Rehabilitation factors following transmetatarsal amputation.

Transmetatarsal amputation (TMA) is a relatively common procedure. The National Center for Disease Statistics reports that in 1991 approximately 32,000 above-knee amputations, 22,000 below-knee amputations, and 10,000 TMAs were performed in the United States.[1] Transmetatarsal amputation was described by McKittrick et al[2] in 1949 as an alternative to below-knee amputation. The development of antibiotic therapy made it possible to control ascending infection and septicemia, and allowed this more distal amputation. When certain surgical criteria are met, most authors and patients believe that a TMA is preferable to a below-knee amputation because it saves the ankle and a portion of the foot and provides a distal weight-bearing residuum.[2,3] In addition, there is evidence indicating that TMA is associated with a lower mortality rate than higher amputations.[4]

Although a TMA often is preferable to a below-knee amputation, there are several risks associated with a TMA. The surgical failure rate of TMA can be defined as the absence of any healing, or the need for a higher amputation, within 3 months of the TMA.[3] Various studies indicate failure rates of TMA of between 17% and 76%.[3,5-7] overall, the failure rate is about 30%.[8] The vascular causes of failure have been well studied, but little research has been conducted on the factors during acute care and rehabilitation of the patient that contribute to a successful outcome.

The purposes of this report are (1) to identify common problems encountered during the rehabilitation of patients with TMA, (2) to identify factors that may contribute to these problems, and (3) to propose methods to manage these problems.

Skin Breakdown

Skin breakdown following TMA is a serious problem that has been recognized by several authors.[2,7,9-13] Sage et al[7] reported 42% of 64 persons with midfoot amputations developed complications following surgery. Twenty-eight percent of the patients displayed early wound failure within 3 weeks of surgery, and 14% experienced ulcerations within the first year after surgery. In a group of 107 patients with TMA, we have noted that 27% developed skin breakdown and 28% required a higher amputation (unpublished data). Most of these complications (48% of the occurrences of skin breakdown and 60% of the higher amputations) occurred within the first 3 months following TMA. The high incidence of skin breakdown and higher amputation soon after TMA emphasizes the need to consider the acute care and early rehabilitation of patients with TMA.

Several factors appear to contribute to a high incidence of skin breakdown following TMA Sanders and Dunlap[6] report that 89% of the patients with diabetes mellitus (DM) and a TMA had sensory neuropathy. Neuropathy is a critical risk factor for patients experiencing unnoticed, repeated trauma that can lead to skin breakdown or primary wound failure.[9,14,15] McKittrick et al[2] reported that in patients with DM and insensitive skin, plantar ulcers reoccur following TMA "in almost all instances" and "in spite of any precautions we have been able to take."

Brand[9] described the gradual loss of sensation in patients with DM and peripheral neuropathy and noted a certain threshold of insensitivity that must be exceeded before patients are at risk for skin breakdown. Numerous investigators[16-18] have identified the 5.07 Semmes-Weinstein monofilament as the best discriminator of the likelihood of neuropathic feet to develop skin breakdown. Patients with intact sensation will experience pain during ambulation soon after TMA surgery and will naturally protect their residuum. Patients with loss of protective sensation may bear weight on open wounds or recent incisions and not receive feedback from pain sensations.[9] These patients are at high risk for skin breakdown or wound failure, and ultimately a higher, less functional amputation.[9,14,15]

Transmetatarsal amputation also results in a substantial loss of weight-bearing areas on the neuropathic foot, thereby increasing the load on remaining tissues.[9] Because the individual's superincumbent body weight has not changed, this reduction in weight-bearing area further increases the pressure and trauma on a foot that is already at risk.

Another factor that may contribute to a high incidence of skin breakdown following TMA is the decreased foot length available to generate a plantar-flexor moment. This decreased foot length causes a shorter available moment arm (the length d in Fig. 1) to generate plantar-flexor moments at the ankle compared with that of subjects without TMA.

In several measurements of full-length feet and TMA residuum, we determined that the available moment arm (distance from ankle joint to distal foot) was 40% to 48% shorter in the TMA residuum compared with the normal foot. Because the plantar-flexor moment is equal to the product of the moment arm and the ground reaction force, a 48% decrease in the moment arm length theoretically would require a 48% increase in the ground reaction force to generate a given plantar-flexor moment (Fig. 1). An increase in the ground reaction force is likely to increase localized pressure at the distal residuum. At the current time, it is unknown whether the ground reaction forces or the ankle moment, or both, are altered following TMA. Additional research is needed to determine the interaction of these factors.

An equinus deformity is another factor that potentially can place increased pressure on the distal residuum and contribute to skin breakdown.[13,19] A loss of dorsiflexion range of motion (ROM) may result in early and prolonged loading of the distal residuum during walking, contributing to skin breakdown.[16,20] Patients with DM and peripheral neuropathy are known to have limitations in joint mobility at the foot and ankle that are associated with high plantar pressures and plantar ulcerations.[16,20 In addition, TMA results in sectioning of the extensor hallucis longus, extensor digitorum longus, and peroneus tertius muscles, which dorsiflex the foot on the ankle. limited joint mobility, the muscular imbalance between the posterior and anterior compartment muscles, and bed rest following surgery may all contribute to a loss in dorsiflexion ROM. Surgery to lengthen the Achilles tendon of patients with DM and an equinus deformity has been shown to help heal chronic ulcers on the distal transmetatarsal residuum.[13] Optimal treatment would focus on prevention of the equinus defonnity.

Decreased Stability

Another problem complicating the rehabilitation of patients with TMA appears to be decreased stability, in weight-bearing activities.[3,10] Patients with DM and neuropathy report more falls and a lower perceived safety level in unusual circumstances than patients with DM alone.[21] Although there is little documentation of the additional disability involved with a TMA, Miller et al[3] report only 45% of patients were independent in gait without an assistive device following successful TMA surgery. Forty percent of the patients required a cane or walker, and 5% were bedridden. Many of our patients complain about instability, especially while bending over or reaching for objects.

Patients with DM and peripheral neuropathy are known to have decreased ankle muscle strength and joint ROM compared with age-matched controls.[22] In addition, many of these patients have multiple medical conditions that make them more susceptible to instability and falling. In our series of 107 patients with TMA (unpublished data), we found a high incidence of DM (77%), hypertension (54%), electrocardiographic abnormalities consistent with coronary artery disease (60%), and prior revascular surgery (51%).

The primary mechanical reason for a lack of stability in this patient group seems to be the decreased moment arm length (ie, short foot) available to generate a plantar-flexor moment at the ankle. Decreased plantar-flexor moments result in a decreased overall support moment during walking.[23] The support moment has been defined as the sum of the hip, knee, and ankle extensor moments and represents a total limb pattern to push away from the ground.[23(pp90-92)] Decreased extensor moments may result in limited stability during walking or other activities. As an example of this decreased stability, consider the subjects leaning forward in Figure 2. A subject without a TMA uses the full foot length to vary the location of the center of pressure (COP) to maintain upward stability.[23(pp93-96)] The toes appear to be an important factor in controlling the COP because they can generate forces at the distal foot. In general, the subject wifl lose stability if the center of gravity (COG) goes beyond the COP.[23] (The COP and COG are located by p and g, respectively, with respect to the ankle joint in Fig. 2).

The subject with a TMA will likely have limited ability to lean or reach forward due to the shortened foot length and active toe flexion available to contain the COG. This limitation in stability would also be expected in an activity such as walking.

Proposed Solutions

Due to the high incidence of skin breakdown soon after surgery, greater efforts are needed to protect the residuum. Clinicians working with this patient group should be aware that sensory neuropathy is an important risk factor.[14-18] Because patients with sensory neuropathy typically do not complain of inability to feel pressure on their foot or residuum, sensation should routinely be tested. The Semmes-Weinstein monofilaments used to measure light pressure sensation have been shown by several independent research groups to provide a quick, reliable method of identifying patients at risk for skin breakdown.[16-18] Patients with decreased sensation require education regarding the effects of peripheral neuropathy and how it may contribute to wound failure.[9]

As many patients with neuropathy seem unable to follow non-weight-bearing precautions,[9,24] devices are needed to help protect the residuum during the patients' return to ambulatory activities. Sanders and Dunlap[6] report a high success rate 83%), using a posterior splint during the early postoperative period of patients with TMA and then fitting patients with a non-weight-bearing, short leg cast for ambulation. A similar approach is used to heal neuropathic wounds with total contact casting. The total contact cast helps protect the residuum and lower leg from trauma and allow ambulation. The total contact cast is well documented in healing chronic neuropathic ulcers.[24-27] A limitation of the total contact cast is that it is not easily removed and the wound cannot be observed except with a complete cast change.

A possible alternative to the total contact cast may be a walking splint[28] or other type of removable clamshell cast or boot. A clamshell boot would immobilize the ankle, protect the wound from trauma, and allow observation of the operative incision. Although the clamshell boot may not provide the complete protection of a total contact cast, it would allow easy wound inspection and be more protective than traditional management techniques such as soft dressings or postoperative cast boots. A rigid, protective clamshell boot may also help to position the foot and prevent equinus deformities.

Another method to attempt to prevent equinus deformities may be early initiation of active exercise of the foot and ankle musculature. As noted, patients with DM and peripheral neuropathy are known to have limited dorsiflexion ROM when compared with age-matched controls.[16] Although there are few data to indicate how patients with DM and limited joint mobility will respond to exercise, it is reasonable for physical therapists to encourage strengthening of the remaining dorsiflexion muscle group. Additional research is needed to determine whether the early use of protective footwear or active exercise can decrease the incidence of skin breakdown or higher amputation foflowing TMA surgery.

Definitive Footwear

Definitive footwear is defined as the patient's prescriptive footwear following complete healing of the primary surgical incision. Some authors[2,29-31] suggest that no formal prosthetic device besides a toe filler and a reinforced sole may be required. Our experience, and that of others,[10] indicates this type of footwear is inadequate for most patients with DM and neuropathy. As illustrated in Figure 3, this footwear often provides inadequate stabilization for the residuum, allowing excessive rotation of the residuum within the shoe. With excessive rotation of the residuum inside the shoe, the heel often pistons out of the shoe, which may result in increased shear forces on the distal residuum between the skin and the toe filler.

Several sources[29-33] recommend placing a steel shank between the inner and outer layers of the sole of the shoe. The rationale for using a steel shank in the sole is to enhance stability of the shoe and reduce shoe distortion.[33] Although a steel shank may provide more stability compared with a shoe without a steel shank, Millstein et al[32] report that 90% of their patients with TMA who were fitted with a custom-made shoe with a steel shank, a toe filler, and molded insoles still complained of footwear problems. This report did not describe the complaints, but our experience with this type of footwear indicates that patients complain of inadequate stabilization of the residuum in the shoe (ie, the residuum pistons out of the shoe).

Some authors[10,11,34] recommend using a custom-made, shortened shoe with a rigid rocker-bottom RRB) sole. One rationale for recommending a short shoe is that it provides an equal lever arm for the residuum and the shoe. Theoretically, an equal lever arm results in an equal plantar-flexor moment for the residuum and the shoe. An equal moment would result in less rotation of the residuum in the shoe when compared with using a full-length shoe. Additional research is needed to determine whether a short shoe reduces rotation of the residuum in the shoe and results in less pressure on the distal residuum than a full-length shoe Fig. 4).

The RRB sole is another device intended to reduce pressure at the distal residuum or foot. In a full-length foot, an RRB sole is designed to reduce forefoot pressures by preventing metatarsophalangeal (MTP) joint extension, especially during push-off.[9,35] Placement of the apex of the rocker is variable, but generally the apex is placed under, or just proximal, to the MTP joints. The subject then is able to roll over the rocker rather than the MTP joints.[9,33,35] Studies on full-length feet have demonstrated that the RRB sole reduces plantar pressures under the metatarsal heads by 20% to 50% when compared with regular footwear.[35,36] Bauman et al[37] investigated the effect of wooden rockers placed on flat wooden sandals for the treatment of patients with insensitive, shortened feet secondary to Hansen's disease. The results of this study indicated that the greatest reduction in distal foot pressures occurred when the rocker was placed proximal to the distal residuum. Because the patient with TMA has no MTP joints, the RRB sole may provide an external mechanism to allow the superincumbant body weight to roll over the forefoot during the push-off phase of gait.[37]

From a theoretical standpoint, a shortened shoe with an RRB sole may help to prevent skin breakdown at the distal residuum, but only at the cost of decreased stability in standing. Decreased stability occurs because of the decreased plantar-flexor moment about the ankle. The reduced ankle plantar-flexor moment is caused by a shortened moment arm that the ground reaction forces can utilize. Figure 4 illustrates the shortened moment arm (d) when using a shortened shoe and RRB sole. The shortened moment arm is a result of the shortened foot, further reduced in length by an RRB sole. This decreased ankle plantar-flexor moment necessarily contributes to instability during weight-bearing activities.

In addition to decreased stability with a shortened shoe, we and others[33,34] have observed that the shortened RRB shoe is not cosmetically acceptable to many patients. Accordingly, even if the shortened RRB shoe has useful functional qualities, these benefits may be lost if shoes are not worn consistently. An ideal solution would be to simultaneously optimize the functional and cosmetic features of definitive footwear.

To solve the concurrent problems of excessive pressures on the distal residuum, decreased stability during walking, and cosmetic acceptance of footwear, we hypothesize that some patients will require a custom-made long shoe (shoe size prior to TMA) with an RRB sole and a polypropylene ankle-foot orthosis (AFO). The AFO could provide total contact support to the residuum and limit motion at the foot and ankle joints (ie, talocrural, subtalar, and midtarsal joints). The need for more proximal stability in TMA has been recognized previously.[11,9,34] We speculate that total contact between the residuum and the socket and immobilization of the joints would distribute forces optimally and should help to reduce peak pressures. These same concepts are used successfully in the fabrication of prosthetic sockets and total contact Casting.[24-27,30,31] Although an AFO would not be expected to provide as much immobilization as total contact casting, the amount of immobilization could be adjusted using traditional orthotic principles as indicated. In addition, the long shoe has the potential to provide a longer lever arm to help generate an ankle plantar-flexor moment compared with a shortened shoe. The AFO should help to transfer the ankle plantar-flexor moment to the knee and contribute to the overall stability and balance of the patient in any weight-bearing activity (Fig. 5).[22] We also hypothesize that the long shoe will be more cosmetically acceptable to patients than a shortened shoe. Additional research is needed to test these hypotheses.

Despite the stability provided by the AFO, we hear some patients complain of instability using an RRB sole. Perhaps these patients could use the custom-made long shoe with an AFO and no RRB sole. A negative factor of using a long shoe and an AFO without the RRB sole may be the limitation in ankle motion and difficulty rolling over the distal foot during the late stance phase of gait.

To overcome the problem of rolling over the distal foot without an RRB sole, patients with TMA may require gait training emphasizing a hip strategy.[22,38] Training a patient to walk using a predominant hip strategy would emphasize using the hip flexor muscles to advance the lower extremity during the late stance phase of gait rather than pushing off with the ankle plantar-flexor muscles.[22,38] Patients would be instructed to decrease their push-off, pull the leg forward from the hip, and decrease their step length.[38] In our laboratory, we have found that instructing patients with DM, peripheral neuropathy, and a recent plantar ulcer to walk with this type of gait pattern reduces forefoot peak plantar pressures an average of 27% compared with walking normally.[38] If subjects reduce their push-off, step length, and walking speed, they can reduce peak plantar forefoot pressures up to 58% compared with subjects walking normally.[39] Brand[9] has been advocating this type of walking pattern to decrease forefoot pressures in patients with DM for many years. Given a full-length shoe and an AFO, we hypothesize that instruction in walking using a hip strategy may enhance stability and reduce peak plantar pressures in patients with a TMA compared with normal ambulation with traditional footwear. Further research is needed to test this hypothesis.

Summary

We believe the weight-bearing residuum of a TMA is an advantage in functional activities when compared with below-knee amputation. There are, however, problems that clinicians should consider during rehabilitation of the patient with TMA. There is a high incidence of skin breakdown following TMA, Many patients complain of instability during walking. Decreased foot length and insensitivity appear to be the primary factors contributing to these problems. Solutions focus on attempting to provide.a substitute for the plantar-flexor lever arm and protection of the insensitive residuum. Clearly, more research is needed in the proper rehabilitation of with TMA.

Acknowledgment

We acknowledge James A Schaaf, PhD, for his assistance with preparation of the figures used in this manuscript.

References

[1] National Hospital Discharge Survey, 1991. Hyattsville, Md: National Center for Health Statistics; 1993. [2 ] McKittrick LS, McKittrick JB, Risley TS. Transmetatarsal amputation for infection or gangrene in patients with diabetes mellitus. Ann Surg. 1949;130:826-842. [3] Miller N, Dardik H, Wolodiger F, et al. Transmetatarsal amputation: the role of adjunctive revascularization. J Vasc Surg. 1991;13: 705-711. [4] Lee JS, Lu M, Lee VS, et al. Lower-extremity amputation: incidence, risk factors, and mortality in the Oklahoma Indian Diabetes Study. Diabetes. 1993;42:876-882. [5] Hodge MJ, Peters TG, Efird WG. Amputation of the distal portion of the foot. South Med J. 1989;82:1138-1142. [6] Sanders LJ, Dunlap G. Transmetatarsal amputation: a successful approach to limb salvage. J Am Podiatr Med Assoc. 1992;82:129-135. [7] Sage R, Pinzur MS, Cronin R, et al. Complications following midfoot amputation in neuropathic and dysvascular feet. J Am Podiatr Med Assoc. 1989;79:277-280. [8] Lynch T, Kanat IO. Transmetatarsal amputation: a literature review and case study. J Am Podiatr Med Assoc. 1991;81:540-544. [9] Brand PW. The diabetic foot. In: Ellenberg M, Rifkin H, eds. Diabetes Mellitus: Theory and Practice. 3rd ed. New Hyde Park, NY: Medical Examination Publishing Co Inc; 1983:829-849. [10] Birke

JA, Sims DS. The insensitive foot. In: Hunt GC, ed. Physical Therapy of the Foot and Ankle. New York, NY: Churchill Livingstone Inc; 1988:143-164. [11] Cavanagh PR, Ulbrecht JS. Biomechanics of the foot in diabetes mellitus. In: Levin ME, O'Neal LW, Bowker JH, eds. Tbe Diabetic Foot. 5th ed. St Louis, Mo: Mosby-Year Book Inc; 1993:223-228. [12] Helm PA, Kowalske KJ. Rehabilitation. In: Levin ME, O'Neal LW, Bowker JH, eds. Tbe Diabetic Foot. 5th ed. St Louis, Mo: Mosby-Year Book Inc; 1993:500. [13] Barry DC, Sabacinski KA, Habershaw GM, et al. Tendo achilles procedures for chronic ulcerations in patients with transmetatarsal amputations. J Am Podiatr Med Assoc. 1993;83: 96-100. [14] Pecoraro RE, Reiber GE, Burgess EM. Pathways to diabetic limb amputation: basis for prevention. Diabetes Care. 1990;13:513-521. [15] Boulton AJM, Betts RP, Franks CI, et al. Abnormalities of foot pressure in early diabetic neuropathy. Diabet Med. 1987;4:225-228. [16] 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. [17] Birke JA, Sims DS. Plantar sensory threshold in the Hansen's disease ulcerative foot. Read at the Proceedings of the International Conference on Biomechanics and Clinical Kinesiology of the Hand and Foot; Madras, India; December 16-18,1985. [18] Holewski JJ, Stress RM, Graf PM, et al. Aesthesiometry: quantification of cutaneous pressure sensation in diabetic peripheral neuropathy. J Rehabil Res Dev. 1988;25:1-10. [19] Parziale JR, Hahn KK. Functional considerations in partial foot amputations. Orthopaedic Review. 1988;17:262-265. [20] Femando DJS, Masson EA, Veves A, Boulton AJM. Relationship of limited joint mobility to abnormal foot pressures and diabetic foot ulceration. Diabetes Care. 1991;14:8-11. [21] Cavanagh PR, Derr JA, Ulbrecht JS, et al. Problems in gait and posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabet Med. 1992;9:469-474. [22] 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;74:299-308. [23] Winter DA. Biomechanics and Motor Control of human Movement. 2nd ed. New York, NY: John Wiley & Sons Inc; 1990. [24] Mueller MJ, Diamond JE, Sinacore DR, et al. Total contact casting in treatment of diabetic plantar ulcers: controfled clinical trial. Diabetes Care. 1989;12:384-388. [25] Sinacore DR, Mueller MJ, Diamond JE, et al. Diabetic neuropathic ulcers treated by total contact casting. Phys Ther. 1987;67:1543-1549. [26] Myerson M, Papa J, Eaton K, Wilson K. The total-contact cast for management of neuropathic plantar ulceration of the foot. J Bone Joint Surg [Am]. 1992;74:261-269. [27] Helm PA, Walker SC, Pullium G. Total contact casting in diabetic patients with neuropathic foot ulcerations. Arch Phys Med Rehabil. 1984;65:691-693. [28] Birke JA, Novick A, Graham S, et al. Methods of treating plantar ulcers. Phys Ther. 1991; 71:116-122. [29] Durham JR, McCoy DM, Sawchuk AP, et al. Open transmetatarsal amputation in the treatment of severe foot infections. Am J Surg. 1989;158:127-130. [30] Sanders GT. Lower Limb Amputation: A Guide to Rehabilitation. Philadelphia, Pa: FA Davis Co; 1986. [31] Shurr DG, Cook TM. Prosthetics and Orthotics. East Norwalk, Conn: Appleton & Lange; 1990. [32] Millstein SG, McCowan SA, Hunter GA Traumatic partial foot amputations in adults: a long-term review. J Bone Joint Surg [Br]. 1988; 70:251-254. [33] Janisse DJ. Pedorthotic care of the diabetic foot. In: Levin ME, O'Neal LW, Bowker JH, eds. The Diabetic Foot. 5th ed. St. Louis, Mo: Mosby-Year Book Inc; 1993:554-572. [34] Coleman WC. Footwear in a management program for injury prevention. In: Levin ME, O'Neal LW, Bowker JH, eds. The Diabetic Foot. 5th ed. St Louis, Mo: Mosby-Year Book Inc; 1993:542-545. [35] Schaff PS, Cavanagh PR. Shoes for the insensitive foot: the effect of a "rocker bottom" shoe modification on plantar pressure distribution. Foot Ankle. 1990;11:129-149. [36] Nawoczenski DA, Birke JA, Coleman WC. Effect of rocker sole design on plantar forefoot pressures. J Am Podiatr Med Assoc. 1988;78: 455-460. [37] Bauman JH, Girling JP, Brand PW. Plantar pressures and trophic ulceration: an evaluation of footwear. J Bone Joint Surg [Br]. 1963;45: 652-673. [38] Mueller MJ, Sinacore DR, Hoogsrate S, Daly L. Effect of hip and ankle walking strategies on peak plantar pressures: implications for neuropathic ulceration. Arch Phys Med Rehabil. In press. [39] Hongsheng Z, Werrsch JJ, Harris GF, et al. Foot pressure distribution during walking and shuffling. Arch Phys Med Rehabil. 1991;72: 390-397.