Efficacy and mechanism of orthotic devices to unload metatarsal heads in people with diabetes and a history of plantar ulcers.

The most common cause for diabetic plantar ulcers is excessive plantar pressure in the presence of sensory neuropathy and foot deformity. Most neuropathic foot ulcers occur beneath the metatarsal heads and are the result of painless trauma caused by excessive plantar pressures during walking. (1-3) Chronic wound treatment failure often leads to serious infection and limb amputation if it is not managed properly. (3) The primary focus of conservative care for the diabetic foot is to protect the foot from excessive pressures and other forms of unnoticed trauma that begin the cascade of events leading to amputation. (3)

Therapeutic footwear and orthotic devices are the primary means of protecting the foot from excessive plantar pressures during walking, (4,5) and some research indicates that therapeutic footwear can reduce the incidence of ulceration in people with diabetes mellitus (DM). (6,7) Although the results of other research (8) question the benefit of footwear in reducing the incidence of skin breakdown, the American Diabetes Association recommends the use of footwear that cushions and redistributes pressure and thereby helps reduce the incidence of skin breakdown and the associated hospitalization, morbidity, and mortality for at-risk patients. (4,5) To this end, Medicare currently reimburses patients with DM and peripheral neuropathy for 1 pair of therapeutic shoes and 3 pairs of accommodative inserts each year.

Total-contact inserts (TCIs) and some forms of metatarsal pads (MPs) are devices commonly used to reduce forefoot pressures. (9-14) The TCI is thought to accommodate deformities and relieve areas of excessive pressure by evenly distributing pressure over the entire plantar surface with the use of moldable materials. (15) An MP often is added to a TCI with the goal of providing additional forefoot pressure relief. (13) Occasionally, this MP is built directly into a custom-made insole. (16) Ashry et al (13) did not find additional pressure reduction with a Plastazote * insole and MP compared with a Plastazote insert alone in a study of people with DM and great toe amputation. Other studies comparing the use of an MP with the use of no MP in subjects with no history of DM, foot impairments, or pain showed significant reductions in plantar pressures, but these results were quite variable and subject specific, (11,12,17) The MP is thought to load the shaft of the metatarsal with the intent of decreasing the stress and soft-tissue compression at the metatarsal head. Bus et al (16) documented the load redistribution by comparing a custom-made insole (which included an MP) with a flat insole but did not report contact area or soft-tissue thickness (STT). Additional research is needed to clarify the efficacy and mechanism of a TCI and a specific MP in reducing plantar pressures during walking in people with DM, peripheral neuropathy, and a history of forefoot ulcers.

The first purpose of this study was to determine the effect of a TCI and an MP on metatarsal head peak plantar pressures (PPP) and pressure-time integrals (PTI). The PPP and the PTI were used as indices of potential trauma to skin. The PPP indicates the highest magnitude of the stress, and the PTI reflects the magnitude of the stress at a specific location over time (ie, 1 stance phase). We hypothesized that, in comparison with a shoe alone, the TCI would reduce the PPP and the PTI and that, in comparison with the TCI, the MP would further reduce pressures. The second purpose of this study was to determine a possible mechanism of pressure reduction for each orthotic component. We hypothesized that the primary mechanism of pressure reduction for the TCI was an increase in contact area between the foot and the supporting surface in the TCI condition compared with the shoe condition. We hypothesized that the primary mechanism of pressure reduction at the metatarsal head with the MP was loading of the soft tissue proximal to the metatarsal head and unloading of the soft tissue over the metatarsal head. Therefore, we expected the STT to be larger under the metatarsal head and smaller under the metatarsal shaft in the MP condition than in the TCI condition. This study incorporates the use of imaging techniques to clarify the location of the MP and pressure sensor with respect to the metatarsal head and to measure the influence of the orthotic devices on soft-tissue compression.

Method

Subjects

Twenty subjects were recruited from the Diabetic Foot Center, Volunteers for Health, the Diabetes Research Training Center at Washington University School of Medicine, and BJC Health System in St Louis, Mo. Criteria for entry into the study were a history of DM, evidence of peripheral neuropathy (inability to sense the 5.07 Semmes-Weinstein monofilament and a vibratory perception threshold of >25 V), a palpable pedal pulse, and a history of a forefoot ulcer. Subjects who were nonambulatory or who had severe midfoot or rear-foot Charcot neuroarthropathy were excluded. Subjects with DM, peripheral neuropathy, and a history of a forefoot ulcer were selected because this group of subjects is most at risk for ulcer recurrence and might benefit from therapeutic footwear and orthotic devices. (4) Subjects with severe neuroarthropathy were excluded because they often require custom-made footwear to fit severe bony deformities. Toe amputations were not an absolute exclusion criteria, and 1 subject had an amputation of the second toe on the tested foot. Four subjects had a single toe amputation on the contralateral side.

Sensation was tested with the 5.07 Semmes-Weinstein monofilament and a Bio-Thesiometer ([dagger]) with established, reliable measures. (18,19) The vibratory perception threshold was defined as the lowest voltage that the subject could perceive on the plantar great toe in a mean of 3 trials. (10,20) A value of 50 V was assigned to subjects who were unable to perceive the voltage even when the maximum amplitude was used. The value (mean [+ or -] SD) obtained for this group was 48.3 [+ or -] 4.1 volts, indicating a severe level of neuropathy. All subjects read and signed the informed medical consent form according to the institutional review board-approved protocol before entrance into the study. Table 1 shows the subject characteristics.

Footwear Conditions

Outcome measures were collected from the subjects in 3 different footwear conditions: shoe only, shoe with TCI, and shoe with TCI and MP. Figure 1 shows the shoe, TCI, and MP used in this study. All subjects wore new SoleTech shoes (style E3010) ([double dagger]) that were fitted by a certified pedorthist or orthotist. This footwear was selected because it is used widely in clinical practice, meets all of the guidelines of the Therapeutic Shoe Bill (Medicare Part B), and can accommodate forefoot deformities and a custom-made orthotic device. In the shoe condition, the shoe was worn with its removable standard 5.0-mm-thick insert made of cross-linked polyethylene foam blended with ethylene vinyl acetate. The standard insert was removed and replaced with the TCI in the TCI and MP conditions.

[FIGURE 1 OMITTED]

To make the TCI, the certified pedorthist or orthotist took a foam impression of the subject's foot to make a positive plaster model of the foot. Foam was compressed 2 to 4 cm to capture the entire impression of the foot and medial longitudinal arch. The TCI was made from a base of 1.27-cm (0.5-in) number 2 Plastazote with a shore value of approximately 35, heightened to include the medial longitudinal arch, and fabricated to fit inside the shoe. The shore value is an indicator of the orthotic material stiffness, and higher values indicate higher stiffness. A 1.27-cm (0.5-in)-thick TCI made from number 2 Plastazote was selected because of the positive outcomes in previous research on subjects with DM and a transmetatarsal amputation. (14) Results indicated that a custom-made shoe with a 1.27-cm (0.5-in) number 2 Plastazote TCI reduced pressures as well as or better than any other condition tested (6 conditions tested (14)). In addition, pilot testing indicated that pressures were reduced substantially when 1.27-cm (0.5-in)-thick material was used but not when 0.64-cm (0.25-in)-thick material was used. One material type (as opposed to multilayer materials) was chosen because methods for fabrication could be standardized easily. All testing was conducted with a new TCI that had not been worn before.

The certified pedorthist or orthotist also provided a standardized MP (according to foot size) that had an adhesive backing and that could be placed in its appropriate location. Pilot testing and our clinical experience suggested that existing prefabricated rubber MPs were not large or stiff enough to make a meaningful reduction in forefoot pressures in this subject population. In addition, the x-ray attenuation of rubber is similar to that of soft tissue, making identification of the MT from computed tomography (CT) data difficult. Our pilot studies indicated that an MP made of cork has a CT value (Hounsfield units) different from that of soft tissue, making it easier to automatically isolate the MP from the foot for the measurements obtained. This material is used clinically to fabricate custom-made MPs and is stiff enough (shore value of 55) to cause tissue deformation. The size of the MP was intended to cover the central 3 metatarsals. The MP is shown in Figure 1, and the dimensions for each MP are shown in Table 2. The certified pedorthist or orthotist also placed on the positive plaster foot mold and TCI a line identifying the metatarsal heads along the contour of the metatarsal phalangeal joints. For this study, the certified pedorthist or orthotist attempted to place the distal aspect of the MP 1 cm proximal (toward the heel) to the line of the metatarsals. Spiral x-ray CT (SXCT) scanning allowed us to determine the placement of the MP in relation to the second metatarsal head.

Plantar Pressure Assessment

Plantar pressures were recorded during walking in all 3 footwear conditions by use of an F-Scan system([section]) and previously validated methods. (21-23) Any excessive callus, which could affect pressure measurements, was trimmed before testing. A new F-Scan pressure sensor was cut to fit the shoe of each subject. Three self-adhering spherical lead markers (1.5 mm) were placed on the forefoot portion of the sensor to enable the precise localization and identification of the sensor and orthotic device on the plantar surface of the foot during SXCT. The pressure sensor was attached to the subject's foot with tape, and a thin sock was placed over the foot to secure the sensor. The sensor was calibrated according to manufacturer guidelines and standardized techniques. (21,22) Data were collected at 50 Hz during 2 walking trials immediately after calibration. A mean of 3 representative steps (not highest or lowest) chosen during the middle portion of 1 walking trial was used for the PPP and PTI variables. The second walking trial was conducted to provide backup data but was not used unless there was a malfunction in the first walking trial. The order of testing for the 3 conditions was determined randomly by use of a prearranged schedule. Subjects were timed as they walked across a 6.1-m (20-ft) walkway but were allowed to walk at their preferred walking speed. Walking trials with speeds that differed by more than 5% between footwear conditions were repeated. The distribution and surface area of plantar pressures during walking were imaged on the computer monitor and used to help train the subject during subsequent simulated loading while seated in the SXCT scanner.

SXCT Scanning and Data Processing

Imaging was performed in a room adjacent to the room used for pressure testing. After acquisition of the pressure data during walking, subjects were immediately positioned on the loading device placed on the SXCT table to practice loading their foot in a manner similar to the way in which the foot was loaded during walking. The reliability and validity of these methods are described in detail elsewhere. (24,25) The subject sat in a modified car seat that could be adjusted so that the selected foot and ankle were positioned to allow only forefoot contact against a board. A digital readout and strain gauge was placed behind the board. The scale was used to measure the load applied to the plantar surface of the foot and to provide feedback to the subject regarding appropriate loading via a handheld digital readout. The back of the scale rested against a rigid acrylic vertical plate. The objective of the scale and loading device was to measure the load applied to the plantar surface while the foot was in a position that simulated PPP during the push-off phase of walking. Previous work indicated that the PPP on the forefoot typically occurs at 80% of the stance phase during walking when only the forefoot is in contact with the ground. (26) Previous work also indicated that pushing at 50% of body weight through a single foot was reasonable for subjects to perform and was a good surrogate for 80% of the stance phase during walking. (24,25) We recorded static PPP in the scanner to check this protocol. The subject's foot was scanned by SXCT in the 3 footwear conditions while the subject pushed against the loading device with 50% of his or her body weight.

Data Reduction and Analysis

First, the pressure sensor and MP were registered (aligned) with the bony anatomy of the foot by use of lead markers as described in detail elsewhere (Fig. 2). (27) Registration of the sensor with the bony anatomy allowed identification of the specific PPP under each metatarsal head. Metatarsal head centers were located and identified from the SXCT image date by use of Analyze software. ** The coordinates identifying the metatarsal head centers were transformed to the coordinate system of the pressure sensor by a previously described method, and reliability testing determined that the mean difference between repeated measures was less than [+ or -]0.11 pixel. (27) The location of the sensor pixel directly under each metatarsal head was entered into custom-made software. A region of interest that measured 7 sensor pixels (4 rows distal, metatarsal head row, and 2 rows proximal) by 3 sensor pixels (1 column of pixels on each side of the metatarsal head location) around the metatarsal head was identified. Therefore, the region of interest contained 21 sensor pixels. The PPP for this region was the highest pressure value within these 21 pixels. The PTI was calculated by summing the pressure values for all 21 pixels over 1 stance phase. The input to this program was an ASCII file containing all frames of pressure data recorded over 3 steps during the middle portion of a walking trial. The means of the PPP and PTI values under each metatarsal head over 3 steps were tabulated and entered into a database.

[FIGURE 2 OMITTED]

The PPP under the midshaft of the second metatarsal was determined by identifying the F-Scan sensor pixel under the second metatarsal head. The shaft of the second metatarsal was chosen because the MP was positioned to have its greatest effect at this location. This column of pixels and the adjacent columns were followed toward the midfoot, and the highest PPP that was outside the metatarsal head region of interest was identified in the MP condition by methods similar to those described above. This sensor pixel location was recorded, and the PPP for this sensor pixel location for each of the other 2 footwear conditions also was recorded. In addition, the distance of the sensor pixel from the metatarsal head center was used as the location to determine the STT under the second metatarsal shaft for each footwear condition.

The contact area between the foot and the shoe or orthotic device at the instant of PPP during walking for each footwear condition was calculated by use of the F-Scan software (version 4.21). All sensors that registered greater than 20 kPa were included in the calculation of the contact area.

The STTs under the center of each metatarsal head and at a location (mean [+ or -] SD) 3.5 [+ or -] 0.11 cm proximal to the second metatarsal head along the metatarsal shaft were determined from the SXCT image data by use of Analyze software and established, reliable methods. (24,28,29) The STT was determined from a sagittal slice through the metatarsal from the outer surface of the skin to the outer surface of the metatarsal for the 3 footwear conditions. A previous study indicated that the mean bias of repeat STT measurements under the metatarsal heads and midshaft region was less than 0.3 mm, with a standard deviation of less than 1.2 mm. (24)

Differences in the outcome measures between footwear conditions were determined for each metatarsal head by use of repeated-measures analysis of variance. When a significant overall effect was determined, paired t tests were conducted to determine whether there were significant differences among individual footwear conditions. The overall alpha level was set at .05.

Results

The PPP and PTI values during walking for each footwear condition are shown in Figure 3. There were significant main effects for PPP and PTI between footwear conditions at each metatarsal head (P<.005). Mean changes in PPP and PTI and individual statistical comparisons between footwear conditions for each anatomic location are shown in Table 3. There were significant individual differences between footwear conditions at each location (P<.05). In summary, compared with the shoe condition, the TCI condition reduced metatarsal head PPP by 19% to 24% and PTI by 16% to 23%. Compared with the TCI condition, the addition of the MP resulted in an additional 15% to 20% reduction in PPP and a 22% to 32% reduction in PTI at the metatarsal heads and a 308% increase in PPP at the midshaft of the second metatarsal.

[FIGURE 3 OMITTED]

The walking speeds (mean [+ or -] SD) were 0.91 [+ or -] 0.20 m/s for the shoe condition, 0.95 [+ or -] 0.21 m/s for the TCI condition, and 0.93 [+ or -] 0.22 m/s for the MP condition. This 4% difference in walking speed was not considered meaningful.

The contact areas (mean [+ or -] SD) between the foot and the footwear at the instant of PPP during walking were 75.1 [+ or -] 25.2, 102.0 [+ or -] 32.9, and 99.4 [+ or -] 30.9 [cm.sup.2] in the shoe, TCI, and MP conditions, respectively (different between conditions at P<.0001). Individual analyses indicated that there was a 27% increase in contact area in the TCI condition compared with the shoe condition (P<.0001) but no difference in contact area between the MP and the TCI conditions.

Figure 3C shows the PPP values obtained at each metatarsal head for each of the footwear conditions during SXCT. The patterns of pressure distribution and the effect of footwear on PPP values obtained during SXCT (Fig. 3C) were similar to the patterns of pressure distribution and the effect of footwear on PPP values obtained during the walking trials (Fig. 3A).

The STT values for each footwear condition are shown in Figure 4. There were significant main effects for changes in STT between footwear conditions at each location (P<.001). The mean changes in STT determined by individual comparisons between footwear conditions at each location are shown in Table 4. There were no significant differences in STT between the TCI condition and the shoe condition at any of the locations. There were, however, significant differences in changes in STT for metatarsal heads 2 to 5 and the second metatarsal midshaft location between the TCI condition and the MP condition (P<.005). Compared with the STT in the TCI condition, in the MP condition, the STT increased 8% to 22% at metatarsal heads 2 to 5 and decreased 14% at the second metatarsal mid-shaft location (Tab. 4).

[FIGURE 4 OMITTED]

The distal aspect of the MP was located (mean [+ or -] SD) 9.6 [+ or -] 6.6 mm away from the center of the second metatarsal head, and the apex (thickest point) of the MP was located 20.6 [+ or -] 6.1 mm away from the center of the second metatarsal head, as determined during SXCT testing (Fig. 2).

[FIGURE 2 OMITTED]

Discussion

Consistent with our hypothesis, both the TCI and the MP had substantial and additive effects in reducing the pressures at the metatarsal heads. Compared with the shoe alone, the TCI reduced metatarsal head PPP and PTI 16% to 24%, and the addition of the MP reduced pressures another 15% to 32% at the metatarsal heads. Therefore, the total amount of PPP and PTI reduction obtained with the TCI and the MP (compared with the shoe alone) was 29% to 47% at the metatarsal heads. These are substantial pressure reductions that would likely make a clinically meaningful difference. (4)

These pressure reductions obtained with a custom-made TCI are consistent with those reported in other studies. Several publications reported a reduction in plantar pressures of between 30% and 48% under selected metatarsal heads for subjects with DM and a history of ulcers or peripheral neuropathy when a custom-made orthotic device was used instead of therapeutic shoes alone. (5,13,30,31) Previous research also documented significant benefits (3% to 21% pressure reductions) of custom-made inserts over flat accommodative inserts. (16,32) Major benefits obtained with TCIs in therapeutic footwear are that they are relatively easy to fabricate and result in consistent pressure reduction.

The results of studies investigating the effect of an MP on metatarsal head plantar pressures are more variable. In 1 of the few studies investigating the effect of an MP on plantar pressures in people with DM and a foot deformity, Ashry et al (13) found a 41% to 55% pressure reduction when a custom-made Plastazote insert was used instead of no insert but no additional pressure reduction when an MP was added. The authors questioned the benefit of the MP but also wondered whether their MP was large enough to be effective. Bus et al (16) also studied people with DM and peripheral neuropathy and reported a 16% reduction in metatarsal head pressure when a custom-made insole that included a built-in MP was used instead of a 0.95-cm-thick flat insole made of PPT ([dagger])([dagger]); however, it is difficult to separate the effect of the MP from the effect of the rest of the custom-made insole. Other studies that have investigated the effect of an MP on plantar pressures in healthy people without foot impairments reported highly variable results ranging from a 28% increase to a 60% decrease in metatarsal head pressures. (11,12,17) All reports emphasized that pressure responses are variable and are dependent upon subject characteristics and MP differences (shape, size, location, and material properties).

Besides reporting the pressure reduction, we investigated the possible mechanisms of the pressure reduction. In this clinical situation, plantar pressure is equal to the weight-bearing and push-off forces divided by the contact area between the foot and the supporting surface. Plantar pressures typically are greatest at the metatarsal heads during the push-off phase of walking (80% of stance) because weight-bearing and push-off forces are greatest and the weight-bearing contact area is smallest (only metatarsal heads and toes are in contact with the ground). (26) Metatarsal head plantar pressures typically are even higher in people with DM and peripheral neuropathy because of forefoot deformities (ie, hammer toes) that reduce the effectiveness of the toes in bearing weight and reducing the area of contact of the forefoot with the floor. (10,33) In addition, soft tissues under the metatarsal heads tend to decrease in thickness (34,35) and increase in stiffness (35,36) in subjects with DM and peripheral neuropathy compared with control subjects without DM and peripheral neuropathy. These mechanical consequences, resulting from the physiological disturbances of DM and peripheral neuropathy, contribute to unnoticed, excessively high plantar pressures that can lead to skin breakdown. (1-3)

The TCI and the MP appear to help compensate for these musculoskeletal and integumentary impairments through 2 different mechanisms. The results of this study indicate that the TCI allowed a significant, 30% increase in contact area at the instant of PPP; this effect reduced PPP and PTI at the metatarsal heads by 16% to 24%. Other researchers reported an increase in con tact area of 5% to 30% during walking (30,32) and an increase of 63% during standing. (37) These results emphasize the effectiveness of the basic orthotic principle of increasing surface area to decrease excessive localized pressures. The stiffness of the TCI, reflected in its shore value, is another important factor in its ability to accommodate deformity and distribute pressures. (15) The shore value of the TCI used here was 35, and the TCI was somewhat stiffer than the plantar soft tissues in people without DM (estimated shore values of 16-21). (35) Research is needed to determine the optimal stiffness of an orthotic device for people who have DM and peripheral neuropathy and who appear to have skin that is stiffer than that of people without DM. (35,36)

The addition of the MP also reduced metatarsal head pressures, but it did not achieve this goal by increasing contact area. Rather, the MP helped to unload the metatarsal heads by loading the soft tissues and bony structures proximal (toward the heel) to the metatarsal heads. This transfer of load is evidenced by the decrease in STT at the metatarsal midshaft and the increase in STT at the metatarsal head (Tab. 4). Presumably, the reduced pressure and reduced soft-tissue compression at the metatarsal head reduce the trauma to the soft tissue in this area.

Although the MP reduced PPP at the metatarsal heads, the risk of using an MP is that pressures transferred to the metatarsal shaft may cause skin breakdown in this area. We have not observed skin breakdown in this area when using an MP, perhaps because of the conservative MP size used. The MP used in this study, however, was made of cork, was relatively stiff (shore value of 55), and was large (Tab. 2). The distal aspect of the MP was approximately 10 mm from the metatarsal heads, and the apex of the MP was approximately 21 mm from the metatarsal head center (Fig. 2). Perhaps because of its stiffness and large size, the MP even helped to reduce plantar pressures at the first and fifth metatarsal heads. Smaller or less stiff pads (felt or foam) may need to be placed closer to the metatarsal head to be effective. (11) In general, the larger or stiffer the MP, the more the load will be transferred to the location under the MP and away from adjacent metatarsal heads. In addition, the greater the load that is transferred, the more the MP has the potential to cause discomfort in a sensate person or skin breakdown in an insensate person at the location of the MP. The MP used in this study was not worn for an extended period of time, and we do not know whether it would cause skin breakdown in the region below it. Any addition of an MP, especially one of the stiffness and size described in this study, to a TCI for a person with peripheral neuropathy should be considered carefully in light of the potential for skin breakdown. We currently are conducting additional analyses to determine the relationship between pressures and soft-tissue deformation at the metatarsal shaft and the optimal placement of an MP to reduce metatarsal head pressures.

[FIGURE 2 OMITTED]

A benefit of this study was that imaging techniques were used to quantify the location of the metatarsal heads with respect to the plantar pressures and the MP. This technology allowed a more specific analysis of the effect of the orthotic device than has been attained previously.

This study shares the limitation of previous studies, however, in that we investigated only a specific type of TCI and MP. Results vary depending on the shape, position, and material properties of the orthotic device components. (11-13,16,17) This problem is particularly apparent in the use of the MP. Combinations of shapes, positions, and material properties are almost endless. In addition, responses likely will vary according to patient populations. These limitations emphasize the need to develop computational models (such as finite-element analysis) to help understand how stresses are distributed through the foot and how these stresses can be distributed optimally with orthotic devices or surgical procedures. Efforts are under way in several research laboratories to develop computational models that could help to optimize the design of orthotic devices and MPs. (35,38-40) The focused designs estimated by the computational models then could be tested experimentally in patient populations. Results such as those obtained in this study can be used to help test the validity of future computational models.

Other limitations of this study are that the STT was measured during SXCT rather than during walking and that plantar pressures are simply a surrogate measure of trauma to the skin. The results in Figure 3C indicate, however, that the effect of footwear on PPP during the loaded SXCT scan was similar to that during walking (Fig. 3A). Besides the measurement of plantar pressures, additional research is needed to determine whether the orthotic devices described in this study can help to reduce the incidence of skin breakdown in patients with DM and peripheral neuropathy. A benefit of studying plantar pressures and STT is that the mechanism of pressure reduction from the orthotic devices can be clarified.

Conclusion

The TCI and MP used in this study had substantial and additive effects at reducing pressures under the metatarsal heads. The TCI reduces excessive pressures at the metatarsal heads by increasing the contact area of weight-bearing forces. Conversely, the MP acts by compressing the soft tissues proximal to the metatarsal heads and relieving compression at the metatarsal heads. These findings can aid in the design of effective orthotic devices to relieve excessive plantar stresses that may contribute to skin breakdown and subsequent amputation in people with DM and peripheral neuropathy.

This article was received September 24, 2005, and was accepted January 3, 2006.

References

(1) Boulton AJM, Betts RP, Franks CI, et al. Abnormalities of foot pressure in early diabetic neuropathy. Diabet Med. 1987;4:225-228.

(2) 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;6:26-33.

(3) Pecoraro RE, Reiber GE, Burgess EM. Pathways to diabetic limb amputation: basis for prevention. Diabetes Care. 1990;13:513-521.

(4) American Diabetes Association. Consensus development conference on diabetic foot wound care. Diabetes Care. 1999;22:1354-1360.

(5) American Diabetes Association. Preventive foot care in diabetes. Diabetes Care. 2004;27:S63-S64.

(6) Edmonds ME, Blundell MP, Morris ME, et al. Improved survival of the diabetic foot: the role of a specialized foot clinic. Q J Med. 1986;60:763-771.

(7) Chantelau E, Haage P. An audit of cushioned diabetic footwear: relation to patient compliance. Diabet Med. 1994;11:114-116.

(8) 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;287:2552-2558.

(9) Brand PW. The diabetic foot. In: Ellenberg M, Rifkin H, eds. Diabetes Mellitus: Theory and Practice. New Hyde Park, NY: Medical Examination Publishing Co Inc; 1983:829-849.

(10) Bauman JH, Girling JP, Brand PW. Plantar pressures and trophic ulceration: an evaluation of footwear. J Bone Joint Surg Br. 1963;45: 652-673.

(11) Hayda R, Tremaine MD, Tremaine K, et al. Effect of metatarsal pads and their positioning: a quantitative assessment. Foot & Ankle International 1994;15:561-566.

(12) Holmes GBJ, Timmerman L. A quantitative assessment of the effect of metatarsal pads on plantar pressures. Foot & Ankle. 1990;11:141-145.

(13) Ashry HR, Lavery LA, Murdoch DP, et al. Effectiveness of diabetic insoles to reduce foot pressures. J Foot Ankle Surg. 1997;36:268-271; discussion 328-329.

(14) Mueller MJ, Strube MJ, Allen BT. Therapeutic footwear can reduce plantar pressures in patients with diabetes and transmetatarsal amputation. Diabetes Care. 1997;20:637-641.

(15) Janisse DJ. Pedorthic care of the diabetic foot. In: Bowker JH, Pfeifer MA, eds. Levin and O'Neal's The Diabetic Foot. Saint Louis, Mo: CV Mosby Inc; 2001:700-726.

(16) Bus SA, Ulbrecht JS, Cavanagh PR. Pressure relief and load redistribution by custom-made insoles in diabetic patients with neuropathy and foot deformity. Clin Biomech. 2004;19:629-638.

(17) Chang AH, Abu-Faraj ZU, Harris GF, et al. Multistep measurement of plantar pressure alterations using metatarsal pads. Foot & Ankle International. 1994;15:654-660.

(18) Diamond JE, Mueller MJ, Delitto A, Sinacore DR. Reliability of a diabetic foot evaluation [published erratum appears in Phys Ther. 1989;69:994]. Phys Ther. 1989;69:797-802.

(19) Maluf KS, Mueller MJ. Novel Award 2002. Comparison of physical activity and cumulative plantar tissue stress among subjects with and without diabetes mellitus and a history of recurrent plantar ulcers. Clin Biomech. 2003;18:567-575.

(20) Caselli A, Armstrong DG, Pham H, et al. The forefoot-to-rear-foot plantar pressure ratio is increased in severe diabetic neuropathy and can predict foot ulceration. Diabetes Care. 2002;26:1066-1071.

(21) Mueller MJ, Strube MJ. Generalizability of in-shoe peak pressure measures using the f-scan system. Clin Biomech. 1996;11:159-164.

(22) Nicolopoulos CS, Anderson EG, Solomonidis SE, Giannoudis PV. Evaluation of the gait analysis FSCAN pressure system: clinical tool or toy? The Foot. 2000;10:124-130.

(23) Pitei DL, Lord M, Foster A, et al. Plantar pressures are elevated in the neuroischemic and the neuropathic diabetic foot. Diabetes Care. 1999;22:1966-1970.

(24) Commean PK, Mueller MJ, Smith KE, et al. Reliability and validity of combined imaging and pressures assessment methods for diabetic feet. Arch Phys Med Rehabil. 2002;83:497-505.

(25) Mueller MJ, Smith KE, Commean PK, et al. Use of computed tomography and plantar pressure measurement for management of neuropathic ulcers in patients with diabetes. Phys Ther. 1999;79: 296-307.

(26) Kelly VE, Mueller MJ, Sinacore DR. Timing of peak plantar pressure during the stance phase of walking: a study of patients with diabetes mellitus and transmetatarsal amputation. J Am Podiatr Med Assoc. 2000;90:18-23.

(27) Hastings MK, Commean PK, Smith KE, et al. Aligning anatomical structure from spiral x-ray computed tomography with plantar pressure data. Clin Biomech. 2003;18:877-882.

(28) Smith KE, Commean PK, Mueller MJ, et al. Assessment of the diabetic foot using spiral computed tomography imaging and plantar pressure measurements. J Rehabil Res Dev. 2000;37:31-40.

(29) Smith KE, Commean PK, Robertson DD, et al. Precision and accuracy of computed tomography foot measurements. Arch Phys Med Rehabil. 2001;82:925-929.

(30) Albert S, Rinoie C. Effect of custom orthotics on plantar pressure distribution in the pronated diabetic foot. J Foot Ankle Surg. 1994;33: 598-604.

(31) Lord M, Hosein R. Pressure redistribution by molded inserts in diabetic footwear: a pilot study. Journal of Rehabilitation Research & Development. 1994;31:214-221.

(32) Tsung BYS, Zhang M, Mak AFT, Wong MWN. Effectiveness of insoles on plantar pressure distribution. J Rehabil Res Dev. 2004;41:767-774.

(33) Mueller MJ, Hastings MK, Commean P, et al. Forefoot structural predictors of plantar pressures during walking in people with diabetes and peripheral neuropathy. J Biomech. 2003;36:1009-1017.

(34) Gooding GA, Stess RM, Graf PM, et al. Sonography of the sole of the foot: evidence for loss of foot pad thickness in diabetes and its relationship to ulceration of the foot. Invest Radiol. 1986;21:45-48.

(35) Thomas VJ, Patil KM, Radhakrishnan S. Three-dimensional stress analysis for the mechanics of plantar ulcers in diabetic neuropathy. Med Biol Eng Comput. 2004;42:230-235.

(36) Klaesner JW, Hastings MK, Zou D, et al. Plantar tissue stiffness in patients with diabetes mellitus and peripheral neuropathy. Arch Phys Med Rehabil. 2002;83:1796-1801.

(37) Kato H, Takada T, Kawamura T, et al. The reduction and redistribution of plantar pressures using foot orthoses in diabetic patients. Diabetes Res Clin Pract. 1996;31:115-118.

(38) Lemmon D, Shiang TY, Hashmi A, et al. The effect of insoles in therapeutic footwear: a finite element approach. J Biomech. 1997;30: 615-620.

(39) Gefen A, Megido-Ravid M, Itzchak Y, Arcan M. Biomechanical analysis of the three-dimensional foot structure during gait: a basic tool for clinical applications. J Biomech Eng. 2000;122:630-639.

(40) Actis RL, Szabo BA, Commean PK, et al. Effect of modeling assumptions in the plantar pressure distribution of the diabetic foot using the p-version of the finite element method. In: Proceedings of American Society of Mechanical Engineering (ASME), International Mechanical Engineering Congress and Exposition (IMECE), Anaheim, Calif. 2004;59400.

* Bakelite Xylonite Ltd, London, England; distributed by Alimed Inc, 297 High St, Dedham, MA 02026.

([dagger]) Bio-Medical Instrument Co, 15764 Munn Rd, Newbury, OH 44065.

([double dagger]) Advanced Orthopedic Footwear, One Derby Square, PO Box 4425, Salem, MA 01970.

([section]) Tekscan Inc, 307 W First St, South Boston, MA 02127-1309.

** Biomedical Imaging Resource, Mayo Clinic, 200 First St SW, Rochester, MN 55905.

([dagger])([dagger]) Langer Inc, 450 Commack Rd, Deer Park, NY 11729.

MJ Mueller, PT, PhD, FAPTA, is Associate Professor, Program in Physical Therapy, Washington University School of Medicine, 4444 Forest Park Blvd, Box 8502, St Louis, MO 63110 (USA) ([e-mail]). Address all correspondence to Dr Mueller.

DJ Lott, PT, MSPT, CSCS, is a doctoral student, Movement Science Program, Program in Physical Therapy, Washington University School of Medicine.

MK Hastings, PT, DPT, ATC, is Assistant Professor, Program in Physical Therapy, Washington University School of Medicine.

PK Commean, BEE, is Senior Research Engineer, Mallinckrodt Institute of Radiology, Washington University School of Medicine.

KE Smith, BS, is Senior Research Engineer, Mallinckrodt Institute of Radiology, Washington University School of Medicine.

TK Pilgram, PhD, is Instructor, Mallinckrodt Institute of Radiology, Washington University School of Medicine.

Dr Mueller, Dr Hastings, Mr Commean, Mr Smith, and Dr Pilgram provided concept/idea/research design. All authors provided writing and data collection. Mr Lott, Mr Commean, Mr Smith, and Dr Pilgram provided data analysis. Dr Mueller provided project management, fund procurement, and facilities/equipment.

The study was approved by the institutional review board at Washington University School of Medicine.

Funding was provided by the National Center for Medical Rehabilitation Research, National Institutes of Health (RO1 HD36895). Mr Lott was supported by PODS I and II awards from the Foundation for Physical Therapy. The authors acknowledge the Prevention and Control Research Core of the Washington University Diabetes Research and Training Center (P60 DK20579) for assistance in subject recruitment. The authors acknowledge Richard Robb and his associates at the Mayo Biomedical Imaging Resource Clinic, Rochester, Minn, for providing the Analyze software.

Table 1.
Subject Characteristics

Characteristic                                        Value (a)


No. of subjects                                   20
Age (y)                                           57.3 [+ or -] 9.3
No. of men/women (% men)                          12/8 (60)
Body mass index ([kg.sup.2])                      32.5 [+ or -] 7.4
No. of subjects with type 1 or type 2 diabetes     5/15
Diabetes duration (y)                             16.1 [+ or -] 10.5
HbA1c (%)                                          8.1 [+ or -] 1.9
Vibratory perception threshold (V)                48.3 [+ or -] 4.1

(a) Reported as [bar.X] [+ or -] SD, unless otherwise indicated.
HbA1c=glycated hemoglobin.

Table 2.
Metatarsal Pad Sizes

Parameter (a)   Small (mm)   Medium (mm)   Large (mm)

A                   9.3          9.8           9.8
B                  11.0         11.0          11.0
C                  50.8         59.5          63.0
D                  65.6         70.2          74.3
E                  44.4         53.1          57.1
Shoe size          5-7          8-10          11-15

(a) Parameter definitions are shown in Figure 1.

Table 3.
Mean Percent Change in Peak Plantar Pressures and Pressure-Time
Integrals by Anatomic Location and Conditions Compared (a)

             % Change Between Conditions

             Peak Plantar Pressures

                             (Shoe + TCI
Metatarsal   (Shoe + TCI)    + MP) -
Location     - Shoe Only     (Shoe + TCI)

Head 1       -19.7 (.002)    -17.2 (.006)
Head 2       -19.0 (<.001)   -16.5 (.002)
Head 3       -22.4 (<.001)   -15.4 (.003)
Head 4       -21.6 (.003)    -15.0 (.002)
Head 5       -24.3 (<.001)   -20.4 (<.002)
Midshaft 2   +38.9 (.055)    +308 (<.001)

             % Change Between Conditions

             Pressure-Time   Integrals

                             (Shoe + TCI
Metatarsal   (Shoe + TCI)    + MP) -
Location     - Shoe Only     (Shoe + TCI)

Head 1       -18.6 (.005)    -24.0 (<.001)
Head 2       -20.1 (<.001)   -28.3 (<.001)
Head 3       -23.3 (<.001)   -31.8 (<.001)
Head 4       -18.3 (.014)    -25.6 (<.001)
Head 5       -16.1 (.036)    -22.4 (.001)
Midshaft 2

(a) Differences in pressure between conditions were tested for
statistical significance with 2-tailed paired t tests, and P
values are shown in parentheses. TCI = total-contact insert,
MP = metatarsal pad.

Table 4.
Mean Percent Change in Soft-Tissue Thickness by Anatomic
Location and Conditions Compared (a)

             % Change Between Conditions

                            (Shoe + TCI
Metatarsal   (Shoe + TCI)   + MP) -
Location     - Shoe Only    (Shoe + TCI)

Head 1       +2.0 (.271)    +3.4 (.113)
Head 2       +4.4 (.096)    +7.8 (.009)
Head 3       +3.8 (.1761)   +12.5 (<.001)
Head 4       +5.3 (.107)    +22.0 (<.001)
Head 5       +3.9 (.314)    +16.3 (<.001)
Midshaft 2   -2.8 (.082)    -14.3 (<.001)

(a) Differences in soft-tissue thickness between conditions were
tested for statistical significance with 2-tailed paired t tests,
and P values are shown in parentheses. TCI = total-contact insert,
MP-metatarsal pad.