Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Effect of a Walking Splint and Total Contact Casts on Plantar Forces

Andrew Novick, MA., P.T.
James A. Birke, M.S., P.T.
Sandra L. Graham, M.S., P.T.
Edward Koziatek, P.T.

Introduction

Neuropathic plantar ulcers of the foot have been successfully managed by a variety of treatment regimens. One of the earliest methods to promote healing was the application of plaster of paris (POP) casts (1). These devices reduce localized pressure to the lesion and immobilize the surrounding joints and soft tissues while preserving functional ambulation. The casts have either been conventional padded casts (2,3,4), or, more recently, total contact casts (TCCs) using a minimal amount of padding over the malleoli, posterior heel, and other bony prominences, with additional padding over the lesion (5,6,7,8,9,10).

In cases where a POP cast is either contraindicated or not accepted by the patient, alternative types of customized footwear have proven to be effective in healing plantar lesions. One such device is the posterior walking splint (5,6). It provides many of the same benefits as the POP cast, but can be easily removed and donned when access to the lesion is required. The POP-molded double-rocker plaster shoe (MDRPS) (4,8,11,12) and patellar-tendon-bearing (PTB) orthoses (13,14) have also been used successfully.

The primary cause of neuropathic plantar ulcers is the cumulative effect of repetitive mechanical stress to vulnerable areas of the insensitive foot, usually over a bony prominence (15,16,17,18,19). This information, combined with the clinical benefits of customized footwear, prompted gait studies to quantify the effect of these devices on plantar forces and pressure. Birke et al. demonstrated a reduction in pressure under the 1st and 3rd metatarsal heads (MTHs) with both a conventional padded POP cast and a TCC (20). Pollard similarly reported a reduction in vertical force under the 1st, 2nd/3rd, 4th and 5th MTHs and diminished mediolateral (ML) and anteroposterior (AP) shear stresses within a POP cast (21). Birke and Nawoczenski examined the effects of short leg and PTB walking orthoses and demonstrated reduced pressures under the 1st, 3rd and 5th MTHs and the heel (22).

As the design of custom footwear evolves with the acquisition of new knowledge and materials, these devices must continue to be studied from clinical and quantitative perspectives. This study examined the walking splint (SPLINT), the TCC with two different terminal devices (the rubber walking heel [TCC/WH] and the cast boot [TCC/CB]), and a conventional leather oxford shoe (SHOE). All of these devices have been proven to be clinically effective in the healing of plantar lesions, but quantitative data to substantiate the clinical results of the walking splint, or to determine which reduces plantar forces most effectively, were not available. Data with the SHOE were also collected to provide standard force measures during gait with which to compare the TCC/ WH, TCC/CB and SPLINT treatment conditions.

The specific purposes of this study were to

1. substantiate the clinical effectiveness of the SPLINT by demonstrating a reduction in force at the primary weight-bearing sites of the foot when compared to the SHOE,
2. determine if there was a significant difference in force measures between the two TCC conditions (TCC/WH and TCC/CB) as compared to the SPLINT and
3. determine if there was a significant difference in force measures within the TCC, between the TCC/WH and TCC/ GB conditions.

Additionally, the TCC/CB and TCC/WH were compared with the SHOE to substantiate the previous findings of Birke et al. and Pollard (20,21).

Methods and Procedures

Subjects

Twenty normal subjects, 11 males and nine females, between the ages of 21 and 63 years, participated in this study. The mean age for all subjects was 36.5 (+/- 10.4) years, with the mean age for males and females being 35.5 (+/-6.9) and 37.7 (+/-13.8) years, respectively. All subjects were free of clinical symptoms in the lower extremity and presented with normal sensation on the plantar aspect of the right foot.

Instrumentation

Capacitive pressure transducers 2 mm thick and 1.5 cms in diameter were calibrated as described by the manufacturer (Hercules Orthoflex Data System, Allegheny Ballistics Lab, Cumberland, Md.). The calibration signal for each transducer was sampled with a Digital Equipment Corp. (DEC) Analog Data Module (ADM) on a DEC Pro3S0 computer (Digital Equipment Corp., Maynard, Mass.). The calibration data was then transferred to a DEC VAX 11/750 for computation of calibration values. Subject data collected during the walking trials were similarly collected on the Pro3SO and analyzed on the VAX. Subject data were sampled at a rate of 60 hertz for a total of 8.35 seconds, the longest interval of time allowed by the ADM software. A four-channel strip-chart recorder was used to visualize the data being sampled by the computer, which were converted to force data by factoring out the surface area of the transducers (Gulton TR 4000A, Gulton Industries, Inc., East Greenwich, R.L.). This was done to eliminate size of the body part as a variable, since the 1.5cm diameter transducers were placed under the larger calcaneus and midfoot sites, and the much smaller metatarsal heads.

Testing Procedures

The four transducers were secured to the plantar aspect of the right foot at the heel, midfoot (at the level of the navicular), 4th MTH, and 2nd MTH to provide a representative sampling of forces along both the length of the foot and across the entire width of the forefoot. Each transducer was held securely in place with double-sided tape between the transducer and the skin, then further secured with Transpore tape overlying the transducer (3M Co., 3M Center, St. Paul, Minn.) Cotton stockinette covered the foot and remained intact for the duration of the study. A second stockinette, which became incorporated into the plaster devices, was applied over the first. This allowed the four transducers and two stockinettes to remain as a constant interface between the foot and inside of all the devices.

Each subject participated in the following four treatment conditions:

1. plaster walking splint (SPLINT),
2. total contact cast with a rubber walking heel (TCC/WH),
3. total contact cast with cast boot (TCC/ CB) and
4. leather oxford shoe (SHOE).

The testing order of these devices was randomly assigned to avoid bias.

Fabrication of the Plaster Devices

Construction of the total contact casts and walking splint was similar to that described by Birke et al. for clinical use (6), but slightly modified to ensure a standard interface between the foot, transducers and plaster. None of the plaster devices incorporated foam padding over a plantar bony prominence, as would be placed over an open ulcer clinically to further reduce force and promote healing.

Total Contact Cast (Figure 1 and Figure 2 )

Modifications to the described method include the following:

1. Webril (Kendall Co., Boston, Mass.) was used to enclose the toes, which are normally covered with Sifoam (Knit Rite, Kansas City, Mo.), and
2. two rolls of Scotchcast fiberglass resin tape (3M Co., 3M Center, St. Paul, Minn.) were applied to allow weightbearing following a 20-minute drying period.

If randomization of the treatment conditions required the TCC/WH to be run first, a layer of plastic food wrap (Dow Consumer Products, Inc., Indianapolis, Ind.) was placed over the plantar surface of the initial layer of fiberglass tape before the rubber walking heel was secured. This allowed easy removal of the heel so that the same total contact cast could be used for both the TCC/ WH and TCC/CB conditions. The cast boot used for the TCC/CB condition contained a neoprene crepe sole incorporating a rocker design, a cut-back heel and a flat sole under the middle third of the device (AliMed, Inc., 90Dedham, Mass.).

Walking Splint (Figure 3)

The walking splint was similarly fabricated according to Birke et al. with the following modifications:

1. the application of Webril under the toes replacing Sifoam and
2. the use of a stockinette instead of Webril to enclose the leg and foot (6).

Data Collection and Analysis

After application of the first device, the subject assumed the starting position and began walking. Data collection commenced after the subject had walked approximately 20 feet to allow the subject to assume his/her normal gait characteristics. Gait data were collected for the allotted 8.35 seconds and completed prior to the end of the walking trial to prevent sampling of data while the subject was decelerating. The same procedure was repeated for each device. All subjects performed two walking trials in each of the four treatment conditions.

The calibration and raw subject data were transferred to the VAX for analysis. The second of the two data sets, which typically presented less noise when the output was viewed from the strip-chart recorder, was used for analysis for most of the treatment conditions. (This difference was due to the subject's being more acclimated to walking in the device after the first trial.) However, in those few cases when more noise was observed in the 2nd trial due to a misstep by the subject during gait or to equipment malfunction, the initial trial was used. Transducer malfunction occurred at one site for two subjects and at two sites for one subject, resulting in the elimination of data sets at these sites for all four treatment conditions for these subjects (a total of 16 out of a possible 640 datasets).

A calibration value for each transducer was initially calculated prior to data collection. The peak force values for all stance phases contained within the total 8.35-second dataset for each treatment condition trial were then identified. The first and last peak values were disregarded, and a mean peak value for each transducer was calculated for each treatment condition trial for each subject. The mean peak values for the SPLINT, TCC/WH and TCC/CB conditions were also compared to that of the standard leather oxford shoe to calculate a percent change in force from the SHOE. An analysis of variance (ANOVA), SAS, Proc GLM, was then performed on the force values.

Results

Means and standard deviations for the force measures are reported in Table 1 . The ANOVA results (Table 2) show significant differences in the force measures among the four treatment conditions at all four sites each having a significance level of p<0.001 A Duncan's multiple range test was performed to determine which treatment condition(s) differed (Table 3) . At each site, the SHOE condition significantly differed from the other three treatments (collectively termed healing footwear), in that force values when wearing the SHOE were greatest at the heel, 4th MTH and 2nd MTH and least at the midfoot site (Table 3) . The TCC/WH, TCC/CB and SPLINT, therefore, all significantly reduced force at these weight-bearing locations of the foot and conversely increased force in the midfoot when compared to the SHOE. The percentage change of force with the healing footwear at each site when compared to the SHOE is presented in Figure 4 with the accompanying letter designation from the Duncan's test.

At the heel and 4th MTH, the significant difference between the SHOE and the other three treatment conditions was the only finding (Table 3) . Additional differences were found between the TCC/WH and the SPLINT at the midfoot, and between the TCC/WH and both the SPLINT and TCC/ CB at the 2nd MTH. In the midfoot, the SPLINT resulted in the maximal force and correspondingly the greatest increase when compared to the SHOE (14.39 lbs., +/- 89.0 percent change). At the 2nd MTH, the TCC/ WH resulted in the least amount of, and greater reduction in, force (14.80 lbs., -59.9 percent change).

Discussion

The fundamental benefit demonstrated by use of the healing footwear was the reduction of force at the primary weight-bearing areas of the foot, which correspond to the most common sites of plantar ulceration and is the basis for the clinical effectiveness in the healing of plantar lesions (23). Additional reduction of force is obtained clinically by the addition of the Sifoam pad over the lesion, acting as an interface between the lesion and the plaster shell to remove all hard weight-bearing surface away from the wound. Figure 4 shows a minimum decrease of -11.6 percent at the heel, -35.9 percent at the 4th MTH, and -45.9 percent at the 2nd MTH for all of the healing footwear conditions. This was achieved by providing total contact to the entire plantar surface of the foot, thereby creating a larger surface onto which weight-bearing forces were distributed. Since weight-bearing forces, or body weight, remain constant, the amount of force applied to any specific unit area on the plantar surface of the foot will be dependent upon the total area available for distribution of the total force. Increasing the total surface area decreases the force exerted on a specified unit area of the foot, which is the concept of pressure, since pressure is defined as the force per unit area.

The healing footwear effectively redistributes force and is demonstrated by the increases of +52.5 percent, +/-69.7 percent, and + 89.0 percent in the midfoot for the TCC/WH, TCC/CB and SPLINT conditions, respectively, when compared to the SHOE (Figure 4) . This area of the foot not normally subjected to large weight-bearing loads is now being used. Additionally, weight-bearing forces are distributed to the lower leg through the proximal extension of the plaster to further unload the forefoot and rearfoot.

To demonstrate the distribution of the weight-bearing load within the plaster devices, Table 4 shows the force value at each location, the total summed force and the percentage of the total summed force at all four sites for that specific treatment condition. Overall, the summed total decreased from a value of 87.2 lbs. for the SHOE condition to 59.1 lbs. for the TCC/WH, 61.6 for the TCC/ CB, and 65.5 for the SPLINT, indicating a reduction in applied force to the foot when the healing footwear is worn. For the SHOE condition, the 2nd and 4th MTH sites are subjected to 37.2 percent and 24.3 percent of the summed load, respectively. These values are reduced to a maximum of 27.7 percent for the 2nd MTH and 21.1 percent for the 4th MTH with the healing footwear, showing a shift from the forefoot to the rearfoot in the magnitude of the percentage of the summed weight-bearing load. Conversely, the percentage of the summed load at the midfoot for the SHOE was 10.3 percent, which increased to a minimum of 20.4 percent for the healing footwear. Surprisingly, although the actual force value at the heel was reduced for all the healing footwear conditions compared to the SHOE, the percentage of the summed load borne by the heel actually increased from 28.2 percent in the SHOE condition to within a range of 30.3 percent to 35.6 percent for the healing footwear. The following is a discussion of the results at each of the four transducer sites.

Heel

The three healing footwear conditions showed significantly lower forces than the SHOE. The mean force values of 21.60 lbs., 20.92 lbs., and 18.69 lbs. for the TCC/WH, SPLINT, and TCC/CB conditions, respectively, translate to a - 11.6 percent, -11.9 percent and -20.6 percent decrease in force when compared to the 24.60 lbs. recorded for the SHOE. The decrease for the TCC/ CB condition (-20.6 percent) was nearly double that for the TCC/WH and SPLINT and could be due to the cushioning properties of the crepe sole. The degree of force reduction at the heel was less than at the forefoot sites and is consistent with existing literature (20).

Midfoot

All three healing footwear conditions showed a significant increase in force when compared to the SHOE. This is due to better utilization of the medial longitudinal arch as a weight-bearing structure. The total contact construction of the healing footwear raises the insole of the device upward against the arch to provide a base of support and allows better contact by which to absorb weightbearing forces. By contrast, the relatively flat insole of the shoe allows the arch to maintain its more elevated position, thereby reducing the load at the midfoot. This is similar in function to the localized relief built into the healing footwear with the placement of Sifoam padding over the lesion. In both situations, the gap created between the plantar aspect of the foot and the underlying footwear serves to minimize contact and reduce weight-bearing forces.

Additionally, differences were found within the healing footwear group. The SPLINT transmitted the greatest amount of force to the midfoot (14.39 lbs.), and was statistically different than the TCC/WH (12.06 lbs.). These values, when compared to the SHOE (8.98 lbs.), correspond to increases of + 89.0 percent for the SPLINT, + 69.7 percent for the TCC/CB, and + 52.5 percent for the TCC/WH.

Construction of the SPLINT and that of the TCC/WH share some basic features, such as the total contact design and the rubber walking heel. When comparing the results between these two devices at the midfoot, the differences can perhaps be explained by how well each is secured to the leg. With the TCC/WH, the rigid plaster completely surrounds the lower leg, reducing extraneous motion between the leg and cast to a minimum. However, the SPLINT is essentially a posterior splint and must be secured to the leg with an elastic wrap. This provides less constraint on both vertical and horizontal motion of the leg inside the SPLINT and could permit greater compressive plantar forces. Shearing forces on the plantar foot would also likely increase, although the system used in this study cannot measure these forces. It must be noted, however, that despite the presumed increased mobility and the measurable differences in the midfoot and 2nd MTH (see below), healing rates for plantar lesions using either the SPLINT or TCC/WH are comparable (5).

4th MTH

The only significant finding was the reduction in force of the healing footwear in comparison with the SHOE. The mean force values for the SPLINT (13.02 lbs.), TCC/CB (13.01 lbs.), and TCC/WH (11.21 lbs.) represent reductions of -35.9 percent, -36.4 percent and -44.8 percent, respectively, when compared to the SHOE (21.20 lbs.), and they are more than twice the magnitude of those for the heel. This greater reduction can be attributed to proportionately larger percentages of the summed load being borne by the heel and midfoot (Table 4) when compared to the maximal load under the forefoot.

2nd MTH

The three healing footwear conditions all showed significant reductions of force when compared to the SHOE. The force values 17.16 lbs., 17.05 lbs., and 14.80 lbs. for the SPLINT, TCC/CB, and TCC/WH, respectively, represent reductions of -45.9 percent, -46.8 percent, and -52.9 percent when compared to the SHOE. The order of decrease is similar to that for the 4th MTH, with the SPLINT showing the least force reduction, followed by the TCC/CB and then the TCC/WH. The lower decrease with the SPLINT can again be explained by the greater degree of movement occurring within the SPLINT compared to the TCC/CB and TCC/WH conditions. The magnitude of reduction under the 2nd MTH ranges between 1.18 and 1.28 times the reduction measured under the 4th MTH for each device.

Additional differences were found between the TCC/WH and the remaining two devices. The 14.80-lb force value for the TCC/WH is significantly less than the values measured for the SPLINT and TCC/CB, and the -52.9 percent force reduction is over 13 percent greater than the force reductions for the SPLINT and TCC/CB. The difference between the TCC/WH and TCC/CB conditions may be due to the rocker design of the TCC/CB, which would allow more direct force application under the forefoot during the latter stages of stance phase as weight is transferred forward over the front of the TCC/CB. The lack of a difference between the SPLINT and TCC/CB could again be explained by movement within the SPLINT, negating the benefit of the rubber walking heel used with the SPLINT.

Summary

When compared to a conventional leather oxford shoe, all three healing footwear devices were effective in reducing force at the primary weight-bearing areas of the foot, while increasing the distribution of force in the midfoot region. The total contact design of the devices more evenly distributed weight-bearing forces throughout the entire plantar surface of the foot and lower leg, including those areas that do not normally bear a large weight-bearing load. This reduces the amount of force applied to a specific unit area of the foot, which is equivalent to reducing the pressure. The use of foam relief over a lesion site during clinical application would further enhance the force reducing effects demonstrated in this study.

Referring to the three original purposes of the study, the following conclusions are presented:

1. a.) The SPLINT significantly reduced the force in both the rearfoot and the forefoot when compared to the SHOE.
   b.) The SPLINT resulted in significantly greater forces distributed to the midfoot than the SHOE (and the TCC/ WH).
2. a.) There was no difference between the SPLINT and TCC/CB at the 2nd MTH, but both were significantly less effective than the TCC/WH.
   b.) There were no significant differences in force reduction between the SPLINT and either the TCC/CB or the TCC/WH at the heel and 4th MTH.
3. The only difference between the TCC/WH and TCC/CB conditions occurred at the 2nd MTH, with the TCC/WH more effectively reducing force at this site.

When considering the effect of the healing devices on the different regions of the foot, these conclusions can be made:

1. The most effective device for reducing force under the forefoot is the TCC/ WH, although the reduction is statistically significant only at the 2nd MTH.
2. Under the heel, no statistically significant differences were found.
3. At the midfoot, the SPLINT and TCC/ CB produced the greatest increase in force measures, with the SPLINT being most effective.

These findings provide quantitative data illustrating significant force reduction of the three devices compared to the SHOE to support their use in the healing of plantar lesions (5). Although some statistical differences were found among the healing footwear, the magnitude of difference between the force values was not very large (less than 2.4 lbs.). Thus, regardless of the type of healing device used (a walking splint or one of the two types of total contact casts), or the type of terminal device on the bottom of the healing footwear (rubber walking heel or cast boot), similar results were obtained.

This information will benefit all clinicians involved in the management of pressure-related neuropathic ulcers by showing the effectiveness of these orthotic devices in reducing plantar pressure.

Andrew Novick, MA., P.T., is a research physical therapist, Paul W. Brand Biomechanics Lab, Rehabilitation Research, Gillis W. Long Hansen's Disease Center, Carville, La. 70721.

James A. Birke, M.S., PT., is chief, Physical Therapy Department, Gillis W. Long Hansen's Disease Center, Carville, La. 70721.

Sandra L. Graham, MS., P.T., is senior staff physical therapist, U.S. Public Health Service Alaska Native Hospital, Anchorage, Alaska 95510.

Edward Koziatek, P.T., is chief, Physical Therapy Department, U.S. Coast Guard Academy, New London, Conn. 06320.

References:

1. Coleman WC, Brand PW, Birke JA. The total contact cast. Journal of the American Podiatry Association 1984;74: 11:548-52.
2. Brindley GW, Cofield RH. Local treatment for neuropathic foot ulceration and osteoarthropathy. Foot and Ankle 1985;5:5:245-50.
3. Pollard JP, LeQuesne LP. Method of healing diabetic forefoot ulcers. British Medical Journal 1983;286:436-7.
4. Pring DJ, Casiebanca N. Simple plantar ulcers treated by below-knee plaster and moulded double-rocker plaster shoe: a comparative study. Leprosy Review 1982;52:261-4.
5. Birke JA, Coleman WC, Koziatek E, Patout CA, Novick A. Healing rates of plantar ulcers in patients with diabetes mellitus and leprosy. Unpublished study 1990.
6. Birke JA, Graham SL, Novick A, Coleman WC. Methods of treating plantar ulcers. Physical Therapy, February 1991;71:2:116-22.
7. Helm PA, Walker SC, Pullium G. Total contact casting in diabetic patients with neuropathic foot ulcerations. Archives of Physical Medicine and Rehabilitation 1984;65:691-3.
8. Kaplan M, Gelber RH. Care of plantar ulcerations: comparing applications, materials and non-casting. Leprosy Review 1988;59:59-66.
9. Sinacore DR, Mueller MJ, Diamond JE, Blair VP, Drury D, Rose S. Diabetic plantar ulcers treated by total contact casting. Physical Therapy 1987;67:10:1543-9.
10. Walker SC, Helm PA, Pullium G. Total contact casting and chronic diabetic neuropathic foot ulcerations: healing rates by wound location. Archives of Physical Medicine and Rehabilitation 1987 ;68:217-21.
11. Diamond JE, Sinacore DR, Mueller Mi. Molded double-rocker plaster shoe for healing a diabetic plantar ulcer. Physical Therapy 1987;67: 10:1550-2.
12. Joseph B, Joshua 5, Fritschi EP. The moulded double-rocker plaster shoe in the field treatment of plantar ulcer. Leprosy Review 1983 ;54:39-44.
13. Gordon V. Treatment of the patient with calcaneal and plantar pressure ulcers of neurogenic etiology. Journal of the American Osteopathic Association 1976;75:525-6.
14. Gristina AG, Nicastro JF, Clippinger F, Rovere GD. Neuropathic foot and ankle patellartendon-bearing orthosis as an adjunct to patient management, Orthopedic Review 1977;6:5 :53-9.
15. Brand PW. The insensitive foot (including leprosy). In: Jahass MH, ed. Disorders of the foot 1982;2: 1266-86.
16. Brand PW. The diabetic foot. In: Ellenberg M, Rifkin H, eds. Diabetes mellitus: theory and practice 3rd Ed. 1983:829-49.
17. Enna CD, Brand PW, Reed JK, Welch D. The orthotic care of the denervated foot in Hansen's Disease. Orthotics and Prosthetics 1976;30: 1:33-9.
18. Geary NPJ, Klenerman L. The insensitive foot, Northwick Park experience. Leprosy Review 1987;58:79-84.
19. McFarlane RM. The treatment of ulcers of the heel. Geriatrics, April, 1965:326-31.
20. Birke JA, Sims DS, Buford WL. Walking casts: effect on plantar pressures. Journal of Rehabilitation Research and Development 1985 ;22:3: 18-22.
21. Pollard JP, LeQuesne LP, Tappin JW. Forces under the foot. Journal of Biomedical Engineering 1983;5:37-40.
22. Birke JA, Nawoczenski DA. Orthopedic walkers: effect on plantar pressures. Clinical Prosthetics and Orthotics 1988;12:2:74-80.
23. Birke JA, Sims DS. Plantar sensory threshold in the ulcerative foot. Leprosy Review 1986 ;57:261.