Risk Factors in the Diabetic Foot by samhodges

VIEWS: 133 PAGES: 16

									Risk Factors in the Diabetic Foot
Recognition and Management

                           The most frequently reported diabetic foot complication is plantar ulceration.
                           Neuropathic fractures occur less often than plantar ulcers but usually result in
                           major structural deformities of the feet. The primary risk factors for plantar
                           ulceration are believed to be loss of protective sensation and the presence of
                           high plantar pressures. Principal etiologic factors in the development of neuro-
                           pathic fractures are hypothesized to be loss of protective sensation and bone
                           demineralization. Major diabetic foot injuries are preventable by the use of
                           comprehensive screening examinations and patient education. Patients at high-
                           est risk of injury should be seen more frequently and receive more extensive
                           therapy. Successful management of plantar ulcerations is dependent on reduction
                           of pressures under the foot and control of infection. Neuropathic fractures require
                           long-term immobilization to promote healing followed by careful monitoring after
                           removal of casts to prevent reinjury. Future research needs include prospective
                           studies on risk factors and validation of treatment techniques.
                           Key Words: Diabetes mellitus; Foot disorders; Orthotics/splints/casts, lower extrem-
                                      ity; Skin ulcer; Wound healing.

   Foot problems can frequently be life               gical fees. Also, as many as 50% to 70%            The issue of foot problems in the
or limb threatening in the diabetic in-               of all nontraumatic lower extremity am-         diabetic patient is a particularly timely
dividual, yet this complication of dia-               putations are performed on diabetic pa-         one in light of House Bill HR 1325
betes mellitus often is not well under-               tients.6 This statistic represents approx-      ("Medicare Payments for Therapeutic
stood by the patient or the family. In                imately 35,000 amputations each year            Shoes for Individuals with Severe Dia-
addition, health care practitioners are               at an estimated initial cost of $10,000         betic Foot Disease"), which received ap-
usually unaware of the importance of                  per case and an annual financial burden         proval during the 100th (1987-1988)
regular foot examinations in diabetic                 to the nation of over $350 million.7 The        congressional session for a demonstra-
patients. One study conducted at the                  real cost, however, is in life itself because   tion project to begin in October 1988
diabetes clinic of a major metropolitan               the prognosis for patients after lower          (Congressman Christopher H. Smith, R-
hospital showed that only 12.3% of the                extremity amputation is poor, with              NJ; personal communication; Septem-
patients were given a foot examination.1              three- and five-year survival rates of          ber 1988). If this bill were to become
These data are consistent with previ-                 only 65% and 41%, respectively.3 Al-            law, a physician would be entitled to
ously reported findings from a general                though these survival statistics may re-        prescribe footwear for a diabetic patient
medical clinic.2                                      flect the presence of concurrent severe         thought to be "at risk" of injury, and
   The extent of the problem is clearly               health problems, the increased physio-          Medicare would reimburse up to a total
demonstrated by the available statistics              logical and psychological stress imposed        of $250 per year. The passage of this bill
on morbidity and mortality. It has been               on the body by an amputation may also           would sharpen the need for quantitative
estimated that 20% to 25% of all hos-                 be a major factor contributing to death.        information concerning the risk factors
pital admissions of diabetic patients are                Data on the exact incidence of inju-         for foot injury and would also require
for foot problems.3 The average length                ries to the foot of the diabetic individual     better data on effective methods of foot-
of stay for such problems is 26 days4;                are difficult to find. In contrast to the       wear intervention.
based on the average 1986 cost of $639                more well-known complications of reti-             In summary, foot problems in the
for a semiprivate hospital bed,5 this                 nopathy, nephropathy, and coronary ar-          diabetic patient are of major impor-
length of stay translates to $16,614 per              tery disease, the foot is not well repre-       tance, yet they have not received due
admission, excluding medical and sur-                 sented in the general diabetes literature.      attention by those who treat diabetic
                                                      For example, the National Institutes of         patients. Neither is there a great volume
   D. Sims, MA, PT, is a doctoral candidate, Center   Health (NIH) publication Diabetes in            of research devoted to the etiology of
for Locomotion Studies, The Pennsylvania State        America8 does not contain a chapter on          foot problems or their prevention. Pro-
University, University Park, PA 16802.                foot problems. This report, however,
   P. Cavanagh, PhD, is Professor and Director,                                                       spective studies are notably absent from
Center for Locomotion Studies, The Pennsylvania       does state that about 15% of diabetic           the literature. This lack of knowledge
State University, University Park, PA 16802 (USA).    individuals have experienced an ulcer           prevents the diabetic foot from receiving
Address correspondence to Dr. Cavanagh.
   J. Ulbrecht, MD, is Consultant Diabetologist
                                                      on the foot or ankle and that the inci-         the attention it deserves in both research
and Adjunct Professor, Center for Locomotion          dence of gangrene (a possible sequela of        and clinical practice. The purpose of this
Studies. His office address is Internal Medicine      ulceration) is 4.5 cases per 1,000 dia-         article is to present the evidence for and
Associates of State College, PC, 3901 S Atherton
St, State College, PA 16801.                          betic individuals.                              quantification of a number of risk fac-

Volume 68 / Number 12, December 1988                                                                                                      1887
tors for injury to the diabetic foot that       TABLE 1
have been suggested in the literature.          Risk Factors in the Diabetic Foot
Recognition and management of inju-
ries to the foot will also be discussed                       Risk Factor                             Relationship to Injury
with an emphasis on plantar lesions and         Loss of protective sensation                   Absent pain-warning system
neuropathic fractures. Future research          High plantar pressures                         Ulcers usually occur at sites of peak
needs will be identified to encourage the                                                        pressure
reader to contribute to the growing body        Autonomic neuropathy                           Dry, noncompliant skin
of knowledge on preventable complica-           Previous ulcer or amputation                   Stress concentration in scar or
                                                                                                 transfer lesions
tions in the diabetic foot.
                                                Foot deformity                                 Increased local pressures
                                                Neuropathic fractures                          Foot instability and dramatically in-
POTENTIAL RISK FACTORS                                                                           creased plantar pressures
                                                Abnormal foot function                         Abnormal load application
   Injuries to the diabetic foot are usually    High activity level                            Increased cumulative stress
classified according to the presumed            Vascular disease                               Devitalized tissue more susceptible
cause of the lesion. The three most fre-                                                         to injury and healing potential is
quently cited causes are neuropathy,                                                             less
mechanical trauma, and vascular insuf-          Inadequate footwear or foot care               Decreased protection from environ-
ficiency. Most injuries to the foot occur         habits                                         ment and increased pressures on
on the plantar surface and, contrary to                                                          sides and top of foot, barefoot
common belief, do not appear to be due                                                           walking, increased pressures
                                                                                                 from thick callus lesions or for-
to vascular insufficiency.9 For these rea-
                                                                                                 eign objects in shoes, nailbed in-
sons, we will focus our attention on                                                             fections, inadequate shoe break-
plantar ulcers and on the nonvascular                                                            in period, delayed detection of
etiologic factors.                                                                               injuries
   The etiology of plantar ulcers in dia-       Vision loss                                    Self-care injuries, delayed detection
betes is now generally believed to be                                                            of injuries, and frequent trauma
multifactorial, encompassing both phys-                                                          during activity
iological and biomechanical factors.10-12       Inadequate diabetic control                    Increased susceptibility to
A variety of possible causes have been                                                           complications
mentioned, either individually or in
combination, yet the definitive study to
identify the relative importance of the         possible risk factor is discussed individ-      gression of the neuropathy as the disease
risk factors remains to be conducted.           ually, it must be appreciated that the          worsens.
Considering the widespread incidence of         status of the foot may change quickly if           Decreased ability to feel pain second-
ulcers in the diabetic foot, it is surprising   an interaction occurs between risk fac-         ary to neuropathy is widely accepted as
that only one comprehensive study has           tors. For example, a sudden loss of sen-        a primary risk factor in the development
considered multiple risk factors. In this       sation in a foot with preexisting high          of injuries in the diabetic foot.13,16 The
work, Boulton and associates compared           pressures may increase the risk of injury       most frequently reported injuries are
a total of 41 patients divided into three       well beyond any simple additive effects.        skin ulcers over the plantar surface of
groups: individuals with diabetic neu-          Potential risk factors are summarized in        the feet resulting from repetitive trauma
ropathy, those without neuropathy, and          Table 1.                                        not recognized by the patient because of
nondiabetic controls.13 Thirteen pa-                                                            decreased sensation.9 Bone injuries, al-
tients in the neuropathic group had a           Loss of Protective Sensation                    though less common, are frequently not
history of ulceration. The authors used            Neuropathy is one of the most com-           detected for similar reasons. Lack of a
a number of neurological, vascular, and         mon long-term complications of diabe-           pain warning system may allow severe
biomechanical measures and found high           tes. Nerve damage occurs in up to 25%           injuries to occur before treatment is
plantar pressures and elevated vibration        of patients with diabetes mellitus after        initiated, because the patient and prac-
perception thresholds to be most related         10 years duration and in up to 50% of          titioner frequently underestimate the
to the presence of plantar ulcers. This         diabetic patients after 20 years dura-          magnitude of the problem.
important study has provided a foun-            tion.14 The etiology is not completely             Numerous tests have been proposed
dation on which future work can build.          understood but is probably related to           for evaluating neuropathy. Nerve con-
In particular, there is a need for carefully    the biochemical disturbances seen in di-        duction measurement is an extremely
controlled prospective studies in which         abetes and to an inherited predisposition       sensitive technique for detecting and lo-
a variety of suspected risk factors are         to neuropathy. Several diabetic neuro-          calizing neuropathic changes.17 The in-
compared with subsequent rates of               pathic syndromes have been described,           formation provided includes conduc-
ulceration.                                     but the most common is the distal,              tion velocity in addition to amplitude
   Despite limitations in the current           symmetrical polyneuropathy that typi-           and duration of the evoked potentials.
knowledge regarding risk factors, the           cally affects the lower extremities first.15    Although nerve conduction measure-
problem-oriented approach described in          A combined sensory, autonomic, and              ments describe the physiological state
this article provides a useful framework        motor neuropathy is characteristic, al-         of the nerve, there has been renewed
for decision making regarding the like-         though the severity of each component           interest in clinical tests of sensation that
lihood of foot injury (see article by Gial-     may vary. The earliest changes occur in         provide a measure of the functional
lonardo in this issue). Although each           the toes, followed by a proximal pro-           consequences of the neuropathy.18 Gel-

1888                                                                                                                PHYSICAL THERAPY
berman et al compared a variety of func­       ent.23 Non-Pacinian mechanoreceptors            of protective sensation. No such criteria
tional tests of sensation with sensory         are more superficial and respond selec­         exist for nerve conduction. Finally,
nerve conduction measurements in a             tively to low frequency (eg, 25 Hz) stim­       measurements of sensation are generally
novel median nerve compression model.19        ulation or "skin flutter."24 In contrast,       less costly and easier to administer than
Threshold tests using vibration and            Pacinian corpuscles are more deeply sit­        nerve conduction measures and, there­
Semmes-Weinstein monofilaments were            uated and are selectively activated by          fore, have a greater chance of finding
found to be the most sensitive methods         higher frequency (eg, 250 Hz) vibra­            routine use in a clinical situation.
for detecting sensory deficits and were        tions. Most of the commercially avail­
                                                                                               High Plantar Pressures
most likely to reveal the progressive          able vibrometers have a fixed frequency
changes seen in chronic neuropathy and         of 120 Hz, which would tend to bias                There is general agreement that, once
compression syndromes.                         VPT measurements in favor of the Pa­            sensation has been lost, the presence of
   Birke and Sims used monofilament            cinian receptor system. This bias is of         high plantar pressures represents a
testing at 132 plantar ulcer sites in 72       great practical importance because high         strong risk factor for the development
patients with Hansen's disease and 45          frequency VPTs may not accurately re­           of plantar ulcers. Although most of the
plantar ulcer sites in 28 diabetic pa­         flect changes in cutaneous sensation typ­       supporting evidence is retrospective
tients.20 No patient could feel instru­        ically thought to be associated with neu­       rather than prospective,13,27,28 the argu­
ments smaller than a nominal value of          ropathic ulcers in the diabetic foot. The       ment that high pressure causes damage
6.10 units (log [10 x buckling force in        value of high frequency testing is not          to tissues is logical. It has also been
milligrams]) at multiple plantar test sites    clear but may be related to neuropathic         shown that, for reasons as yet unknown,
including the area adjacent to the ulcer.      changes associated with bone injuries.          diabetic patients with neuropathy tend
The ability to perceive the 5.07 mono­         New measurement systems for vibratory           to have higher plantar pressures than
filament was defined as a protective           sensation are emerging that will provide        either nondiabetic controls or diabetic
level of sensation. Application of this        both variable amplitude and variable            patients without neuropathy.29 Thus, a
definition has been shown to discrimi­         frequency stimuli (Lawrence Lee, Sen-           knowledge of pressure-distribution meas­
nate well between diabetic patients with       sortek, Inc, Clifton, NJ; personal com­         urement is important in the assessment
clinical evidence of neuropathy and no         munication; November 1987). These               of the diabetic foot.
history of plantar ulceration and dia­         new measures will allow the use of                 Force and pressure are often used in­
betic patients with neuropathy and             widely separated frequencies, such as 25        correctly as synonymous terms. Force is
previous or active ulceration.21 These         and 250 Hz, for testing.                        entirely independent of the area over
measurements were performed retro­                 Thermal perception has recently been        which it is applied. Pressure is the ap­
spectively but nevertheless suggest that       suggested to be a sensitive method of           plied force divided by its area of appli­
threshold sensation testing may indeed         detecting early neuropathic changes.25 A        cation. The SI (International System of
be useful in a prospective sense to iden­      quantitative measurement system,                Units) unit of pressure is the pascal,
tify those individuals who are at risk for     Thermal Sensitivity Tester ,* is now            which is defined as 1 N.m - 2 (1.45 ×
ulceration.                                    commercially available. The device con­          10-4 psi). Thus, the same force (eg, 1.2
   Boulton and associates attempted to         sists of two thermal plates that may be         times the body weight that can occur
relate vibration perception thresholds         varied independently to produce tem­            under one foot during the single-support
(VPTs) to risk of foot ulceration.13 These     perature differences between plates as          phase of gait) can result in widely differ-
investigators used a 120-Hz, fixed-            small as 0.1°C over a 50°C range. The           ent forefoot pressures, depending on
frequency vibrometer and found that            testing protocol for this device has been       such factors as the anatomy of the foot,
VPT differentiated between diabetic pa­        described in detail by Arezzo et al,26 but      the type of footwear worn, and the bio-
tients with neuropathy, nonneuropathic         no data relating thermal perception to          mechanics of the gait pattern.
diabetic patients, and nondiabetic con­        risk of plantar ulceration are available.          A variety of methods are available to
trols. Despite significant statistical find­       Although a relationship exists be­          study pressure distribution. These meth-
ings, there was considerable overlap in        tween nerve conduction and sensation            ods range from an examination of foot-
the patient score distributions, which         measurements, there is certainly not a          wear or insoles, which gives some infor-
indicated a high degree of variability in       1:1 correspondence. Functional tests of        mation on the history of peak pressures,
the VPT measurements. A follow-up              sensation allow patients to be followed         to the use of sophisticated devices that
analysis of diabetic patients with neu­        for a longer period of time before a            allow the measurement and display of
ropathy and ulceration showed a VPT            ceiling (maximum value) is reached on           the instantaneous pressure during
of ≥35 V (12.25-mm displacement) to            the respective scale. This rationale alone      ground contact.30,31 Although most de-
be related to ulceration.                      is a strong reason to use clinical meas­        vices monitor the interface between the
   Preliminary studies by another inves­       urements of sensation in addition to            bare foot and the surface of the instru-
tigator (J. A. Birke; Gillis W. Long Han­      sensory nerve conduction studies be­            ment, experiments are being conducted
sen's Disease Center, Carville, La; un­        cause the latter are very sensitive to early    with instruments to measure the pres-
published data; October 1987) have also        neuropathic changes but rapidly lose the        sure between the foot and the shoe,32
shown highly variable VPT at 120 Hz            power to detect further impairment.             which is the critical interface in the pa-
in diabetic individuals with a known           Additional justification would be the           tient's everyday life. Typical presenta-
history of foot ulceration. A probable         established relationship between mono­          tions of pressure data for barefoot step-
explanation is that both Pacinian and           filament and vibratory thresholds and          ping are shown in Figure 1. Time-series
non-Pacinian (primarily Meissner's cor­        injury, or what has been defined as loss        plots (Fig. 1A) provide valuable infor-
puscles) receptors mediate the detection                                                       mation about the loading pattern of the
of vibration,22 and the activation of            *Sensortek, Inc, 154 Huron Ave, Clifton, NJ
                                                                                               foot. A peak-pressure summary diagram
these two systems is frequency depend­         07013.                                          (Fig. 1B) shows the highest pressures

Volume 68 / Number 12, December 1988                                                                                                1889
Fig. 1. (A) Time-series plot of the instantaneous plantar pressure from a diabetic patient during stepping. Contour intervals are 75 kPa. (B)
Summary diagram depicting the highest peak pressures that occurred at any time during the support phase of gait for the same patient. Contour
intervals are 100 kPa.

experienced by each region of the foot          underlying tissues in a more concen-             factor for the development of foot le-
during the entire contact interval, re-         trated manner than in the normal foot.           sions in diabetic patients. Disturbances
gardless of the time of occurrence.             A fourth point, and perhaps the most             of the autonomic function may result
   There is some controversy over the           important, is that there is likely to be         in decreased sweating, arteriovenous
establishment of a criterion pressure           interaction between plantar pressure,            shunting, edema, and increased bone
above which it is likely that injury to         the degree of peripheral neuropathy, and         blood flow. Denervation of the sweat
the tissue will occur. Boulton et al13 have     other less well-defined factors in the           glands will result in decreased sweating
suggested a value of 1 MPa (measured            etiology of plantar ulcers. Thus, mod-           of the skin and increased risk of plantar
between the foot and the floor), but            erate pressures and dense neuropathy             ulceration.36 The basis for this assertion
ulcers have been observed in patients           may be more damaging than high pres-             is that dry skin is thought to be less
with considerably lower plantar pres-           sures and mild neuropathy.                       compliant than moist skin and, there-
sures.33 The problem of establishing a             An example of a plantar pressure dis-         fore, more likely to crack. In one study,
criterion value is complicated by nu-           tribution in a patient who has experi-            100% of the feet of patients with a his-
merous factors. First, the variability in       enced a plantar ulcer is shown in Figure         tory of plantar ulceration exhibited de-
the normal population is quite large.34         2. A comparison of the location of the           creased sweating, as measured by an
The usual approach of defining the              ulcer with the peak-pressure plot pro-           increase in electrical skin resistance.13
mean pressure in a large ulcer-free group       vides fairly convincing evidence of the          Increased arteriovenous shunting is an-
of subjects, plus or minus two standard         association between ulceration and high          other mechanism by which autonomic
deviations, produces a range that is al-        pressure. In contrast, the pressure distri-      neuropathy may contribute to the de-
most certain to include individuals who         bution in Figure 3 is from a patient with        velopment of foot lesions in diabetic
are at risk for the development of le-          higher pressures and similar loss of sen-        patients.37 There is extensive evidence
sions. Second, although most instru-            sation, but this individual has not ex-          to suggest that autonomic denervation
ments measure only the pressure at right        perienced a plantar ulcer. These exam-           leads to opening of arteriovenous chan-
angles to the contact surface (the "nor-        ples emphasize the importance of a mul-          nels, mostly in the skin but also at other
mal" stress in mechanical terms), the           tifactorial approach to the evaluation of        deeper sites. This shunting may decrease
tissue is sensitive to stress in all direc-     risk in the diabetic foot because we cur-        skin capillary bed perfusion despite an
tions. Pollard et al have suggested that        rently know of no single predictor of the        overall increase in blood flow in the
the "shear" stress in particular could be       tendency to ulcerate.                            foot, thus adversely affecting tissue via-
important.35 Third, the effect of the                                                            bility. Peripheral edema is another pos-
pressure will differ depending on the                                                            sible consequence of arteriovenous
                                                Autonomic Neuropathy
previous history of a particular foot. For                                                       shunting,37 which may be relevant in the
example, prior ulceration may have left           The presence of autonomic neuropa-             pathogenesis of ulceration. Neuropathic
fibrous tissue that will transmit force to      thy is thought to be an important risk           fractures are probably the most devas-

1890                                                                                                               PHYSICAL THERAPY
Fig. 2. (Left) Foot photograph from a patient with loss of protective sensation, hallux rigidus, and an ulcer (indicated by arrow) under the great
toe. (Right) Diagram depicting peak plantar pressures during late support phase of gait for the same patient. Contour intervals are 125 kPa.

Fig. 3. (Left) Foot photograph from a patient with loss of protective sensation, marked equinus deformity, heavy callus formation (indicated by
arrow), and absent history of ulceration. (Right) Diagram depicting peak plantar pressures during late support phase of gait for the same patient.
Contour intervals are 125 kPa.

Volume 68 / Number 12, December 1988                                                                                                        1891
                                                              PLANTAR ULCER

                                                                    Wound Care

                                                            No                       Yes

                            Walking Casts                         Bed rest or NWB
                                                                                                      Modified Footwear
                             or Splints                             Ambulation

                      No                                    No                                                            No
                              >60 days                                 Healed                             Healed

                                     Yes                                    Yes                                 Yes


                                                                  Interim Footwear


                                                                 Definitive Footwear

Fig. 4.   Plantar ulcer treatment protocol (NWB = nonweight-bearing). (Adapted from Birke and Sims.101)

tating consequence of diabetic neurop-           autonomic nervous system is complex,            Clarke have suggested that there is a
athy, and a leading theory of the                comprising sympathetic and parasym-             "typical" sequence of autonomic loss in
pathogenesis of these fractures impli-           pathetic, adrenergic and cholinergic,           diabetic patients.39 Decreased sweating
cates increased bone blood flow second-          and afferent and efferent pathways.             in the feet and impotence are often
ary to autonomic neuropathy.37,38                Many body systems are under some                viewed as occurring early after the onset
   There are many methods for evaluat-           autonomic control; perhaps the most             of diabetes, but conflicting data have
ing autonomic neuropathy; however,               noteworthy are the cardiovascular, gas-         also been presented suggesting that
their relative utility and practical impor-      trointestinal, urogenital, and cutaneous        changes in the leg occur late after
tance, particularly with respect to the          (sweating) systems. Testing one pathway         onset.40
foot, are poorly defined. This confusion         may reflect little on the function of an-          The integrity of the autonomic nerv-
has contributed to the limited under-            other. Indeed, it is reasonable to expect       ous system is often assessed by measure-
standing of the role of autonomic                that dropout of autonomic pathways at-          ment of cardiovascular reflexes such as
neuropathy as a risk factor for the de-          tributable to diabetic neuropathy will          heart rate and blood pressure responses
velopment of diabetic foot lesions. The          not be uniform, although Ewing and              to different challenges, but these reflexes

1892                                                                                                               PHYSICAL THERAPY
are of limited relevance to the foot be-       scar tissue that develops as a result of      explanation is likely because heel fat-
cause they reflect changes in both car-        the normal healing process is noncom-         pad thickness was also noted to be de-
diac and peripheral circulation function.      pliant. Stresses on the bottom surface of     creased in diabetic patients. Signifi-
Peripheral contributions to the cardio-        the foot that would ordinarily be dissi-      cantly less fat was found underneath the
vascular reflexes can be isolated partially    pated in the normal, mobile soft tissues      heads of the first and second metatarsals
by using temporary veno-occlusive              become concentrated in the adherent           in diabetic patients with plantar ulcers
plethysmography.41 Total limb blood            scar.11 Special attention must be given       when compared with a nondiabetic con-
flow, vascular resistance, and venous          to recently healed ulcers to prevent          trol group. Diabetic patients without
tone can be measured at rest and in            reoccurrences.                                plantar ulcers had intermediate fat-pad
response to a variety of stimuli. Leg             Transfer (iatrogenic) lesions may oc-      thicknesses, which were less than the
blood flow is increased in patients with       cur following amputation. For example,        control group but not significantly dif-
autonomic neuropathy, and the respon-          removal of the great toe may result in        ferent from the group with ulcers. These
siveness of the local vascular bed to          an ulceration beneath the first metatar-      findings add another dimension to the
stimuli is decreased.37 Blood flow can         sal head. Loads that were formerly            anecdotal reports implicating foot struc-
also be measured using Doppler tech-           shared by the great toe and first meta-       ture as a risk factor for plantar ulcera-
niques. A "pulsatility index" based on         tarsal head must be accepted by the first     tion. Skeletal deformities alone may not
Doppler measurements has been sug-             metatarsal head alone, thus often pro-        be sufficient to cause ulceration without
gested as an indicator of the fast diastolic   ducing an injury. Careful planning of         the concurrent atrophy or displacement
flow seen in the leg in patients with          surgical procedures will help to mini-        of the plantar fat pads.
diabetic neuropathy.37                         mize the possibility of stress-transfer le-      In addition to atrophy of muscle and
   Other methods of assessing auto-            sions. Much research is still needed in       adipose tissue, changes in connective
nomic function in the lower extremities        this area, particularly addressing the ef-    tissue as a result of diabetes may also
include measurement of sweating and            fects of specific surgical procedures on      play a role in the development of de-
skin temperature. Several methods for          plantar pressure distribution.                formities leading to plantar lesions. In
the measurement of sweating exist,39 but                                                     the presence of hyperglycemia, both col-
perhaps the most accurate is the recently      Foot Deformity                                lagen and keratin are known to undergo
devised method of Low et al called the            Diabetes mellitus is widely believed to    nonenzymatic glycosylation, the spon-
quantitative sudomotor axon reflex test        cause structural alterations in the foot      taneous attachment of free glucose.47,49
(Q-SART).42 In this procedure, local           that increase the risk of ulceration by       Nonenzymatic glycosylation results in
sweating is induced by acetylcholine           altering the plantar pressure distribution    increased cross-linking and changes in
iontophoresis, and the amount of mois-         or compromising the fit of footwear.          the mechanical properties of the tissues
ture produced is measured using a              Claw toes, plantarly prominent metatar-       and in their enzymatic digestion by col-
special apparatus. This direct test of         sal heads, intrinsic muscle weakness,         lagenase.50 The build-up of rigid keratin
autonomic function in the foot is easily       acquired rear-foot varus, and equinus         on the plantar aspect of the foot may
performed in a clinical setting. Ther-         deformity of the ankle are frequently         have a damaging effect on underlying
mometry may also be useful for indi-           mentioned as being typical findings in        tissue and, therefore, may be a predis-
rectly evaluating changes in blood flow        the diabetic foot.3,44,45 All of these de-    posing factor for ulceration. Limited
attributable to autonomic neuropathy.          formities are presumably related to mus-      joint mobility is common in diabetic
In general, the baseline temperatures in       cle weakness secondary to neuropathy.         patients and is currently believed to be
the feet of patients with diabetic neurop-     Anterior displacement of the submeta-         secondary to nonenzymatic glycosyla-
athy are warmer, which suggests in-            tarsal fat pads has also been cited as a      tion of periarticular collagenous struc-
creased blood flow at rest as a result of      common structural change related to           tures.51,52 This factor, in part, may
arteriovenous shunting, and the re-            claw-toe deformities.46 Deformities un-       explain the high incidence of hallux
sponse to stimuli are blunted, presum-         related to diabetes that may predispose       limitus and first-ray stiffness in these
ably because of inability to further in-       the foot to injury include hallux abducto     patients, which often results in ulcera-
crease blood flow.40,43                        valgus, forefoot varus, plantar-flexed        tion under the great toe or first metatar-
   In summary, studies attempting to           first ray, and primary rear-foot varus.       sal head.
evaluate the contribution of autonomic         Despite common beliefs, all of these              Foot deformities that have been iden-
neuropathy to the pathogenesis of dia-         observations must be classified as anec-      tified as potentially relevant to the de-
betic foot lesions should use measures         dotal findings because we could not lo-       velopment of plantar ulcers should be
of autonomic dysfunction in the foot           cate any experimental investigations          quantified. Routine measurements
(local sweating, thermography, or blood        documenting a relationship between            should include an assessment of claw
flow). The outcomes of these tests are         joint deformities and foot lesions.           toes, hallux limitus, first-ray stiffness,
more likely to be related to foot injuries        Recently, some quantitative evidence       forefoot varus, rear-foot varus, subtalar
than testing of more proximal functions        has been presented by Gooding et al,          joint stiffness, limitation of ankle dorsi-
(cardiovascular reflexes).                     who demonstrated that there is typically       flexion, and foot morphology. Claw toes
                                               a loss of thickness in the submetatarsal      and hallux mobility may be measured
Previous Ulcer or Amputation                   fat pads in the diabetic foot.10 No men-      using standard goniometric technique
  Clinical observations suggest that the       tion was made regarding the presence of       with the foot in a weight-bearing posi-
most likely site for plantar ulceration in     claw-toe deformities, so it is unclear        tion. This position has been suggested
an insensitive foot is at the location of a    whether the observed changes were pri-        to increase the reliability of measure-
previous ulcer. This fragility is probably     mary loss of soft tissue, changes second-      ments of digital deformities.53,54 First-
due to two factors: 1) The newly formed        ary to toe deformities, or perhaps a com-      ray mobility may be quantified using
epithelium is extremely thin, and 2) the       bination of these two factors. The latter     the method described by Rodgers and

Volume 68 / Number 12, December 1988                                                                                             1893
Cavanagh.55 The remainder of the clin-      in abnormal weight-bearing patterns.          metatarsal fractures. Plantar ulceration
ical examination, with the exception of     The generalized osteoporosis that is as-      complicated by osteomyelitis of the
foot morphology, has been described in      sociated with diabetes mellitus is also       metatarsal heads may also contribute to
detail by Smith et al.56 A one-half body    believed to be important. Increased           fractures in the forefoot.61
weight footprint can yield useful infor-    blood flow to bone secondary to auto-            Clinically, neuropathic fractures
mation regarding foot morphology. The       nomic (sympathetic) neuropathy has            should be suspected in all patients with
concept of an "arch index," which can       been implicated recently as a probable        signs of inflammation (increased
be used to quantify the footprint, is a     cause of foot bone decalcification.59,60      warmth and swelling) of the foot in the
means of classifying feet into "low-        The mechanism is believed to be in-           absence of an open wound. Thermo-
arch," "normal-arch," and "high-arch"       creased osteoclastic activity stimulated      graphy64 and bone scans65-67 may be use-
groups.57 Other special evaluative pro-     by the accelerated blood flow. Hypere-        ful in detecting neuropathic destruction of
cedures that may be useful include          mia associated with infection may have        bone not detectable by conventional ra-
weight-bearing radiographs and ultra-       a similar effect.61 Other factors, includ-    diography. Bone destruction should be
sonographic measurement of submeta-         ing bone demineralization attributable        assumed to be secondary to neuropathy
tarsal fat-pad thickness.                   to bed rest or immobilization, may also       unless bone cultures are available because
   It would appear reasonable to com-       contribute to neuropathic fracture.           neuroarthropathy is indistinguishable
bine the various measures of foot de-          The incidence of neuropathic frac-         from osteomyelitis on radiographs.
formity mentioned above—and others          tures is much less than that of plantar
that may later appear to be relevant—       ulcers, approximately 1 in 800 diabetic       Abnormal Foot Function
into a "foot-deformity index." Such an      individuals.58 Neuropathic fractures oc-
index would contain weighted contri-        cur most frequently in the tarsometatar-         It is surprising how often a foot may
butions from those deformities that can     sal region followed by the metatarsopha-      appear, from physical examination, to be
be related to ulceration by prospective     langeal joints and lastly the ankle           free of significant deformity yet may func-
studies. Because the needed information     and subtalar joints.62 Two general cate-      tion in a manner that increases the risk of
is not currently available, an interim      gories of neuropathic bone changes are        injury. Although a static examination
alternative would be to simply assign       recognized in the diabetic foot: 1) those     gives some indication of the causes of
unweighted ranks to the various foot        attributable to neuroarthropathy (hy-         abnormal function, a dynamic examina-
deformities and use the sum as a com-       pertrophic changes) and 2) those attrib-      tion reflects the compensation for any
posite index. This procedure would un-      utable to osteolysis (atrophic changes).      misalignment or restriction in motion.
doubtedly result in some loss of            Hypertrophic changes involve the prox-        The patient's chosen pattern of compen-
sensitivity.                                imal joints of the foot and ankle,            sation and the magnitude of any abnor-
                                            whereas atrophic changes are usually          mal compensatory movements are fre-
Neuropathic Fractures                       confined to the distal metatarsals and        quently not predictable on an a priori
   A neuropathicfracture may be defined     phalanges.63                                  basis. For these reasons, the patient's gait
as a disruption of the bone in the ab-         Neuroarthropathy is characterized by       should be studied routinely during a com-
sence of overt trauma in an individual      primary involvement of the joints and         prehensive examination. During a visual
who has peripheral neuropathy. The im-      periarticular structures with associated      kinematic assessment, attention should be
portance of neuropathic fractures is that   reactive periosteal new bone growth.          directed successively at the hips, knees,
they may produce gross foot deformi-        Neuropathic fractures involving the tar-      ankles, rear feet, forefeet, and toes from
ties. These structural changes usually      sus are the most devastating. Harris and      both frontal and sagittal views. Common
cause dramatically increased plantar        Brand outlined five patterns of tarsal        gait problems that may result in foot in-
pressures, resulting in a major risk of     disintegration in the insensitive foot.61     juries can often be detected by observa-
ulceration.                                 Destruction of the medial arch was re-        tional kinematic analysis (Tab. 2), but
   The etiology of neuropathic fractures    ported to be the most common type of          equivocal findings should not be over-
is incompletely understood.38 Classi-       injury. Typical radiographic changes          stated because these procedures are only
cally, these fractures are viewed as oc-    were described as dorsal compression of       moderately reliable.68 Patients should use
curring in the diabetic foot primarily      the navicular followed by involvement         whatever walking aids they normally use,
because of the loss of sensation. Minor     of the head of the talus. Late changes        and unless contraindicated by the risk of
trauma is hypothesized to produce mi-       involved a total collapse of the midfoot      damage to the foot, they should be ob-
crofractures that are not recognized be-    producing the typical rocker bottom            served both barefoot and in preferred foot-
cause of loss of pain sensation. Pro-       foot deformity.                               wear. Great care must be taken during
longed vigorous activity without the            Osteolysis refers to resorption of bone   barefoot walking to make sure that no
benefit of a pain warning system then       in the distal metatarsals and phalanges,       sharp objects that could cause foot injury
leads to further bony destruction.58 The    which produces a characteristic "pencil-      are present in the environment.
process often is not noted until marked     point" appearance of the metatarsal               A variety of quantitative methods are
bony destruction has occurred. On rare       shaft. The pathogenesis of these changes      available to assess the dynamic motion of
occasions, the patient will remember a      is not clear but may be related to re-        the lower extremity. These methods in-
specific minor injury that occurred sev-    peated minor fractures with unsuccess-        clude conventional video recording, high-
eral weeks before the onset of a grossly     ful attempts at healing resulting in con-     speedfilmor video, and the use of optoe-
deformed foot. Another factor likely to      centric bone loss.38 Although diabetic        lectronic devices.69,70 Although protocols
contribute to the pathogenesis of neu-       osteolysis initially does not involve the     for the use of these devices in research are
ropathic fractures is impairment of mus-    joints and periarticular structures, this     well established, specific clinical protocols
cular function (strength and propriocep-     process may secondarily involve the           relevant to foot pathology remain to be
tion) secondary to neuropathy resulting      metatarsophalangeal joints following          validated.

1894                                                                                                         PHYSICAL THERAPY
High Activity Level                             TABLE 2
                                                Common Gait Problems and Their Relationship to Injury in the Diabetic Foot
   Although it seems reasonable to sup-
pose that the risk of ulceration would be                       Problem                                   Relationship to Injury
increased by vigorous physical activity,        Excessive pronation                              Increased shearing on medial forefoot
there are no systematic studies in the lit-     Excessive supination                             Increased lateral forefoot pressures
erature relating the specific risk of foot                                                         and lateral instability
injury to activity level. LaPorte et al have    Midfoot weight-bearing                           Neuropathic destruction of tarsal
studied the relationship between some of                                                           bones
the other complications of Type I diabetes      Ankle equinus                                    Increased forefoot loading and func-
and chronic physical activity.71 Some oc-                                                          tional lower extremity length dis-
cupations can clearly be identified as in-                                                         crepancy
                                                Abductory twist                                  Increased forefoot shear loads
creasing the risk of direct trauma to the       Marked asymmetry in gait                         Increased stress on preferred side
foot, but the subtle changes that result        Impaired balance and navigation                  Increased risk of falling and self-injury
from increased cumulative loading (which        Excessive out-toeing or in-toeing                Increased stress on medial or lateral
might be defined as the product of inten-                                                          border of the foot
sity and frequency) are more difficult to       Drop foot deformity                              Instability and increased risk of ankle
quantify.                                                                                          sprains and fractures, toe injuries
   Devices to measure the number of foot-                                                          attributable to inadequate floor
strikes during a given time period are                                                             clearance, increased forefoot pres-
occasionally used in postcoronary care to                                                          sures from secondary equinus
assess patient activity levels, but, to our
                                                Amputations or shortened foot                    Increased pressure over distal end of
knowledge, they have not been used for                                                             foot from a shoe that is too long,
diabetic patients. Force transducers have                                                          blistering from rubbing on toe fillers
been used to monitor the loads applied to                                                          in footwear
the lower extremity during walking.72 An
insole device called a "limb load monitor"
is placed inside the patient's shoe, and an     Wounds located on nonweight-bearing               vascular screening test is the segmental
audible tone sounds when a given load is        areas of the feet, decreased local skin tem-      systolic pressure test.74 Systolic pressures
exceeded. Thus, the patient who, for ex-        perature, and delayed healing of lesions          are measured at the arm and ankle using
ample, may be recovering from joint-            are all possible signs of vascular compro-        a standard pneumatic cuff and an ultra-
replacement surgery learns not to over-         mise. In addition, a primary presentation         sound Doppler unit. The pressures are
load the extremity beyond the preset            of gangrene is obviously attributable to          usually expressed as a ratio of ankle to
value, which is adjustable as recovery pro-     vascular disease.                                 arm values (ankle-brachial index) for the
gresses.73 A similar device, in which plan-        Circulatory abnormalities in the dia-          same side of the body. Wound healing
tar pressure is sensed and fed back, would      betic foot are quite complex. Functional          has been reported to be likely if the ankle
seem to have application for the diabetic       alterations of the circulation secondary to       systolic pressure is >80 mm Hg or the
foot.                                           autonomic neuropathy may be very im-              ankle-brachial index is X).45,79,80 but no
   In our clinic, we routinely ask a number     portant in the pathogenesis of foot lesions       conclusive statements can be made re-
of questions to characterize the patient's      but are still poorly understood. The role         garding the circulation distal to the level
activity level. This evaluation includes        of diabetes as a risk factor for atheroscle-      of measurement (F. W. Wagner, personal
generating a log of a typical day (eg, hours    rosis is well established. Diabetic individ-      communication, March 1988). These cri-
spent standing, sitting, walking, sleeping),    uals tend to have a similar frequency of          teria are considered to be useful because
recording relevant habits (eg, use of stairs,   proximal (iliac and femoral) atheroscle-          most major amputations are performed
kitchen work), and forming a profile of         rotic lesions as nondiabetic individuals.75       because of large-vessel involvement.58
the patient's occupational and recrea-          In contrast, they tend to have more disease       Vascular lesions may be localized by
tional activities. From this composite, a       in thetibialand peroneal arteries and less        measuring the systolic pressures in multi-
subjective impression emerges, which, if        in the arteries of the foot.76-78 A frequent      ple limb segments.
nothing more, allows identification of          misconception about circulation in the               Occasionally (5%-10% of patients),
those patients with activity levels that        diabetic foot concerns "small-vessel dis-         ankle systolic pressures may not be
place them at high risk. Much research          ease." Microcirculatory disease is often          obtainable or may be greatly elevated
needs to be completed to quantify the           blamed for diabetic foot ulcers, yet there        because of the inability to occlude a non-
 specific effect of different regimens of ac-   is little evidence that this condition exists.    compliant artery.79 This fact has produced
tivity on the diabetic foot.                    A thickening of the basement membrane             some controversy regarding the validity of
                                                surrounding muscle capillaries is a well-         these measurements,77,81 but a recent
                                                established lesion in diabetes, but a similar     study has shown that ankle indexes may
Vascular Disease
                                                lesion is not found in the skin.76,77 Even        be predictive of healing as long as the
   Vascular disease is viewed too fre-          in muscle, the functional significance of         results are interpreted cautiously, disre-
quently as the major pathogenetic factor        this thickened basement membrane is in            garding values >1.5.82 The difficulty in
in the development of diabetic foot le-         question. The concept of small-vessel dis-        interpreting ankle pressures has led to the
sions. We believe loss of sensation to be       ease should be abandoned because the              investigation of the predictive value of
more important, but the possible role of        arteries of the foot are often free of athero-    great-toe pressures. Absolute systolic toe
vascular disease cannot be ignored. Dia-        sclerotic changes and surgical revasculari-       pressures of >30 mm Hg are thought to
betic foot lesions are unlikely to heal if      zation is frequently successful.76-78             be consistent with healing.83
the circulation is markedly impaired.74            The most commonly used noninvasive                Segmental plethysmography may also

Volume 68 / Number 12, December 1988                                                                                                         1895
be used to provide an estimate of the           ening is usually the result of wear              fortunately, standard test instruments of
regional circulation.84 Gibbons et al           compression of soft orthotic materials,          this type are currently unavailable.
showed that in 100 patients undergoing          such as polyethylene foams, which causes
forefoot amputation, a strong pulse vol-        "trapping" of the toes, thus restricting the     Vision Loss
ume recording (pneumatic plethysmog-            normal forward movement of the foot in              Although generally not thought to be
raphy) was the only noninvasive method          the shoe during walking.                         related to foot problems, the impairment
to correctly predict healing.81 Other non-          Poor foot-care habits appear to dramat-      of vision—which is a frequent complica-
invasive testing methods include                ically increase the risk of foot injury. Pa-     tion of diabetes mellitus—can have severe
ultrasound85 and laser Doppler flowme-          tient education programs should strive to        consequences for the foot. Data are not
try.86 Ultrasound provides information          eliminate behaviors that can lead to in-         available concerning the incidence of foot
aboutflowpatterns in arteries; laser Dop-       jury. The most important rule is that pa-        injury during walking in patients with
pler flowmetry, still a new technique, pro-     tients with insensitive feet must never          vision loss, but Johnson and Keltner have
vides information about blood flow in           walk barefoot. Barefoot walking greatly          shown that patients with bilateral periph-
skin capillaries. Transcutaneous partial        increases the probability of an injury be-       eral scotomas are three times more likely
pressure of arterial oxygen measurement         cause of increased stress on the plantar         to be involved in an automobile accident
has also been used to evaluate skin per-        surface and penetrating wounds from              than those with normal vision.92 It seems
fusion. This technique has been shown in        sharp objects such as a discarded insulin        likely that the statistics for foot injury
a recent study to be superior to Doppler        needle. Sandals may be worn inside the           during ambulation would be similar, par-
pressures in the assessment of local limb       house as an alternative to shoes.                ticularly because obstacles to locomotion
perfusion.87 The patient should be referred                                                      are sensed with peripheral vision and pe-
for follow-up arteriography whenever                Common problems associated with ne-
                                                glect or improper foot care are increased        ripheral scotomas are rarely recognized by
there is doubt about the results of nonin-                                                       the individual (H. L. Liebowitz, The
vasive testing, particularly if a distal        plantar pressures attributable to thick cal-
                                                lus lesions, nail-bed infections, and crack-     Pennsylvania State University; personal
revascularization procedure is being                                                             communication; June 1988).
considered.                                      ing of dry skin. Daily inspection of the
                                                 feet and shoes is essential for prevention         Beyond the risk of trauma during lo-
Inadequate Footwear or                           of injuries. If the patient is not capable of   comotion, injury may occur during at-
Foot-Care Habits                                 adequate self-examination, the help of a        tempts at foot self-care, particularly nail
   Few experiments have been conducted           spouse or companion may be needed.              care. The problem is frequently com-
to determine the effects of different foot-      Brief soaking or bathing of the feet fol-       pounded by the presence of mycotic nails
wear or foot-care conditions on the risk         lowed by application of oil, which acts as      that are difficult to trim with ordinary nail
of foot injury. This lack of research is         an evaporative barrier, will replace mois-      clippers. A patient with impaired vision
unfortunate because these risk factors are       ture lost from the skin and prevent crack-      will also be unable to perform an adequate
clearly modifiable. As a result, the infor-      ing.88 Any opening in the skin provides a       self-inspection of the foot. Thus, early
mation contained in this section repre-          route for infection in the foot.                signs of damage usually are not detected
sents the opinions of the authors based                                                          because 1) they cannot be felt because of
                                                    The primary benefit of developing a
upon their experience, supplemented                                                              loss of sensation secondary to neuropathy
                                                 systematic approach to personal foot care
by the few reports that do exist in the                                                          and 2) they cannot be seen because of
                                                 is that problems are usually detected be-
literature.                                                                                      poor vision. In all of these problems re-
                                                 fore they progress to major injuries. One
                                                                                                 lated to vision, enlisting the help of a
   It is important to recognize that al-         study performed in a large teaching hos-
                                                                                                 spouse or companion who has the sensory
though shoes are generally believed to           pital showed a 50% reduction in the num-
                                                                                                 modalities that the patient has lost is es-
prevent injuries, they can also increase the     ber of amputations in their diabetic
                                                                                                 sential. Routine testing of vision will pro-
risk of foot injury. If protective sensation     patients following the introduction of a
                                                                                                 vide the clinician with knowledge of the
is lost because of neuropathy, poorly de-        comprehensive foot-care educational pro-
                                                                                                 patient's visual status, which is important
signed or improperly fitting shoes may           gram.89 Prevention of diabetic foot com-
                                                                                                 for the prevention of foot injuries.
cause serious foot problems. The most            plications is also highly dependent on the
common footwear-related injuries are le-         knowledge of professional staff and the
                                                                                                 Inadequate Diabetic Control
sions on the toes and sides of the forefoot.     level of cooperation between medical dis-
Claw-toe deformities may result in pres-         ciplines involved in the treatment of the          Neuropathy and vascular disease, both
sure over the dorsal aspect of the proximal      patient.90-91 The most common methods           likely pathogenetic factors in the devel-
interphalangeal joints in a shoe with a toe      of evaluating the effectiveness of a patient    opment of diabetic foot lesions, are two
box that is too shallow. A narrow toe box        education program are direct observation        of the late complications of diabetes. Al-
will increase the pressures over the sides       of patient behavior during clinic visits,       though the hypothesis that good blood
of the forefoot and may result in ischemic       questionnaires, and frequency counts of         glucose control over many years can pre-
necrosis of the skin if the stress is applied    some event such as number of ulcers or          vent the late complications of diabetes is
continuously for several hours. The use of       amputations per unit of time. Although          an old one, it still cannot be considered
thick bandages inside a shoe can convert         all of these methods may be useful, ad-         proven. Nevertheless, the evidence in fa-
a well-fitting shoe into one that is danger-     ministering a validated questionnaire be-       vor of this hypothesis is quite compel-
ously narrow. Distal toe lesions are usually     fore and after an educational program           ling,93 and it is reasonable to recommend
related to a shoe that is too short. Insoles     would seem to be the most reasonable            good long-term control of diabetes to pa-
or sandals that are deeply molded in the         approach to obtain quantitative prospec-        tients in an attempt to prevent the late
toe region will cause an effective shorten-      tive information regarding patient knowl-       complications. Good long-term control
ing of the footwear, which externally ap-        edge. This evaluation should be followed        cannot be easily achieved without fre-
pears to be of adequate length. This short-      up by measurements of compliance. Un-           quent home self-monitoring of either

1896                                                                                                                PHYSICAL THERAPY
blood or urine glucose. Home self-moni-           therapy, such as total contact casting, have    on the individual's assumed level of risk.
toring of blood sugar has become possible         been shown to reduce healing times for          A higher level of risk requires more de-
over the last 10 years and has made a             plantar ulcers97,98 and to relieve plantar      tailed patient education, strict adherence
major impact on management of diabetes.           pressures,35,99,100 there is a lack of infor-   to the use of proper footwear, and more
Capillary blood is obtained by pricking a         mation on footwear management and               frequent follow-up visits.
finger, and the blood is then applied to an       treatment of neuropathic fractures. Much           All patients should be instructed in the
enzyme-impregnated strip, the color of            of the material in this section, therefore,     daily management of their blood glucose
which changes in proportion to the blood          currently cannot be supported by direct         levels as well as an overview of possible
sugar. Reflectance meters are available to        experimental results. Authoritative opin-       complications related to diabetes. Individ-
provide a numerical readout of the blood          ions have been cited, when available, to        uals with loss of protective sensation in
sugar, or alternatively, the blood sugar can      support treatment recommendations. De-          the feet should receive specific training in
be estimated visually by comparing the            spite a lack ofrigorousexperimental vali-       techniques of self-inspection, routine
color of the strip to that of a chart. Meas-      dation, the approach we suggest represents      skin care, and proper selection of foot-
urement of glycohemoglobin (hemoglo-              an organized scheme that will allow prac-       wear. All educational materials should be
bin A1c) or HbA1c, which provides an              titioners to apply treatments system-           written in clear, concise language. Specific
estimate of the average blood sugar over          atically so that their effects can be           guidelines for foot-care instruction are
a 6- to 10-week period, should be per-            evaluated.                                      outlined in Table 4. Videotaped instruc-
formed several times a year to confirm                                                            tion is preferred so that a consistent mes-
the results of home self-monitoring.              Prevention                                      sage may be presented at every viewing.
   Recent advances in biochemical re-                                                             This videotaped instruction may be sup-
search have strengthened the argument in             Regular and comprehensive screening          plemented by written materials and dis-
favor of "tight" control of diabetes as a         examinations are mandatory for early            cussion. A compendium of available edu-
means of preventing the late complica-            identification ofriskfactors in the diabetic    cational materials was recently produced
tions of the disease. A major theory in this      foot. Although the exact importance of          by the NIH.102 The best time to present
area involves the polyol pathway.94 Hy-           the various risk factors remains to be de-      information to the patient is usually fol-
perglycemia leads to accumulation of tis-         termined, it would appear reasonable to         lowing a treatment success because the
sue sorbitol, a product of the polyol path-       attempt a "blanket coverage" of all prob-       practitioner's credibility is enhanced. Be-
way. This accumulation is hypothesized            able risk factors when treating a diabetic      havior modification often requires a great
to contribute to tissue damage either             patient. Relevant historical information        deal of repetitive instruction over a series
through direct osmotic effects94 or through       should include a determination of the pa-       of visits. Modular instructional units will
changes in myoinositol metabolism and             tient's current neurological status with re-    allow easy review of only the needed in-
subsequent alterations in the polyphos-           spect to strength and sensation, previous       formation to prevent boredom.
phoinositide-signaling pathways.95 All of         plantar ulcers or amputations related to           Proper footwear is one of the most
the biochemical consequences of in-               diabetes, foot-care habits, and level of ac-    important factors in the prevention of
creased polyol-pathway way flux in dia-           tivity.                                         ulceration. Foot orthotic devices and
betes can be prevented by a new class of             Footwear should be inspected for             shoes should be matched to the level of
drugs, the aldose reductase inhibitors. Sev-      proper style and fit. The feet should be        risk of injury (Tab. 5). If deformities or
eral large-scale clinical trials of these drugs   examined for any skin lesions, signs of         high plantar pressures are combined with
are underway, and the aldose reductase            inflammation, or nonspecific swelling. A        a loss of protective sensation (risk category
inhibitors will likely become important in        clinical biomechanical foot examination         2), careful footwear management is re-
the management or, more likely, preven-           should be performed to document joint           quired to prevent wounds. If excessive
tion of the late complications of diabetes        ranges of motion and the presence of any        pressures cannot be reduced adequately
including foot problems.                          deformities. Neurological evaluation            by the use of special footwear, surgical
   There are also some theoretical reasons        should include measurements of sensation
to recommend good blood glucose control           thresholds for monofilaments and vibra-
in the presence of an acute foot lesion.          tion, strength of muscles innervated by
Inadequate control probably interferes            peroneal and tibial nerves, and evaluation      TABLE 3
with white-cell function, which is impor-         of deep tendon reflexes. Other special pro-     Modified Foot Risk Categories8
tant in combating infection.96 Also,              cedures that should be performed rou-
wound healing might be impaired in the            tinely include recording of ankle-brachial        Category
setting of very poor diabetes control, per-       indexes, gait observation, vision checking,
                                                  weight-bearing radiography, and measure-              0      Protective sensation intact
haps because of glycosylation of structural                                                             1      Loss of protective sensation
proteins.47,48                                    ment of glycosylated hemoglobin. Plantar
                                                                                                        2      Loss of protective sensation
                                                  pressure distribution and arch indexes also
                                                                                                                 and high plantar pressure
                                                  are extremely useful data if the equipment                     or deformity
MANAGEMENT OF THE                                 is available.                                         3      Loss of protective sensation
HIGH-RISK FOOT                                       Upon completion of a comprehensive                          and history of ulcer
                                                  screening examination, the patient should             4      Loss of protective sensa-
   Much research remains to be con-               be categorized according to the level of                       tion, history of ulcer, and
ducted to determine the etiology of foot          risk of foot injury. We have modified                          high plantar pressure or
problems in the diabetic patient. The lack        the risk categories originally suggested by                    deformity
of data is perhaps even more serious con-         Birke and Sims101 to include plantar                  5      Neuropathic fracture
cerning the effectiveness of various treat-       pressure-distribution measurements (Tab.          a
                                                                                                     Modification of risk categories originally
ment protocols. Although some forms of            3). Patient management should be based          suggested by Birke and Sims.101

Volume 68 / Number 12, December 1988                                                                                                     1897
correction of deformities may be indicated
on a prophylactic basis, particularly for      TABLE 4
toe deformities. The level of patient activ-   Patient Foot Education Guidelines
ity may also need to be adjusted to prevent    Self-inspection
skin irritation.                               1. Inspect all surfaces of feet daily for signs of skin irritation
   Patient follow-up should be adjusted        2. Feel for areas of increased temperature
according to the paaient'slevel of foot risk   3. Palpate plantar surface for tenderness
and current clinical status. Those individ-    4. Inspect socks and shoes before donning for worn areas or foreign objects
uals with intact protective sensation do       5. Replace worn socks and avoid use of obsolete footwear
                                               6. Report problems early to a medical professional
not require frequent detailed foot exami-
                                               Skin care
nations. Patients with loss of protective
                                               1. Remove excess callus by lightly sanding before bathing rather than cutting with blades
sensation should be seen annually for a        2. Hydrate skin daily by brief soaking or bathing followed by application of an emollient
complete foot examination, although            3. Trim nails straight across monthly; thickened nails may require professional assistance
more frequent informal checks are advis-       4. Superficial wounds should be kept clean and rested by removing weight from foot
able. If special footwear is required to       5. Report large or slowly healing (no signs of healing within 2-3 days) wounds to a medical
accommodate high pressures or deformi-            professional
ties, the patient should be examined at        Footwear selection
least every six months to ensure proper        1. A low-wedge crepe-soled shoe with a high rounded toe box, moldable upper (leather or
maintenance of shoes and foot orthoses.           nylon mesh), and adjustable closure (straps or laces) is the preferred style
More frequent follow-up visits are rec-        2. Fit shape and size of feet
                                               3. Choose shoes with extra depth if insoles are to be added
ommended if there has been a recent
                                               4. Never walk barefoot, including inside the house
change in neurological status such as loss     5. Break in new shoes slowly (<2 hours a day for first week)
of sensation or muscle weakness because        6. Repair or replace worn insoles or shoes
the patient may be more vulnerable to
injury during this time.
                                               TABLE 5
Plantar Injuries                               Footwear Recommendations by Risk Category
   An organized team approach is                            Risk Categorya                      Recommended Footwearb
required to heal ulcers and prevent reul-
                                                        0                           Shoes of proper style and fit
ceration (see article by Mueller and Dia-               1                           Soft nonmolded insoles and extra-depth shoes
mond in this issue). A four-level treat-                2                           Custom-molded foot orthoses, extra-depth shoes,
ment protocol (Fig. 4) was suggested by                                               and accommodative modifications
Birke and Sims.101 Proper wound eval-                   3                           Custom-molded foot orthoses, extra-depth shoes,
uation and care is the first level. The                                               and rigid rocker sole for forefoot ulcers
pressure under the wound must then be                   4                           Custom-molded foot orthoses, extra-depth shoes,
decreased for the wound to heal. Interim                                              rigid rocker sole, and accommodative
footwear should be provided in the im-                                                modifications
mediate posthealing period to protect                   0
                                                            Forefoot fracture       Custom-molded foot orthoses, extra-depth shoes,
the foot. Finally, definitive footwear is
                                                                                      and rigid rocker sole
selected based on the severity of the                       Midfoot fracture        Custom-molded shoes and foot orthoses, rigid
problem.                                                                              rocker sole, and reinforced medial shoe counter
   Recognition of plantar injuries. Plan-                   Rear-foot fracture      Fixed ankle boot or patellar tendon-bearing or-
tar injuries may be divided broadly into                                              thosis with surgical boot, custom-molded foot
two categories: 1) preulceration and 2)                                               orthoses, cushioned heel, rigid rocker sole, and
ulceration. We recognize a "preulcer" as                                              lateral heel flare
any undesirable change in the intact             a
                                                     See Table 3 for definitions of foot risk categories.
plantar skin. Any one or combination             b
                                                     Footwear should not be used as a primary treatment for healing ulcers or fractures.
of the following conditions would con-
stitute a preulcerous change:                  progress into a complete ulcer. Early              great toe, first metatarsal head, and fifth
 1. An increase in the skin temperature        detection by careful inspection and tem-           metatarsal head.20,28,104 During exami-
    locally of 2°C as compared with the        perature measurements is needed so that            nation, the index of suspicion should be
    contralateral foot (difference in skin     appropriate intervention can be effected           greatest for these areas.
    temperature) in the absence of sub-        to prevent such a progression. The most              Untreated foot ulcers usually result in
    stantial prior physical activity.          practical instrument for monitoring skin           deep infections with destruction of ten-
2. Intraepidermal maceration (a boggy          temperature in a clinical setting is a             don, bone, ligament, or joint. Wagner
    texture of the skin resulting from ex-     hand-held infrared thermometer or a                has outlined five different ulcer grades,
    travasation of fluid such as would be      thermocouple.103                                   depending on the severity of the lesion.80
    seen in an intact blister).                   A plantar ulcer is defined as an open           This grading scheme is useful for sur-
3. Hematoma or callus with underlying          injury resulting in a partial- or full-thick-      gical management of foot wounds but
    hematoma.                                  ness skin wound. Greater than 90% of               does not provide enough levels to dis-
   The importance of recognition of            all plantar ulcers have been reported to           criminate between wounds in the first
preulcerous changes is that, if left un-       occur in the forefoot region.20104 The             three categories, which are often man-
treated, such changes would be likely to       most common sites of ulceration are the            aged nonsurgically.105 We suggest an al-

1898                                                                                                                 PHYSICAL THERAPY
                                                                                            ing is the preferred method of obtaining
Proposed Foot Lesion Grading System Compared with Wagner's Classification System80          culture samples because superficial
                                                                                            swabbing rarely leads to representative
  Lesion Grades Recommended by Wagner               Proposed Grading System                 information about deeply infecting or-
0—Intact skin                                    0—Absent skin lesions                      ganisms.111 Use of proper culture tech-
                                                 1 —Dense callus lesions but no             nique is extremely important in the
                                                     preulcer or ulcer                      planning of antibiotic therapy.
                                                 2—Preulcerative changes                       Methods of decreasing pressure dur-
1 —Superficial ulcer involving skin only         3—Partial thickness (superfi-              ing healing. There are basically three
                                                     cial) ulcer
                                                                                            approaches that may be taken to reduce
2—Deep ulcer involving tendon, bone, liga-       4—Full-thickness (deep) ulcer
    ment, or joint                                   but no involvement of
                                                                                            the pressure on the bottom surface of
                                                     tendon, bone, ligament,                the foot during healing. The first
                                                     or joint                               method is the use of modified footwear
                                                 5—Full-thickness (deep) ulcer              such as a sandal with a relief cut out of
                                                     with involvement of ten-               the foot bed under the ulcerated area or
                                                     don, bone, ligament, or                a half-length sandal that eliminates all
                                                     joint                                  forefoot pressures. This approach
3—Localized infection (abscess or osteomy-       6—Localized infection (ab-                 should only be used with preulcers or
    elitis in foot)                                  scess or osteomyelitis) in             superficial ulcers. A second technique
                                                     foot                                   for deeper wounds is the use of bed rest
                                                 7—Proximal spread of infec-
                                                                                            or strict nonweight-bearing ambulation.
                                                     tion (ascending cellulitis
                                                     or lymphadenopathy)
                                                                                            Although this may be the ideal method
4—Gangrene of forefoot only                      8—Gangrene of forefoot only                for healing plantar wounds, it is often
5—Gangrene of majority of foot                   9—Gangrene of majority of                  unrealistic to expect patients to adhere
                                                     foot                                   strictly to such a regimen. A third, more
                                                                                            practical, approach is the use of con-
                                                                                            trolled weight-bearing devices. Walking
ternative classification system as shown     deep to prevent abscess formation. The         casts have been shown to be extremely
in Table 6. This new system emphasizes       opening of narrow, deep ulcers should          effective in reducing the forefoot pres-
the importance of early identification of    be enlarged slightly, followed by loosely      sures. 35-100 The mean healing time for
foot problems in the prevention of plan-     packing the wound with gauze to ensure         a plantar ulcer treated by casting is 5 to
tar ulcers.                                  that the deeper layers heal first.             6 weeks.97 The technique of applying a
   Wound care. All wounds should be             All foot ulcers should be considered        total contact walking cast has been de-
evaluated thoroughly, including explo-       to be infected. Superficial infections are     scribed in detail previously.112 All casts
ration using a blunt, sterile instrument     probably best managed by local care.           applied to insensitive feet should be
such as a nasal probe. The dimensions        Deep infections usually require the use        changed frequently to prevent second-
of the wound should be measured, and         of systemic antibiotics and possible sur-      ary abrasions attributable to looseness.
the lesion graded according to the depth     gical incision and drainage. The consen-       The recommended changing schedule is
and tissues involved (Tab. 6). Wound         sus currently favors extensive local           within one week of initial casting and
tracings on sterile acetate film may be      opening of infected tissue rather than         then at two-week intervals until healing
more useful than diameter measure-           attempting drainage through minimal            is complete. A posterior splint with a
ments if the borders are irregular. Serial   incisions and drains.108 There is also         forefoot relief pad and a walking heel is
drawings may be compared visually or         general agreement that complete non-           an alternative to casting that will allow
digitized electronically to obtain area      weight-bearing is essential to allow heal-     more frequent inspection of the skin.
measurements. Photographs provide            ing of a foot that has a deep infection.       The details of construction have been
another means of documenting wound           Rapid proximal spread of infection may         described elsewhere.101 We strongly rec-
location and size. A cardinal sign of a      occur if the patient is allowed to con-        ommend the use of crutches with all
deep ulcer that has penetrated a joint       tinue weight-bearing on the foot.16 The        walking devices to further reduce the
capsule is the production of a clear fluid   use of topical hyperbaric oxygen has           stresses on the foot. In the event that an
with bubbles. Leakage of the joint fluid     been suggested by some to be of value          individual has bilateral plantar ulcers,
often delays wound healing and may           in healing foot wounds,109 but in a re-        the most serious wound should be im-
require a partial synovectomy to elimi-      cent well-controlled study, this form of       mobilized with casting initially and
nate a chronic draining sinus.               therapy was shown to be totally ineffec-       modified footwear applied to the oppo-
   Necrotic tissue should be removed         tive on diabetic foot ulcers.110               site foot.
from the wound by blunt debridement.            The criteria for specific wound-care           Revascularization. If wounds fail to
Callus surrounding the wound should          measures are poorly defined, but a             heal after appropriate measures have
be trimmed neatly with a sharp blade         reasonable indication for the use of an-       been taken to reduce the stress on the
because its presence may prevent the         tibiotics is the presence of pronounced        foot, revascularization surgery may be
growth of new epithelium.106,107 Perox-      cellulitis.108 Diabetic foot infections con-   required. Noninvasive vascular tests will
ide may be used to cleanse the wound,        tain an average of five organisms, with        provide preliminary information re-
followed by the application of a topical     Staphylococcus aureus being the most           garding the probable location of arterial
antibacterial agent such as povidone-        frequently found pathogen.111 Approxi-         obstructions. If operable lesions are de-
iodine and a thin gauze dressing. Ideally,   mately two thirds of wounds are infected       tected on arteriography, measures to re-
the wound should be as wide as it is         by anaerobes. Superficial wound scrap-         duce plantar stress should be concomi-

Volume 68 / Number 12, December 1988                                                                                            1899
tant with revascularization surgery.           between feet. Dependent swelling in the       mobilization is downgraded. Finally, de-
    Immediate post-healing period. The         feet is usually greatest in late afternoon    finitive footwear is selected in the late
actions taken immediately following            or evening. Shoe measurements should          stage.
wound healing are critical for the pre-        be performed at midday to obtain a               Early stage (0-3 months). All sus-
vention of reulceration. Arigid,custom-        median fit. Sufficient time should be         pected fractures should be immobilized
molded sandal with a rocker sole design        allowed following injury or surgery for       with a nonweight-bearing below-knee
should be applied to the foot after cast       foot swelling to stabilize before selecting   cast for a period of two weeks regardless
removal. Advantages of an open sandal          definitive footwear. Size differences be-     of the initial radiographic findings be-
include accommodation of swelling,             tween feet are common. Shoes should           cause clinical signs of neuropathic frac-
greater protection of the plantar surface      be mismatched (two pairs required) if         tures may precede radiological findings
because the foot bed can be made               the difference exceeds one full size.         by several weeks.64 Three-phase bone
thicker than a shoe insole, and ease of           The most commonly used footwear            scans may also be useful in differentiat-
modification to relieve local stress. Elas-    for the diabetic patient is extra-depth       ing between bone and soft tissue le-
tic stockings should routinely be used to      shoes. An alternative to a traditional        sions.65-67 If follow-up radiographs are
control swelling after the removal of a        extra-depth shoe is a good quality run-       negative for fracture, the patient may
cast. Careful monitoring of the skin tem-      ning shoe with a removable insole.115         return to a normal level of activity; oth-
perature and swelling will provide ad-         Soft insoles or custom-made foot or-          erwise, casting should be continued un-
vance warning of impending reulcera-           thoses may be placed in this type of          til initial healing is complete. In general,
tion or neuropathic fracture caused by         footwear without danger of causing in-        the initial casting period should be 1.5
temporary weakening of the bones sec-          creased pressures over the top and sides      to 2 times the duration of immobiliza-
ondary to osteoporosis associated with         of the feet. A rigid rocker sole modifi-      tion typically required to heal a similar
immobilization.113 Activity level of the       cation is typically used following ulcer      fracture in an individual with normal
patient should be controlled strictly in       healing. This shoe modification has           sensation in the foot.11 This lengthened
accordance with the temperature and            been shown to be effective in reducing        period of immobilization is required to
girth measurements. Distances walked           forefoot pressures during walking.116         compensate for the lack of a pain feed-
and approximate percentage of weight-          Badly deformed feet may require cus-          back system, which would normally
bearing on the foot should be recorded         tom-made footwear. Footwear recom-            cause self-imposed protection during
in a daily log.                                mendations by foot risk category are          walking such as limping.58 The first half
    Definitive footwear. The preferred         outlined in Table 5. A detailed descrip-      of this chosen period of immobilization
shoe style is a straight or semicurved last    tion of foot orthoses and shoe modifi-        should be strict nonweight-bearing fol-
with a low-wedge crepe sole, soft insert,      cations was provided by Reed and Ther-        lowed by partial weight-bearing with
moldable upper, and adjustable closure         iot.117 In general, we would advise that      crutches for the remainder of immobi-
(see article by McPoil in this issue). In      patients be provided with two pairs of        lization. Casts should be changed at two-
general, straight-laced shoes will have        shoes and one pair of sandals. This pre-      to three-week intervals to minimize the
greater width and height in the toe box,       caution will provide some variation in        possibility of secondary abrasions.101 In
which will minimize pressures over the         the weight-bearing patterns on the feet       selected cases, primary surgical fixation
 sides and top of the forefoot. A wedge        to prevent repetitive loading of the same     of the fractures may be indicated, fol-
 crepe sole may decrease foot stress by        areas and ensure that backup footwear         lowed by immobilization by casting.113
providing softness and less movement           is always available.                              Intermediate stage (3-6 months). The
 of the foot because this design is inher-        New footwear should always be bro-         extent of immobilization is decreased in
 ently more stable than a conventional         ken in slowly to prevent injuries attrib-     the intermediate stage to provide more
 sole.114 Soft compliant insoles will dis-     utable to fitting problems. In general,       relative movement of the body segment.
 tribute the loads on the plantar surface       new shoes should not be worn more            Fractures involving the metatarsals may
 of the foot in both space and time,           than two hours a day for the first week       be managed with a rigid, custom-
 thereby decreasing peak pressures.            followed by four hours a day maximum          molded sandal. Fracture bracing should
 Leather or nylon mesh uppers will mold        the second week. Wearing time should          be used for immobilization of midfoot
 easily to the foot during the initial wear-   ideally be distributed between the morn-      and rear-foot fractures. In addition, a
 ing to provide an improved fit as op-         ing and afternoon rather than continu-        patellar tendon-bearing orthosis may be
 posed to noncompliant plastics. Strap or       ous wearing of new shoes. Minor shoe         useful for managing rear-foot fractures.
 lace closures allow for adjustment to          adjustments may be required during           Crutches are recommended as an addi-
 accommodate foot swelling or stretching       break-in to improve fit and can usually       tional precaution, especially during the
 of the uppers.                                be performed by a shoe repair shop.           first four to six weeks following cast
     Proper shoe fit is as important as the                                                  removal. Elastic stockings that provide
 style of the footwear. The most com-          Neuropathic Fractures
                                                                                             graduated support of 25 to 30 mm Hg
 monly used equipment to measure shoe             A staged approach is needed to suc-        are strongly recommended for the con-
 size is the Brannock scale. This device       cessfully manage neuropathic fractures.       trol of swelling. The patient should be
 will give an approximate indication of        The most important aspect of treatment        allowed to progressively increase the ac-
 the required length and width of the          is early detection of bone injuries before    tivity level during the intermediate stage
 shoe. The proper size is then determined      the development of a major deformity.         with careful surveillance of the previous
 by trying several length and width com-       This early detection is followed by an        fracture site using daily temperature and
 binations smaller and larger than the         initial period of immobilization or sur-      girth measurements. All measurements
 suggested size. Some patient variables        gery followed by immobilization. The           should be compared to baseline record-
 that may affect shoe fit are time of day,     patient then enters an intermediate stage     ings taken at the time the cast was
 recent surgery or injury, and asymmetry       of management where the degree of im-         removed. If these clinical signs are un-

1900                                                                                                           PHYSICAL THERAPY
stable (measurements show a steady in-                      Chicago, IL, American Hospital Association,                al: Vertical forces acting on the feet of diabetic
                                                            1987, p 204                                                patients with neuropathic ulceration. Br J Surg
crease in magnitude with increasing                    6.   Warren R, Kihn RB: A survey of lower extrem-               68:608-614,1981
level of activity), immobilization in a                     ity amputations for ischemia. Surgery 63:107-        29.   Boulton AJM, Betts RP, Franks CI, et al:
                                                            120,1968                                                   Abnormalities of foot pressure in early diabetic
cast may be necessary for an additional                7.   Levin ME: Pathophysiology of diabetic foot                 neuropathy. Diabetic Medicine 4:225-228,
one to two weeks. The degree of im-                         lesions. In Davidson JK (ed): Clinical Diabetes            1987
mobilization may need to be readjusted                      Mellitus: A Problem-Oriented Approach. New           30.   Duckworth T, Boulton AJM, Betts RP, et al:
                                                            York, NY, Thieme Medical Publishers Inc,                   Plantar pressure measurements and the pre-
continuously until the bone tissue can                      1986, pp 383-396                                           vention of ulceration in the diabetic foot. J
accept relatively normal loads without                 8.   National Diabetes Data Group (NDDG): Dia-                  Bone Joint Surg [Br] 67:79-85, 1985
signs of irritability.                                      betes in America. Bethesda, MD, US Dept of           31.   Cavanagh PR, Rodgers MM, liboshi A: Pres-
                                                            Health and Human Services, NIH Publication                 sure distribution under symptom-free feet dur-
   Late stage (6-12 months). Definitive                     No. 85-1468, 1985                                          ing barefoot standing. Foot Ankle 7:262-276,
footwear is chosen in the late stage. Elas-            9.   Jauw-Tjen L, Brown AL: Normal structure of                 1987
                                                            the vascular system and the general reactive         32.   Schaff P, Kirsch D, Hauser W, et al: Eine
tic stockings for control of swelling                       changes of the arteries. In Fairbairn JF, et al            Gerateentwicklung zur Messung der Druck-
should be continued indefinitely. Fore-                     (eds): Peripheral Vascular Diseases. Philadel-             verteilung unter der FuBsohle im Schuh und
foot fractures can usually be managed                       phia, PA, W B Saunders Co, 1972, pp 45-62                  deren Anwendbarkeit in der Diabetologie. Ak-
                                                      10.   Gooding GAW, Stess RM, Graf PM, et al:                     tuelle Endokronologie Stoffwechsel 7:129-
successfully with an extra-depth shoe                       Sonography of the sole of the foot: Evidence               135,1986 (German)
containing a soft insole and a rigid                        for loss of foot pad thickness in diabetes and       33.   Cavanagh PR, Hennig EM, Rodgers MM, et
rocker sole modification. Custom-made                       its relationship to ulceration of the foot. Invest         al: The measurement of pressure distribution
                                                            Radiol 21:45-48, 1986                                      on the plantar surface of diabetic feet. In
footwear is usually required for midfoot              11.   Brand PW: The diabetic foot. In Ellenburg M,               Whittle M, Harris D (eds): Biomechanical
and rear-foot fractures. Occasionally,                      Rifkin H (eds): Diabetes Mellitus: Theory and              Measurement in Orthopaedic Practice. Lon-
                                                            Practice, ed 3. New Hyde Park, NY, Medical                 don, England, Oxford University Press, 1985,
surgical excision of plantar prominences                    Examination Publishing Co Inc, 1983, pp 829-               pp 159-168
may be necessary if the patient has de-                     849                                                  34.   Rodgers MM: Plantar Pressure Distribution
veloped a marked rocker-bottom foot                   12.   Delbridge L, Ctercetko G, Fowler C, et al: The             Measurement During Barefoot Walking: Nor-
                                                            aetiology of diabetic neuropathic ulceration of            mal Values and Predictive Equations. Doctoral
deformity following a midfoot collapse.                     the foot. Br J Surg 72:1-6, 1985                           Dissertation. University Park, PA, The Penn-
This should be followed by application                13.   Boulton AJM, Hardisty CA, Betts RP, et al:                 sylvania State University, 1985
of a custom-molded shoe with a rigid                        Dynamic foot pressure and other studies as           35.   Pollard JP, LeQuesne LP, Tappin JW: Forces
                                                            diagnostic and management aids in diabetic                 under the foot. J Biomed Eng 5:37-40, 1983
rocker sole. Rear-foot fractures require                    neuropathy. Diabetes Care 6:26-33, 1983              36.   Frykberg RG: Podiatric problems in diabetes.
more extreme footwear intervention.                   14.   Pirart J: Diabetes mellitus and its degenera-              In Kozak GP, et al (eds): Management of
Typically, a fixed ankle boot or patellar                   tive compliations: A prospective study of                  Diabetic Foot Problems. Philadelphia, PA,
                                                            4,400 patients observed between 1947 and                   W B Saunders Co, 1984, pp 45-67
tendon-bearing orthosis is necessary for                    1973. Diabetes Care 1:168-188, 252-263,              37.   Edmonds ME: The neuropathic foot in diabe-
an indefinite period of time. After a                       1978                                                       tes: Part I. Blood flow. Diabetic Medicine
                                                      15.   Greene DA, Brown MJ: Diabetic polyneurop-                  3:111-115, 1986
minimum of one year, an attempt may                         athy. Seminars in Neurology 7:18-29, 1987            38.   Brooks AP: The neuropathic foot in diabetes:
be made to progress to an oxford style                16.    Brand PW: Management of the insensitive                   Part II. Charcot's neuroarthropathy. Diabetic
rigid shoe if the ankle joint alignment                     limb. Phys Ther 59:8-12, 1979                              Medicine 3:116-118, 1986
                                                      17.   Johnson EW: Practical Electromyography.              39.   Ewing J, Clarke BF: Diabetic autonomic neu-
and ligamentous stability are adequate.                     Baltimore, MD, Williams & Wilkins, 1980                    ropathy: Present insights and future pros-
A return to the more protective foot-                 18.    Dyck PJ: Quantification of cutaneous sensa-               pects. Diabetes Care 9:648-658, 1986
wear is mandatory at the first signs of                     tion in man. In Dyck PJ, et al (eds): Peripheral     40.   Fushimi H, Inoue T, Nishikawa M, et al: A new
                                                            Neuropathy. Philadelphia, PA, W B Saunders                 index of autonomic neuropathy in diabetes
inflammation.                                               Co, 1975, pp 465-479                                       mellitus: Heat-stimulated thermographic pat-
                                                      19.   Gelberman RH, Szabo RM, Williamson RV, et                  terns. Diabetes Res Clin Pract 1:103-107,
SUMMARY                                                     al: Sensibility testing in peripheral nerve                1985
                                                            compression syndromes. J Bone Joint Surg             41.   Zelis R, Mansour EJ, Capone RJ, et al: The
   We have presented an overview of the                     [Am] 65:632-638, 1983                                      cardiovascular effects of morphine. J Clin In-
                                                      20.    Birke JA, Sims DS: Plantar sensory threshold              vest 54:1247-1258, 1974
risk factors for the development of dia-                    in the ulcerative foot. Lepr Rev 57:261-267,         42.   Low PA, Caskey PE, Tuck RR, et al: Quanti-
betic foot lesions. The potential rele-                     1986                                                       tative sudomotor axon reflex test in normal
                                                      21.    Holewski JJ, Stess RM, Graf PM, et al: Aes-               and neuropathic subjects. Ann Neurol
vance and methods of quantification of                      thesiometry: Quantification of cutaneous                   24:573-580, 1983
each risk factor were discussed. In ad-                     pressure sensation in diabetic peripheral neu-       43.   Archer AG, Roberts VC, Watkins PJ: Blood
dition, prevention of foot lesions as well                  ropathy. J Rehabil Res Dev 25:1-10, 1988                   flow patterns in painful diabetic neuropathy.
                                                      22.   Talbot WH, Darian-Smith I, Kornhuber HH, et                Diabetologia 27:563-567, 1984
as treatment of lesions, should they oc-                    al: The sense of flutter-vibration: Comparison       44.   Habershaw G, Donovan JC: Biomechanical
cur, were outlined. The need for ex-                        of the human capacity with response patterns               considerations in the diabetic foot. In Kozak
panded research regarding the etiology                      of mechanoreceptive afferents from the mon-                GP, et al (eds): Management of Diabetic Foot
                                                            key hand. J Neurophysiol 31:301-334, 1968                  Problems. Philadelphia, PA, W B Saunders
and treatment of diabetic foot problems               23.    Gescheider GA, Joelson JM: Vibrotactile tem-              Co, 1984, pp 32-44
has been emphasized throughout this                         poral summation for threshold and supra-             45.   Jacobs RL, Karmody A: The diabetic foot. In
                                                            threshold levels of stimulation. Perception and            Jahss MH (ed): Disorders of the Foot. Phila-
article.                                                    Psychophysics 33:156-162, 1983                             delphia, PA, W B Saunders Co, 1982, vol 2,
                                                      24.    Dellon LA: Touch sensibility in the hand. J               pp 1377-1397
REFERENCES                                                  Hand Surg [Br] 9:11-13, 1984                         46.   Coughlin MS: Mallet toes, hammer toes, claw
                                                      25.    Guy RJC, Clark CA, Malcolm PN, et al: Eval-               toes, and corns: Causes and treatments of
  1. Bailey TS, Yu HM, Rayfield EJ: Patterns of             uation of thermal and vibration sensation in               lesser-toe deformities. Postgrad Med 75:191 -
     foot examination in a diabetes clinic. Am J            diabetic neuropathy. Diabetologia 28:131-                  198, 1984
     Med 78:371-374, 1985                                    137,1985                                            47.   Schnider SL, Kohn RR: Glycosylation of hu-
  2. Cohen SJ: Potential barriers to diabetic care.   26.    Arezzo JC, Schaumburg HH, Laudadio C, et                  man collagen in aging and diabetes mellitus.
     Diabetes Care 6:499-500, 1983                          al: Thermal sensitivity tester: Device for quan-           J Clin Invest 66:1179-1181, 1980
  3. Levin ME, O'Neal LW: The Diabetic Foot. St.            titative assessment of thermal sense in dia-         48.    Delbridge L, Ellis CS, Robertson K, et al:
     Louis, MO, C V Mosby, Co, 1983                         betic neuropathy. Diabetes 35:590-592, 1986                Nonenzymatic glycosylation of keratin from
  4. Krall LP: Education: A treatment for diabetes.   27.    Stokes IAF, Faris IB, Hutton WC: The neuro-               the stratum corneum of the diabetic foot. Br
     In Marble A, et al (eds): Joslin's Diabetes            pathic ulcer and loads on the foot in diabetic             J Dermatol 112:547-554, 1985
     Mellitus, ed 12. Philadelphia, PA, Lea & Febi-          patients. Acta Orthop Scand 46:839-847,             49.    Brownlee M, Vlassara H, Cerami A: Nonen-
     ger, 1985, pp 466-481                                   1975                                                      zymatic glycosylation and the pathogenesis
  5. Hospital Statistics: Data from the American      28.    Ctercetko GC, Dhanendron M, Hutton WC, et                 of diabetic complications. Ann Intern Med
     Hospital Association 1986 Annual Survey.                                                                          101:527-537,1984

Volume 68 / Number 12, December 1988                                                                                                                             1901
50. Hamlin Cr, Kohn RR, Luschin JH: Apparent              74. Wagner FW: Treatment of the diabetic foot.                ATPase in the pathogenesis of diabetic com-
    accelerated aging of human collagen in dia-               Compr Ther 10:29-38, 1984                                 plications. N Engl J Med 316:599-606, 1987
    betes mellitus. Diabetes 24:902-904, 1975             75. Janka HU, Stadl E, Mehnert H: Peripheral            96.   Casey Jl: Host defense and infections in dia-
51. Shinabarger Nl: Limited joint mobility in adults          vascular disease in diabetes mellitus and its             betes mellitus. In Ellenburg M, Rifkin H (eds):
    with diabetes mellitus. Phys Ther 67:215-                 relation to cardiovascular risk factors: Screen-          Diabetes Mellitus: Theory and Practice, ed 3.
    218, 1987                                                 ing with Doppler ultrasonic technique. Diabe-             New York, NY, Medical Examination Publish-
52. Brink SJ: Limited joint mobility as a risk factor         tes Care 3:207-213, 1980                                  ing Co Inc, 1983, pp 667-678
    for diabetes complications. Clinical Diabetes         76. LoGerfo FW, Coffman JD: Vascular and mi-            97.   Pollard JP, LeQuesne LP: Method of healing
    5:121,123-127,1987                                        crovascular disease of the foot in diabetes. N            diabetic forefoot ulcers. Br Med J [Clin Res]
53. Ingram AS: Anterior poliomyelitis. In Cren-               Engl J Med 311:1615-1619, 1984                            286:436-437,1983
    shaw AH (ed): Campbell's Operative Ortho-             77. LoGerfo FW: Vascular disease, matrix abnor-         98.   Sinacore DR, Mueller MJ, Diamond JE, et al:
    paedics, ed 6. St. Louis, MO, C V Mosby Co,               malities, and neuropathy: Implications for limb           Diabetic plantar ulcers treated by total contact
    1980, vol 2, p 1535                                       salvage in diabetes mellitus. J Vase Surg                 casting: A clinical report. Phys Ther 67:1543-
54. Fromherz WA: Examination. In Hunt GC (ed):                5:793-796,1987                                            1549,1987
    Physical Therapy of the Foot and Ankle. New           78. Hurley JJ, Auer Al, Hershey FB, et al: Distal       99.   Birke JA, Sims DS, Buford WL: Walking casts:
    York, NY, Churchill Livingstone Inc, 1988, pp             arterial reconstruction: Patency and limb sal-            Effects on plantar foot pressures. J Rehabil
    59-90                                                     vage in diabetics. J Vase Surg 5:796-802,                 Res Dev 22:18-22, 1985
55. Rodgers MM, Cavanagh PR: A Device for the                 1987                                               100.   Helm PA, Walker SC, Pullium G: Total contact
    Measurement of First Ray Mobility. Proceed-           79. Raines JK, Darling BC, Buth J, et al: Vascular            casting in diabetic patients with neuropathic
    ings of the North American Congress on Bio-               laboratory criteria for the management of pe-             foot ulcerations. Arch Phys Med Rehabil
    mechanics. Montreal, Quebec, Canada, Au-                  ripheral vascular disease of the lower extrem-            65:691-693,1984
    gust 25-27, 1986, pp 205-206                              ities. Surgery 79:21-29, 1976                      101.   Birke JA, Sims DS: The insensitive foot. In
56. Smith LS, Clarke TE, Hamill CL, et al: The            80. Wagner FW: The dysvascular foot: A system                 Hunt GC (ed): Physical Therapy of the Foot
    effects of soft and semi-rigid orthoses upon              for diagnosis and treatment. Foot Ankle 2:64-             and Ankle. New York, NY, Churchill Living-
    rearfoot movement in running. Podiatric                   122, 1981                                                 stone Inc, 1988, pp 133-168
    Sports Medicine 76:227-233, 1986                      81. Gibbons GW, Wheelock FC, Hoar CS: Pre-             102.    National Diabetes Information Clearinghouse:
57. Cavanagh PR, Rodgers MM: The arch index:                  dicting success of forefoot amputations in                Foot Care and Diabetes: Selected Annota-
    A useful measure from footprints. J Biomech               diabetic patients by noninvasive testing. Arch            tions. Bethesda, MD, US Dept of Health and
    20:547-552,1987                                           Surg 114:1034-1036, 1979                                  Human Services, NIH Publication No. 81-
58. Edmonds ME: The diabetic foot: Pathophysi-            82. Edmonds ME, Gilbey S, Walters HW, et al:                  1870,1981
    ology and treatment. Baillieres Clin Endocrinol           Improved survival of the diabetic ischaemic        103.    Bergtholdt HT: Temperature assessment of
    Metab 15:889-916, 1986                                    foot. Diabetic Medicine 2:506A, 1985                      the insensitive foot. Phys Ther 59:18-22,
59. Cundy TF, Edmonds ME, Watkins PJ: Osteo-              83. Barnes RW, Thornhill B, Nix L, et al: Predic-             1979
    penia and metatarsal fractures in diabetic neu-           tion of amputation wound healing: Roles of         104.    Sabato S, Yosipovitch Z, Simkin A, et al:
    ropathy. Diabetic Medicine 2:461-464, 1985                Doppler ultrasound and digit photoplethys-                Plantar trophic ulcers in patients with leprosy:
60. Edmonds ME, Clarke MD, Newton S, et al:                   mography. Arch Surg 116:80-83, 1981                       A correlative study of sensation, pressure and
     Increased uptake of bone radiopharmaceuti-           84. Gibbons GW, Campbell DR: Noninvasive di-                  mobility. Int Orthop 6:203-208, 1982
    cal in diabetic neuropathy. Q J Med 572:843-              agnostic studies. In Kozak GP, et al (eds):        105.    Dannels EG: A preventive metatarsal osteot-
    855, 1985                                                 Management of Diabetic Foot Problems. Phil-               omy for pre-ulcers in American Indian dia-
61. Harris JR, Brand PW: Patterns of disintegra-              adelphia, PA, W B Saunders Co, 1984, pp                   betics. J Am Podiatr Med Assoc 76:33-37,
    tion of the tarsus in the anaesthetic foot. J             91-96                                                      1986
     Bone Joint Surg [Br] 48:4-16, 1966                   85. Kempczinski RF, Yao JST: Practical Nonin-          106.   Jones RO: Ulceration in the neurotrophic foot
62. Sinha S, Munichoodappa CS, Kozak GP: Neu-                 vasive Vascular Diagnosis. Chicago, IL, Year              of Hansen's Disease. J Am Podiatr Assoc
     roarthropathy (Charcot joints) in diabetes mel-          Book Medical Publishers Inc, 1982                         72:299-303, 1982
     litus. Medicine 51:191-210, 1972                     86. Cochrane T, Sheriff SB, Boulton AJM, et al:        107.    Peacock EE, Van Winkle W: Wound Repair,
63. Tuccio AT, Wertheimer SJ: Noninfectious os-                Laser Doppler flowmetry: In the assessment               ed 2. Philadelphia, PA, W B Saunders Co,
     seous alterations in the diabetic foot. J Foot           of peripheral vascular disorders? A preliminary            1976
     Surg 24:154-157, 1985                                    evaluation. Clin Phys Physiol Meas 7:31-42,        108.    Gibbons GW: The diabetic foot: Amputations
64. Sandrow RE, Torg JS, Lapayowker MS, et al:                 1986                                                     and drainage of infections. J Vase Surg
    The use of thermography in the early diagno-          87. Hauser CJ, Klein ST, Mehringe CM, et al:                  5:791-793,1987
     sis of neuropathic arthropathy in the feet of            Assessment of perfusion in the diabetic foot       109.    Diamond E, Forst MB, Hyman SA, et al: The
     diabetics. Clin Orthop 88:31-33, 1972                     by regional transcutaneous oximetry. Diabe-              effect of hyperbaric oxygen on lower extrem-
65. Eymontt MJ, Alavi A, Dalinka MK, et al: Bone              tes 33:527-531, 1984                                       ity ulcerations. J Am Podiatr Assoc 72:180-
     scintigraphy in diabetic osteoarthropathy. Ra-       88. Boeker MJ, Leu MM: A study of the effects                  185,1982
     diology 140:475-477, 1981                                of hydration and emollients on the feet of the     110.    Leslie CA, Sapico FL, Ginunas VJ, et al: Ran-
66. Park HM, Wheat LJ, Siddiqui AR: Scinto-                    aged. J Am Podiatr Assoc 68:402-418, 1978                domized controlled trial of topical hyperbaric
     graphic evaluation of diabetic osteomyelitis:        89. Davidson JK, Alogna M, Goldsmith M, et al:                oxygen for treatment of diabetic foot ulcers.
     Concise communication. J Nucl Med 23:569-                Assessment of program effectiveness at                     Diabetes Care 11:111 -115, 1988
     573,1982                                                  Grady Memorial Hospital, Atlanta, GA. In Stei-    111.    Mclntyre KE: Control of infection in the dia-
67. Visser HJ, Jacobs AM, Oloff L, et al: The use              ner G, Lawrence PA (eds): Educating Diabetic              betic foot: The role of microbiology, immuno-
     of differential scintigraphy in the clinical diag-        Patients. New York, NY, Springer-Verlag,                  pathology, antibiotics and guillotine amputa-
     nosis of osseous and soft tissue changes                  New York Inc, 1981, p 329                                 tion. J Vase Surg 5:787-790, 1987
     affecting the diabetic foot. J Foot Surg 23:74-      90. Rosenqvist U, Carlson A, Luft R: Evaluation        112.    Coleman WC, Brand PW, Birke JA: The total
     85, 1984                                                  of comprehensive program for diabetes care                contact cast: A therapy for plantar ulceration
68. Krebs DE, Edelstein JE, Fishman S: Reliability             at primary health-care level. Diabetes Care               on insensitive feet. J Am Podiatr Med Assoc
     of observational kinematic gait analysis. Phys            11:269-274,1988                                           74:548-552, 1984
     Ther 65:1027-1033, 1985                              91. Deeb LC, Pettijohn FP, Shirah JK, et al: Inter-    113.    Warren G: Tarsal bone disintegration in lep-
69. Atha J: Current techniques for measuring mo-               ventions among primary-care practitioners to              rosy. J Bone Joint Surg [Br] 53:688-695,
     tion. Applied Ergonomics 15:245-257, 1984                 improve care for preventable complications of             1971
70. Krag MH: Quantitative techniques for the                   diabetes. Diabetes Care 11:275-280, 1988          114.    Edwards CA: Orthopedic Shoe Technology.
     analysis of gait. Automedica 6:85-97, 1985           92. Johnson CA, Keltner JL: Incidence of visual                Muncie, IN, Precision Printing, 1981
71. LaPorte RE, Cruickshanks KJ, Cavender DE,                  field loss in 20,000 eyes and its relationship    115.    Soulier SM: The use of running shoes in the
     et al: Pittsburg insulin-dependent diabetes               to driving performance. Arch Ophthalmol                   prevention of plantar diabetic ulcers. J Am
     mellitus morbidity and mortality study: Physi-            101:371-375,1983                                          Podiatr Med Assoc 76:395-400, 1986
     cal activity and diabetic complications. Pedi-       93. Shade DS, Santigo JV, Skyler JS, et al: Dia-       116.    Sims DS, Birke JA: Effect of rocker placement
     atrics 78:1027-1033, 1986                                 betes Mellitus: Effects of Intensive Treatment            on plantar foot pressures (Abstract). Proceed-
72. Wannstedt GT, Craik RL: Clinical evaluation                on Long-Term Complications. New York, NY,                 ings of the Twentieth Annual Meeting of the
     of a sensory feedback device: The limb load               Medical Examination Publishing Co Inc, 1983,              US Public Health Service Professional Asso-
     monitor. Bulletin of Prosthetic Research 29:8-            pp 88-112                                                 ciation, Atlanta, GA, April 9-12,1985, p 53
     38, 1979                                             94. Cogan DG, Kinoshita JH, Kador PF, et al:           117.     Reed JK, Theriot S: Orthotic devices, shoes
73. Gapsis JJ, Grabois M, Borrell RM, et al: Limb              Aldose reductase and complications of dia-                and modifications. In Hunt GC (ed): Physical
     load monitor: Evaluation of a sensory feed-               betes. Ann Intern Med 101:82-91, 1984                     Therapy of the Foot and Ankle. New York,
     back device for controlled weight bearing.           95. Greene DA, Lattimer SA, Sima AAF: Sorbitol,                NY, Churchill Livingstone Inc, 1988, pp 285-
     Arch Phys Med Rehabil 63:38-41, 1982                      phosphoinositides, and sodium-potassium-                  313

1902                                                                                                                                     PHYSICAL THERAPY

To top