Effect of Achilles tendon lengthening on ankle muscle performance in people with diabetes mellitus and a neuropathic plantar ulcer.

One of the most serious complications involving patients with diabetes mellitus (DM) and associated loss of protective sensation is the development of recurrent ulcers on the plantar surface of the foot. (1-3) Limited ankle dorsiflexion (eg, equinus deformity) has been implicated as a contributing factor in recurrent ulceration, presumably because this deformity prevents the leg from rolling over the foot during the late stance phase of gait, resulting in excessive plantar pressures. (4-7) Excessive plantar pressures result in tissue breakdown and delayed wound healing. (1,8)

Total-contact casting (TCC) is a common method used to manage plantar ulcers in people with DM. The effectiveness of TCC is believed to be due primarily to a reduction in plantar pressures at the ulcer site. (9,10) Although TCC is effective at healing ulcers initially, (11-15) the rate of reulceration following cast removal is high. (16-18) Tendo-Achilles lengthening (TAL) has been performed in this population with the rationale that surgical lengthening of the Achilles tendon will increase ankle dorsiflexion range of motion (DF-ROM), reduce plantar pressures, and prevent skin breakdown. (5-7,19) Our recent controlled clinical trial (20) indicated that risk reduction for short-term (7 months) and long-term (2.1 years) ulcer recurrence was 75% and 53%, respectively, for subjects who received TAL and TCC compared with those who received TCC alone. Studies also have indicated that the TAL resulted in substantial increases in ankle DF-ROM (9[degrees]-19[degrees]) (5,6,20) and short-term (7 months) reductions in forefoot peak plantar pressures. (6,20) Tendo-Achilles lengthening also affects ankle muscle performance, (7,20) presumably because of the acute change in length-tension relationships of the gastrocnemius and soleus muscles. (21,22) Plantar-flexor peak torque has been reported to decrease about 21% to 32% 8 weeks following TAL. (7,20) A reduction in ankle muscle performance following TAL could be especially problematic for people with DM because typically ankle muscle performance is already compromised in this population. (23-26) Both plantar-flexor peak torque (27-29) and passive torque (29) have been reported to be reduced in people with DM and positively correlated with gait measures, such as walking speed, (29) plantar-flexor moment, (28,29) and plantar-flexor stiffness. (9) A further reduction in ankle muscle performance from a TAL procedure could have a substantially negative impact on the walking ability of individuals with DM, who also have loss of protective sensation.

In a previously published article on our investigation of the effect of TAL on wound healing, (20) we briefly reported on the effect of TAL on plantar-flexor peak torque. We acknowledge that the plantar-flexor peak torque and DF-ROM data in the current article were published in our previous article. (20) The purposes of this article are to expand upon the previous report and to describe the effects of TAL on the torque-generating behavior of the ankle muscles in individuals with DM and peripheral neuropathy. Specifically, we will report on concentric plantar-flexor and dorsiflexor muscle peak torque, peak torque angle, passive plantar-flexor torque at 0 degrees of dorsiflexion, and maximal DF-ROM. Based on our previously published case report, (7) we hypothesized that peak plantar-flexor torque and passive torque (passive torque at 0[degrees] of dorsiflexion) would be reduced initially after surgery (only one group had surgery), but would recover to the baseline level within 8 months. We also speculated that the angle of plantar-flexor peak torque would shift into more DF-ROM following TAL.

Materials and Methods

This study was part of a larger randomized controlled clinical trial. (20) Subjects were randomly assigned to participate in 1 of 2 groups. The TAL group received the TAL procedure and TCC. The TCC group received TCC alone. Tests were conducted at 3 time points. The first test (pretest) was conducted an average ([+ or -]SD) of 10 [+ or -] 16 days before the intervention. The second test (initial posttest) occurred an average of 17[+ or -]29 days after conclusion of primary treatment (which included immobilization with TCC for both groups) and healing of the plantar forefoot wound. The third test (8-month posttest) occurred 8[+ or -]2 months alter wound closure. The period of 8 months was chosen because we speculated that the ankle muscles would have an adequate time to rehabilitate and we could carefully monitor the time period when ulcers are most likely to reoccur.

Subjects were considered for inclusion in this controlled clinical trial if they had a history of DM, loss of protective sensation (unable to sense a 5.07 Semmes-Weinstein monofilament on a least one location on the plantar surface of the foot (30)), maximal passive DF-ROM of 5 degrees or less, and a recurrent or nonhealing forefoot ulcer (Wagner scale grade II (31)). A limitation of 5 degrees of DF-ROM was chosen because most authors believe that at least 10 degrees of DF-ROM is required for normal ambulation. (32) A recurrent or nonhealing ulcer was defined as at least the second occurrence of a plantar ulcer or previous failure to heal a plantar ulcer with the use of TCC.

Subjects were excluded for consideration if they would not benefit from a TAL procedure (ie, were nonambulatory), had a history of cerebrovascular accident or other neurological problem complicating their rehabilitation, had a history of hindfoot Charcot fractures, had an ankle-arm index <0.45 (to rule out severe vascular problems), or were unable to tolerate the anesthesia required for TAL. We did not exclude midfoot or forefoot Charcot deformities or partial foot amputations. Additionally, subjects were excluded from this portion of the project if data on muscle performance were missing for any of the 3 testing sessions.

Randomization began in 1998 and was stopped in 2002. An a priori power analysis was conducted to predict the number of subjects needed for the plantar-flexor muscle performance outcome variables. The literature contained only studies on the effect of a TAL procedure on DF-ROM (maximal dorsiflexion angle). Based on these studies, effect size was estimated conservatively at 50%. (5,6) Being conservative with our expectations, we estimated that a sample size of 60 people would allow detection of a 25% effect size with a power of .80 and the alpha level at .05. (33) Because the effect size of intervention was greater than anticipated for all outcome measures, testing was terminated in 2002 with the 64 subjects described in our previous article (20) and the subset of 29 subjects reported in this article. A subset of subjects from the previous study (20) was used for this study for several reasons. The primary wound healing outcomes reported in our previous article (20) (percentage of wounds healed and percentage of wounds that reoccurred) were frequency-type data and required a greater number of subjects to achieve adequate power compared with the ratio data reported in this article. Therefore, the original study was designed to conduct extensive testing (ie, muscle performance measures) on a subset of subjects because of the cost and time involved in additional testing. Finally, only the data for those subjects who had measurements available for all 3 testing sessions could be included in the statistical analysis for this report.

Subjects were recruited from the Diabetic Foot Center at Barnes Jewish Hospital associated with Washington University School of Medicine in St Louis, Mo. Informed consent was obtained from all subjects who agreed to participate using a form approved by the Institutional Review Board at Washington University. Subjects were randomly assigned to the TAL group or the TCC group using a prearranged schedule. (20) Once a subject agreed to participate, he or she was referred to the patient coordinator for the study, who assigned the subject according to the prearranged schedule and arranged all testing sessions.

Twenty-nine subjects met the study inclusion criteria and agreed to participate. Fifteen subjects (3 female, 12 male) were randomly assigned to the TAL group, and 14 subjects (4 female, 10 male) were randomly assigned to the TCC group. Semmes-Weinstein monofilament sensory testing and a hemoglobin [A.sub.1c] (Hb [A.sub.1c]) blood test were conducted to characterize the subjects. Methods of sensory testing followed a previously described reliable technique. (4,30) Subject characteristics for each group are described in Table 1. Randomization methods were successful because there were no differences between groups in any subject characteristic listed (P>.05). Overall, subjects were 55[+ or -]10 ([bar.X][+ or -]SD) years of age and predominantly male (22 male, 7 female), with type 2 DM (21 subjects with type 2 DM, 8 subjects with type 1 DM) for a duration of 19[+ or -]12 ([bar.X][+ or -]SD) years. All subjects had severe peripheral neuropathy and lacked protective sensation as evidenced by a history of a plantar ulcer and the inability to sense the 5.07 Semmes-Weinstein monofilament on at least one location on the plantar surface of the foot. (30)

Methods of treatment have been described in detail previously. (20) Briefly, all subjects received identical treatment except for the TAL procedure and initial weight-bearing status as described below. All necrotic tissue and callus surrounding the ulcer were sharply debrided. The wound was covered with a dry gauze dressing. The subjects in the TAL group were positioned supine on the operating table, and intravenous sedation was administered. Local anesthesia was injected, and 3 hemisections were made in the Achilles tendon using the Hoke triple hemisection technique. (34) Then the surgeon firmly pushed the plantar forefoot into dorsiflexion until the foot could be brought into about 10 degrees of dorsiflexion, based on a visual estimation. The orthopedic physician was careful to avoid excessive force that might cause complete transection or over-lengthening of the tendon. No sutures were used to close the 3 tenotomy sites, and a dry gauze 4 x 4 dressing was applied and held in place with a sterile cotton wrap.

After the TAL procedure, subjects were immobilized with TCC to reduce forefoot pressure, to facilitate plantar wound healing, and to protect the ankle, foot, and tendon during the healing process. (9) The cast was applied as described previously, (35) except the distal end of the toe box was left open and a standard rocker cast shoe was used rather than a walking heel. The cast was applied to the lower leg with the ankle joint in a neutral position (ie, 0[degrees] of dorsiflexion). The cast was initially changed after 1 week and was subsequently changed every 2 to 3 weeks for at least 6 weeks or until complete healing of the forefoot ulcer. Partial weight bearing was allowed in the cast immediately, and after the first week the subject progressed to full weight bearing but was asked to limit his or her activities as much as possible. After casting, the involved foot was placed in a padded diabetic pressure-relief walking boot (DH Pressure-Relief Walker *) for 1 to 4 weeks until the subjects felt stable enough to walk with their extra-depth shoes with custom-molded inserts that were prescribed using published recommendations. (36) Subjects participated in a home exercise program provided by a physical therapist as described below.

Subjects in the TCC group were treated with a total contact cast using identical methods as the TAL group except that subjects were allowed to fully bear weight immediately after initial application of the cast. The ankle was positioned as close to neutral as possible, and the cast was changed every 2 to 3 weeks until the plantar ulcer was healed. Subjects then were instructed to wear their extra-depth shoes with custom-molded inserts. (36) There was no difference in days immobilized with TCC between the 2 groups (Tab. 1).

After treatment with TAL or TCC, all subjects were instructed in a home exercise program by a physical therapist using Thera-Band ([dagger]) to provide resistance to musculature around the ankle. The exercise program included use of red Thera-Band (moderate resistance), progressing to green Thera-Band (heavy resistance) to resist ankle plantar-flexion, dorsiflexion, inversion, and eversion movements. Subjects completed 3 sets with 10 repetitions in each set, one time per day, 3 to 5 days per week. (37)

Concentric plantar-fiexor and dorsiflexor muscle peak torque were measured as an indicator of active ankle muscle performance. Passive plantar-flexor muscle performance was characterized by passive plantar-flexor torque at 0 degrees of dorsiflexion. To determine if the range through which the plantar-flexor muscles develop active and passive torque was altered by TAL, the concentric peak torque angle and maximal dorsiflexion angle were measured. All muscle performance measurements were obtained using a Kin-Com isokinetic dynamometer (software version 4.06 ([double dagger])). Methods with established reliability have been described previously. (26) Briefly, intraclass correlation coefficients (ICCs) for active and passive muscle performance measures were calculated from 3 trials obtained in a single session. Using 34 subjects, ICC (3,1) values ranged from .97 to .98. (26)

For the concentric tests, the Kin-Com was set in the isokinetic mode and the gravity correction procedure was performed on the empty ankle apparatus. The foot was not included in the gravity correction because the plantar-flexor muscles exert passive resistance against the footplate and the weight of the foot was assumed to be negligible (~1.5% of body weight). (38) The subjects were positioned supine on the Kin-Com with the foot strapped to the ankle apparatus and the knee stabilized at 10 degrees of flexion. (28) The axis of the dynamometer was aligned with the axis of the ankle joint. The testing speed was 60[degrees]/s, which is comparable to the ankle angular velocity during the stance phase of walking. (23) Subjects were allowed 3 to 5 submaximal practice plantar-flexion contractions to become acquainted with the resistance and speed of movement. For plantar-flexion peak torque, the foot was placed in a position of maximum dorsiflexion and the subjects were instructed to push as hard and as fast as possible through full available range of motion. For dorsiflexion peak torque, the foot was placed in a position of maximum plantar flexion and subjects were instructed to pull up using the same guidelines. Subjects were allowed to rest between repetitions. The maximum peak torque of the 3 trials and the angle of this peak torque were recorded.

For the passive torque measurements, electromyography (EMG) (CGS-67 Multichannel Electromyographic System ([section])) was used on the first 10 subjects to verify that the plantar-flexor muscles were not actively contracting. Surface electrodes with attached preamplifiers were applied over the belly of the tibialis anterior muscle, the gastrocnemius muscle, and the soleus muscle (distal to the gastrocnemius muscle belly and lateral to the Achilles tendon). The raw signal was collected and high-pass filtered at 40 Hz, creating a frequency response of 40 to 4,000 Hz. The KinCom settings were the same as for the concentric tests. Subjects were instructed to relax their leg muscles, and the ankle was positioned in maximal plantar flexion. The Kin-Com apparatus then moved the ankle joint from plantar flexion into maximal dorsiflexion. Subjective complaints, increased EMG activity, and limb movement in the apparatus were monitored carefully during the procedure. If increased EMG activity (ie, above baseline) or a break in the torque curve was viewed on the oscilloscope, the subjects were instructed to relax and the procedure was repeated. After testing 10 subjects, use of EMG was eliminated because ankle muscle activity could always be predicted by a break in an otherwise smooth torque curve. Three trials of passive torque and angle data were collected. Passive torque at 0 degrees of dorsiflexion and maximal dorsiflexion angle were recorded for each trial. For both variables, the average value of the 3 trials was used for statistical analysis.

A 2 (group) x 3 (times of testing) repeated-measures analysis of variance (ANOVA) was used to determine differences for each of the muscle performance measures. Follow-up t tests using the error terms from the ANOVA were used for post hoc comparison on variables found to have a group x time interaction. The alpha level for all analyses was set at .05.

Results

There were group x time interactions for concentric plantar-flexor peak torque, concentric plantar-flexor peak torque angle, passive torque at 0 degrees of dorsiflexion, and maximal dorsiflexion angle (P<.05), indicating the TAL procedure affected these variables over time differently than TCC alone (Tab. 2). There were no differences in pretreatment measurements between groups for any of the outcome measures (P>.05).

Subjects in the TAL group showed a 31% decrease ([bar.X][+ or -]SE) in concentric plantar-flexion peak torque following surgery (35[+ or -]3 to 24[+ or -]3 N*m, initial posttest versus pretest, P<.05, Tab. 2, Fig. 1), but the peak torque value returned to pretest level 8 months after treatment (24 [+ or -] 3 to 34 [+ or -] 4 N*m, initial posttest to 8-month posttest P<.05, Tab. 2). Subjects in the TCC group showed no changes in concentric peak torque across the testing times (P>.05, Tab. 2, Fig. 1).

[FIGURE 1 OMITTED]

The angle of concentric plantar-flexor peak torque moved 16 degrees into dorsiflexion following treatment with TAL (initial posttest versus pretest, P<.05, Tab. 2) and did not change 8 months after treatment in the TAL group compared with the initial posttest values (8-month posttest versus initial posttest, P>.05, Tab. 2). The angle of concentric plantar-flexor peak torque moved 4 degrees into dorsiflexion following immobilization with TCC (initial posttest versus pretest, P<.05, Tab. 2). At the 8-month follow-up, the angle of concentric plantar-flexor peak torque for the TCC group was not different than the pretest and initial posttest values (P >.05), but was less than values for angle of concentric peak torque in the TAL group (P<.05, Tab. 2).

Subjects in the TAL group showed a 64% reduction in passive torque at 0 degrees of dorsiflexion following surgery (18 [+ or -] 2 to 6 [+ or -] 2 N*m, initial posttest versus pretest, P<.05, Tab. 2, Fig. 2). At 8 months after surgery, passive torque at 0 degrees increased to 60% of the presurgery level (6 [+ or -] 2 to 10 [+ or -] 2 N*m, initial posttest to 8-month posttest, P=.05, Tab. 2, Fig. 2), but the 8-month value was still different than the presurgery level. Subjects in the TCC group showed no differences in passive torque at 0 degrees of dorsiflexion across the testing times (P >.05, Tab. 2, Fig. 2).

There were no differences in dorsiflexion peak torque across groups or time (P>.05, Tab. 2, Fig. 3). The group x time interaction for angle of dorsiflexor peak torque, however, approached significance (P=.06, Tab. 2). Angle of dorsiflexor peak torque moved into greater dorsiflexion (- 38[degrees] to - 27[degrees]) following treatment with TAL (initial posttest versus pretest, P<.05, Tab. 2). Angle of dorsiflexor peak torque moved about 8 degrees back into more plantar flexion 8 months after surgery compared with initially after surgery (8-month posttest versus initial posttest, P<.05), resulting in a final value that was not different from the pretest value (P >.05). There was no change in angle of dorsiflexor peak torque in the TCC group across time (P >.05).

Maximal dorsiflexion angle moved more than 10 degrees into dorsiflexion following the TAL procedure (initial posttest versus pretest, P<.05, Tab. 2, Fig. 4) and remained at that level 8 months after surgery (8-month posttest versus initial posttest, P>.05, Tab. 2, Fig. 4). Subjects in the TCC group showed no differences in maximal dorsiflexion angle across the testing occasions (P>.05, Tab. 2, Fig. 4).

Discussion

As hypothesized, TAL had a dramatic effect on plantar-flexor muscle performance. Subjects undergoing the TAL procedure had a 31% reduction in concentric plantar-flexor peak torque after surgery and immobilization, but the peak torque returned to the baseline level after 8 months. In addition, the concentric peak torque angle moved 16 degrees into dorsiflexion and remained unchanged 8 months after surgery and immobilization, suggesting a shift in the length-tension relationship of the plantar-flexor muscles. (39) The findings of reduced peak concentric torque and a shift of the peak torque angle are consistent with the computer-simulated findings of Delp et al. (21) The authors reported that a simulated TAL in subjects with combined gastrocnemius and soleus muscle contracture resulted in a 37% reduction in the magnitude of the total plantar-flexor torque and a shift of the peak toward dorsiflexion.

Despite the reduction in plantar-flexor torque after surgery and immobilization, torque values returned to the baseline level 8 months after immobilization. The improvement in plantar-flexor peak torque in the 8 months alter surgery and immobilization may be related to the subjects' return to walking and the progressive resistance exercise program. We do not know how adherent subjects were to this home exercise program.

Perhaps greater improvements could be made with a more structured or supervised exercise program. Although concentric plantar-flexor peak torque returned to the baseline level after 8 months, the angle at concentric plantar-flexor peak torque was no different at 8 months than it was initially after surgery and immobilization. These results suggest that the range of motion through which the plantar flexors develop active torque may be permanently altered following TAL.

The changes in passive plantar flexor muscle performance following TAL were similar to the changes in active muscle performance. Passive torque at 0 degrees decreased 64% following TAL and immobilization, but increased to 60% of the baseline level at 8 months after surgery and immobilization. In addition, maximal dorsiflexion angle increased by more than 10 degrees initially after treatment and remained unchanged at the 8-month time point. As with active muscle performance, the effects of TAL on passive muscle performance were similar to those reported in the computer simulation study of Delp et al. (21) The simulation predicted a substantial decrease in passive plantar-flexor torque and a shift in the onset of passive torque toward dorsiflexion. These changes in passive muscle performance reinforce the possibility that the length-tension relationship of the plantar-flexor muscles is altered by TAL, leading to a shift (toward dorsiflexion) in the range through which these muscles develop torque.

Although passive torque decreased dramatically following TAL, it increased to 60% of the baseline level within 8 months. Similar to that of concentric peak torque, the improvement in passive torque development may be directly related to the increase in activity level (walking, progressive strengthening exercises). If the increase in muscle force was accompanied by increased muscle cross-sectional area, myofibrillar structures responsible for passive tension generation. (40-41) could be increased as well, resulting in an increase in passive torque generation for a given joint angle. Chleboun et al (42) reported greater passive muscle stiffness (torque/angle) and muscle volume in a group of men who trained regularly with weights compared with untrained men, suggesting a positive relationship between the active and passive torque-generating abilities of a muscle.

The initial changes in active and passive plantar-flexor muscle performance following TAL have considerable implications for walking and standing in people with DM who also have loss of protective sensation. Concentric (28,29) and passive (29) plantar-flexor torque have been shown to be predictive of plantar-flexor moments (torque) during gait; therefore, a reduction in concentric and passive torque, as noted initially after surgery and immobilization, could impair gait performance. The results of a single-subject study by Hastings et al (7) support this statement. The authors reported a 68% decrease in the peak plantar-flexor moment during walking in a patient after TAL and immobilization.

Although not the focus of this report, we were interested in the correlation between plantar-flexor peak torque and walking speed as an indicator of walking ability. Walking speed was determined by using a stopwatch to time subjects as they walked 15.2 m (50 ft). In the current study, Pearson product moment correlation coefficients (r) between concentric plantar-flexor peak torque and walking speed ranged from .36 to .56 (P<.05) across the 3 testing occasions when subject groups were combined (n=29). In addition, we noticed that some subjects showed instability at the ankle and knee during walking soon after surgery and subsequent immobilization. This instability appeared to improve over time as the muscle performance improved. We currently are investigating quantitatively the effect of surgery on patients' functional limitations and perceived disability. In light of these results and observations, a potential compromise in walking ability initially after treatment with TAL must be considered, especially in this population, who generally have deficits in walking abilities before surgery. (23,29,43)

An encouraging finding from this study was that, at 8 months after treatment, both active and passive muscle performance improved, suggesting that muscles of individuals with sensory neuropathy may have the ability to adapt to increased demands within a "new" range of motion. More research is needed to determine the ability of people with DM and peripheral neuropathy to increase active and passive muscle performance using a progressive resistive exercise program.

Dorsiflexor muscle performance showed minimal changes following TAL and immobilization. There was no change in concentric dorsiflexion peak torque, although the angle of peak torque moved 11 degrees into more DF-ROM. Although the group x time interaction for angle of peak torque was not significant at the P=.05 level (P=.06), a post hoc power analysis of this variable indicated that the observed power was .55. Given the clinically meaningful change in range of motion of 11 degrees, a larger sample would likely have resulted in a group X time interaction for angle of peak torque. The finding of a shift in the angle of peak torque toward dorsiflexion suggests that the dorsiflexor muscles were able to adapt to a new range by maintaining the same level of concentric torque generation. Such a shift may be beneficial because the peak torque is closer to an angle where the person would be expected to use the muscle during standing or walking. It is important to note, however, that this shift may be temporary. Alter 8 months, the angle of peak torque moved 8 degrees back in the plantar-flexion direction.

Somewhat surprisingly, the TCC group showed no changes in concentric peak torque, passive torque at 0 degrees of dorsiflexion, or maximal dorsiflexion angle after 5 weeks of immobilization with TCC. There are a number of possible reasons that may explain the maintenance of active and passive muscle performance in the TCC group. First, the subjects remained weight bearing the entire time of immobilization. Although not measured, weight-bearing forces and muscle contractions likely continued during the immobilization period. In addition, the negative effects of immobilization may have been offset by the positive effects of wound healing and reduced edema in the lower leg. During initial testing, all subjects had an open plantar wound. Although none of the subjects complained of pain during testing, they may have been reluctant to perform maximally resisted plantar flexion. Furthermore, the cast was changed every 1 to 3 weeks, and subjects were encouraged to move their ankles.

Subjects who received TAL were given slightly different weight-bearing precautions than those who received only TCC. The TAL group was instructed to remain partial weight bearing for 1 week after surgery, whereas the TCC group was allowed to be full weight bearing. Both groups were allowed to be full weight bearing during the remainder of the immobilization period. We did not monitor weight-bearing status in either group. We do not believe that this minor difference in precautions would explain the differences in ankle muscle performance reported in this article.

Conclusion

This is the first study to prospectively evaluate the effects of TAL on active and passive muscle performance in subjects with DM and a neuropathic ulcer. The results indicate that TAL led to a temporary decrease in active (peak concentric torque) and passive (passive torque at 0[degrees] of dorsiflexion) plantar-flexor muscle performance. The TAL also resulted in sustained increases in the angle of peak concentric torque and maximal dorsiflexion angle. If treatment with TAL is being considered to reduce ulcer recurrence, care must be taken to monitor or address the initial compromise of plantar-flexor muscle performance. Further study is needed to determine the effects of TAL on functional limitations and disability in people with DM and a neuropathic plantar ulcer.

Table 1.

Subject Characteristics for Tendo-Achilles Lengthening (TAL) and
Total-Contact Casting (TCC) Groups (a)

                                 Group

                                     TAL (n=15)           TCC (n=14)

Age (y) (b)                        55 [+ or -] 9       54 [+ or -] 10
Duration of DM (y) (b)             21 [+ or -] 10      18 [+ or -] 14
BMI (b)                          33.8 [+ or -] 5.8   31.8 [+ or -] 6.8
Hb [A.sub.1c] (%) (b)             8.8 [+ or -] 1.7    8.9 [+ or -] 2.0
No. of previous ulcers              5 [+ or -] 5        3 [+ or -] 2
Days immobilized in a cast (b)     44 [+ or -] 17      35 [+ or -] 18
Male/female (c)                  12/3                 10/4
Type 1/type 2 DM (c)              3/12                 5/9

(a) Values for continuous variables are means and standard deviations.
DM=diabetes mellitus, BMI=body mass index, Hb[A.sub.1c]=hemoglobin
[A.sub.1c].

(b) No difference between groups (P>.05 using 1 test between
2 independent means).

(c) No difference between groups (P>.05 using chi-square statistic).

Table 2.

Means (Standard Errors) for Ankle Muscle Performance Variables for
Tendo-Achilles Lengthening (TAL) and Total-Contact Casting (TCC)
Groups (a)

                              Group         n (c)          Pretest

PFPT (N*m)                    TAL           15              35 (3)
                              TCC           14              38 (4)
PFPT angle (b) ([degrees])    TAL           15              -7 (9)
                              TCC           14              -4 (5)
Zero torque (N-m)             TAL           10              18 (2)
                              TCC           10              13 (2)
DFPT (N*m)                    TAL           14              11 (3)
                              TCC           11               8 (3)
DFPT angle (b) ([degrees])    TAL           14             -38 (2)
                              TCC           11             -36 (2)
Max DF angle (b)
  ([degrees])                 TAL           14               0 (2)
                              TCC           14               4 (2)

                                                            P Value for
                                                            Group x
                              Initial       8-Month         Time
                              Posttest      Posttest        Interaction

PFPT (N*m)                     24 (3) (d)    34 (4) (e)     <.05
                               42 (4)        42 (4)
PFPT angle (b) ([degrees])      9 (4) (d)     7 (4) (d)     <.05
                                0 (6) (d)    -2 (6)
Zero torque (N-m)               6 (2) (d)    10 (2) (d,e)   <.05
                               13 (2)        15 (2)
DFPT (N*m)                     10 (2)        11 (2)         >.05
                                8 (3)        10 (3)
DFPT angle (b) ([degrees])    -27 (4) (d)   -35 (2) (e)     =.06
                              -37 (4)       -37 (2)
Max DF angle (b)
  ([degrees])                  11 (1) (d)    11 (1) (d)     <.05
                                6 (1)         6 (1)

(a) PFPT=concentric plantar-flexor peak torque, Zero torque=passive
plantar-flexor torque at 0 degrees of dorsiflexion, DFPT=concentric
dorsiflexor peak torque, Max DF angle=maximal dorsiflexion angle.

(b) Negative values indicate plantar flexion; positive values
indicate dorsiflexion.

(c) Subject number varies because only data for subjects who
successfully completed all 3 tests were included in the data
analysis.

(d) Significantly different from pretest values (P<.05).

(e) Significantly different from initial posttest values (P<.05).

* Royce Medical Co. 742 Pancho Rd, Camarillo, CA 93012.

([dagger]) The Hygenic Corporation, 1245 Home Ave, Akron, OH 44310.

([double dagger]) Chattecx Corp, 4717 Adams Rd. PO Box 489, Hixson, TN 37343.

([section]) Therapeutics Unlimited Inc, 2835 Friendship St, Iowa City, IA 52240.

References

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

(2) Quebedeaux TL, Lavery LA, Lavery DC. The development of foot deformities and ulcers after great toe amputation in diabetes. Diabetes Care. 1996;19:165-167.

(3) Rosenbloom AL, Silverstein JH, Lezotte DC, et al. Limited joint mobility ill childhood diabetes mellitus indicates increased risk for microvascular disease. N Engl J Med. 1981;305:191-194.

(4) Mueller MJ, Diamond JE, Delitto A, Sinacore DR. Insensitivity, limited joint mobility, and plantar ulcers in patients with diabetes mellitus. Phys Ther. 1989;69:453-462.

(5) Lin SS, Lee TH, Wapner KL. Plantar forefoot ulceration with equinus deformity of the ankle in diabetic patients: the effect of tendo-Achilles lengthening and total contact casting. Orthopedics. 1996;19:465-475.

(6) Armstrong DG, Stacpoole-Shea S, Nguyen H, Harkless LB. Lengthening of the Achilles tendon in diabetic patients who are at high risk for ulceration of the foot. J Bone Joint Surg Am. 1999;81:535-538.

(7) Hastings MK, Mueller MJ, Sinacore DR, et al. Effects of a tendo-Achilles lengthening procedure on muscle function and gait characteristics in a patient with diabetes mellitus. J Orthop Sports Phys Ther. 2000;30:85-90.

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

(9) Birke JA, Sims DS, Buford WL. Walking casts: effect on plantar toot pressures. J Rehabil Res Dev. 1985;22:18-22.

(10) Shaw JE, Hsi WL, Ulbrecht JS, et al. The mechanism of plantar unloading in total contact casts: implications for design and clinical use. Foot Ankle Int. 1997;18:809-817.

(11) Armstrong DG, Nguyen HC, Lavery LA, et al. Off-loading the diabetic foot wound: a randomized clinical trial. Diabetes Care. 2001;24: 1019-1022.

(12) Mueller MJ, Diamond JE, Sinacore DR, et al. Total contact casting in treatment of diabetic plantar ulcers: controlled clinical trial. Diabetes Care. 1989;12:384-388.

(13) Sinacore DR, Mueller MJ, Diamond JE, et al. Diabetic plantar ulcers treated by total contact casting: clinical report. Phys Ther. 1987;67: 1543-1549.

(14) Helm PA, Walker SC, Pullium G. Total contact casting in diabetic patients with neuropathic foot ulcerations. Arch Phys Med Rehabil. 1984;65:691-693.

(15) Myerson M, Papa J, Eaton K, Wilson K. The total-contact cast for management of neuropathic plantar ulceration of the toot. J Bone Joint Sung Am. 1992;74:261-269.

(16) Sinacore DR. Total contact casting for diabetic neuropathic ulcers. Phys Ther. 1996;76:296-301.

(17) Apelqvist J, Larsson J, Agardh CD. Long-term prognosis for diabetic patients with foot ulcers. J Intern Med. 1993;233:485-491.

(18) Helm PA, Walker SC, Pullium GF. Recurrence of neuropathic ulceration following healing in a total contact cast. Arch Phys Med Rehabil. 1991;72:967-970.

(19) Berry DC, Sabacinski KA, Habershaw GH, et al. Tendo Achilles procedures tot chronic ulcerations in the outcome of diabetic foot ulcers. J Am Podiatr Med Assoc. 1993;83:96-100.

(20) Mueller MJ, Sinacore DR, Hastings MK, et al. Effect of Achilles tendon lengthening on neuropathic plantar ulcers: a randomized clinical trial. J Bone Joint Sung Am. 2003;85:1436-1445.

(21) Delp SL, Statler K, Carroll NC. Preserving plantar flexion strength after surgical treatment for contracture of the triceps surae: a computer simulation study. J Orthop Res. 1995;13:96-104.

(22) Kaufman KR, Shaughnessy WJ. Computer simulation of surgical treatment for equinus deformity in cerebral palsy. J Appl Biomech. 2000; 16:70-79.

(23) Mueller MJ, Minor SD, Sahrmann SA, et al. Differences in the gait characteristics of patients with diabetes and peripheral neuropathy compared with age-matched controls. Phys Ther. 1994;74:299-313.

(24) Andersen H, Poulsen PL, Mogensen CE, Jakobsen J. Isokinetic muscle strength in long-term IDDM patients in relation to diabetic complications. Diabetes. 1996;45:440-445.

(25) Salsich GB, Mueller MJ, Sahrmann SA. Passive ankle stiffness in subjects with diabetes and peripheral neuropathy versus an age-matched comparison group. Phys Ther. 2000;80:352-362.

(26) Salsich GB, Brown M, Mueller MJ. Relationships between plantar flexor muscle stiffness, strength, and range of motion in subjects with diabetes-peripheral neuropathy compared to age-matched controls. J Orthop Sports Phys Ther. 2000;30:473-483.

(27) Salsich GB, Mueller MJ. Relationships between measures of function, strength and walking speed in patients with diabetes and transmetatarsal amputation. Clin Rehabil. 1997;11:60-67.

(28) Mueller MJ, Minor SD, Schaaf JA, et al. Relationship of plantar-flexor peak torque and dorsiflexion range of motion to kinetic variables during walking. Phys Ther. 1995;75:684-693.

(29) Salsich GB, Mueller MJ. Effect of plantar flexor muscle stiffness on selected gait characteristics. Gait Posture. 2000;11:207-216.

(30) Diamond JE, Mueller MJ, Delitto A, Sinacore DR. Reliability of a diabetic foot evaluation. Phys Ther. 1989;69:797-802.

(31) Wagner FW Jr. Treatment of the diabetic foot. Compr Ther. 1984;10: 29-38.

(32) Perry J. Gait Analysis: Normal and Pathological Function. Thorofare, NJ: Slack Inc; 1992.

(33) Cohen J, Cohen P. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. 2nd ed. London, United Kingdom: Lawrence Erlbaum Associates; 1983.

(34) Campbell WC. Campbell's Operative Orthopaedics. St Louis, Mo: Mosby; 1998.

(35) Sinacore DR, Mueller MJ. Total contact casting in the treatment of neuropathic ulcers. In: Bowker JH, Pfeifer MA eds. The Diabetic Foot. St Louis, Mo: Mosby; 2001:301-320.

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

(37) Kisner C, Colby LA. Therapeutic Exercise: Foundations and Techniques. Philadelphia, Pa: FA Davis Co; 1990.

(38) Winter DA. Biomechanics and Motor Control of Human Movement. 2nd ed. New York, NY: John Wiley & Sons Inc; 1990.

(39) Sharrard WJ, Bernstein S. Equinus deformity in cerebral palsy: a comparison between elongation of the tendo calcaneus and gastrocnemius recession. J Bone Joint Surg Br. 1972;54:272-276.

(40) Magid A, Law DJ. Myofibrils bear most of the resting tension in frog skeletal muscle. Science. 1985;230:1280-1282.

(41) Wang K, McCarter R, Wright J, et al. Regulation of skeletal muscle stiffness and elasticity by titin isoforms: a test of the segmental extension model of resting tension. Proc Natl Acad Sci (USA). 1991;88: 7101-7105.

(42) Chleboun GS, Howell JN, Conaster RR, Giesey JJ. The relationship between elbow flexor volume and angular stiffness at the elbow. Clin Biomech. 1997;12:383-392.

(43) Cavanagh PR, Derr JA, Ulbrecht JS, et al. Problems with gait and posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabet Med. 1992;9:469-474.

GB Salsich, PT, PhD, is Assistant Professor, Department of Physical Therapy, Saint Louis University, 3437 Caroline St, St Louis, MO 63104 (USA) ([e-mail]). Address all correspondence to Dr Salsich.

MJ Mueller, PT, PhD, FAPTA, is Associate Professor and Director of the Applied Biomechanics Laboratory, Program in Physical Therapy, Washington University School of Medicine, St Louis, Mo.

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

DR Sinacore, PT, PhD, is Associate Professor, Program in Physical Therapy, Washington University School of Medicine.

MJ Strube, PhD, is Professor, Department of Psychology, Washington University.

JE Johnson, MD, is Associate Professor, Chief, Foot and Ankle Service, Department of Orthopaedic Surgery, Washington University School of Medicine.

Dr Salsich, Dr Mueller, Dr Sinacore, and Dr Johnson provided concept/idea/research design. Dr Salsich and Dr Mueller provided writing. Dr Mueller and Dr Hastings provided data collection, and Dr Salsich, Dr Hastings, and Dr Strube provided data analysis. Dr Mueller provided project management and fund procurement. Dr Johnson provided subjects. All authors provided consultation (including review of manuscript before submission). The authors acknowledge Jennifer Henry for patient coordination and data management and the Prevention and Control Research Core of the Washington University Diabetes Research Training Center, P60 DK 20579, for help with subject recruitment.

This study was approved by the Institutional Review Board at Washington University.

Funding was provided by National Center for Medical Rehabilitation Research, National Institutes of Health, RO1 HD 36802.

This article was received February 13, 2004, and was accepted July 19, 2004.