Relationship of foot deformity to ulcer location in patients with diabetes mellitus.

Relationship of Foot Deformity to Ulcer Location in Patients with Diabetes Mellitus Root et al [1] and others [2-4] have described clinically observed patterns of plantar callus formations, which they attribute to increased plantar stresses characteristic of certain foot abnormalities. These authors theorize that different types of foot abnormalities cause specific patterns of weight bearing and subsequent stresses on the plantar surface of the foot, resulting in identifiable patterns of callus formation.

Forefoot varus deformity has been defined as a midtarsal joint abnormality in which the forefoot is in an inverted position in relation to the bisector of the posterior aspect of the calcaneus when the subtalar joint (STJ) is in a neutral position. [1-4] These authors indicate that, for the medial aspect of the forefoot to make contact with the ground during stance, compensation must occur at the triplanar STJ axis. Subtalar joint pronation permits the medial forefoot to bear weight. This condition has been referred to as compensated forefoot varus. [1-4] The characteristics of this type of foot abnormality are a flattened arch, a prominent navicular tuberosity, and an everted calcaneus; this condition is commonly called "flatfoot." [3] In the presence of a compensated forefoot varus, some authors [1,3,4] speculate the peroneus longus muscle is unable to stabilize the first ray (first metatarsal and medial cuneiform) and the first ray becomes hypermobile. During gait, compensatory pronation at the STJ causes increased medial foot weight bearing. A transfer of increased medial weight bearing to the second or third metatarsal heads is believed to occur because the first metatarsal is hypermobile. Root et al [1] and others [2-4] indicate that keratomas will develop under the central metatarsal heads as a result of increased loading and abnormal shear over an excessive period of time.

A different weight-bearing pattern will occur if the STJ has inadequate range of motion and is unable to compensate for the forefoot varus deformity. Uncompensated forefoot varus occurs if the STJ is unable to pronate enough to bring the medial forefoot into full contact with the ground. [1-3] In an uncompensated forefoot varus, primary weight-bearing forces remain lateral, placing excessive stresses on the lateral side of the foot. It has been suggested that limited STJ eversion associated with callus under the fifth metatarsal head is a good clinical indicator of this uncompensated state. [1-4]

Forefoot valgus has been defined as a midtarsal joint abnormality in which the forefoot is in an everted position in relation to the bisector of the posterior calcaneus when the STJ is in a neutral position. [1-4] With this foot abnormality, the medial aspect of the forefoot (particularly the first metatarsal head) will load before the lateral forefoot, resulting in supinatory compensation at the STJ and a lateral shift of weight-bearing forces to the fifth metatarsal head. This condition is in contrast to forefoot varus in which the lateral aspect of the forefoot (particularly the fifth metatarsal head) will load before the medial forefoot, resulting in a pronatory compensation at the STJ and a medial shift of weight-bearing forces. [2,3] The weight-bearing patterns of a forefoot valgus will be somewhat different, depending on whether the forefoot valgus is rigid, flexible, or accompanying a plantar-flexed first ray. [3,4] If the forefoot valgus is rigid or occurs in conjunction with a plantar-flexed first ray, callus formation would be expected under the first or fifth metatarsal head. [3,4] The pattern of callus formation observed with a flexible forefoot valgus is variable, dependent on the amount of pronatory compensation available at the STJ. [4]

Callus formations rarely develop into an ulcer if the patient has normal sensation. Patients with diabetes, however, may have a neuropathy that compromises their sensation and makes them more susceptible to ulceration. [5,6] Several authors [5-9] have emphasized the role of decreased sensation and concurrent increased, repetitive mechanical pressures as principal causative factors of ulceration. Several studies [7,10,11] have indicated that patients who are unable to sense a 5.07 Semmes-Weinstein (SW) monofilament lack protective sensation and are at greater risk for ulceration.

A variety of foot deformities may increase mechanical pressure on the plantar aspect of the foot in patients with diabetes. Neuropathy can lead to a typical foot deformity including claw toes, prominent metatarsal heads, distal migration of the metatarsal fat pad, and equinus deformity. [6,8,12] Mueller et al [7] and others [13,14] have documented limited joint mobility in the feet of diabetic patients, which may contribute to increased plantar pressures by making the foot less flexible and less tolerant of changing positions. Patients with diabetes can also develop neuropathic fractures of the foot, often referred to as a Charcot's joint or Charcot's foot. [6] Radiographic changes are most often seen in the tarsal or metatarsophalangeal joints (Fig 1). Joint destruction can be quite extensive, leading to severe joint deformity. When neuropathic fractures involve the tarsal joints, the plantar region may become convex, a condition often referred to as a "rocker-bottom" foot (Fig 2). [6,15] Excessive pressure and possible ulceration would be expected in the midfoot region. [6]

In addition to acquired deformities, patients with diabetes may have forefoot varus or valgus abnormalities. These abnormalities may allow increased, repetitive pressure in the previously described patterns. If these patterns exist, a higher incidence of ulceration would be expected in the areas where pressure is theoretically increased. For example, a patient who developed a plantar ulcer in the presence of diabetes, insensitive feet, and an uncompensated forefoot varus would be expected to show the ulcer at the fifth metatarsal head (Fig 3). The theory of callus formation specific to certain foot types is based on the clinical observations of several authors. [1-4] To our knowledge, no research supports this theory. In addition, we are aware of no research to indicate an association or relationship between type of foot deformity and location of ulcers in patients with diabetes.

The purpose of this study was to determine whether a relationship exists between foot deformity and the location of ulcers in patients with diabetes mellitus (DM) and associated insensitivity of the foot. The following specific hypotheses were tested: 1) Patients with Charcot's joint changes at the midfoot would most often have plantar ulcers at the midfoot; 2) patients with a compensated forefoot varus would most often have ulcers at the plantar surface of the second, third, or fourth metatarsal head; and 3) patients with an uncompensated forefoot varus or forefoot valgus would most often have ulcers at the plantar surface of the first or fifth metatarsal. These foot abnormalities were chosen because they have been reported in the literature as associated with specific areas of increased pressure [1-4,6] and because we believe we could reliably identify them. [16]

Method

Subjects

Subjects were drawn consecutively from the outpatient physical therapy department at the Irene Walter Johnson Institute of Rehabilitation and the Diabetic Food Center at Washington University School of Medicine (St Louis, Mo). Criteria for inclusion in the study were that the patient be diagnosed with insulin-dependent diabetes mellitus (IDDM) or non-insulin-dependent diabetes mellitus (NIDDM), have a documented history of a neuropathic plantar ulcer on the midfoot or forefoot, and have one of the following deformities: 1) Charcot's joint of the midfoot, 2) a forefoot varus (Compensated or uncompensatede of at least 2 degrees, or 3) a rigid forefoot valgus of at least 2 degrees. Patients with a partial foot amputations, ulceration on the toes or rear foot (calcaneus), or an ulcer resulting from a single incident of trauma (eg, stepping on a tack) were excluded from the study.

Forty-two feet of 40 patients (12 female, 28 male) with foot ulcers were examined. Thirty-eight patients with unilateral lesions and 2 patients with bilateral plantar ulcers met the inclusion criteria. The mean age of the patients was 54.6 years (s = 9.8). The mean duration since diagnosis of DM was reported as 16.2 years (s = 7.3). Eleven patients were classified as have IDDM because they had an early age at onset ([is less than] 35 years of age) and were prone to developing ketoacidosis. Most of the remaining patients received insulin therapy to improve glycemic control (at the discretion of their primary physician). Semmes-Weinstein monofilaments of various stiffnesses were used to quantify sensation. [7,10,11,16] Severely decreased sensation was defined as the inability to sense a 5.07 SW monofilament. Absent sensation was defined as the inability to sense a 6.10 SW monofilament. [7,16] Fourteen of the feet showed severely decreased cutaneous sensation, and 28 of the feet showed absent sensation. All subjects signed an informed consent form prior to participating in the study.

Procedure

All measurements of STJ ROM, STJ neutral, and forefoot--to--rear-foot relationship were taken by the same two physical therapists (MJM or JED). The testers have had extensive experience taking the measurements on diabetic patients and developing intertester consistency. In a previous report, Diamond and colleagues [16] described their training sessions and how they subsequently obtained three sets of measurements on 31 patients with DM. To assess interrater reliability, patients were randomly tested by one of the two testers (MJM or JED) and next by the other tester. To assess intrarater reliability, the patient was measured a third time by one of the testers, again determined randomly. Testers were in separate rooms when the measurements were taken. Intrarater and interrater intraclass correlation coefficients (ICCs) for the measurements obtained by these testers were deemed to be acceptable, and standard errors of the means were relatively low (Tab 1). [16] The ICC(2,1) formula was used to document the reliability of two testers with extensive training and to generalize the results to testers with similar training. [17] Table 1 includes the standard errors of the means for goniometric measurements to aid in the interpretation of the reliability of a single score. The standard error of the mean can be useful in estimating the range of values within which a patient's "true" value may lie. [16,18]

All measurements were taken with the patient in the prone position with the foot and ankle overhanging the end of the table, as described in previous reports. [7,16] Goniometric measurements were taken with a plastic goniometer with 6-in (*1) arms and the scale marked in 2-degree increments. Passive STJ ROM was assessed by measuring maximal calcaneal inversion and eversion from anatomical zero. [2,7,16] One arm of the goniometer was aligned with the bisector of the distal one third of the lower leg, and one arm was aligned with the bisector of the calcaneus. The examiner positioned the goniometer with one hand and maximally inverted or everted the calcaneus with the other. The STJ-neutral position was determined by grasping the lateral forefoot, passively dorsiflexing the foot at the talocrural joint, and pronating the forefoot at the transverse tarsal joint until a soft end-feel was encountered. A soft end-feel is encountered when further dorsiflexion at the talocrural joint is first restricted by the pull of the triceps surae muscle group. The foot was then passively pronated or supinated at the STJ until the head and neck of the talus were felt equally at either side of the talonavicular joint. [2,16,19] The calcaneal position was noted by the same technique used for measuring STJ inversion and eversion. With the STJ in the neutral position, the forefoot--to--rear-foot relationship was determined. The goniometer was placed with one arm paralle to the plane of the metatarsal heads and the other arm aligned perpendicular to the bisection of the calcaneus (Fig 4). [2,16]

Foot deformities were classified by the primary investigator (MJM) according to operational definitions as 1) Charcot's foot, 2) compensated forefoot varus, or 3) uncompensated forefoot varus or forefoot valgus. The deformity was classified as Charcot's foot if bony prominences existed at the plantar midfoot, a condition sometimes described as a rocker-bottom foot (Fig 2). [6,15] These feet also showed radiographic evidence of Charcot's changes at the talonavicular or calcaneal cuboid joints, as determined by an orthopedic physician (VPB) (Fig 1). The deformities of uncompensated forefoot varus and forefoot valgus were combined because the previously reported areas of stress and callus formation overlap for the two deformities. [3,4] The deformity was classified as a compensated forefoot varus if the foot demonstrated a forefoot varus deformity of at least 2 degrees and a greater amount of STJ eversion than the forefoot varus deformity. The deformity was classified as an uncompensated forefoot varus or forefoot valgus if the foot demonstrated less STJ eversion than the amount of forefoot varus or at least 2 degrees of a rigid forefoot valgus. For example, a foot deformity demonstrating 4 degrees of forefoot varus and 10 degrees of STJ eversion would be classified as a compensated forefoot varus. A foot deformity demonstrating 4 degrees of forefoot varus but 0 degrees of STJ eversion would be classified as an uncompensated forefoot varus.

Location of ulcer was specified in one of three areas on the plantar foot. The first area was the midfoot, defined as the area between the midtarsal joints and the tarsometatarsal joints. The second area was defined as the second, third, or fourth metatarsal. The third area was defined as the first and fifth metatarsals. The third are was not separated into two distinct areas because the callus formation of a forefoot valgus is expected under the first or fifth metatarsal head. [3,4] Combining these two areas (first and fifth metatarsal heads) into one area is consistent with the a priori hypothesis of this study. Given the inclusion criteria for this study, these three locations and foot deformities were mutually exclusive and exhaustive categories.

Data Analysis

The Pearson chi-square test of association was used to determine whether a relationship existed between foot deformity and location of ulcer. [20] The alpha level of significance for all comparisons was established at .05.

Results

Six of the 7 patients with a Charcot's joint exhibited ulceration of the midfoot. Nine of the 18 patients with a compensated forefoot varus showed ulceration at the second, third, or fourth metatarsal head. Fifteen of the 17 patients with an uncompensated forefoot varus or forefoot valgus showed ulceration at the first or fifth metatarsal head (Tab 2). The Pearson chi-square test of association indicated a significant relationship between the type of foot deformity and location of ulcer ([X.sup.2] = 42.12, df = 4, p < .001). Because neither forefoot deformity demonstrated ulceration in the midfoot, a 2 x 2 Pearson chi-square test was performed using only the forefoot abnormalities and their respective ulcer locations. This test indicated a significant relationship between forefoot deformity and location of ulcer ([X.sup.2] = 5.93, df = 1, p < .025).

Discussion

These results indicate that a significant statistical relationship existed between these foot classifications and location of ulcer. Six of the 7 patients with Charcot's feet demonstrated ulceration at the midfoot. The patient who was the exception had an ulcer at the lateral plantar aspect of the fifth metatarsal, apparently caused by ill-fitting shoes. Left untreated, acute Charcot's joint changes of the midfoot may cause a complete collapse of the arches of the midfoot, resulting in the rocker-bottom appearance of the foot and an increased risk for ulceration (Fig 2). Predisposing factors for developing these often unnoticed neuropathic fractures are prolonged bed rest, immobilization, and infection in the presence of a sensory neuropathy. [5,15] Patients with acute Charcot's feet experience localized edema and warmth in the foot, but little or no pain. Radiographs are required to confirm the diagnosis (Fig 1). Treatment should consist of prompt immobilization with total-contact casting. [21] We have found that early intervention appears to minimize bony destruction. Patients may require surgical revision or amputation to correct severe deformities and prevent recurrent ulceration, but we have had success in rehabilitating some patients such as the patient whose foot is shown in Figure 2 with accommodative footwear alone.

All of the patients classified with a forefoot deformity demonstrated ulceration under the metatarsal heads. We combined the deformities of uncompensated forefoot varus and forefoot valgus and location of the first and fifth metatarsal heads, based on the rationale of previous reports and theories described in the literature, [1-4] indicating there is overlap in the location of where callus formation might be expected. Looking at these deformities separately, 6 to 9 patients with an uncompensated forefoot varus showed ulceration at the fifth metatarsal head and 3 patients showed ulceration at the first metatarsal head. Six of the 8 patients with a forefoot valgus showed ulceration at the first metatarsal head, one at the second metatarsal head, and one at the third metatarsal head. Eight of the 9 patients classified with a compensated forefoot varus with ulceration at the first or fifth metatarsal head showed ulceration at the first metatarsal head. This high incidence of ulceration at the first metatarsal head is inconsistent with the theory that the first ray is hypermobile with a compensated forefoot varus. [1-4] Certainly, other factors influence the location of increased pressure at the forefoot besides a forefoot varus or valgus deformity, particularly in patients with DM. [7,12-14] Many of the patients in this study had limited joint mobility, claw toes or hammertoes, distal migration of the plantar metatarsal fat pad, prominent metatarsal heads, or a plantar-flexed first ray. The ulcers that were located inconsistent with the hypotheses in this study seemed to occur in patients with severe deformities of these types. These deformities appeared to provide the greatest source of pressure contributing to ulceration, irrespective of classification of forefoot varus or valgus. In addition to foot abnormalities, changes at the tibia, knee, and chip can influence weight-bearing forces on the plantar aspect of the foot. [2]

Therse results provide a connecting link between the theory stating a primary cause of plantar ulceration is decreased sensation with the imposition of repetitive increased mechanical pressures [5-9] and the theory stating callus formation is dependent on foot type. [1-4] All of the patients in this study had severely decreased or absent sensation as measured by SW monofilaments. The noted foot abnormalities appeared to allow increased localized pressure that was repeated with each step of the patient. Because the ulcers were in a pattern associated with those predicted, the results support the theory of characteristic callus formation (as a result of increased localized pressure) associated with certain foot classifications. [1-9]

As noted previously, the number of foot classifications evaluated in this study was limited. The classifications chosen were those most definable and representing the crux of the hypotheses. [22] Further research is needed to identify other relationships between type of deformity and location of increased plantar pressure points. This study did not actually measure increased pressure. Increased pressure was inferred by the location of the ulcer. Although studies have been conducted to assess pressure at various parts of the diabetic and the nondiabetic foot, [9] further research is needed to quantify the pressure distribution as it relates to foot type or available foot and ankle ROM during gait. In addition, deformities were not quantified beyond classification as one of three foot deformity types. The severity of each deformity was not considered in establishing a relationship. Additional research is required to quantify the degree of deformity and the relationship to ulcer formation. Furthermore, this study was retrospective, evaluating patients who already had an ulcer. A prospective study would help to assess the predictive validity of this stu dy.

Clinical Implications

These results have several implications for the treatment of the diabetic foot. Rigid orthoses should not be used to improve the alignment of foot abnormalities because the forces required to make corrections may cause ulceration on insensate feet. In general, foot abnormalities and deformities should be accommodated with the foot positioned as close to the STJ-neutral position as possible. One reason total-contact casting may be successful in healing plantar ulcers is that loads are spread over an increased area and the foot is immobilized in a position as close to STJ neutral as possible. [21,23-25]

Proper footwear for an insensitive foot often requires an insole using a combination of several materials. Firmer material can be used to improve the alignment of a forefoot varus or valgus, but the firm material must be covered with softer materials, as described by Coleman [26] and Lockard. [27] As as example, the patient whose foot is shown in Figure 4 had an uncompensated forefoot varus that appeared to allow increased forces at the fifth metatarsal head, contributing to the ulcer shown in Figure 3. Appropriate footwear for this patient consisted of extra-depth shoes with a rigid rocker-bottom sole. [25,26] His custom-fabricated orthotic device consisted of a firm support for the medial forefoot (also called a medial forefoot post), which was covered with 3/16-in molded Plastazote[R]. (*2) Footwear and orthotic recommendations according to foot deformity, sensitivity, and history of ulceration have been proposed to assist the clinician in prescription of appropriate footwear. [8,26] Additional research is required to determine the effects of exercise and various types of footwear on plantar pressures so that optimal treatment strategies can be devised to treat and prevent foot problems.

Conclusion

A significant relationship existed between foot classification and the location of ulcers in this group of 40 patients with DM and insensitive feet. The results of this study support the hypotheses 1) that decreased sensation, coupled with increased, repetitive pressures, is a primary cause of plantar ulcers and 2) that certain foot types are associated with characteristic patterns of pressure distribution and callus formation. We believe patients with insensitive feet ad with the types of foot deformity described in this article should be fit with orthoses and footwear that accommodate their respective deformity in a position as near to STJ neutral as possible with the goal of preventing plantar ulceration.

(*1) 1 in = 2.54 cm.

(*2) Bakelite Xylonite Ltd, London, England, distributed by AliMed Inc, 297 High St, Dedham, Ma 02026.

References

[1] Root ML,, Orien WP, Weed JH. Clinical Biomechanics: Normal and Abnormal Function of the Foot. Los Angeles, Calif: Clinical Biomechanics Corp; 1977-2

[2] McPoil TG, Brocato RS. The foot and ankle biomechanical evaluation and treatment. In Gould JA, Davies GJ, eds. Orthopaedic and Sports Physical Theraphy. St Louis, Mo: CV Mosby Co; 1985;2:313-341

[3] Gray GW. When the Feet Hit the Ground Everything Changes: Program Outline and Prepared Notes--A Basic Manual. Toledo, Ohio: American Physical Rehabilitation Network; 1984

[4] Tiberio D. Pathomechanics of structural foot deformities. Phys Ther. 1988;68:184--1849

[5] Brand PW. The diabetic foot. In Ellenberg M, Rifkin H, eds. Diabetes Mellitus: Theory and Practice. 3rd ed. New York, NY: Medical Examination Publishing Co Inc; 1983:829-849

[6] Levin ME. The diabetic foot: pathophysiology, evaluation, and treatment. In Levin ME, O'Neal LW, eds. The Diabetic Foot. 4th ed. St Louis, Mo: CV Mosby Co; 1988:1-50

[7] Mueller MJ, Diamond JE, Delitto A, et al. Insensitivity, limited joint mobility, and plantar ulcers in patients with diabetes mellitus. Phys Ther. 1989;69:453-462

[8] Sims DS Jr, Cavanagh PR, Ulbrecht JS. Risk factors in the diabetic foot: recognition and management. Phys Ther. 1988;68:1887-1902

[9] Boulton AJM, Hardisty CA, Betts RP, et al. Dynamic foot pressure and other studies as diagnostic and management aids in diabetic neuropathy. Diabetes Care. 1983;6:26-33

[10] Birke JA, Sims DS Jr. Plantar sensory threshold in the Hansen's disease ulcerative foot. Read at the Proceedings of the International Conference on Biomechanics and Clinical Kinesiology of Hand and Foot; December 16-18, 1985; Madras, India

[11] Holewski JJ, Stess RM, Graf PM, et al. Aesthesiometry: quantification of cutaneous pressure sensation in diabetic peripheral neuropathy. J Rehabil Res Dev. 1988;25:1-10

[12] Holewski JJ, Moss KM, Stess R. Prevalence of foot pathology and lower extemity complications in a diabetic outpatient clinic. J Rehabil Res Dev. 1989;26:35-44

[13] DElbridge L, Perry P, Marr S, et al. Limited joint mobility in the diabetic foot: relationship to neuropathic ulceration. Diabetic Med. 1988;5:333-337

[14] Birke JA, Cornwall MW, Jackson M. Relationship between hallux limitus and ulceration of the great toe. Journal Orthopaedic and Sports Physical Therapy. 1988;10:172-176

[15] Birke JA, Sims DS Jr. The insensitive foot. In Hunt GC, ed. Physical Therapy of the Ankle and Foot. New York, NY: Churchill Livingstone Inc; 1988;133-168

[16] Diamond JE, Mueller MJ, Delitto A, et al. Reliability of a diabetic foot evaluation. Phys Ther. 1989;69:797-802

[17] Lahey MA, Downey RG, Saal FE Intraclass correlations: there's more there than meets the eye. Psychol Bull. 1983;93:586-595

[18] Anastasi A. Psychological Testing. 6th ed. New York, NY: Macmillan Publishing Co; 1988:133-135

[19] Elveru RA, Rothstein JM, Lamb RL, et al. Methods for taking subtalar joint measurements: a clinical report. Phys Ther. 1988;68:678-682

[20] Hays WL. Statistics. 4th ed. New York, NY: Holt, Rinehart and Winston Inc; 1988:768-786

[21] Mueller MJ, Diamond JE, Sinacore Dr, et al. Total contact casting in the treatment of diabetic plantar ulcers: controlled clinical trial. Diabetes Care. 1989;12:384-388

[22] Rosenthal R, Rosnow RL. Essentials of Behavior Research: Methods and DAta Analysis. New York, NY: McGraw-Hill Book Co; 1984:1-13

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

[24] Walker SC, Helm PA, Pullium G. Total contact casting and chronic diabetic neuropathic foot ulcerations: healing rates by wound location. Arch Phys Med Rehabil. 1987;68:217-221

[25] Mueller MJ, Diamond JE. Biomechanical treatment approach to diabetic plantar ulcers: a case report. Phys Ther. 1988;68:1917-1920

[26] Coleman WC. Footwear in a management program of injury prevention. In Levin ME, O'Neal LW, eds. The Diabetic Foot. 4th ed. St Louis, Mo: CV Mosby Co; 1988:293-309

[27] Lockard MA. Foot orthoses. Phys Ther. 1988;68:1866-1873

M Mueller, MHS, PT, is Instructor, Program in Physical Therapy, and Clinical Specialist, Irene Walter Johnson Institute of Rehabilitation, Washington University School of Medicine, 509 S Euclid Ave, St Louis, MO 63110 (USA). Address all correspondence to Mr Mueller.

S Minor, PhD, PT, is Instructor, Program in Physical Therapy, Washington University School of Medicine.

J Diamond, MHS, PT, is Supervisor, Department of Physical Theraphy, Irene Walter Johnson Institute of Rehabilitation.

V Blair, MD, is Medical Director, Diabetic Foot Center, and Assistant Professor of Orthopedic Surgery, Department of Orthopedics, Washington University School of Medicine.