ELEVATED PLANTAR PRESSURES have been associated

Transcription

1 253 Foot Pressures During Level Walking Are Strongly Associated With Pressures During Other Ambulatory Activities in Subjects With Diabetic Neuropathy Katrina S. Maluf, PT, PhD, Robert E. Morley Jr, DSc, Edward J. Richter, MSEE, Joseph W. Klaesner, PhD, Michael J. Mueller, PT, PhD ABSTRACT. Maluf KS, Morley RE, Richter EJ, Klaesner JW, Mueller MJ. Foot pressures during level walking are strongly associated with pressures during other ambulatory activities in subjects with diabetic neuropathy. Arch Phys Med Rehabil 2004;85: Objective: To assess the relationship between foot pressures measured during level walking and other types of ambulatory activity in subjects with diabetes mellitus (DM) and peripheral neuropathy (PN). Design: Descriptive survey with repeated measures. Setting: University medical center. Participants: Convenience sample of 16 ambulatory subjects with DM and PN. Interventions: Not applicable. Main Outcome Measures: Peak pressure and pressure-time integral (PTI) at the heel, great toe, first metatarsal head (MT1), and third metatarsal head (MT3) during level walking, ramp climbing, stair climbing, and turning at a self-selected speed. Results: Peak pressure and PTI during level walking correlated highly with pressures during ramp climbing (r range,.85.97) and turning (r range,.75.96) at all regions examined and with pressures during stair climbing at MT1 and MT3 (r range,.84.91). Correlations between pressures during level walking and stair climbing were moderate at the great toe (r range,.46.57) and poor at the heel (r range,.12 to.06). With few exceptions, pressures during ramp climbing, stair climbing, and turning were less than (P.008) or equal to pressures during level walking. We found no activity-related differences in peak pressure or PTI independent of the effects of preferred walking speed. Conclusions: Results support the clinical evaluation of peak pressure during level walking as an efficient method to screen for maximum levels of stress on the foot as patients with DM and PN perform their daily activities. Key Words: Activities of daily living; Diabetic foot; Rehabilitation by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation From the Department of Kinesiology and Applied Physiology, University of Colorado, Boulder, CO (Maluf); Electrical Engineering Department, Washington University, St. Louis, MO (Morley, Richter); and Program in Physical Therapy, Washington University School of Medicine, St. Louis, MO (Klaesner, Mueller). Supported by the Foundation for Physical Therapy and National Center for Medical Rehabilitation Research (grant nos. R01 HD 36576, 2T32HD ). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the author(s) or on any organization with which the author(s) is/are associated. Reprint requests to Michael J. Mueller, PT, PhD, Program in Physical Therapy, Washington University School of Medicine, Campus Box 8502, 4444 Forest Park Blvd, Rm 1101, St. Louis, MO , muellermi@msnotes.wustl.edu /04/ $30.00/0 doi: /j.apmr ELEVATED PLANTAR PRESSURES have been associated with an increased risk of skin breakdown in patients with diabetes mellitus (DM) and peripheral neuropathy (PN). 1-3 Consequently, plantar pressure assessment has become an important tool for the treatment and prevention of chronic foot ulcers, which develop in approximately 10% of people with diabetes. 4 Standard methods of pressure assessment quantify pressure within a selected region of the foot as a patient walks in a straight line across a level surface. 5 Analysis of pressures during level walking, however, may not accurately reflect activity-related variations in foot pressure that occur throughout the day. Studies have shown that plantar pressures in healthy persons are influenced by changes in walking speed 6,7 and by the type of activity being performed Compared with walking barefoot on a level surface, 1 study 9 found that pressure under the heel was reduced when descending stairs and that pressure in the metatarsal region was elevated during a pivot turn. Others 8 have reported similar results for pressures measured inside the shoe during level walking compared with stair climbing and walking in a circle. Walking on inclined surfaces has been shown to increase forefoot loading and to decrease rear foot loading, whereas the opposite effect has been observed for walking on declined surfaces. 10 The combined results of these studies suggest that plantar pressures measured during level walking may not accurately reflect pressures during other ambulatory activities. Observations in healthy persons have led to speculation that the assessment of foot pressures during level walking alone may be insufficient to determine cumulative foot trauma and the effect of therapeutic interventions for patients with foot pathology. 8 However, findings from healthy persons may not apply to patients with foot pathology. The ability to generalize findings to patients with DM and PN is particularly questionable, because of the presence of gait impairments, limited joint mobility, 14,15 and structural foot deformities 16,17 that could load similar regions of the foot across all types of weightbearing activity. The primary purpose of our study was to assess the relationship between foot pressures measured during level walking and other types of ambulatory activity in patients with DM and PN. A second purpose was to determine the effect of preferred walking speed on activity-related differences in foot pressure. Compared with walking on level surfaces, we expected to observe higher forefoot pressures when subjects with DM and PN ascended stairs and ramps because of an increase in the amount of ankle mobility and force of push-off needed to achieve these tasks. However, we also expected regional foot pressures during level walking to be strongly associated with pressures during other types of ambulatory activity in this patient population.

2 254 ACTIVITY-RELATED FOOT PRESSURES, Maluf Table 1: Characteristics of Study Population Sex Men 12 (75) Women 4 (25) Age (y) Body mass index (kg/m 2 ) Type of DM Type 1 4 (25) Type 2 12 (75) Duration of DM (y) Previous plantar ulcer Yes 7 (44) No 9 (56) Foot deformity Hammer toe deformity 9 (56) Hallux valgus 10 (63) Dorsiflexion ROM (deg) 3 (5) 7 to10 VPT (V) Hb A 1c (%) NOTE. Values are n (%) or mean standard deviation (SD) and range. Abbreviations: Hb A 1c, glycosylated hemoglobin; ROM, range of motion. METHODS Participants Subjects with DM were recruited from outpatient medical clinics and community health fairs in the St. Louis region. Subjects were included in the study if they were found to be at risk of developing a plantar ulcer due to a loss of protective sensation, defined as the inability to detect a 5.07 Semmes- Weinstein monofilament on the plantar foot. 18 Vibration perception thresholds 18,19 (VPTs) were used to quantify the magnitude of sensory impairment for eligible subjects and were not used as an inclusion criterion for the study. Exclusion criteria included an inability to walk on level surfaces, stairs, and ramps without an assistive device; amputation of the tested extremity (excluding the lesser toes); or any lower-extremity surgery performed 1 year before testing. Subjects were also excluded if their shoe size did not match our inventory of footwear (see below). Sixteen eligible subjects provided informed consent in accordance with study procedures approved by the institutional review board at Washington University School of Medicine. Characteristics of the study population are presented in table 1. Participants included 12 men and 4 women, with a mean age of years and a body mass index of kg/m 2. The reported time since diagnosis of type 1 (25%) or type 2 (75%) DM ranged from 6 months to 44 years. Forty-four percent of subjects reported a history of plantar ulcers, and most subjects had 1 or more foot deformities identified by clinical examination. On average, subjects had limited ankle dorsiflexion range of motion (ROM; 3 5 ) and slightly elevated glycosylated hemoglobin (Hb A 1c ; 7.6% 1.7%). Test Procedures A brief clinical examination was performed to identify structural foot deformities and to assess ankle ROM and the degree of sensory impairment present in our study population. The presence or absence of foot deformity was assessed by visual inspection and palpation, as described by Bresater et al. 20 Sensory impairment was quantified with a biothesiometer a to measure VPT at the great toe. 18,20 Passive ankle dorsiflexion ROM was measured using a handheld goniometer, with the subject positioned supine and the knee extended. 21 Additionally, Hb A 1c was measured as an index of blood glucose control. After clinical examination, standardized inserts were constructed for each subject using pressure sensors from the inshoe multisensory data acquisition device 22,23 (IMD). The IMD is a portable electronic device capable of monitoring pressures at 4 discrete locations on the plantar foot during prolonged activity. The device uses commercially available pressure sensors b (sensor size: mm) that can be placed at any location in the insole. Construction, accuracy, and reliability of a prototype device have been described. 22,23 Continued development of the IMD has included the addition of calibration software c using 2-point linear interpolation based on a maximum nonlinearity of.42% within the test range of 0 to 62.5N/cm 2. The accuracy of calibration software was assessed by applying a series of known external loads corresponding to a pressure range of 5 to 35N/cm 2. Pressure data from the IMD were found to correlate highly with known pressure values (r.997), with less than 5% mean absolute error. 24 Standardized inserts were constructed for each subject by marking the midpoint of the heel, great toe, first metatarsal head (MT1), and third metatarsal head (MT3) with a grease pencil. Subjects then were asked to stand and shift their weight forward and back, with their foot placed comfortably inside an extra-depth shoe with a 1 4-in (.64-cm) white plastazote #2 insert. A mm window was cut out of the plastazote #2 insert at the site of each bony landmark identified by the transfer of grease pencil marks onto the insert. Pressure sensors were placed within each window of the insert such that the bottom of each sensor was in direct contact with the relatively rigid EVA midsole. Sensors were secured with adhesive, and a layer of 1 8-in (.32-cm) pink plastazote #1 was placed over the top of the insert. Placement of pressure sensors between the midsole and a thin ( 1 8-in) layer of plastazote was designed to minimize the amount of orthotic material in regions of the foot that were monitored during testing. The thicker ( 1 4-in) layer of plastazote surrounding each sensor was used to help prevent movement of the sensors, and to provide additional protective cushioning for regions of the foot that were not monitored during testing. Inserts were constructed using new materials for each subject, and pressure sensors were connected to the data storage unit of the IMD, which remained strapped to the lateral calf during data collection. All subjects wore standardized polyester socks and extra-depth shoes (Drew Duplex, d women s sizes 7, 9, 11; Drew Doubler, d men s sizes 10, 11, 13), which were purchased new and were recycled between subjects. Pressure data were sampled at a frequency of 56.25Hz from the foot with the most recent history of ulceration or from a foot selected at random in subjects without a history of foot ulcers. The finite sampling and storage capacity of the IMD limited data collection to 4 discrete regions of the foot. Three regions of the forefoot were selected to represent the most common sights of ulceration observed in our clinical practice. The heel was selected so that patterns of forefoot and rear foot loading could be compared with findings from other studies All subjects performed the following activities in randomized order while wearing the IMD: walking in a straight line across a 15-m level walkway (2 trials); ascending (RU) and

3 ACTIVITY-RELATED FOOT PRESSURES, Maluf 255 descending (RD) a 7.3-m, 7% grade ramp (3 trials); ascending (SU) and descending (SD) a flight of 13 steps (19cm height 27cm depth, 3 trials); and walking clockwise and counterclockwise around the perimeter of a circle 1m in diameter to simulate turning with the foot positioned on the inside (T-I) and outside (T-O) of the turn (5 continuous rotations in each direction). Subjects were instructed to walk at a comfortable speed to simulate their activity in the home and community. All activities were performed on an identical concrete surface, and subjects were encouraged to rest as often as needed between trials. Subjects were permitted to use a handrail for balance during stair climbing, but otherwise did not use an assistive device. Stair-climbing trials were repeated when subjects were observed to use their upper extremities for weight support rather than balance. All activity trials were timed using a stopwatch, and walking speed was calculated by dividing the total distance traveled by the time required to complete each trial. The distance traveled during level walking and ramp climbing was determined by measuring the length of the level walkway and ramp, respectively. The total distance traveled during stair climbing was calculated by adding the horizontal (stair depth by number of stairs) and vertical (stair height by number of stairs) lengths of the stairwell. The distance traveled by the inside foot and the outside foot during turning was calculated as 5 times the circumference of a circle whose radius was equal to the distance between each foot and the center of a circle 1m in diameter drawn on the floor. Data Processing and Analysis Data from the IMD were downloaded to a host computer for processing. After excluding the first and last steps from each activity trial, the mean peak pressure and pressure-time integral (PTI) recorded at the heel, great toe, MT1, and MT3 were calculated across all trials for each activity, using custom software written in Matlab. 22,e The relationship between regional pressures measured during level walking and each of the other activities was quantified using Pearson correlation coefficients. Overall differences in mean peak pressure and PTI were analyzed separately, using a 7 (activity) by 4 (region) repeated-measures analysis of variance (ANOVA). Post hoc analyses were conducted using a 1-factor repeated-measures ANOVA for paired comparisons between regional pressures during level walking and each of the other activities. Differences in the preferred walking speed for each activity were assessed using a 1-factor repeated-measures ANOVA, and post hoc analyses were repeated with preferred walking speed entered as a covariate to assess the effect of speed on activityrelated differences in peak pressure and PTI. A Bonferroniadjusted significance criterion of P less than.008 was used for all post hoc paired comparisons. RESULTS Peak pressures during level walking correlated highly with peak pressures during ramp climbing (RD r range,.95.97; RU r range,.88.96) (fig 1) and turning (T-O r range,.93.96; T-I r range,.75.88) (fig 1) for all regions of the foot and with peak pressures during stair climbing for the forefoot (SD r.84 and r.87, SU r.86 and r.91 at MT1 and MT3, respectively) (fig 1). Peak pressures during level walking and stair climbing correlated only moderately at the great toe (SD r.46, SU r.57) (fig 1) and did not correlate at the heel (SD r.06, SU r.12) (fig 1). Relationships between PTI during level walking and other types of ambulatory activity were characterized by a similar pattern of results (RD r range,.85.97; RU r range,.90.94; SD r range,.04 [heel] to.88 [MT1]; SU r range,.04 [heel] to.84 [MT1]; T-O r range,.89.96; T-I r range,.79.92). A significant activity-by-region interaction was found for peak pressure (P.001) (fig 2) and PTI (P.001) (fig 3). Compared with level walking, post hoc analyses showed that peak pressures at the heel and great toe were significantly lower during stair climbing (P.003) and turning (P.001), peak pressure at MT1 was significantly lower when descending a ramp (P.007) and turning with the inside foot (P.001), and peak pressure at MT3 was significantly lower when turning (P.004) and descending a ramp (P.001). The only significant increase in peak pressure occurred under MT3 when ascending a ramp (P.003). Compared with level walking, PTI was significantly lower at the great toe when turning with the inside foot (P.001) and at the heel when stair climbing (P.001) and descending a ramp (P.002). PTI was significantly higher at MT1 and MT3 during stair climbing (P.007) and at the heel when ascending a ramp (P.001) and turning with the inside foot (P.001). Stair climbing and turning activities were performed at a significantly slower speed than level walking (P.001) (table 2). Differences in peak pressure and PTI between level walking and all other activities were no longer significant after using analysis of covariance to adjust for differences in preferred walking speed. DISCUSSION The purpose of our study was to assess the relationship between regional foot pressures during level walking and other types of ambulatory activity in subjects with DM and PN. We found that pressures during level walking correlated highly with pressures during ramp climbing and turning for all regions of the foot examined in this study. Pressures during stair climbing were strongly associated with pressures during level walking at MT1 and MT3 but not at the great toe or heel. We also found that peak pressures during level walking were greater than or equal to peak pressures during other types of activity, with the exception of peak pressure at MT3 when ascending a ramp. Given the limited resources of clinical and laboratory environments, these findings support the assessment of peak pressure during level walking as an efficient method to screen for maximum levels of stress on the foot as patients with DM and PN perform their daily activities. More caution must be used when generalizing PTI results to activities other than level walking, because of elevated PTI at the forefoot during stair climbing and at the heel during ramp climbing and turning. Contrary to expectation, we found that peak pressures during stair climbing, ramp climbing, and turning were often lower than peak pressures during level walking in subjects with DM and PN. The most likely explanation for this finding is that patients with DM and PN walk more slowly when presented with tasks that are more challenging than walking across a straight, level surface. We found no significant difference in peak pressure or PTI between any of the activities in this study after adjusting for differences in preferred walking speed. These findings suggest that walking speed has an important effect on foot pressures during self-paced ambulation in the community. Walking on inclined surfaces has been shown to increase forefoot loading and decrease rear foot loading in healthy subjects, whereas the opposite effect has been observed for walking on declined surfaces. 10 Our results show a similar trend for ramp climbing in subjects with DM and PN, although only differences in forefoot peak pressure and heel PTI were significant. Larger differences in pressure might have been

4 Fig 1. Correlations between (A) peak pressures during level walking and ramp climbing, (B) level walking and stair climbing, and (C) level walking and turning at the heel, great toe (GT), MT1, and MT3. Similar correlations were observed for PTIs measured during level walking and other types of ambulatory activity (data not shown).

5 ACTIVITY-RELATED FOOT PRESSURES, Maluf 257 Fig 2. Peak pressures at (A) the heel, (B) great toe, (C) MT1, and (D) MT3 during level walking compared with descending a ramp, ascending a ramp, descending stairs, ascending stairs, and turning with the foot positioned on the outside and inside of the turn. *Significantly different from level walking (P<.008). observed for steeper gradients, such as those examined in other studies 10 ; however, very steep gradients are likely to be avoided by patients with DM and PN when walking in the community. Two previous comparisons 8,9 of peak pressure during level walking and stair climbing in healthy subjects found a reduction in peak pressure at the heel during stair climbing but no difference in peak pressure at the forefoot between these activities. We observed similar results for peak pressure during stair climbing in subjects with DM and PN. Unlike peak pressure, however, we found that PTI was significantly increased at the forefoot during stair climbing, possibly because of a longer duration of foot-to-floor contact associated with a slower walking speed during stair climbing. Previous studies 8,9 have reported an increase in peak pressure and PTI for several regions of the foot during turning compared with level walking. Rozema et al 8 noted, in addition, that peak pressures during level walking and turning were only weakly associated at the great toe (r.56) and MT1 (r.31) in healthy subjects. In contrast with these findings, we observed a strong association between peak pressures during level walking and turning for all regions of the foot in patients with DM and PN. We also found that peak pressure and PTI during turning were significantly lower than or equal to pressures during level walking with only 1 exception (ie, elevated PTI at the heel of the outside foot). Subjects in our study performed the turning task at a significantly slower speed than level walking (43 46m/min vs 69m/min), whereas subjects in an earlier study 8

6 258 ACTIVITY-RELATED FOOT PRESSURES, Maluf Fig 3. PTIs at (A) the heel, (B) great toe, (C) MT1, and (D) MT3 during level walking, compared with descending a ramp, ascending a ramp, descending stairs, ascending stairs, and turning with the foot positioned on the outside and inside of the turn. *Significantly different from level walking (P<.008). performed the turning task at a higher speed than level walking (90m/min vs 61m/min). These differences in walking speed may help explain discrepancies between our results and other observations of pressures during turning. Traditional methods of pressure assessment do not account for individual differences in weight-bearing activity that may contribute to cumulative tissue stress and risk of tissue injury. By combining the results of traditional pressure assessment with activity data from an accelerometer, recent efforts have been made to estimate cumulative stress on the plantar forefoot as patients with DM and PN perform their daily activities. 25 These estimates of cumulative tissue stress assume that PTIs measured during level walking in the laboratory accurately reflect pressures during ambulation in the home and community. Results from this study support the use of PTI during level walking, to provide a reasonable estimate of cumulative forefoot stress in patients with DM and PN who perform limited stair climbing. Ours is the first study to show a strong association between foot pressures during level walking and other activities performed at a comfortable walking speed by patients with DM and PN; however, certain limitations should be recognized. First, pressures were sampled over a relatively small area in 4 discrete regions of the foot that were selected on the basis of commonly observed sites of ulcer formation in patients with DM and PN. The standardized placement of pressure sensors allowed us to assess regional differences in foot pressure across several activities performed by subjects at high risk of devel-

7 ACTIVITY-RELATED FOOT PRESSURES, Maluf 259 Activity Level walk Ramp down Ramp up Stair down Stair up Turn (outside foot) Turn (inside foot) Table 2: Preferred Walking Speed Walking Speed* (m/min) NOTE. Values are mean SD and 95% confidence interval. *P.001 (ANOVA); P.001 vs level walk oping foot ulcers. Because of the inability of many subjects to recall the precise location of their previous ulcers and because of the heterogeneity of local foot deformities in our study population, we did not attempt to isolate high-risk areas of the foot for individual subjects. Although the standardized location of 1 or more sensors did correspond to high-risk areas of the foot for most of our subjects (table 1), additional studies are needed to clarify the relationship between pressures during level walking and other types of ambulatory activity for regions of local foot deformity and sites of previous ulceration. A second limitation is that we did not quantify differences in shear that were likely to be higher for activities such as ramp climbing and turning compared with level walking. Additionally, it is unclear to what extent the plastazote insert influenced the pressures measured in our study. Although use of a cushioned insert is likely to reduce absolute foot pressures, it seems unlikely that the relative magnitude of pressure reduction attributable to the insert would differ between activities performed with the same insert. Pressure measurements also may have been affected by nonlinear measurement error, because the linearity of sensor output was not assessed for pressures exceeding 62.5N/cm 2. This type of error would be expected to inflate correlations among pressures in the nonlinear measurement range; however, most pressures in this study were within the confirmed range of linearity. Finally, our study examined only a limited number of factors that could influence foot pressures during ambulation in the community. Other studies are needed, for example, to determine how assistive devices and therapeutic footwear affect foot pressures during activities other than level walking. CONCLUSIONS Our study evaluated the relationship between foot pressures during level walking and other types of ambulatory activity in patients with DM and PN who were at high risk for developing foot ulcers. Peak pressure and PTI during level walking correlated strongly with pressures during other activities, with the exception of pressures at the great toe and heel during stair climbing. Because of the reduced speed of stair climbing and turning compared with level walking, peak pressures during these activities were lower than or equal to peak pressures during level walking. These results support the clinical evaluation of peak pressure during level walking as an efficient method to screen for maximum levels of stress on the foot as patients with DM and PN perform their daily activities. Findings from our study also have implications for research using pressures during level walking to estimate cumulative tissue trauma 25 and risk of plantar tissue injury in patients with DM and PN. Acknowledgments: We gratefully acknowledge Chrysta Lloyd and Jay Kuruvi for assistance with data processing. 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: Stess RM, Jensen SR, Mirmiran R. The role of dynamic plantar pressures in diabetic foot ulcers. Diabetes Care 1997;20: 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: Moss SE, Klein R, Klein BE. The prevalence and incidence of lower extremity amputation in a diabetic population. Arch Intern Med 1992;152: Cavanagh PR, Ulbrecht JS. Clinical plantar pressure measurement in diabetes: rationale and methodology. Foot 1994;4: Kernozek TW, LaMott EE, Dancisak MJ. Reliability of an in-shoe pressure measurement system during treadmill walking. Foot Ankle Int 1996;17: Zhu H, Wertsch JJ, Harris GF, Alba HM. Walking cadence effect on plantar pressures. Arch Phys Med Rehabil 1995;76: Rozema A, Ulbrecht JS, Pammer SE, Cavanagh PR. In-shoe plantar pressures during activities of daily living: implications for therapeutic footwear design. Foot Ankle Int 1996;17: Lundeen S, Lundquist K, Cornwall MW, McPoil TG. Plantar pressures during level walking compared with other ambulatory activities. Foot Ankle Int 1994;15: Grampp J, Willson J, Kernozek T. The plantar loading variations to uphill and downhill gradients during treadmill walking. Foot Ankle Int 2000;21: Katoulis EC, Ebdon-Parry M, Lanshammar H, Vileikyte L, Kulkarni J, Boulton AJ. Gait abnormalities in diabetic neuropathy. Diabetes Care 1997;20: Courtemanche R, Teasdale N, Boucher P, Fleury M, Lajoie Y, Bard C. Gait problems in diabetic neuropathic patients. Arch Phys Med Rehabil 1996;77: Mueller MJ, Minor SD, Sahrmann SA, Schaaf JA, Strube MJ. Differences in the gait characteristics of patients with diabetes and peripheral neuropathy compared with age-matched controls. Phys Ther 1994;74: Fernando DJ, Masson EA, Veves A, Boulton AJ. Relationship of limited joint mobility to abnormal foot pressures and diabetic foot ulceration. Diabetes Care 1991;14: Delbridge L, Perry P, Marr S, et al. Limited joint mobility in the diabetic foot: relationship to neuropathic ulceration. Diabet Med 1988;5: Robertson DD, Mueller MJ, Smith KE, Commean PK, Pilgram T, Johnson JE. Structural changes in the forefoot of individuals with diabetes and a prior plantar ulcer. J Bone Joint Surg Am 2002; 84: Mueller MJ, Minor SD, Diamond JE, Blair VP. Relationship of foot deformity to ulcer location in patients with diabetes mellitus. Phys Ther 1990;70: Armstrong DG, Lavery LA, Vela SA, Quebedeaux TL, Fleischli JG. Choosing a practical screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern Med 1998;158: Armstrong DG, Hussain SK, Middleton J, Peters EJ, Wunderlich RP, Lavery LA. Vibration perception threshold: are multiple sites of testing superior to single site testing on diabetic foot examination? Ostomy Wound Manage 1998;44:70-4, Bresater LE, Welin L, Romanus B. Foot pathology and risk factors for diabetic foot disease in elderly men. Diabetes Res Clin Pract 1996;32: Diamond JE, Mueller MJ, Delitto A, Sinacore DR. Reliability of a diabetic foot evaluation. Phys Ther 1989;69:

8 260 ACTIVITY-RELATED FOOT PRESSURES, Maluf 22. Maluf KS, Morley RE, Richter EJ, Klaesner JW, Mueller MJ. Monitoring in-shoe plantar pressures, temperature, and humidity: reliability and validity of measures from a portable device. Arch Phys Med Rehabil 2001;82: Morley RE, Richter EJ, Klaesner JW, Maluf KS, Mueller MJ. In-shoe multisensory data acquisition system. IEEE Trans Biomed Eng 2001;48: Maluf KS. Preventing ulcer recurrence in patients with diabetes: experimental application of the Physical Stress Theory [dissertation]. St. Louis (MO) Washington Univ; p Maluf KS, Mueller MJ. Novel Award 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 (Bristol, Avon) 2003;18: Suppliers a. BioMedical Instrument Co, Munn Rd, Newbury, OH b. Firma Paromed Vertriebs GmbH, Hubertushof-Heft 8, Neubeuern, Germany. c. TÜV Product Service, Niederlassung München, Ridlerstr 65, D München, Germany. d. Drew Shoe Corp, 252 Quarry Rd, Lancaster, OH e. The MathWorks, 3 Apple Hill Dr, Natick, MA