step-tostep outcomes. The lower coefficients of variation (SD/mean) for these parameters shown in Table 2.1 also indicate this. Nevertheless, among the selected parameters, peak pressure is the most commonly reported and clinically most relevant parameter in assessment of the diabetic foot. Therefore, the results for peak pressure were leading in the recommendations made. This priority given in the interpretation to peak pressure was also one of the reasons to exclude the deviant results on force-time integral in three right forefoot regions in the recommendations made. Another reason was that the coefficient of variation for force-time integral across the range of footsteps only showed a minimal change (Figure 2.1E), suggesting that the coefficients at 12 footsteps were quite similar to those of the other pressure parameters. This indicates that the deviant results on force-time integral had no significant impact on the overall study findings. Probably for the same reason, differences 37

40 Chapter 2 were found in the required numbers of steps between the left and right feet (Table 2.3). Although these differences were sometimes large, for example for contact area in the hallux region (2 footsteps required for the left foot, 9 for the right foot), the coefficients of variations at these step numbers were not very different between left and right. Moreover, left-right differences for the primary parameter, peak pressure, were small. Therefore, no specific conclusion should be drawn from the discrepancies between the left and right foot. The results indicate that more footsteps are required to obtain reliable and valid in-shoe plantar pressure data in neuropathic diabetic patients than to obtain reliable data in healthy subjects (8 steps per foot) 8. However, the current study showed reliable data after only 3 steps per foot, but valid data after 12 steps per foot. Kernozek et al. did not report on an individual patient-based validity analysis. Furthermore, data collection methods were different between studies: treadmill walking at a controlled speed versus overground walking at a comfortable speed. Finally, neuropathic patients are generally different from healthy subjects in presence of foot deformity, gait and balance abnormalities, and footwear use, which may also affect in-shoe pressure data consistency in a different way 9, 10. Therefore, a valid comparison between these two studies is not possible. Most gait laboratories are confined in space. Therefore, data from multiple walking trials are often collected to obtain a representative number of steps per foot for data analysis. In this case, speed of walking has to be standardised between trials, where some trials may be excluded because speed is outside the set range. Furthermore, data from multiple walking trials needs further processing to obtain an ensemble average for all footsteps. Efficiency of data collection and analysis would improve if only one walking trial containing 12 steps per foot would be collected. This would require a setting for unobtrusive walking along a straight path of about 25 meters, and a mobile and telemetric system that can measure over such a distance. Because a step or trial effect was not shown in this study, such an effect also seems unlikely in a longer walking trial of 25 meters. A few limitations apply to this study. First, the criterion level of 10% for the calculated coefficient of variation and the use of 20 footsteps as reference protocol were arbitrarily chosen. Different choices may have resulted in different outcomes. However, as Figure 2.1 shows for the pressure parameters, the decline in coefficient of variation curves reached some kind of plateau at a level of approximately 10% between 10 and 15 footsteps. Therefore, in hindsight, the 10% criterion level and the 20-step reference protocol seem to be adequately chosen. Secondly, an age-matched healthy control group was not included which limits the assessment of the specific effect of diabetic neuropathy on data reliability and validity. Such an assessment was, however, beyond the scope of this study, in which we primarily aimed to establish recommendations for testing individual neuropathic diabetic patients in a clinical setting. Further research is needed to compare patients with healthy subjects. Finally, the 38

41 Number of required steps for representative in-shoe pressure data outcomes are specific for high-risk diabetic patients being tested in custom-made footwear. Although the data may be externally valid for other subgroups of patients (e.g. non-neuropathic, wearing standard footwear, or with active ulceration offloaded unilaterally), this can not be determined from the current study. Nevertheless, in-shoe plantar pressure analysis seems most relevant for evaluating therapeutic footwear in high-risk neuropathic patients for preventative purposes. 2 Conclusion This study showed that 12 midgait steps per foot were required to obtain reliable and valid in-shoe dynamic plantar pressure data in neuropathic diabetic patients wearing custommade therapeutic footwear. This finding provides directions for the use of in-shoe plantar pressure assessment for clinical practice and research purposes. Furthermore, the results contribute to the standardization of protocols on foot pressure measurements, a topic that has gained recent interest within the International Foot and Ankle Biomechanics community ( Based on the study findings, we recommend that 12 midgait steps per foot are collected when dynamic in-shoe plantar pressures are measured in neuropathic diabetic patients wearing custom-made therapeutic footwear. If space and equipment allows, data is preferably collected within one walking trial to improve efficiency. There may be reasons to collect more than 12 steps per foot, for example to improve statistical power. However, we suggest that this is considered in relation to need, possible patient burden, and efficiency. We further recommend that authors report the (minimal) number of footsteps collected for in-shoe plantar pressure analysis. 39

42 Chapter 2 References 1. 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(10): 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(5): Boulton AJ, Kirsner RS, Vileikyte L. Clinical practice. Neuropathic diabetic foot ulcers. N Engl J Med 2004;351(1): Bus SA, Valk GD, van Deursen RW et al. The effectiveness of footwear and offloading interventions to prevent and heal foot ulcers and reduce plantar pressure in diabetes: a systematic review. Diabetes Metab Res Rev 2008;24 (Suppl 1):S162-S Paton J, Bruce G, Jones R, Stenhouse E. Effectiveness of insoles used for the prevention of ulceration in the neuropathic diabetic foot: a systematic review. J Diabetes Complications 2011;25(1): Garrow AP, van Schie CH, Boulton AJ. Efficacy of multilayered hosiery in reducing in-shoe plantar foot pressure in high-risk patients with diabetes. Diabetes Care 2005;28(8): Owings TM, Woerner JL, Frampton JD, Cavanagh PR, Botek G. Custom therapeutic insoles based on both foot shape and plantar pressure measurement provide enhanced pressure relief. Diabetes Care 2008;31(5): Kernozek TW, LaMott EE, Dancisak MJ. Reliability of an in-shoe pressure measurement system during treadmill walking. Foot Ankle Int 1996;17(4): Ducic I, Short KW, Dellon AL. Relationship between loss of pedal sensibility, balance, and falls in patients with peripheral neuropathy. Ann Plast Surg 2004;52(6): Katoulis EC, Ebdon-Parry M, Lanshammar H, Vileikyte L, Kulkarni J, Boulton AJ. Gait abnormalities in diabetic neuropathy. Diabetes Care 1997;20(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(7): Apelqvist J, Bakker K, van Houtum WH, Schaper NC. Practical guidelines on the management and prevention of the diabetic foot: based upon the International Consensus on the Diabetic Foot (2007) Prepared by the International Working Group on the Diabetic Foot. Diabetes Metab Res Rev 2008;24 (Suppl 1):S181-S

43 Number of required steps for representative in-shoe pressure data 13. Bus SA, van Deursen RW, Kanade RV et al. Plantar pressure relief in the diabetic foot using forefoot offloading shoes. Gait Posture Bus SA, Ulbrecht JS, Cavanagh PR. Pressure relief and load redistribution by custommade insoles in diabetic patients with neuropathy and foot deformity. Clin Biomech 2004;19(6): Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1): Altman DG. Practical statistics for medical research. Chapman and Hull, London

44 42

45 Chapter 3 Offloading effect of therapeutic footwear in patients with diabetic neuropathy at high risk for plantar foot ulceration Diabetic Medicine Dec; 29(12): Reprinted with permission from John Wiley & sons MLJ Arts R Waaijman M de Haart R Keukenamp F Nollet SA Bus 43

46 Chapter 3 Abstract Aims Custom-made therapeutic footwear is often prescribed to patients with diabetic neuropathy, foot deformity and a healed plantar foot ulcer. Offloading these feet is important to prevent ulcer recurrence. The aim was to evaluate the offloading effect of custom-made footwear in these patients. Methods In 171 patients with diabetic neuropathy (336 feet) with foot deformity and a recently healed plantar foot ulcer, plantar pressures walking barefoot and inside new custom-made footwear were measured. At the previous ulcer location and at locations of highest barefoot pressure due to the deformity, in-shoe pressures were compared with non-deformed feet. The footwear was considered effective in offloading when in-shoe peak pressure at these locations was <200 kpa. Results Mean in-shoe peak pressures ranged between 211 and 308 kpa in feet with forefoot deformity (vs kpa in non-deformed feet) and between 140 and 187 kpa in feet with midfoot deformity (vs. 112 kpa in non-deformed feet). Offloading was effective in 61% of all feet with deformity, 81% of feet with midfoot deformity, 44% of feet with forefoot deformity, and 62% of previous ulcer locations. Inter-subject variability in measured in-shoe plantar pressure was large. Conclusions Offloading of custom-made footwear is often not sufficiently achieved in high-risk diabetic feet with deformity. Highest offloading success rates were seen at known high-risk locations such as previous ulcer locations and Charcot feet, the lowest success rates in forefoot deformities. Together with the large inter-subject variability in pressure outcomes, this emphasizes the need for evidence-based prescription and evaluation procedures to assure adequate offloading. 44

47 Offloading effect of therapeutic footwear Introduction Diabetic foot ulceration imposes a large burden on the patient and the health care system, with increased risk for infection and amputation, lower health-related quality of life, and high treatment costs 1-4. In the absence of protective sensation, elevated plantar pressure is causative of plantar foot ulcers 5, 6, and is strongly associated with foot deformity 7-9. Patients with these conditions are often prescribed with custom-made therapeutic footwear. This footwear aims to prevent ulceration by reducing peak pressures at high-risk plantar foot sites through pressure redistribution. Although custom footwear is widely used and assumed to reduce ulcer risk, the scientific evidence is still meagre 10. Among other factors, this may be related to the lack of knowledge about the offloading effect of this footwear. In clinical practice, footwear evaluation is still heavily based on the experience of the clinical team and a trial-and-error approach, using foot ulceration as most informative outcome for footwear success and clinical decision making. Quantitative methods are, however, available to help in footwear evaluation and to improve the interpretation of footwear offloading. 3 Studies on footwear offloading in high-risk patients with diabetic neuropathy and foot deformity are scarce. One study showed higher in-shoe plantar pressures in a cohort of patients with diabetes and different foot deformities when compared to patients without foot deformity 9, but this was tested in usual footwear that was not further specified and the study did not include a dedicated sample of high-risk patients. A more recent study measured forefoot in-shoe plantar pressures in a group of high-risk patients who remained healed from previous ulceration for a longer period of time while wearing therapeutic footwear, and found a mean peak pressure value of approximately 200 kpa 11. Although this pressure level should not be interpreted as an individual peak pressure threshold for ulcer-free survival, it may be used as reference to assess success in footwear offloading when an in-shoe pressure measuring device with similar specifications is used. It is currently unclear to what extent custom-made footwear for these high-risk patients with diabetic neuropathy and foot deformity meets this criterion. Therefore, the aim of this observational study was to investigate to what extent offloading by means of custom-made footwear results in peak plantar pressures below 200 kpa in at-risk foot areas affected by foot deformity in patients with diabetic neuropathy and a history of plantar foot ulceration. Furthermore, the study aimed to assess determinants of in-shoe offloading success. We hypothesized to find insufficient overall offloading success rates (i.e. < 80%), based on the high ulcer recurrence rates still found in patients with diabetes 12, and the lowest success rates (i.e. highest in-shoe peak pressures) in feet with the most severe foot deformities. 45

48 Chapter 3 Patients and methods Subjects A total of 171 consecutive patients with diabetic neuropathy (contributing 336 feet), who presented at the outpatient diabetic foot clinics of 10 Dutch hospitals between january 2008 and oktober 2010 and fullfilled the inclusion criteria, participated in this study. The study was part of the DIAFOS trial (trial register ID NTR1091)*. Baseline patient characteristics are shown in Table 3.1. All patients had a healed plantar foot ulcer within 18 months prior to the assessment. Loss of protective sensation due to neuropathy was verified by the inability to sense the pressure of a 10g Semmes-Weinstein monofilament at one or more of three plantar foot sites tested: hallux, first metatarsal head, and fifth metatarsal head, or the vibration of >25 Volt measured at the dorsal hallux using a Bio-thesiometer (Bio-Medical Instrument Company, Newbury, Ohio) 6. Patients were excluded when they were unable to walk unaided for 100m or when they had bilateral amputation proximal to the metatarsal bones. Written informed consent was obtained from each patient prior to the start of the study and all procedures were approved by the medical ethics committee of the Academic Medical Center in Amsterdam, the Netherlands. Foot deformity Each foot was physically examined by one of three trained researchers (MA, RW, or RK) for presence of deformity. Additionally, photographs of each foot were taken in a loaded and unloaded position using a standardized protocol. These photographs were assessed for presence of deformity and forefoot amputation level (i.e. digit, ray, or transmetatarsal) by two teams of two trained observers who reached consensus on outcome. The presence of deformity was primarily diagnosed based on photographic assessments, except for prominent metatarsal heads and limited joint mobility, which were diagnosed based on the physical examination. Hallux valgus was defined as lateral deviation of the hallux with respect to the first metatarsal bone, hammer toes as hyperflexed inter-phalangeal joints with plantar floor contact of the top of the toes in a loaded position, claw toes as hyperextended metatarso-phalangeal joints with hyperflexed inter-phalangeal joints, pes cavus as a high medial foot arch, pes planus as a flattened medial foot arch, prominent metatarsal heads as palpable bony prominences, and Charcot foot as visible dislocation of the midfoot joints, in some cases resulting in a rocker bottom deformity. The presence of Charcot foot and the level of amputation were verified in the medical file of the patient. Range of motion in the first metatarso-phalangeal joint was measured using goniometry in a supine position and limited joint mobility was considered present when the range of motion was smaller than one standard deviation below the mean of all feet. Finally, feet without foot deformity were labelled as non-deformed feet. 46

49 Offloading effect of therapeutic footwear Footwear All patients were prescribed with new therapeutic footwear, which included fully customized footwear (i.e. custom insoles in custom-made shoes, N=146) or semi-customized footwear (i.e. custom insoles in extra-depth shoes, N=25). In each of the 10 participating centres, footwear was prescribed by a rehabilitation specialist and manufactured by an orthopaedic shoe technician, both with minimally 4 years of experience in diabetic foot treatment. Shoe lasts were generally created from plaster cast or from foam impressions with geometrical foot measures. Blueprints were used to specify at-risk regions to be targeted. Although not enforced by any protocol, footwear design mostly followed the Delphi-based algorithm published by Dahmen et al. 13. Custom insoles consisted of multi-layered materials, mostly with a cork base added with micro-cork and a midlayer of ethylene vinyl acetate based multiform. Local softening in the insole was frequently added and corrective elements (e.g. metatarsal pad or bar), were occasionally incorporated. The insoles were finished with a leather, PPT (Langer, Inc., Deer Park, New York, USA) or Plastazote (Zotefoams plc, Croydon, UK) top cover. Shoe outsoles mainly consisted of stiffened rubber or Poron and a roller configuration. 3 Instrumentation Barefoot dynamic plantar pressures were measured using an EMED-X platform (Novel GmbH, Munich, Germany) which consists of capacitance-based sensors in a spatial resolution of 4 sensors per cm 2 that were sampled at 70 Hz. In-shoe dynamic plantar pressure was measured using the Pedar-X system from Novel. This system comprises flexible 2 mm thick insoles, available in different length and width sizes, containing a matrix of 99 capacitancebased sensors in a spatial resolution of approximately 2 sensors per cm 2, with each sensor sampling at 50 Hz. The insoles were placed inside the shoe between the sock and insole. Each insole was calibrated prior to data collection according to the manufacturer s specifications. Procedures For barefoot pressure assessment, at least 4 walking trials were collected per subject using a 2-step approach to the platform 14. In-shoe plantar pressure was assessed while walking repeatedly along a walkway of minimally 10 meter length. It was assured that pressures from at least 12 midgait steps per foot were collected 15. Patients walked at a self-selected comfortable speed, which was measured using a stopwatch, and was controlled between trials (± 5% variation allowed). All patients wore thin seamless socks which were provided during the measurements. 47

50 Chapter 3 Table 3.1. Baseline patient s characteristics and walking speed for in-shoe pressure measurement. Variable Outcome Number of patients 171 Number of (non-amputated) feet 336 Gender (male / female) 140 / 31 Age (years) 62.8 (10.2) Diabetes type (Type 1 / Type 2) 49 / 122 Diabetes duration (years) 14.0 ( ) Body mass index (kg/m 2 ) 30.9 (6.0) HbA1C (mmol/mol) 64 (42) HbA1C (%) 7.6 (1.4) Vibration perception threshold (V) 45 (11) Walking speed (m/s) 1.03 (0.28) Data are expressed as N, mean (SD) or median (inter-quartile range). Data analysis Novel multimask software (Version ) was used to analyze the pressure data. Footsteps showing major deviances in the ground reaction force curves were removed manually. For both the barefoot and in-shoe pressure distribution diagrams, masks were created per foot to divide the foot into 10 anatomical regions: lateral and medial heel, lateral and medial midfoot, metatarsal 1, metatarsal 2/3, metatarsal 4/5, hallux, toes 2/3, and toes 4/5. Additionally, the previous ulcer location was masked. For each region, the mean peak pressure over the multiple steps per foot was calculated 16. All feet were grouped according to type of deformity (claw toes, hallux valgus, Charcot foot, etc). For each deformity group, one region-of-interest was defined based on where the mean measured barefoot peak pressure in that group was highest and therefore considered most at risk for ulceration. Such a region-of-interest could cover one or two anatomical regions, such as metatarsal heads 2-5 in the case of claw toes (Table 3.2). In each group of feet with a particular deformity, other deformities could be present as long as these other deformities did not have the same region-of-interest as the primary deformity. This means that feet with multiple deformities could be included in more than one deformity group. Per deformity, footwear offloading was assessed only at the region-of-interest. Additionally, footwear offloading was assessed in all patients at the previous ulcer location. Footwear offloading was assessed in three ways. First, in-shoe peak pressure was compared between each deformity group and the no-deformity group. Secondly, the relative peak 48

51 Offloading effect of therapeutic footwear pressure reduction between in-shoe and barefoot was calculated for each deformity group and the no-deformity group. Third, the percentage of cases with in-shoe peak pressure below 200kPa (defining success in offloading) was calculated, based on a previous report indicating a 200kPa threshold as effective for ulcer-free survival in the forefoot 11. Offloading success scores were considered insufficient when this percentage was lower than 80%, and sufficient when above 80%. Statistical analysis Statistical analyses were carried out using PASW statistics version 18 (SPSS inc., Chicago, USA), and if not specified otherwise, a significance level p < 0.05 was used. Independent sample t-tests were used to compare in-shoe peak pressure between each deformity group and the no-deformity group. Pearson correlation coefficients were calculated to determine the association between barefoot and in-shoe peak pressure at the previous ulcer location. Between participating centres, differences in in-shoe plantar pressures were assessed with one-way analysis of variance and post-hoc testing, while differences in offloading success were assessed with Kruskal-Wallis tests. Univariate logistic regression analyses of offloading success at the previous ulcer location were performed including the variables: vibration perception threshold, body mass index, type of foot deformity, barefoot peak pressure, walking speed, type of footwear, and participating centre (p < 0.10). Significant factors were included in a multivariate logistic regression model in a backwards stepwise fashion.a 3 Figure 3.1. Histogram showing the barefoot peak pressure (grey bars) and in-shoe peak pressures (black bars) at the previous ulcer location for each of the 147 subjects with a non-amputated previous ulcer location. The horizontal dashed line represents the 200kPa threshold used to define offloading succes in this study. Data are ranked by measured in-shoe peak pressure. In 17 cases (12%), the peak pressure saturation level of the EMED platform (1275kPa) was reached. 49