Preventive Foot Care in People With Diabetes foot care … · of Family Medicine, 1110 W Michigan St., Long Hospital, Room 200, Indiana University, Indianapolis, IN 46202. E-mail:

R e v i e w s / C o m m e n t a r i e s / P o s i t i o n S t a t e m e n t s

T E C H N I C A L R E V I E W

Preventive Foot Care in People WithDiabetesJENNIFER A. MAYFIELD, MD, MPHGAYLE E. REIBER, PHD, MPHLEEJ. SANDERS, DPM

DENNIS JANISSE, CPEDLEONARD M. POGACH, MD

Diabetic foot ulcers and lower-extrem-ity amputations are serious andexpensive complications that befall

up to 15% of people with diabetes duringtheir lifetime. Relatively simple and inex-pensive interventions may decrease theamputation rate up to 85% (1-5). ThisTechnical Review was undertaken toupdate the evidence for various preventiveand therapeutic modalities to decrease dia-betic foot ulcers and lower-extremityamputations and to develop the clinicalpractice recommendations for preventivefoot care of people with diabetes.

The Technical Review is organized intothree parts: 1) major demographic andcomorbidity risk factors, 2) foot-related riskfactors, and 3) preventive strategies. Theclinical implications of this review willappear in a separate American DiabetesAssociations (ADA) Position Statement onpreventive care of the foot (6a). This Tech-nical Review covers only the care of thenonulcerated foot. The assessment andmanagement of diabetic foot ulcers andCharcot arthropathy including surgicalmanagement, will be covered in a laterreview.

This Technical Review is based on orig-inal research published in the peer-reviewed literature in the English language.

The research quality was graded using amodification of the system used by the U.S.Preventive Services Task Force (APPENDIX)

(6b,7). Preference was always given tostudies that provided patient-oriented out-comes or clinically relevant care, ratherthan disease-oriented outcomes. For exam-ple, a study that predicted foot ulcer riskwas preferred over a study that predicted adecrease in nerve conduction velocity.

ImportanceAlthough only 3% of the population havediagnosed diabetes, half of all nontraumaticlower-extremity amputations in the U.S.occur in people with diabetes (8). Theannual age-adjusted amputation ratebetween 1980 and 1990 varied from 5.1 to8.1 per 1,000 people with diabetes, but thenumber of amputations increased 29% overthat decade to 54,000 in 1990 (9). Non-traumatic lower-extremity amputationoccurred in 1.9% of all hospital discharges ofdiabetic people between 1983 and 1990(10). About half of the amputations are ofthe toes or foot; the other half (43-65%) areamputations at the transtibial (below-the-knee) or transfemoral (above-the-knee) level(8). The prevalence of amputation in theU.S. in 1989 was 2.8% for people with dia-betes (8).

From the Department of Family Practice QAM.), Bowen Research Center, Indiana University, Indianapolis,Indiana; the Health Services Research Center (G.E.R.), Veterans Affairs Puget Sound Health Care System, Seat-tle, Washington; the Veterans Affairs Medical Center (L.J.S.), Lebanon, Pennsylvania; the Department of Phys-ical Medicine and Rehabilitation (D.J.), Medical College of Wisconsin, Milwaukee, Wisconsin; and theVeterans Affairs Medical Center ( L.P), East Orange, New Jersey.

Address correspondence and reprint requests to Jennifer A. Mayfield, Bowen Research Center, Departmentof Family Medicine, 1110 W Michigan St., Long Hospital, Room 200, Indiana University, Indianapolis, IN46202. E-mail: [email protected].

L.J.S. has received honoraria from Novartis and Eli Lilly and serves as a member of the advisory panelsfor Boehringer Mannheim-Lilly Alliance Amputations Prevention Initiative, the Department of VeteransAffairs, the American Podiatric Medical Association, and Pfizer. J.M. serves on an advisory panel for Merck-Medeco and has served on advisory panels for Eli Lilly and Boehringer Mannheim.

Abbreviations: AAI, ankle-arm index; MAC, medial arterial calcinosis; OR, odds ratio; pack-year, smok-ing one pack of cigarettes per day for one year.

A table elsewhere in this issue shows conventional and Syste"me International (SI) units and conversionfactors for many substances.

Approximately 85% of all amputationsare preceded by a nonhealing foot ulcer(11,12). Foot ulcers affect up to 15% of allpeople with diabetes sometime in their life-time (13). The annual incidence of self-reported foot ulcers in diabetic peopleranges from 2.4 to 2.6% (14), and theprevalence of foot ulcers ranges from 4 to10% (15-18), reflecting differences in pop-ulations and wound management strategiesacross geographic regions. Almost 6% of allU.S. hospital discharges with a diabetesdiagnosis between 1983 and 1990 includeda lower-extremity ulcer diagnosis; 46% of allulcer hospitalizations were in people withdiabetes (10). The direct costs for care offoot ulcers was estimated to be $145 millionin 1986 (19). An extensive review of theepidemiological literature on diabetic footulcers and amputation is available in Dia-betes in America (8) (also at http://diabetes-in-america.s-3.com/default.htm).

DEMOGRAPHIC ANDCOMORBIDITY RISKFACTORS— The risk factors for dia-betic foot ulcers and lower-extremity ampu-tation are similar and will be describedtogether in this review and noted separatelywhen they differ. Also, the risk factors aresimilar for people with type 1 (1DDM) andtype 2 (NIDDM) diabetes, so data on thesepopulations are reported together andnoted separately only when they differ. (Allreferences to the type of diabetes have beenupdated to the preferred terminology of"type 1" and "type 2.") These risk factorsinclude demographic characteristics,comorbid conditions, and foot pathology.Multivariable analyses of risk factors areshown in Table 1.

Age and duration of diabetesThe risk of ulcers and amputation increasestwo- to fourfold with both age and durationof diabetes (8,14,20). In the U.S., between1989 and 1993, the prevalence of amputa-tions was 1.6% for diabetic people age18-44 years, 2.4% for people age 45-64years, and 3.6% for people age ^65 years(8). In Wisconsin, the prevalence of ampu-tation was 2.4% for people with diabetesonset before 30 years, and 4.4% for peoplewith older onset diabetes (14). Age and

DIABETES CARE, VOLUME 21, NUMBER 12, DECEMBER 1998 2161

Technical Review

Table 1—Comparison of significant risk factors for amputation in multivariable models

Characteristicsand risk factors

Design

Data collection

Sample size (n)Analysis

Measure

Male versus femaleNonwhite versus whiteDiabetes duration

(per x years)Age (per x years)Glycemic control

(per x units)Insulin useRetinopathyRenal diseaseProteinuriaHypertension (per

mmHgofdBPorsBP)Cholesterol

SmokersStrokeHeart diseaseNeuropathy

Vibratory senseMACPeripheral vasculardisease

Foot deformitiesUlcer historyNo outpatient

education

Mosset al. (14)

Wisconsin Wisconsin(early onset) (late onset)

Cohort

Self-report,exam

1,210 1,780Logistic

regression

Nelsonet al. (20),

PimaIndians

Cohort

Exam

4,399Mantel-

Haenszel

Study and population

Selby andZhang (22),

HMO(San Francisco"

Nestedcase-controlChart review

150 + 278Logistic

regression

OR Incidence rate ratio OR

NS 2.8NS —NS 1.8 per 10

2.0 per 10 NS1.4 per 1.5 per

2% GHb 2% GHb— —

1.4per2stepst NS— —NS 4.3

2.1 per 10 dBP NS

— —

NS NS— —— —— —

— —— —— —

— —

10.5 4.6— —

AdjustedNA

Adjusted

Adjusted—

—2.12.2—NS

NS

NS——

2.0 patellarreflex

—4.8—

———

MatchedNSNS

Matched1.75 per

glucose score*—

3.68—NS

1.02 per 1 sBP

NS

NS2.7NS

4.05

———

———

Lehtoet al. (23),

type 1I (Finland)

Cohort

Exam

1,044Cox

regression

RR

Adjusted—

2.2 for 9

Adjusted2.5 for HbA!

>13.4%—3.6NS1.3NS

1.8 for>6.2 mmol/1

NS—NS

4.3 Achillesreflex2.7—

3.9 absent pulses2.1 femoral bruit

———

Lee et al. (27),type 2

(Oklahoma Indians)Men Women

Cohort

Exam

332 543Cox proportional

hazard model

Rate ratio

NA NANA NA

1.05 per 1 1.08 per 1

NS NS1.08 per NSmmol/1

2.56 NS3.19 3.33NS NSNS NS

1.15 per 1.28 per10 sBP 10 dBP

NS 1.18 permmol/1

NS NS— —NS NS— —

— —— —— —

— —— —— —

Reiberet al. (25),

Maleveterans

Case-control

Chart review,exam

80 + 236Logistic

regression

OR

NANS

Adjusted

AdjustedAdjusted

—AdjustedAdjusted

——

6.4 for HDL<1.3umol/l

NS———

12.9——

——

16.5

Mayfieldetal. (21),

PimaIndians

Case-control

Chart review

63 + 183Logistic

regression

OR

6.5NA

1.4 per 5

1.3 per 51.6 per

50 mg/dl—

4.6f4.6=fNSNS

NS

NS4.6?4.6*

2.1 anydiagnosis

——

2.1 anydiagnosis

2.12.1—

Most ORs or RRs compare with a baseline group without the feature. dBP, diastolic blood pressure; NA, not applicable; sBP, systolic blood pressure; RR, relativerisk. Data from Nelson et al. (20) are adjusted for age/sex. *Glucose score = 50 mg/dl fasting glucose or 68 mg/dl random glucose; T2 steps in grade of retinopa-thy; fretinopathy, renal disease, stroke, and heart disease were combined.

duration of diabetes are highly correlated. Inpopulations with accurate ascertainment ofthe onset of diabetes (i.e., Pima Indians withtype 1 diabetes), duration of diabetesbecomes the predominant predictor of ulcerand amputation risk (14,20,21). The rela-tionship of the duration to prevalence ofulceration and amputation appears to besimilar for people with type 1 and type 2diabetes (14).

SexMale sex has been associated with 1.6increased risk of ulcers (8,14) and 2.8- to

6.5-fold higher risk of amputation(14,20-22) in most studies of people withtype 2 diabetes. No difference between thesexes was reported for people with type 2diabetes in Finland (23) or people withtype 1 diabetes (14). The mechanism of theincreased risk for men has yet to beinvestigated.

Race and socioeconomic statusAnalysis of hospital discharges for diabeticfoot ulcers suggest no difference by race(8). A twofold higher risk of amputationhas been described for Hispanics and

blacks as compared to whites (8,24) and upto a fourfold higher rate in the Pima Indians(20). However, an evaluation of a Californiahealth maintenance organization found nodifference by race (22), suggesting that theobserved differences may be due to socioe-conomic issues or a lack of access to healthcare.

Social factorsThe lack of "social connectedness" (dennedas living alone, no visits from a friend or rel-ative in the past month, no attendance atsocial or religious gatherings, and personal

2162 DIABETES CARE, VOLUME 21, NUMBER 12, DECEMBER 1998

Technical Review

life dissatisfaction) is associated with a 2.1-to 3.8-fold higher risk of amputation (25).People presenting with a foot ulcer weremore likely to live alone or be from a lowersocial class (26).

Glycemic controlPoor glycemic control increases the risk ofneuropathy (see below) and amputation(14,20,22,23,27). The increased risk foramputation is observed even when neu-ropathy is controlled in multivariableanalysis. An HbA! level >13.4% was asso-ciated with a 2.2 relative risk of amputationin a Finnish population with type 2 dia-betes (23), and a 50 mg/dl increase in themean random glucose in the Pima Indianswas associated with a 1.6 odds ratio (OR)for amputation (21). Only one study hasreported an association of glycemia withulcers. A 1.4-1.5 increased risk of self-reported ulcers was associated with a gly-cated hemoglobin increase of 2% in peoplewith both type 1 and type 2 diabetes inWisconsin; however, the models did notcontrol for neuropathy (14).

Comorbid conditionsThe risk of diabetic foot disease is associ-ated with diabetic complications in otherorgan systems, including micro vasculardisease (e.g., diabetic retinopathy renal dis-ease) and macro vascular disease (e.g., coro-nary artery disease) (14,17,20-23,25,27).

Patient education and self-carepracticesThe lack of patient education on foot carehas been associated with a 3.2 increasedrisk of amputation (25). In a survey ofpatients with ulcers, only 29% previouslyconsidered they were at risk for foot prob-lems, compared with 59% of the controlsubjects without ulcers, although 30% inboth groups reported they had been giveninformation on foot care (26). Approxi-mately 70% of both groups had not hadtheir foot remeasured for footwear in thepast 10 years.

Tobacco and alcohol useCigarette smoking is a major risk factor forperipheral vascular disease (a major riskfactor for amputation) and amputation innondiabetic people. In diabetic people, theevidence for a relationship between tobaccoand ulcers or amputation is variable.Tobacco use has been associated withmicrovascular disease (e.g., retinopathy,nephropathy) in people with diabetes (28)

and cardiovascular disease (29). Most stud-ies of people with diabetes have failed toshow an association of cigarette smokingwith an increased risk of macrovasculardisease (30), peripheral vascular disease(31-33), diabetic foot ulcers (34,35), oramputation (22,23,25,31). However, a fewstudies show a weak relationship betweensmoking and peripheral vascular disease,ulcers, or amputation risk. A clinic-basedstudy found an association of smoking withproximal (pelvic, femoropopliteal) periph-eral vascular disease, but not with distaldisease (below-the-knee) in people withdiabetes (36). A population-based cohortstudy in Wisconsin of people with type 1diabetes age ^18 years found an associa-tion between ulcers and 10+ pack year his-tory of smoking (OR 1.3) and currentsmokers (OR 2.3); however, these findingswere of borderline statistical significance(CI included 1.0), and no increased riskwas found for smokers with type 2 diabetesin the same population (14). A study ofpeople with diabetes in three English com-munities found more peripheral vasculardisease (OR 2.16) in smokers comparedwith nonsmokers (37).

Alcohol consumption has been associ-ated with an increased risk of foot ulcers inBritish men with diabetes and impotence(38). Although one study found an associ-ation between CAGE scores and neuropa-thy (39), most studies have not foundalcohol to be a major risk factor for neu-ropathy (37,40), diabetic foot ulcers (35), oramputation (22,23,25). (The CAGE screen-ing instrument is a brief questionnaire andstands for cut down, annoyed by criticism,guilty about drinking, eye-opener drinks.)

FOOT PATHOLOGY ANDASSESSMENT— The major foot-related conditions that increase the risk ofulcers and amputations are peripheral neu-ropathy, altered biomechanics, peripheralvascular disease, and skin pathology, aswell as a history of foot ulcers.

Peripheral neuropathyPeripheral neuropathy is defined as symp-toms and/or signs of peripheral nerve dys-function in people with diabetes afterexclusion of other causes (41). The preva-lence of neuropathy depends on the defin-ition and population investigated. Theprevalence of neuropathy increases withage, duration of diabetes, presence ofmicrovascular complications, and poorglycemic control (42-45). Macrovascular

risk factors (i.e., hypertension, hyperlipi-demia) do not appear to be associated withthe risk of neuropathy (39,44). The mostcommon form of clinical neuropathy, distalsymmetric sensorimotor polyneuropathy,affects up to 50% of people who have haddiabetes 5:15 years (37,43,44) and affectsthe motor and sensory modalities in a"stocking-glove" pattern. Autonomic neu-ropathy frequently develops concurrently.

Peripheral neuropathy is associatedwith an 8- to 18-fold higher risk of ulcera-tion (34,35) and a 2- to 15-fold higher riskof amputation (8,20-23). Peripheral neu-ropathy is thought to be the underlyingpathophysiological alteration leading toCharcot arthropathy. The increased risk forthese adverse outcomes is impartedthrough several different mechanisms.First, the loss of protective sensations thatinclude pain, pressure, and temperature,removes the signals of damaging stimuli orconditions. Second, the motor componentof polyneuropathy results in atrophy of theintrinsic muscles (interosseous, lumbri-cals), resulting in a flexion deformity, whichcreates areas of increased pressure underthe metatarsal heads and tips of the toes.Third, the peripheral sympathetic auto-nomic neuropathy that often accompaniespolyneuropathy causes dyshidrosis and dryskin, which can readily crack. Autonomicneuropathy may also be involved with arte-riovenous shunting leading to altered skinand bone perfusion (46,47).

The onset of the loss of protective sen-sation is usually insidious and may progressat different rates in different types of nerves(sensory [proprioception, touch, vibration,pain, temperature], motor, and auto-nomic). Abnormalities of nerve functionand symptoms are only moderately corre-lated (38,48,49), with no one modalityclearly serving as a gold standard to predictadverse patient outcome. Various methodsto aggregate symptoms and signs have beendeveloped and used for research purposes(49,50). An American Diabetes Associationconsensus conference (41) on diabetic neu-ropathy recommended that at least onemeasure from clinical symptoms, clinicalexamination, electrodiagnostic studies,quantitative sensory testing, and autonomicfunction testing should be used to definediabetic neuropathy for research purposes.

Electrophysiological tests (e.g., nerveconduction studies) have moderate sensitiv-ity for nerve dysfunction but detect subclin-ical disease, with little clinical or prognosticsignificance (51). On the other hand, the tra-


Technical Review

ditional clinical assessment of peripheralneurological function (e.g., reflexes, pin-prick, vibration with a tuning fork, and lighttouch to cotton wisp) are highly subjectiveand have poor interobserver reproducibility(K < 0.75) (52). Tests of autonomic functionare not standardized and have poor repro-ducibility and only fair predictive ability forulcers or amputation (53).

Psychophysical somatosensory thresh-old tests for vibration and light touch pro-vide the best discrimination in the clinicalsetting to identify the loss of protective sen-sation. The vibratory perception thresholdcan be determined using an electronic tun-ing fork that vibrates at 120 Hz. The ampli-tude of vibration is varied from 10 to 50volts with the Bio-thesiometer (Bio-medicalInstruments, Newbury, OH) or the HorwellNeurothesiometer (Scientific LaboratorySupplies, Nottingham, U.K.). A moresophisticated instrument, the Vibrameter(Somedic, Stockholm), adjusts for the pres-sure of the probe on the skin, and testsfrom 0.05- to 399-um amplitude (54). Thedecrease in vibratory perception (i.e.,increased threshold) is highly predictive ofsubsequent ulceration. People with ulcerswere almost 11 times more likely to havevibratory thresholds >25 V than peoplewithout ulcers (55). In a prospective studyof people without a history of ulcers, thosewith vibratory threshold >25 V had a 6.8odds of developing an ulcer over the next 4years, compared with those with vibrationthresholds <25 V (34). The Bio-thesiome-ter costs several hundred dollars and takesat least 5 to 10 min to perform. Otherthreshold tests, generally used in a researchsetting, evaluate temperature (hot and cold)(56) and pain, using pinprick thresholds(57). One prospective study found the riskof minor skin lesions increased withimpaired thermal sensitivity (58). The useof the "yes/no" method of limits for sensorythresholds (59) has been shown to be fasterand more reliable than the traditional twoalternative forced-choice method (60).

The Rydel-Seiffer tuning fork (Ge-brueder Martin, Tuttlingen, Germany) pro-vides a simpler and less expensivesemiquantitative measure of vibratory sense(54). This 128-Hz tuning fork is calibratedon an arbitrary scale of 0-8 by movingweights at the extremities of the prongs. In aselected high-risk population, 95% of thepeople with ulcers had scores of ^ 4 . Thegraduated tuning fork was highly correlatedwith the electronic vibrometer (r = — 0.90, P< 0.001), and had a coefficient of variation

of 6-8% in the same day by different exam-iners (61) and 24% over several weeks timeby a single examiner (54). Results from inex-perienced examiners correlated well withthe experienced examiner (r = 0.87, P <0.001). The tuning fork can be carried in apocket, is easy and painless to use, and costsapproximately $100. However, no prospec-tive data are available linking the graduatedtuning fork to later ulceration or amputation.

The Semmes-Weinstein monofilamentassesses the threshold for light touch/pres-sure in a semiquantitative fashion (62).This instrument, a nylon fiber several cen-timeters long embedded in a handle, capi-talizes on the unique physical properties ofa buckling column to produce a quantifi-able pressure (63). A range of monofila-ments are available. The monofilaments areidentified by a number that equals the log-arithm of 10 times the force in milligramsrequired to bow the monofilament. Inter-and intra-rater agreement K values are 0.72and 0.83 (52). Testing of the plantar surfaceof the toes and metatarsal heads providesmost of the discriminatory ability; the dor-sum and heel provide little additional infor-mation (64). Two studies found the 5.07monofilament (10 g) correlated best withthe presence or history of an ulcer (64,65),while two other studies suggested the 4.21monofilament (1 g) to be a better discrim-inator (66,67). The proposed cutoffs fordefining an insensitive foot range from oneto three out of six sites, providing a sensi-tivity of 0.84-1.00 and specificity of0.77-1.00 to predict a current or futureulcer. In a prospective, population-basedstudy, the inability to appreciate a 5.07monofilament in one or more standard-ized sites on the foot was associated with anincreased risk of ulceration (9.9 [4.8-21.0];OR [95% CI]) and amputation (17[4.5-95]) (68).

Self-reported symptoms of neuropathy(pain, numbness, tingling) affect 30-40% ofall people with diabetes. In a U.S. popula-tion survey, 28% reported numbness and27% reported pain or tingling (45). About20% of the diabetic population with periph-eral neuropathy denned by multiple meas-ures reported painful symptoms (43). Painis associated with measures of small-fiberneuropathy (69) but has only fair correla-tion with large-fiber neuropathy and footulceration (38,70). Thus, the presence orabsence of symptoms should not be used toassess the risk of ulcers or amputation (70).

Painful diabetic neuropathy tends toimprove within 1-2 years (71-73), espe-

cially if the onset was associated with arecent metabolic disturbance (72); how-ever, the longer the pain has been present,the more likely it is to persist (73-75).Signs of large fiber neuropathy (vibration,ankle jerks, pain sensation) tend to remainstable or worsen over time (71,74). Theimpact of improved glycemic control onsymptomatic neuropathy remains contro-versial and is reviewed elsewhere (76).

Altered biomechanicsAn increased risk of ulceration and ampu-tation has been associated with alterationsin the normal biomechanics of the foot,including increased plantar pressure, bonyabnormalities, and limited joint mobility.An extensive review of the role of biome-chanics in the diabetic foot was recentlypublished (77).

Increased plantar pressure has beenassociated with recent or current ulcers(65,78,79) and with risk of amputation. Ina prospective study, plantar ulcers devel-oped in 35% of the people with static pres-sures > 12.3 kg/cm2 measured on an opticalpedobarograph, while no ulcers developedin people with pressures <12.3 kg/cm2

(80). Pressures > 10 kg/cm2 (98.1 kPa) havebeen associated with increased ulceration(81). Static plantar pressure can be meas-ured using a Harris mat or a polytechnicmodified force plate; dynamic pressures canbe measured with a pedobarograph, instru-mented shoes, and in-shoe or in-sole pres-sure transducers; shear stress can bemeasured with in-shoe transducers. Normalranges and risk level cutoffs have not beenstandardized because different devices givedifferent results under the same foot, andthe transducers wear quickly with repeatedmeasurements. Furthermore, plantar pres-sure patterns appear to change over time(82). Dynamic and shear pressures areprobably more informative than static pres-sures, but are more difficult to measure. Anextensive review of pressure measurementmethodologies is available (83).

The causes of the increased plantarpressure in diabetic people includeincreased body mass, structural alterationsof the bone and connective tissue, limitedjoint mobility, changes in skin and callusformation, and changes in posture and gait.Body weight is associated with <14% ofthe variance of peak plantar pressure in dia-betic people (84).

Increased plantar pressure is oftennoted in the presence of bony deformities.Up to half of all people with diabetes have


Technical Review

a hammer toe or claw toe deformity(15,16,85) created by motor neuropathy,which causes atrophy of the interosseousand lumbricals muscles, thus leading tounopposed hyperextension of the toes (86).This hyperextension deformity exposes themetatarsal heads to increased pressure,resulting in higher rates of ulceration, par-ticularly under the first metatarsal headand the great toe (~50% of all plantarulcers) (65,87). Migration of the plantar fatpad anteriorly with the flexor tendonsintensifies the local pressure (85). The footpad is thinner in people with diabetes, thusincreasing the problems with bony defor-mities causing increased pressure (88).Increased metatarsal head pressure ishighly associated with the presence of neu-ropathy (89). Increased pressure can alsooccur with common foot deformities, suchas bunions and hallux rigidus, and after alower-extremity amputation (90).

Limited joint mobility, first described inthe hands (91) is associated with increasedpressure, current ulcer, and a history of pastulcers (92-95). The limitation developsover time in people with diabetes because ofglycosylation of the skin, soft tissue, andjoints (91,96,97). Diabetic people with lim-ited joint mobility at the subtalar and firstmetatarsophalangeal joint had higher plan-tar pressures than those with normal mobil-ity, and were more likely to have a history ofulcers (93,94). Ankle dorsiflexion <5°, andsubtalar joint motion <30° are significantlyassociated with plantar ulceration (92).However, the reliability of measurement ofthese joints was fairly low when assessed by14 physical therapists (intra-rater reliabilitycoefficient of 0.79-0.90 and inter-rater reli-ability of coefficient of 0.22-0.05) (98), sug-gesting limitations in the clinical utility ofjoint mobility assessment.

People with diabetes and neuropathyare more likely to have gait abnormalities(99), postural instability, and sway (100)and are 15 times more likely to suffer sometype of injury during ambulation thanthose without neuropathy (101).

Bony deformities or limited joint mobil-ity alone may not be sufficient to place aperson at increased risk of ulcers. Rather,the combination of altered biomechanicswith peripheral neuropathy appears tocause the increased risk. In a study of peo-ple with limited joint mobility, all of thepeople who had a history of ulcers also hadneuropathy (93). In another study thatcompared foot conditions in people witheither diabetes or rheumatoid arthritis, sim-

ilar rates of bony deformity were noted;however, the people with diabetes hadsignificantly more neuropathy (102). Of thepeople with diabetes, 32 had a history ofulcers, while none of those with rheumatoidarthritis had an ulcer history. In a prospec-tive study, limited joint mobility measuredby the "praying hand" method (91) wasassociated with a threefold higher rate offoot ulceration; however, this associationdisappeared when abnormal vibratory sensewas taken into account (55). Bony defor-mities from any cause were associated witha twofold higher risk of amputation in thePima Indians and were an independent riskfactor, even after controlling for demo-graphic risk factors, neuropathy, peripheralvascular disease, and history of ulcers (21);however, the retrospective chart reviewprobably had an incomplete ascertainmentof patients' neuropathy status.

Alterations in skin properties may alsocontribute to altered biomechanics andincreased pressures. The skin in people withdiabetes is less pliable, because of the nonen-zymatic glycosylation of collagen and ker-atin, which creates cross-linkage within thetissue (103). Glycosylation of keratin fromthe sole of the foot was increased in peoplewith diabetes and was strongly correlated toglycemic control. Callus formation is a nor-mal physiological response to chronic directpressure and shear stress on the skin. Littleinformation is available to support the com-mon belief that callus formation is morecommon in people with diabetes. A smallcase-control study found no increasedprevalence of calluses, compared with thegeneral population (16), but a study of peo-ple with limited joint mobility causingincreased pressure under the first metatarsaljoint found increased callus formation ascompared with those without increasedpressure (105). A callus may function as aforeign body at the skin surface and furtherincrease localized pressure by up to 29%(106). Hemorrhages and early ulcers canform underneath callus and are commonlymistaken for plantar warts by naive examin-ers (107). Hemorrhagic ulcers were found in27% of the ulcers in neuropathic feet in oneclinic series (2). Calluses were associatedwith an 11-fold higher risk of ulcers andwere more predictive of ulceration thanincreased plantar pressure alone in a clinicpopulation (29).

Increased pressure alone does notappear to cause tissue damage; rather, it isthe cumulative action of moderate levels ofpressure applied repeatedly over time that

results in tissue damage. With ambulation,the foot experiences repetitive low-stressimpact, which accumulates into tissue dam-age leading to ulceration. PW Brand, a pio-neer in the research and management of theneuropathic foot, hypothesized that the per-son with intact sensation subconsciouslyperceives inflammation from repetitive low-level trauma and unconsciously alters thegait to redistribute the pressure (108). Theperson with a loss of protective sensationdoes not appreciate the increasing damageoccurring to the foot, and continues to trau-matize the same tissue. He tested this theoryusing a runner with normal sensation,measuring thermal patterns periodically.Increased temperature was initially noted atpoints of maximum pressure, but the ther-mal pattern changed over time, suggestingthat the runner had altered his gait (108).This same change was not seen in peoplewith neuropathy. Another investigator wasunable to demonstrate an increased vari-ability in the gait pattern between neuro-pathic and non-neuropathic patients usingdynamic gait measurement; however, thesubjects were observed for only 15 min ofwalking, which may not be sufficienttrauma to cause tissue damage and promptgait alterations (77). Further investigation isneeded to establish the mechanism bywhich increased pressure leads to ulcersand amputation.

The severity and location of the tissuetrauma in an insensitive foot may be identi-fied by temperature differentials across thefoot (108). In the normal foot, the highesttemperatures occur in the medial arch in a"butterfly" pattern, with a mean temperatureof 25.7 ± 2.1 to 27.2 ± 2.1°C (109,110).When tissue is exposed to repetitive,increased focal pressure, the local tempera-ture increases (111). Increased temperaturehas been described over prominentmetatarsal heads (112). In a 3-year prospec-tive study of people with symptomatic neu-ropathy, areas with elevated temperatureexperienced an increased incidence of plan-tar ulceration (109).

However, several aspects of diabetic footpathology limit the usefulness of tempera-ture assessment of the neuropathic foot. Inpeople with peripheral neuropathy, markedarteriovenous shunting increases the bloodflow in the skin (46). This alteration, whichmay be due to involvement of the auto-nomic nerves, increases the overall temper-ature of the foot and ameliorates the abilityto increase local blood flow. Diabetic peoplewith sensory and painful neuropathy had a


Technical Review

Table 2—Percentage of ulcers and amputations associated with specified pivotal or precipitatingevents

Ulcers Amputation

Edmonds Apelqvist Pecoraro Mayfieldetal. (150) etal. (87) etal. (11) etal. (21)

Population (n)

Accidental cutsShoe traumaMai perforans*Thermal traumaDecubituslatrogenicVascular occlusionParonychia and

dermatologic conditionsUnknown/unspecifiedEdemaMiscellaneous

386 ulcers in239 patients

547

12

3213

314patients

184012

6

16

4

80veterans

836

383

14

12

63 PimaIndians

268

3851577

0*Also called a repetitive stress ulcer. Usually a new ulcer located on the plantar surface, but could includeulcers on the dorsum of toes or lateral border of shoes. Some authors combined this category with shoetrauma.

mean foot temperature 2-7°C warmer, com-pared with diabetic people without neu-ropathy or nondiabetic people (110,112).Conversely, foot temperature does notincrease in people who have significantperipheral vascular disease (111).

A skilled clinician can detect a changeof ~2°C using the hand (108). Quantitativetechniques to measure temperature includethermistors, liquid-crystal contact thermog-raphy, thermography detectors, and infraredthermography. The reproducibility of ther-mography over 3-6 months had a correla-tion coefficient of 0.82 (112). A 2°difference has been suggested as clinicallysignificant, indicating an inflammatoryprocess leading to ulceration or Charcotarthropathy (108). A prospective study ofpeople with healed ulcers and no peripheralvascular disease found a mean 3.6°C differ-ence between the foot with the prior ulcerand the contralateral foot at the visit justbefore a subsequent ulceration (113). Fur-ther studies are needed on the effectivenessof prospective temperature screening forimpending ulcers and Charcot arthropathy.

Although quantitative plantar pressuremeasurement is being used increasingly inthe clinical setting to customize footwear,no studies to date have demonstrated theeffectiveness of this modality over the tra-ditional clinical exam, which uses clues ofbony deformities, erythema, increasedwarmth, and callus to identify areas ofincreased pressure (114).

Peripheral vascular diseasePeripheral vascular disease, defined as ath-erosclerosis of the peripheral blood ves-sels, is 2-3 times more likely to develop inpeople with diabetes than in the generalpopulation (115). The prevalence dependson the definition and population surveyed.Peripheral vascular disease, defined as atleast one absent pulse, was noted in 15% ofpeople at 10 years and 45% of people at 20years after diagnosis of diabetes (116). Inthe University Group Diabetes Program(UGDP) study, the cumulative incidenceof nonpalpable dorsalis pedis pulse, inter-mittent claudication, and arterial calcifica-tion was 35, 38, and 61% (men) and 38,24, and 32% (women), respectively, afterdiabetes diagnosis (117). The age-adjustedincidence per thousand of claudication forpeople with diabetes in the FraminghamStudy was 12.6 and 8.4 for men andwomen, respectively (115). Peripheral vas-cular disease is highly correlated with ageand duration of diabetes (115-117). Con-troversial associations include smoking (seeearlier discussion) and glycemic status(31,117,118). Peripheral vascular diseaseaffects the femoral and iliac arteries in asimilar rate in people with and withoutdiabetes; however, those with diabetes aremuch more likely to have involvement ofthe peroneal and tibial vessels and to sparethe vessels in the foot (119,120).

Peripheral vascular disease is an infre-quent precipitating event (~5-7%) for

ulcers or amputations (11,21,87) (Table2). However, peripheral vascular diseaseplays a major role in delayed wound heal-ing and gangrene and is a contributing fac-tor to almost half of the amputations (11).

Traditionally, the gold standard for thediagnosis of peripheral vascular disease hasbeen the angiogram. In the clinical setting,the ratio of blood pressures of the lowerextremity to arm pressure, called ankle-armindex (AAI) or ankle-brachial index (ABI), iseasier to obtain and has fair predictive valuefor delayed wound healing (121) andamputation (122). The lower normal limithas been variably defined as 0.94-0.97(123,124). Critical ischemia includes anAAI <0.5. (125). An AAI <0.9 is 95% sen-sitive and almost 100% specific in detectingangiogram-positive disease (126). The vari-ability of AAI can be attributed mainly tobiological variability, and to a lesser degree,observer variation. The European commu-nity prefers the absolute blood pressure ofthe ankle and toe instead of the AAI (125).

The symptoms of peripheral vasculardisease are claudication or rest pain. Clau-dication is defined as pain in the calf thatdevelops upon walking and is relievedwithin 10 min of rest (127). Rest pain isdefined as pain that occurs at rest and isrelieved by dependent positioning of thelegs. Chronic critical ischemia has recentlybeen defined as persistently recurringischemic rest pain requiring regular anal-gesia for more than 2 weeks, with an anklesystolic pressure ^50 mmHg and/or a toesystolic pressure of ^30 mmHg (125)(chronic critical ischemia also includesobvious tissue loss such as gangrene). Clau-dication and rest pain may be more difficultto assess in the person with diabetesbecause of the frequent coexistence ofnocioception or hyperesthesia caused byneuropathy. In addition, the distal locationof the vascular lesions is less likely to pro-duce classic symptoms. Claudication, asdefined by the Rose criteria (98), has apoor sensitivity (9-20%) but excellentspecificity (96-99%) to predict AAI <0.8in a nondiabetic population (128). Thesymptoms of claudication had a sensitiv-ity/specificity of 22/96% against a goldstandard of blood pressure, treadmills, andDoppler studies in 458 volunteers withdiabetes (129). In a population of veteranswith diabetes, the inability to walk one cityblock had a sensitivity/specificity of0.5/0.87 to predict an AAI <0.5 (33).

The clinical exam for early peripheralvascular disease remains an inexact art.


Technical Review

Detection of the dorsalis pedis and posteriortibial artery by palpation is greatly affectedby room temperature, biological variation,and provider skill. The dorsalis pedis iscongenitally absent in up to 12% of thewhite population and <5% of the blackpopulation (130,131). The probability ofagreement on an absent pulse betweenexperienced examiners has been reported torange from 0.49 to 0.59 (132,133) but canbe improved from a K of 0.3-0.6 to 0.6-0.7with training and practice (134,135). Thesensitivity/specificity of a decreased orabsent posterior tibial pulse was 71/91% foran AAI ^0.9 in a nondiabetic population(128). In another study, palpable pedalpulses were always present when toe bloodpressure was >40 mmHg and AAI >0.5;patients who lacked palpable pulses in bothfeet had an AAI <0.9 (136). Pulse palpationin 458 volunteers with diabetes had a sen-sitivity/specificity of 67/69%, using a goldstandard of blood pressures, treadmills, andDoppler studies (129). Simple clinicalexams or signs with poor predictive valueinclude capillary refill time, foot skin cool-ness, diminished lower limb hair growth,and blue/purple foot skin color (33). Testsof reactive hyperemia provided no moreaccuracy than the AAI and were not toler-ated by half of the subjects in one study(123). The best ancillary vascular testappeared to be venous filling >20 s, whichhad a sensitivity/specificity of 22/94% topredict an AAI ^0.5 (33). Stress testingmay provide additional information to theAAI, but has low reproducibility (137).

The vascular assessment of people withdiabetes is further complicated by the pres-ence of medial arterial calcinosis (MAC) orMonckebergs sclerosis. In MAC, the medialwall of the blood vessels become calcified,producing a "lead-pipe" condition, whichmay increase blood pressure readings, and atextreme levels, prevent blood pressureassessment because the vessel becomes non-compressible (138). MAC is observed radi-ographically in 60-80% of people withdiabetes for < 10 years (20,117). The con-dition is progressive, developing first in thefeet and moving proximally, and is highlyrelated to age, duration of diabetes, loss ofvibration perception, and poor glycemiccontrol (139,140). MAC has been noted inboth diabetic and nondiabetic people withsympathetic dennervation (141). MAC isgenerally associated with increased anklesystolic blood pressures and AAI (r = 0.40,r = 0.35, respectively) (140). Although AAI>1.3 is highly specific for MAC, MAC is

present at normal ranges of AAI and in up toone third of diabetic people with AAI ^1.0(140,142). In view of the frequency of MACin patients with diabetes, segmental pres-sures, and therefore AAI 5:1.0, may be arti-ficially elevated and may require furtherevaluation.

Toe pressures measured with plethys-mography may be a more accurate reflec-tion of blood supply to the foot (143) andmay provide better prediction of woundhealing than ankle pressure or AAI(122,144,145). However, toe pressuresmeasured with pulse-volume recordingsare thought to have poor reproducibility(137), especially in the presence of auto-nomic neuropathy (146). The impact ofMAC on the accuracy of blood pressuremeasurements and the preferred vascularassessment measured in the person withdiabetes remains controversial.

Finally, several studies suggest that skinperfusion is an important and independentpredictor of ulceration and wound healingin diabetes (35,121). In a case-controlstudy of a veterans population, multivariateanalysis revealed the most important riskfactors for ulceration were transcutaneousoxygen <30 mmHg (OR 58 [95% CI5.08-658]), absence of Achilles tendonreflexes (OR 6.5), and being insensate to5.07 monofilament (OR 18) (35).

After consideration of these many diffi-culties in vascular evaluation, experts at arecent consensus conference recommendedthat AAI be used in addition to claudicationhistory and pulse palpation for screeningpeople with diabetes. They recommendedthat a screening AAI be performed for allpeople with type 1 diabetes ^35 years oldor with ^20 years duration of diabetes,and for all people with type 2 diabetes andaged >40 years (147). The value of theserecommendations has yet to be demon-strated in a population-based study.

The gold standard for confirmation ofvascular disease has been the angiogram,but this test is expensive, invasive, and maycause renal and anaphylactic complications.Noninvasive tests using Doppler wave formanalysis, duplex ultrasound scanning, andnuclear magnetic resonance can provideinformation on the flow and pressures ateach segment of the lower extremity withless risk to the patient, but cannot provideas much information on collaterals or runoffbefore graft placement (148). The value ofnoninvasive laboratory testing in the screen-ing, diagnosis, and management of periph-eral vascular disease needs further study.

Ulcers and other skin pathologyUlcers are defined as any break in the cuta-neous barrier (in clinical studies the breakmust extend through the full thickness ofthe dermis). The site of ulceration is highlyrelated to the associated foot pathology andprecipitating event (65,87). In the neuro-pathic foot experiencing increased plantarpressure, ulcers are more common underthe tips of the toes and plantar surface ofthe metatarsal heads, especially under thefirst and fifth, and under any bony defor-mity. In the ischemic foot, ulcers are morelikely to occur on the tips of the toes andlateral border of the foot.

Approximately 5% of people with dia-betes for <20 years reported a current orpast foot ulcer (37). A randomized con-trolled trial found that the history of anulcer increased the risk of ulceration 13-fold(58). In another prospective study, the riskof foot ulceration was more highly associ-ated with a history of a previous ulcer (56.8[5.08-658]; OR [95% CI]) than with callus(OR 11) or plantar pressure >10 kg/cm2

(OR 4.7) (35). A history of a prior footulcer is associated with a 2- to 10.5-foldhigher risk of amputation (14,21,25,27).

Foot ulcers recurred in 30-40% peryear of people who returned to their ownusual shoes, thus highlighting the impor-tant role increased pressure and footwearplay in the development of ulcers (2,149).A number of minor skin conditions arethought to increase the risk of ulceration oramputation. These include dry skin (16),edema (2), improper nail care (17), andingrown toenails (87,150). In a prospectivestudy, dry skin, ingrown nails, improperlytrimmed nails, edema, and onychomycosiswere not associated with an increased riskof ulceration in the following year, but fun-gal dermatitis was associated with a three-fold increased risk of a minor, nonulceratedskin lesion (58).

Causal pathwaysEach of the risk factors discussed above arerarely the sole cause of amputation. Thecausal pathway to amputation includesmany component causes. These may includepathophysiological components (ischemia,neuropathy, infection, faulty wound heal-ing), pathological components (ulceration,gangrene), and environmental components(minor trauma). In one case series, absentprotective sensation was a component causein 82%, ischemia in 46% of the cases, andulceration complicated by failure to heal waspresent in 72% of the cases (11).


Technical Review

Pivotal or precipitating eventsIn almost all amputations, an event can beidentified that sets off the cascade of con-sequences, which culminates in amputa-tion. This event, often called the "pivotalevent" or "precipitating event," has beenidentified and categorized in several caseseries of amputations (Table 2). Shoe-related trauma is the most frequent eventleading to ulcers and amputation. Many ofthe events listed in Table 2 could be pre-vented by proper protective foot wear andfoot care practices.

Risk stratificationRisk classification systems aggregate riskconditions with similar prognosis or treat-ment requirements. No formal risk classifi-cation system has been developed to predictdiabetic foot ulceration, and few multivari-ate analyses of ulcer risk have been pub-lished. In one descriptive study, ulcers weremore highly associated with a vibratorythreshold >25 V than with low ankle bloodpressure or elevated plantar pressure (55).In a case-control study of a veterans popu-lation, the most important risk factors forulceration were absence of Achilles tendonreflexes (OR 6 .5), being insensate to 5.07monofilament (OR 18), and transcutaneousoxygen <30 mmHg (58 [5.08-658]; OR[95% CIl) (35). Several risk classificationsystems have been published to predict therisk of amputation (21,114,151-153). Mostsystems were based on expert opinion, butone system was later validated in a prospec-tive fashion (68) and another was developedusing multivariate analyses of retrospectivechart review (21). Although each systemassigns different weights to the various riskfactors, all incorporate the same basic riskfactors of peripheral neuropathy, peripheralvascular disease, bony deformities, evidenceof increased plantar pressure, and a historyof ulcers.

PREVENTIVEINTERVENTIONS—Diabetic footcare programs can decrease the rate of ulcersand amputations by 44 to 85% according toone randomized controlled trial (1) and anumber of pre-post-design studies (2-5).These programs have usually included athorough foot risk assessment, callus andnail care, customized footwear, wound care,and patient education provided by a multi-disciplinary team with a special interest andexpertise in the foot. Most of the interven-tions also incorporated system changes ofstandardized records, chart reminders,

patient-tracking data, and increased clinicaccess. These studies implemented the var-ious interventions simultaneously, making itdifficult to assess the relative impact of eachcomponent to the overall success of the pro-gram. The evidence for specific interven-tions is reviewed below.

Glycemic controlIn the Diabetes Complications and ControlTrial (DCCT), people with type 1 diabeteswho achieved near-normal glycemic con-trol experienced a 69% reduction in thesubclinical neuropathy and a 57% reduc-tion in clinical neuropathy, as comparedwith the control subjects who received theusual treatment and who had higher levelsof glycemia (118). The specific impact ofnormalizing blood glucose, lipids, or bloodpressure on the development of peripheralvascular disease in people with either typeof diabetes is untested.

Foot examsExamination of the foot is an obvious, fun-damental step to identifying certain footrisk factors that can be modified, thusreducing the risk of ulceration and ampu-tation. Little standardization exists on whatconstitutes an adequate foot exam, and nodata are available to support either theeffectiveness or optimum frequency of footexams.

The most rudimentary exam consistsof a visual scan of the skin surface forbreaks in the cutaneous barrier, increasedwarmth, and callus formation. The visualscan at each contact with a health careprovider can identify new, unsuspected footlesions. In a community survey of diabeticpeople, 9% of foot ulcers were unknown tothe patient (18). The ritual of a foot exam ateach visit may emphasize the importance ofthis activity to the patient and might pro-vide a teaching opportunity to reinforceself-care, but this has yet to be evaluatedrigorously.

A comprehensive examination of thefoot, including an assessment of the neuro-logical, vascular, and biomechanical statusis fundamental to identifying patients withrisk factors and to implementing interven-tions. The optimum interval for screeningasymptomatic diabetic people has not beenstudied. A 1-year interval has been sug-gested by experts, based on the natural his-tory of neuropathy and peripheral vasculardisease. The optimum interval for assess-ment of the person with high-risk condi-tions has also not been evaluated.

Approximately 40-60% of people withdiagnosed diabetes have received a footexam within the past year (154-157). Footexams are more likely to be performed ifthere are medical record reminders (1) or ifthe patients socks and shoes are removedbefore the physician enters the exam room(158). Primary care providers currently rec-ognize the prognostic significance of footulcers and are twice as likely to referpatients with foot ulcers for podiatric footcare and education, compared withpatients with bony deformities or periph-eral neuropathy (159). When health careproviders are provided with education andpractice guidelines, the foot examinationtends to be more complete and podiatricreferral is more likely (1).

Peripheral neuropathyThe primary preventive strategy for neu-ropathy is to maintain glycemic control atas near-normal levels as possible (118).Secondary prevention involves early detec-tion of foot risk factors so that preventivefootwear and patient education can be pro-vided (see below). No tertiary preventivestrategy is yet available; that is, no treat-ment is currently available in the U.S. toreverse neuropathy. A number of pharma-cological agents, however, are under devel-opment and assessment. A completediscussion on the medical management ofperipheral neuropathy is beyond the scopeof this paper.

Altered biomechanicsManagement of the person with abnormalbiomechanics depends primarily ondebridement of callus and nails and on pro-vision of footwear tailored to the specificpathology (see FOOTWEAR section below).Conservative management of increasedplantar pressure involves debridement ofcallus and footwear modification. Removalof the callus under the forefoot using ascalpel or abrasive device can reduce meanpeak pressures on the foot by 29% (106).The rate of callus formation can also bedecreased by appropriate footwear (seebelow). Physical therapy has been suggestedfor limited joint mobility and balance prob-lems but has not been evaluated for effec-tiveness. Surgical modification of the bonydeformities of the foot, including resectionof the metatarsal head, have been associatedwith decreased pressures and improvedulcer healing (160,161), but have not beenrigorously evaluated against other manage-ment strategies.


Technical Review

Peripheral vascular diseasePeople in the general population withsymptoms of claudication rarely progressedto amputation over 10-15 years (1.6 to1.8%) (162,163), but the rate of progres-sion appears to be somewhat higher in dia-betic populations (15-20%) (117).

Patients with claudication have beensuccessfully treated with a graduated exer-cise program (164,165). However, someproviders are now advocating angioplastyor vascular bypass (166). Patients withsevere rest pain that qualifies as criticalischemia have been treated with amputa-tion in the past, but now may be offeredvascular bypass or angioplasty. People withclaudication or rest pain in the general pop-ulation treated with bypass and angioplastyreported improved functional status in sev-eral descriptive studies (167). A study ofVeterans Administration general popula-tion patients with claudication or rest painrandomized to angioplasty or bypasssurgery found equal improvement in thequality of life and functional status (168). Asubanalysis of patients with diabetes wasnot possible because of the small samplesize. The few studies that have reportedoutcomes separately for the diabetic popu-lation note more complications and adverseoutcomes as compared with the nondia-betic population. Institutional-basedreports suggest a decrease in amputationrates after the introduction of angioplastyand vascular bypass to the pedal vessels(169,170). Epidemiological studies of largepopulations have not detected a decrease inthe number of amputations since the adop-tion of these surgical approaches (171).

The value of early detection and inter-vention for peripheral vascular disease in theasymptomatic person with diabetes remainsunknown. A complete discussion on themedical management of peripheral vasculardisease is beyond the scope of this paper.

Abnormal skin conditions andfoot ulcer historyExperts suggest that patients should betaught to avoid foot soaks, to dry the footthoroughly, and to use skin moisturizersliberally, but little research is available on theeffectiveness of these self-care practices. In asurvey of self-care practices, dry skin andmaceration were found as frequently in peo-ple who reported soaking their feet as inthose who did not (172). Experts also sug-gest that the nails should be trimmed to thecontour of the toe and extend 2 mmbeyond the nail bed. The edges should be

gently filed to prevent catching on socksand bedding. A health care professionalshould trim nails that are exceptionallythick or deformed or for people withperipheral vascular disease or peripheralneuropathy. However, little research is avail-able to support these recommendations.

Tinea pedis can be treated effectivelywith good hygiene and topical antifungalagents. Until recently, the only cure foronychomycosis (mycotic nails) was nailablation, but now itraconazole andterbinafine provide culture cure rates of70-80% and cosmetic improvement in-50% (173,174). Itracanazole may inter-act with sulfonylureas, cisapride, statindrugs, warfarin, digoxin, and calcium-channel blockers, while terbinafine doesnot. The high recurrence rates are probablydue to immunological factors. Because ofthe high cost of the medications and highrecurrence rate, their cost-effectiveness isquestionable and has yet to be explored forpeople with diabetes. Pulse dosing proto-cols, which use medications only a fewdays per month and thus reduce cost, areavailable for itraconazole (175) and areunder investigation for terbinafine.

FootwearWhen fitted properly, footwear can reduceabnormal pressures, reduce the formationof callus and ulcers, and protect the footfrom external trauma. The compositionand design of the sole and insert can affectpressures and callus formation. In onestudy of people with diabetes, peak pres-sures on the sole were highest when walk-ing barefoot and were significantly loweredwhen measured in properly selected andfitted shoes (176). Plantar pressure in themidfoot was decreased by 31-51% in ten-nis shoes with a firm rubber sole, com-pared with a flat, flexible sole with nocushion (177,178). Rocker and wedge solemodifications can reduce the pressureunder the metatarsal heads up to 30%(177,179). Athletic running shoesdecreased plantar callus formation in agroup of people with diabetes, reducing theneed for debridement by threefold andreducing the number of people complain-ing of painful callus from 70 to 9% (180).Cushioned inserts of viscoelastic polymerdecreased focal pressure by about half(181). A customized molded insert allowsmaximum foot and insole contact, reduc-ing focal areas of increased pressure morethan a flat insert (182). For diabetic peoplewith a transmetatarsal amputation, a cus-

tom-made full-length shoe with a total con-tact insert and a rocker bottom sole hasbeen shown to improve functional mobilityand reduce plantar pressure better thanregular footwear or a short shoe (183). Anumber of materials are available and varyin their compressibility and durability(184). Thick, padded socks also decreasepeak forefoot pressure and the area underthe time-pressure curve by a mean of 25and 29%, respectively (185), and were wellaccepted by patients. However, severalpatients developed ulcers from wearing thesocks in shoes that did not accommodatethe extra thickness (186).

Tovey (187) has summarized the prin-ciples of proper footwear selection.Footwear should relieve areas of excessiveplantar pressure, reduce shock and shear,and accommodate, stabilize, and supportdeformities. Shoes should fit both the footshape and size. The first metatarsopha-langeal joint should be accommodated inthe widest part of the shoe and the lengthshould allow 3/8 to 1/2 inch between theend of the shoe and the longest toe. Theshoe should have sufficient room in the toearea and over the instep. Shoes with lacescan adjust for edema and deformities. Theheel should fit snugly without unduemotion. In certain feet, limitation of jointmobility may add to stability and painreduction. People with balance problemsmay benefit by footwear with wide, lowheels to improve stability and assistivedevices such as canes.

Therapeutic footwear, selected accord-ing to Toveys principles, has been shown tobe effective in the prevention of ulcerationfor people at high risk. People with neu-ropathy and a history of prior ulcerationwere randomized to either a depth shoewith insert padding per Toveys specifica-tions made by a major Italian shoe manu-facturer or to their own usual footwear.After 1 year, the foot ulcer relapses weresignificantly lower (28 vs. 58%, P = 0.009)in the specially padded shoe (188).

The effectiveness of footwear is heavilydependent on footwear acceptability anduse. In a cohort of high-risk patients whowere provided therapeutic footwear (pro-tective cushioned footwear with paddedinsoles), patients who wore their shoes>60% of the time reduced the ulcerrelapse rate by >50%, compared withpatients who wore their shoes for less time(149). Patients who wore the shoes werefree of ulcer recurrence for up to 20months, whereas 38% of patients who did


Technical Review

Table Al—Type of evidence, study designs, and inferences for technical review

Type

I-AI-BII-A1I-BII-CIII

IVXY

Study design

Randomized controlled trial, crossover trialsControlled trial, nonrandomizedCohort, case-controlTime series, pre-post studies, repeated panelCross-sectional population-based dataDescriptive studiesCase series, case reportsExpert opinion and consensus opinionMeets all of the screening and diagnosis criteria*Meets part of the screening and diagnosis criteria*

Inferences

Causation, efficacy of treatmentor risk modification

Association

Hypothesis generation

*Screening and diagnostic test criteria: 1) gold standard, preferably a patient outcome that matters rather thana disease; 2) definitions of the test and the outcomes are clear and easily reproduced; 3) test characteristics:sensitivity, specificity, reproducibility provided; 4) subjects: population-based preferred over selected popu-lations; and 5) tests that provided results early in the disease course are preferred to ones that can only beused late in the course, especially if effective early intervention is possible.

not use the shoes experienced another footulcer within 1 year. These high-use patientswere also more likely to receive debride-ment of callus and nails (2.0 SD [1.7-2.3]vs. 1.3 [0.9-1.7]) times per month (P <0.05), suggesting a selection bias. After 40months, this protective effect leveled off,and 54% of the high-use patients versus100% of the low-use patients had experi-ence ulcer relapses, suggesting that thefootwear may have lost its ability to absorbshock after 2 years. The cosmetic accept-ability of the extra-depth shoe and rockerbottom limits the effectiveness of thera-peutic footwear. In a study of 85 high-riskdiabetic patients who were prescribed ther-apeutic footwear, only 49 patients woretheir prescribed shoes to a subsequentclinic visit. Those who failed to purchase orwear the prescribed shoe cited the unat-tractive appearance of the shoes (189).

Most people with diabetes and neu-ropathy alone can wear commercialfootwear such as walking shoes and athleticshoes that meet Tovey's principles. Peoplewith hammer toes, bunions, or evidence ofhigh plantar pressure will need extradepth,depth-inlay, or custom shoes that provide aslightly wider and deeper toe box. A numberof styles are available from specialty shoe dis-tributors for less than $200. Commercialand custom insoles may be needed to redis-tribute the pressure, and sole modifications,including a rocker or wedge, may be neededfor people with limited joint mobility Cus-tom-molded shoes are usually required forpeople with severe deformities such Charcotarthropathy, severe arthritis, or postamputa-tion. These shoes are fabricated over a plas-

ter replica of the patients foot by a specialistand are often very expensive.

Often, people with neuropathy pur-chase shoes that are too small to have theusual sensation of footwear pressure.Experts suggest that before shoe shopping,the patient should stand on a piece of paperand draw an outline of the foot and use thisfor comparison with the prospective shoe.Patients with evidence of elevated pressure,neuropathy, bony deformities, or a historyof an ulcer should have their footwear fittedby a professional (i.e., pedorthist or otherfoot care professional) (176). The effective-ness of quantitative pressure measurementsystems to customize footwear has yet to beevaluated.

Patient educationMost foot care intervention programs haveincluded patient education along with otherinterventions, making it difficult to distin-guish the relative impact of education in theoverall success (1-5). Studies of lecture-style educational sessions on foot care haveshown short-term improvements in knowl-edge, but few changes in foot care practicesand foot conditions (190). Educationalinterventions using motivational techniques(191) or skill-based education (192) tend toshow more changes in self-care behaviors.In a randomized trial of a county hospitalpopulation, patients received an intensiveevaluation and education session, rein-forced with behavioral contracts andreminders, and their doctors received edu-cation and chart reminders (1). The controlpopulation received the usual care andusual provider education. After 1 year,

patients were more likely to report appro-priate self-care of the foot behaviors, includ-ing inspection of feet and shoes, washing offeet, and drying between toes. However,other behaviors, including testing of bathtemperature and reporting foot problemswere not significantly altered. Patients withthe intervention had fewer serious footlesions (OR 0.41) including ulcers. Anotherstudy randomized primary care sites tousual care or group education provided bythe general practitioner. The patient groupshad similar demographic and foot pathol-ogy at baseline. At 6 months, the interven-tion group had significant reductions incallus (49 vs. 82%), fewer minor skinpathology (49 vs. 65%), and improper nailtrimming (27 vs. 92%) (193). Both of thesestudies highlight the effectiveness of com-bining provider and patient education.

Only one randomized controlled studyhas been conducted of foot education asthe sole intervention. Veterans from a high-risk foot clinic were randomized to "usualeducation" or a 1-h slide lecture showingulcers and amputations followed by a sim-ple, one-page instruction sheet to takehome. After 2 years, people receiving theeducational session had a threefolddecrease in ulceration (P < 0.005) andamputation rates (P < 0.0025) (194). Lit-tle is known about the effectiveness of edu-cation over longer periods of time or forlow-risk populations.

The patients ability to conduct an ade-quate self-exam of the foot may be limitedby poor vision, obesity, poor mobility, orcognitive problems. One study found 71%of the patients in a high-risk foot clinichad poor vision (195). Another studyfound that 39% of the elderly diabetic peo-ple studied were unable to reach their toesto remove a simulated lesion that had beenapplied to the foot and that only 14% of theelderly subjects could respond appropri-ately to a plantar lesion (196). Patientsshould be evaluated for these types of lim-itations, and if present, family members ornursing services should be recruited toprovide foot care and surveillance.

Little research has been conducted onthe content of patient education. Frequentsurveillance of the foot and early notifica-tion of health care professionals is sug-gested. No inexpensive systems for homemonitoring of temperature have been devel-oped. Experts suggest that patients shouldbe taught to palpate the feet with theirhands to detect increased warmth, but theeffectiveness of this practice has not been


Technical Review

Table A2—Type of evidence for selected foot care examinations and management strategies in diabetic people without a current ulcer

Evaluation method or management strategy Type of evidence References

Foot screening and surveillanceFrequency and type of exam

Low-risk patienttHigh-risk patientt

Comprehensive foot examinationHistoryPeripheral neuropathy

LOPS

Other neuropathyAltered biomechanicsIncreased focal pressure

PVD

Skin pathology

Risk conditionsPrevention of:

NeuropathyPVD

Neuropathy, evidence of increasedplantar pressure, or a history of plantarulcers without bony deformity

Moderate bony deformitiesToe deformitiesBunions

Marked bony deformities(i.e., Charcot foot, amputation)

Limited joint mobility (MTP or ankle)Callus

Nail pathologyThick, mycotic, or painfulIngrown or mycoticMycotic

PVDAsymptomaticClaudication

Rest pain (critical ischemia)Ulcer historyMinor skin abnormalities

Dry skinFungal infections

Therapeutic footwear

Patient educationLow-risk patienttHigh-risk patienttFrequency of educationEducation content:

(knowledge and self-management skills)

Educational assessment

Comprehensive annual examComponents of comprehensive exam every 3-6 months

History of neuropathy, vascular disease or treatment, prior ulcer, or amputation

Semmes-Weinstein monofilament 5.07 (10 g) or threshold testsfor vibration

Assessment of motor strength, proprioception, gait, and stabilityAssessment of bony deformities, limited joint mobility, prior amputationErythema, callus formation, pre-ulcer (i.e., hemorrhage under callus)Temperature assessment by palpation or thermometer

Useful in people with deformity or LOPS aloneNot useful in people with PVD or autonomic neuropathy

Symptoms of claudication or rest painPalpation of dorsalis pedis and posterior tibial pulsesVisual inspection for breaks in the cutaneous barrier

(i.e., ulcers) and minor skin abnormalities (thickenedor ingrown nails, fissures, fungal infection, erythema, edema, dry skin)

Glycemic levels maintained as near normal as possibleSmoking cessationAthletic shoe or walking shoe with cushioned soles or inserts,

adequate toe box, and lace-up vamp

Extra-depth shoe ± insertExtra wide shoeMolded shoe to accommodate footSurgical modification of foot to fit footwearRocker sole ± insertDebridementCushioned soles and inserts

DebridementExcisionOral antifungal medications

ObservationExercise therapyBypass surgery or angioplastyDistal bypass, angioplasty, or amputationDetermine probable etiology and treat

EmollientsTopical agentsShould be fitted by a person with expertisein therapeutic footwear

Patient education on foot care and footwear (possibly useful)Patient education on foot care and footwear (beneficial)AnnualImpact of neuropathyFoot hygieneToenail trimming (low-risk foot)Soaking feet, prohibitionSkin moisturizersFootwear selectionAvoidance of foot traumaDaily surveillance of the footWhen to consult medical carePatients ability to conduct self-foot exam and respond

to high-risk conditions

IVIV

Il-A

I-A, X

II-A, 111111

II-A, ll-B

II-A111

Il-C, XX

II-A

14,21,25,35,27,58

34,35,54,55,61,64,65,68

99-10193-95

103-106

109,112,113110,112

33,125,128,12933,128,134-H6

2,16,17,58,87,150

1-AII-AI-AIII

1-AIIII-All-BI-AIl-B1-A

IVIVI-A

11-A1-A

1-A, ll-Bll-B, 111

IV

IVIV111

1-AI-AIVIV111IVIVIV1-AIl-AIVIVIII

11829

176-178180,181

182,183,188

181,182160,161177,179

106180

*t,

173-175

117,162,163164,165166-168168-171

••'

*

176

190-193194

19.3*

17611,21,87,150

195.196

*No references provided for recommendations based on expert opinion (IV). tHigh-risk (one or more of the following conditions): LOPS, evidence of increasedpressure, limited joint mobility, bony deformity including amputation, PVD, or a current or past foot ulcer; low-risk: none of the listed conditions. LOPS, loss ofprotective sensation; PVD, peripheral vascular disease.


Technical Review

evaluated. Experts also suggest that peoplewith neuropathy should also be advised tobreak in the new shoes gradually to mini-mize the formation blisters and ulcers.

ExerciseExercise is an important management toolfor people with diabetes that must be mod-ified for individuals with high-risk foot con-ditions. Experts recommend non-weight-bearing or low-impact exercise, such asswimming or bicycle riding. These types ofexercises will provide cardiovascular andmetabolic benefits but may not providecalcium retention for prevention of osteo-porosis.

Economic implicationsThe economic impact of foot-related prob-lems in people with diabetes is considerable.However, cost has been difficult to assessbecause of the rapid changes in health careover the past 15 years. For example, in 1985,a hospital in southern California charged$23,500 for a vascular bypass and $24,700for an amputation (197) but in 1990, theDeaconess Hospital in Boston, Massachu-setts charged on average $15,981 for a vas-cular bypass and $18,341 for an amputation(198). This may reflect geographic differ-ences in charges, but also reflects the shift toshorter length of stay and cost-containmentefforts. In 1992, the average hospital reim-bursement for a lower-extremity amputa-tion from the Medicare program was$10,969, compared with $26,940 from pri-vate insurers. During this same period, reha-bilitation was moved from the acute carehospital to rehabilitation facilities and nurs-ing homes, and was reimbursed in 1992 at arate of $7,000 to $21,000 per person.

In contrast, the costs for the preventivestrategies listed above are negligible. Footexams take less than 5 min of a primary careproviders time and monofilaments cost lessthan $25. Patient education and debride-ment of the nails and callus are generallyreimbursed and have a low cost. In 1998,insoles off the shelf cost $5 to $20, custominsoles cost up to $200, depth shoes ($150to $200), and custom-molded shoes up to$500. Medicares evaluation of therapeuticfootwear for diabetic people with high-riskconditions found the program to be costneutral (199). No formal cost-benefit analy-ses or modeling of preventive foot care hasbeen published yet, but would be valuable.

SUMMARY — A number of effective,low-cost strategies are available to identify

and treat the person at risk for diabeticfoot ulcers and lower-extremity amputa-tion. These strategies must be more widelyadopted by all diabetic care providers tomaintain the integrity and function of thelower limb, and thus improve the quality oflife for people with diabetes.

APPENDIX

Methods for the Technical Review onfoot careThis Technical Review is based on originalresearch published in English peer-reviewed literature. The articles wererestricted to human subjects and excludedabstracts and unpublished work. Emphasiswas given to clinically important outcomes,especially those that matter to patients,rather than disease-oriented outcomes.

The type of evidence for studies onprognosis, risk factor identification, andinterventions was scored using a modifica-tion of the U.S. Preventive Services TaskForce Criteria (Table Al). The highest levelof evidence was used for scoring; however,if only one small study was available, thenext lower level was included in the desig-nation.

Interventional studies under the con-trol of the investigator were graded as levelI. Randomized controlled trials and well-designed crossover studies were coded as I-A. Controlled, nonrandomized studies,coded as II-1 in the U.S. Preventive TaskForce classification system, were graded asI-B because the quality and inferences weremore similar to I-A studies.

Level II involved various observationalstudy designs that permit the association ofa risk or treatment with outcomes, but can-not confirm causation or efficacy of treat-ment. Analytic studies (i.e., case-control,cohort data) were coded as II-A. Data frompopulation-based time series data, such asvital statistics or utilization data (i.e., hos-pital discharge records) were coded as II-B.Population-based cross sectional data (i.e.,National Health Interview Study) werecoded as II-C.

Studies involving descriptive cases,case series, or comparisons of case series,even if the study called one group "con-trols," were coded as level III. Evidenceentirely based on expert opinion was iden-tified as level IV Both of these categories ofdata are used for hypothesis generation.

Data for the clinical management ofthe foot in diabetic patients without a cur-rent ulcer are given in Table A2.

Screening and diagnostic tests do not fitinto the U.S. Preventive Services Task ForceClassification system, so were evaluated dif-ferently. All tests had to have a gold standardand a clear outcome, preferably a patient-oriented rather than disease-oriented out-come. An example of this would be thepreference for a screening test that predictedthe risk of ulceration over a test that pre-dicted a nerve conduction value <2 SDs.The evaluation had to provide clear defini-tions for the test and outcomes, and providedetails on the test characteristics, includingsensitivity, specificity, and reproducibility.The more generalized the population, thebetter; or the replication of the study wasalso counted. Studies of test characteristicsthat met all of the criteria were coded as (X);those that met at least half, but not all, of thecriteria were coded as (Y).

References1. Litzelman DK, Slemenda CW, Langefeld

CD, Hays LM, Welch MA, Bild DE, FordES, Vinicor F: Reduction of lower extrem-ity clinical abnormalities in patients withnon-insulin-dependent diabetes mellitus.Ann Intern Med 119:36-41, 1993

2. Edmonds ME, Foster A, Watkins PJ: Cancareful foot care in the diabetic patient pre-vent major amputation? In limb Salvageand Amputation far Vascular Disease. Green-halgh RM, Jamieson CW, Nicolaides AN,Eds. Philadelphia, WB Saunders, 1988, p.407-417

3. Davidson JK, Alogna M, Goldsmith M,Borden J: Assessment of program effec-tiveness at Grady Memorial Hospital -Atlanta. In Educating Diabetic Patients.Steiner G, Lawrence PA, Eds. New York,Springer, 1981, p. 329-348

4. Runyan JW Jr, Vander Zwaag R, JoynerMB, Miller ST: The Memphis DiabetesContinuing Care Program. Diabetes Care3:382-386, 1980

5. Assal JP Muhlhauser I, Pernet A, Gfeller R,Jorgens V Berger M: Patient eduation as thebasis for diabetes care in clinical practiceand research. Diabetologia 28:602-613,1985

6a. American Diabetes Association: Preventivecare of the foot in people with diabetes(Position Statement). Diabetes Care 21:2178-2179,1998

6b. U.S. Preventive Task Force: Guide to Clini-cal Preventive Services, Report of the U.S.Preventive Services Task Force. Baltimore,MD, Williams & Wilkins, 1996

7. Woolf SH, Diguiseppi CG, Atkins D,Kamerow DB: Developing evidence-basedclinical practice guidelines: lessons learnedby the U.S. Preventive Services Task Force.Ann Rev Public Health 17:511-538, 1996


Technical Review

8. Reiber GE, Boyko EJ, Smith DG: Lowerextremity foot ulcers and amputations indiabetes. In Diabetes in America. 2nd ed.Harris MI, Cowie CC, Stern MP, Boyko EJ,Reiber GE, Bennett PH, Eds. Washington,DC, U.S. Govt. Printing Office, 1995 (NIHpubl. no. 95-1468)

9. Centers for Disease Control: Diabetes Sur-veillance, 1993. Atlanta, GA, U.S. Depart-ment of Health and Human Services, 1993

10. Preston SD, Reiber GE, Koepsell TD.Lower Extremity Amputations and ImpatientMortality in Hospitalized Persons With Dia-betes: National Population Risk Factors andAssociations. PhD thesis. University ofWashington, 1993

11. Pecoraro RE, Reiber GE, Burgess EM:Pathways to diabetic limb amputation:basis for prevention. Diabetes Care13:513-521, 1990

12. Larsson J: Lower extremity amputation indiabetic patients. PhD thesis. Lund Uni-versity, 1994

13. Palumbo JP, Melton LJ: Peripheral vasculardisease and diabetes. In Diabetes in Amer-ica. Harris M, Hamman R, Eds. Washing-ton, U.S. Govt. Printing Office, 1985, p.XV 1-21

14. Moss SE, Klein R, Klein BEK: The preva-lence and incidence of lower extremityamputation in a diabetic population. ArchIntern Med 152:610-616, 1992

15. Rosenqvist U: An epidemiological surveyof diabetic foot problems in StockholmCounty 1982. Acta Med Scand 687(Suppl.):55-60, 1984

16. Borssen B, Bergenheim T, Lithner F: Theepidemiology of foot lesions in diabeticpatients aged 15-50 years. Diabet Med7:438-444, 1990

17. Neil HAW, Thompson AV, Thorogood M,Fowler GH, Mann JI: Diabetes in theelderly: the Oxford Community DiabetesStudy. Diabet Med 6:608-613, 1989

18. Walters DP, Gatling W, Mullee MA, HillRD: The distribution and severity of dia-betic foot disease: a community study withcomparison to a non-diabetic group. Dia-bet Med 9:354-358, 1992

19. Huse DM, Oster G, Killen AR, Lacey MJ,Colditz GA: The economic costs of non-insulin-dependent diabetes mellitus. J AmMedAssoc 262:2708-2713, 1989

20. Nelson R, Gohdes D, EverhartJ, HartnerJ,Zwemer F, Pettitt D, Knowler W: Lower-extremity amputations in NIDDM: 12-yrfollow-up study in Pima Indians. DiabetesCare 11:8-16, 1988

21. Mayfield JA, Reiber GE, Nelson RG,Greene T: A foot risk classification systemto predict diabetic amputation in Pimaindians. Diabetes Care 19:704-709, 1996

22. Selby JV, Zhang D: Risk factors for lowerextremity amputation in persons with dia-betes. Diabetes Care 18:509-516, 1995

23. Lehto S, Pyorala K, Ronnemaa T, Laakso

M: Risk factors predicting lower extremityamputations in patients with NIDDM.Diabetes Care 19:607-612,1996

24. Lavery LA, Ashry HR, vanHoutum W,Pugh JA, Harkless LB, Basu S: Variation inthe incidence and proportion of diabetes-related amputations in minorities. DiabetesCare 19:48-52, 1996

25. Reiber GE, Pecoraro RE, Koepsell TD: Riskfactors for amputation in patients withdiabetes mellitus. Ann Intern Med117:97-105, 1992

26. Masson EA, Angle S, Roseman P Soper C,Wilson I, Cotton M, Boulton AJM: Dia-betic foot ulcers: do patients know how toprotect themselves? Practical Diabetes6:22-23, 1989

27. Lee J, Lu M, Lee V, Russel D, Bahr C, LeeE: Lower extremity amputation: incidence,risk factors, and mortality in the Okla-homa Indian Diabetes Study. Diabetes42:876-882, 1993

28. Chaturvedi N, StephensonJM, Fuller JH,The EURODIAB IDDM ComplicationsStudy Group: The relationship betweensmoking and microvascular complicationsin the EURODIAM IDDM ComplicationsStudy. Diabetes Care 18:785-792,1995

29. Smoking and Health: A Report of the SurgeonGeneral. Office of Smoking and Health,U.S. Department of Health, Educaiton,and Welfare. Washington, DC, U.S. Govt.Printing Office, 1979

30. Janka HU, Ziegler AG, Standl E, MehnertH: Daily insulin dose as a predictor ofmacrovascular disease in insulin treatednon-insulin-dependent diabetics. DiabeteMetab 13:359-64, 1987

31. West KM, Ahuja MMS, Bennett PH,Czyzyk A, Mateo de Acosta O, Fuller JH,Grab B, Grabauskas V, Jarrett RJ, Kosaka K,Keen H, Krolewski AS, Miki E, Schiack V,Teuscher A, Watkins PJ, Stober JA: Therole of circulating glucose and triglycerideconcentrations and their interactions withother "risk factors" as determinants of arte-rial disease in nine diabetic populationsamples from the WHO MultinationalStudy. Diabetes Care 6:361-369,1983

32. World Health Organization MultinationalStudy of Vascular Disease in Diabetics,Diabetes Drafting Group: Prevalence ofsmall vessel and large vessel disease in dia-betic patients from 14 centres. Diabetologia28:615-640, 1985

33. Boyko EJ, Ahroni JH, Davignon DSV, Pri-geon RL, Smith DG: Diagnostic utility ofthe history and physical examination forperipheral vascular disease among patientswith diabetes mellitus. ] Clin Epidemiol50:659-668, 1997

34. Young MJ, Breddy JL, Veves A, BoultonAJM: The prediction of diabetic neuro-pathic foot ulceration using vibration per-ception thresholds: a prospective study.Diabetes Care 17:557-560, 1994

35. McNeely MJ, Boyko EJ, Ahroni JH, StenselVL, Reiber GE, Smith DG, Pecoraro RE:The independent contributions of diabeticneuropathy and vasculopathy in footulceration. Diabetes Care 18:216-219,1995

36. Janka HU, Standl E, Mehnert H: Periph-eral vascular disease in diabetes mellitusand its relation to cardiovascular risk fac-tors: screening with the Doppler ultra-sonic technique. Diabetes Care 3:207-213,1980

37. Kumar S, Ashe HA, Fernando DJS, TsigosC, Young RJ, Ward JD, Boulton AJM: Theprevalence of foot ulceration and its con-e-lates in type 2 diabetic patients: a popula-tion-based study. Diabet Med 11:480-484,1994

38. Young RJ, Zhou YQ, Rodriguez E, PrescottRJ, Ewing DJ, Clarke BF: Variable rela-tionship between peripheral somatic andautonomic neuropathy in patients withdifferent syndromes of diabetic polyneu-ropathy. Diabetes Care 35:192-197, 1986

39. Adler AI, Boyko EJ, Ahroni JH, Stensel V,Forsberg RC, Smith DG: Risk factors fordiabetic peripheral sensor)' neuropathy.Diabetes Care 20:1162-1167, 1997

40. Franklin GM, Shetterly SM, Cohen JA,Baxter J, Hamman RF: Risk factors for dis-tal symmetric neuropathy in NIDDM. Dia-betes Care 17:1172-1177, 1994

41. American Diabetes Association, AmericanAcademy of Neurology: Report and rec-ommendations of the San Antonio Con-ference on Diabetic Neuropathy. DiabetesCare 11:592-597, 1988

42. Pirart J: Diabetes mellitus and its degener-ative complications: a prospective study of4,400 patients observed between 1947and 1973. Diabetes Care 1:168-262,1978

43. Dyck PJ, Kratz KM, Karnes JL, Litchy WJ,Klein R, Pach JM, Wilson DM, O'Brien PC,Melton LJ: The prevalence by staged sever-ity of various types of diabetic neuropathy,retinopathy, and nephropathy in a popu-lation-based cohort: the Rochester Dia-betic Neuropathy Study. Neurology43:817-824, 1993

44. Franklin GM, Kahn LB, Baxter J, MarshallJA, Hamman RF: Sensory neuropathy innon-insulin-dependent diabetes mellitus:the San Luis Valley Diabetes Study. Am JEpidemiol 131:633-643, 1990

45. Harris M, Eastman R, Cowie C: Symptomsof sensory neuropathy in adults withNIDDM in the U.S. population. DiabetesCare 16:1446-1452, 1993

46. Edmonds ME, Roberts VC, Watkins PJ:Blood flow in the diabetic neuropathicfoot. Diabetologia 22:9-15, 1982

47. Corbin DOC, Young RJ, Morrison DC,Hoskins P, McDicken WN, Housley E,Clarke BF: Blood flow in the foot,polyneuropathy and foot ulceration in dia-betes mellitus. Diabetologia 30:468-473,


Technical Rexiew

198748. Dyck PJ, Bushek W, Spring EM, Karnes JL,

Litchy WJ, O'Brien PC, Service FJ: Vibra-tory and cooling detection thresholdscompared with other tests in diagnosingand staging diabetic neuropathy DiabetesCare 10:432-440, 1987

49. Dyck PJ, Karnes JL, Daube J, O'Brien PSFJ:Clinical and neuropathological criteria forthe diagnosis and staging of diabeticpolyneuropathy. Brain 108:861-880,1985

50. Dyck PJ: Evaluation procedures to detect,characterize, and assess the severity of dia-betic neuropathy. Diabet Med 8:48-51,1991

51. Dyck PJ: Quantitating severity of neu-ropathy. In Peripheral Neuropathy. 3rd ed.Dyck PJ, Thomas PK, Griffin JW, Low PA,Poduslo JF, Eds. Philadelphia, WB Saun-ders, 1993, p. 686-697

52. Maser RE, Nielsen VK, Bass EB, Manjoo Q,Dorman JS, Kelsey SF, Becker DJ, OrchardTJ: Measuring diabetic neuropathy, assess-ment and comparison of clinical examina-tion and quantitive sensory testing.Diabetes Care 12:270-275, 1989

53. Neumann C, Branchtein L, Schmid H:Severe autonomic neuropathy: how manysymptoms? Diabetes Care 18:133-134,1995

54. Liniger C, Albeanu A, Bloise D, Assal J:The tuning fork revisited. Diabet Med7:859-864, 1990

55. Boulton AJM, Kubrusly DB, Bowker JH,Gadia MT, Quintero L, Becker DM, SkylerJS, Sosenko JM: Impaired vibratory per-ception and diabetic foot ulceration. Dia-bet Med 3:335-337, 1986

56. Fruhstorfer H, Lindblom U, Schmidt WG:Method for quantitative estimation of ther-mal thresholds in patients. J Neural Neuro-surg Psychiatry 39:1071-1075, 1976

57. Chan AW, MacFarlane LA, Bowsher D,Campbell JA: Weighted needle pinpricksensory thresholds: a simple test of sensoryfunction in diabetic peripheral neuropathy.JNeurol Neurosurg Psych 55:56-59, 1992

58. Litzelman DK, Marriott DJ, Vinicor F:Independent physiological predictors offoot lesions in patients with NIDDM. Dia-betes Care 20:1273-1278, 1997

59. Gerr FE, Letz R: Reliability of a widelyused test of peripheral cutaneous vibrationsensitivity and a comparison of two testingprotocols. BrJInd Med 45:635-639,1988

60. Sekuler R, Nash D, Armstrong R: Sensi-tive, objective procedure for evaluatingresponse to light touch. Neurology23:1282-1291, 1973

61. Thivolet C, El Farkh J, Petiot A, SimonetC, Tourniaire J: Measuring vibration sen-sations with graduated tuning fork. Dia-betes Care 13:1077-1080, 1990

62. Bell-Krotoski JA: Sensibility testing: cur-rent concepts. In Rehabilitation of the Hand:

Surgery and Therapy. 3rd ed. Hunter JM,Schneider LH, Mackin EJ, Callahan AD,Eds. St. Louis, MO, CV Mosby, 1989, p.109-128

63. Levin S, Pearsall G, Ruderman RJ: VonFrey's method of measuring pressure sen-sibility in the hand: an engineering analy-sis of the Weinstein-Semmes pressureaesthesiometer. J Hand Surg 3:211-216,1978

64. HolewskiJJ, Stess RM, Graf PM, GrunfeldC: Aesthesiometry: quantification of cuta-neous pressure sensation in diabeticperipheral neuropathy. J Rehabil Res Dev25:1-10, 1988

65. Birke JA, Sims DS: Plantar sensory thresh-old in the ulcerative foot. Lepr Rev57:261-267, 1986

66. Sosenko JM, Kato M, Soto R, Bild DE:Comparison of quantitative sensory-threshold measures for their associationwith foot ulceration in diabetic patients.Diabetes Care 13:1057-1061, 1990

67. Kumar S, Fernando DJS, Veves A, KnowlesEA, Young MJ, Boulton AJM: Semmes-weinstein monofilaments: a simple, effec-tive and inexpensive screening device foridentifying diabetic patients at risk of footulceration. Diabetes Care 13:63-68, 1991

68. Rith-Najarian SJ, Stolusky T, Gohdes DM:Identifying diabetic patients at high riskfor lower-extremity amputation in a pri-mary health care setting. Diabetes Care15:1386-1389,1992

69. Tsigos C, White A, Young RJ: Discrimina-tion between painful and painless diabeticneuropathy based on testing of largesomatic nerve and sympathetic nervefunction. Diabet Med 9:359-365, 1992

70. Veves A, Manes C, Murray HJ, Young MJ,Boulton AJM: Painful neuropathy and footulceration in diabetic patients. DiabetesCare 16:1187-1189, 1993

71. Mayne N: The short-term prognosis indiabetic neuropathy. Diabetes 17:270-273,1968

72. Young RJ, Ewing DJ, Clarke BF: Chronicand remitting painful diabetic polyneu-ropathy. Diabetes Care 11:34-40, 1988

73. Benbow SJ, Chan AW, Bowsher D, Mac-Farlane LA, Williams G: A prospectivestudy of painful symptoms, small-fibrefunction and peripheral vascular disease inchronic painful diabetic neuropathy. Dia-bet Med 11:17-21, 1994

74. Bischoff A: The natural course of diabeticneuropathy: a follow-up. Horm Metab Res9:98-100, 1980

75. Boulton AJM, Armstrong WD, ScarpelloJHB, Ward JD: The natural history ofpainful diabetic neuropathy: a 4 yearstudy. Postgrad Med} 59:556-559, 1983

76. Boulton A: What causes neuropathic pain?J Diabetes Complications 6:58-63, 1992

77. Cavanagh PR, Simoneau GG, Ulbrecht JS:Ulceration, unsteadiness, and uncertainty:

the biomechanical consequences of dia-betes mellitus. J Biomechanics 26:23-40,1993

78. Ctercteko GC, Dhanendran M, HuttonWC, Lequesne LP: Vertical forces acting onthe feet of diabetic patients with neuro-pathic ulceration. Br J Surg 68:608-614,1981

79. Veves A, Van Ross ER, Boulton AJ: Footpressure measurements in diabetic andnondiabetic amputees. Diabetes Care15:905-907, 1992

80. Veves A, Murray HJ, Young MJ, BoultonAJM: The risk of foot ulceration in diabeticpatients with high foot pressure: a prospec-tive study. Diabetologxa 35:660-663, 1992

81. Boulton AJM, Hardisty CA, Betts RP,Franks CI, Worth RC, Ward JD, Duck-worth T: Dynamic foot pressure and otherstudies as diagnostic and management aidsin diabetic neuropathy. Diabetes Care6:26-33, 1983

82. Boulton AJM, Betts RP, Franks CI, WardJD, Duckworth T: The natural history offoot pressure abnormalities in neuropathicdiabetic subjects. Diabetic Res 5:73-77,1987

83. Cavanagh PR, Hewitt FG Jr, Perry JE: In-shoe plantar pressure measurement: areview. Foot 2:185-194, 1992

84. Cavanagh PR, Sims DS Jr, Sanders LJ:Body mass is a poor predictor of peakplantar pressure in diabetic men. DiabetesCare 14:750-755, 1991

85. Coughlin MS: Mallet toes, hammer toes,claw toes, and corns: causes and treat-ments of lesser toe deformities. PostgradMed 75:191-198, 1984

86. Habershaw G, Donovan JC: Biomechani-cal considerations of the diabetic foot. InManagement of Diabetic Foot Problems.Kozak GR Hoar CS, Rowbotham JL,Wheelock FC, Gibbons GW, Campbell D,Eds. Philadelphia, WB Saunders, 1984, p.32-44

87. Apelqvist J, Larsson J, Agardh CD: Theinfluence of external precipitating factorsand peripheral neuropathy on the devel-opment and outcome of diabetic footulcers. J Diabetes Complications 4:21-25,1990

88. Gooding GAW, Stess RM, Graf PM: Sonog-raphy of the sole of the foot: evidence forloss of foot pad thickness in diabetes andits relationship to ulceration of the foot.InvestRadiol 21:45-48, 1986

89. Boulton AJM, Betts RP, Franks CI, NewrickPG, Ward JD, Duckworth T: Abnormalitiesof foot pressure in early diabetic neuropa-thy. Diabet Med 4:225-228,1987

90. Lavery LA, Lavery DC, Quebedeax-Farn-ham TL: Increased foot pressures aftergreat toe amputation in diabetes. DiabetesCare 18:1460-1462, 1995

91. Rosenbloom AL, Silverstein JH, LezotteDC, Riley WJ, Maclaren NK: Limited joint


Technical Review

mobility in childhood diabetes mellitusindicates increased risk of microvasculardisease. NEnglJMed 305:191-194, 1982

92. Mueller MJ, Diamond JE, Delitto A, Sina-core DR: Insensitivity, limited joint mobil-ity, and plantar ulcers in patients withdiabetes mellitus. Phys Ther 69:453-462,1989

93. Fernando DJS, Masson EA, Veves A, Boul-ton AJM: Relationship of limited jointmobility to abnormal foot pressures anddiabetic foot ulceration. Diabetes Care14:8-11, 1991

94. Birke JA, Franks BD, Foto JG: First rayjoint limitation, pressure, and ulceration ofthe first metatarsal head in diabetes melli-tus. Foot Ankle Int 16:277-284, 1995

95. Delbridge L, Perry P, Marr S, Arnold N,Yue DK, Turtle JR, Reeve TS: Limited jointmobility in the diabetic foot: relationshipto neuropathic ulceration. Diabet Med5:333-337,1988

96. Grgic A, Rosenbloom AL, Weber FT, Gior-dano B, Malone JI, Shuster JJ: Joint con-tracture: common manifestation ofchildhood diabetes mellitus. J Pediatr88:584-588, 1976

97. Shinabarger NI: Limited joint mobility inadults with diabetes mellitus. Phys Ther67:215-218, 1987

98. Elveru RA, Rothstein JM, Lamb RL:Goniometric reliability in a clinical set-ting. Phys Ther 68:672-677, 1988

99. Mueller MJ, Minor SD, Sahrmann SA,Schaaf JA, Strube MJ: Differences in thegait characteristics of patients with dia-betes and peripheral neuropathy com-pared with age-matched controls. PhysTher 74:299-313, 1994

100. Simoneau GG, Ulbrecht JS, Derr JA,Becker MB, Cavanagh PR: Postural insta-bility in patients with diabetic sensoryneuropathy. Diabetes Care 17:1411-1421,1994

101. Cavanagh PR, Derr JA, Ulbrecht JS, MaserRE, Orchard TJ: Problems with gait andposture in neuropathic patients withinsulin-dependent diabetes mellitus. Dia-bet Med 9:469^-74, 1992

102. Masson EA, Hay EM, Stockley I, Veves A,Betts RP, Boulton AJM: Abnormal footpressures alone may not cause ulceration.Diabet Med 6:426-428,1989

103. Delbridge L, Ellis CS, Robertson K,Lequesne LP: Non-enzymatic glycosyla-tion of keratin from the stratum corneumof the diabetic foot. Br J Dermatol112:547-554, 1985

104. Gibbs RC, Boxer MC: Abnormal biome-chanics of feet and their cause of hyper-keratoses. ] Am Acad Dermatol 6:1061-1069,1982

105. Murray HJ, Young MJ, Hollis S, BoultonAJM: The association between callus for-mation, high pressures and neuropathy indiabetic foot ulceration. Diabet Med

13:979-982, 1996106. Young MJ, Cavanagh PR, Thomas G, John-

son MM, Murray H, Boulton AJM: Theeffect of callus removal on dynamic plan-tar foot pressures in diabetic patients. Dia-betes Med Suppl 9:55-57,1992

107. Rosen RC, Davids MS, Bohanske LM,Lemont H: Hemorrhage into plantar callusand diabetes mellitus. Cutis 35:339-341,1985

108. Brand PW: Repetitive stress in the devel-opment of diabetic foot ulcers. In The Dia-betic Foot. 4th ed. Levin ME, Davidson JK,Eds. St. Louis, MO, Mosby 1988, p. 83-90

109. Benbow SJ, Chan AW, Bowsher DR,Williams G, Macfarlane IA: The predictionof diabetic neuropathic plantar foot ulcer-ation by liquid-crystal contact thermogra-phy. Diabetes Care 17:835-839, 1994

110. Archer AG, Roberts VC, Watkins PJ: Bloodflow patterns in painful diabetic neuropa-thy. Diabetologia 27:563-567, 1984

111. Clark RP, Goff MR, Hughes J, KlenermanL: Thermography and pedobarography inthe assessment of tissue damage in neuro-pathic and atherosclerotic feet. Thermology3:15-20, 1989

112. Chan AW, MacFarlane IA, Bowsher DR:Contact thermography of painful diabeticneuropathic foot. Diabetes Care14:918-922, 1991

113. Armstrong DG, Lavery LA, Liswood PJ,Todd WF, Tredwell JA: Infrared dermalthermometry for the high-risk diabeticfoot. Phys Ther 77:169-177, 1997

114. Birke JA, Sims DS: The insensitive foot. InPhysical Therapy of the Foot and Ankle. HuntGC, Ed. New York, Churchill Livingstone,1988,p. 133-168

115. Kannel WB, McGee DL: Diabetes and car-diovascular disease: the FraminghamStudy. J Am Med Assoc 241:2035-2038,1979

116. Melton LJ III, Macken KM, Palumbo PJ,Elveback LR: Incidence and prevalence ofclinical peripheral vascular disease in apopulation-based cohort of diabeticpatients. Diabetes Care 3:648-654, 1980

117. Kreines K, Johnson E, Albrink M, Knat-terud GL, Levin ME, Lewitan A, NewberryW, Rose FA: The course of peripheral vas-cular disease in non-insulin-dependentdiabetes. Diabetes Care 8:235-243, 1985

118. Diabetes Control and Complications TrialResearch Group: The effect of intensivetreatment of diabetes on the developmentand progression of long-term complica-tions in insulin-dependent diabetes melli-tus. NEnglJ Med 329:977-986, 1993

119. Strandness DE Jr, Priest RE, Gibbons GE:Combined clinical and pathologic study ofdiabetic and nondiabetic peripheral arter-ial disease. Diabetes 13:366-372, 1964

120. Conrad MC: Large and small artery occlu-sion in diabetics and nondiabetics withsevere vascular disease. Circulation

36:83-91,1967121. Pecoraro RE, AhroniJH, Boyko EJ, Stensel

VL: Chronology and determinants of tis-sue repair in diabetic lower-extremityulcers. Diabetes Care 40:1305-1313,1991

122. Holstein P: The distal blood pressure pre-dicts healing of amputations on the feet.Acta Orthop Scand 55:227-233, 1984

123. Hiatt WR, Marshall JA, Baxter J, SandovalR, Hildebrandt W, Kahn LR, Hamman RI::Diagnostic methods for peripheral arterialdisease in the San Luis Valley DiabetesStudy. J Clin Epidemiol 43:597-606, 1990

124. Carter SA: Clinical measurement of sys-tolic pressures in limbs with arterial exclu-sive disease. JAMA 207:1869-1874,1969

125. European Society of Vascular Surgery:Chronic Critical Leg Ischaemia. EurJ VaseSurg 6:1-3, 1992

126. Fowkes F: The measurement of athero-sclerotic peripheral arterial disease andepidemiologic surveys. Int J Epidemiol17:248-254, 1988

127. Rose GA, Blackburn H, Gillum RE, PrineasRJ: Cardiovascular Survey Methods. Geneva,Switzerland, World Health Org., 1982(WHO Monograph Series, no. 56)

128. Criqui MH, Fronek A, Klauber MR, Bar-rett-Connor E, Gabriel S: The sensitivity,specificity, and predictive value of tradi-tional clinical evaluation of peripheral arte-rial disease: results from noninvasivetesting in a defined population. Circulation71:516-522, 1985

129. Marinelli MR, Beach KW, Glass MJ, Pri-mozichJE Strandness DEJr: Noninvasivetesting versus clinical evaluation of arterialdisease. J Am Med Assoc 241:2031-2034,1979

130. Barnhorst DA, Barner HB: Prevalence ofcongenitally absent pedal pulses. N EnglJMed 278:264-265,1968

131. Silverman JJ: Incidence of palpable dor-salis pedis and posterior tibial pulsationsin soldiers. Am Heart] 32:82-87, 1946

132. LudbrookJ, Clarke AM, McKenzieJK: Sig-nificance of absent ankle pulse. Br MedJ(No. 5294): 1724-1726, 1962

133. Meade TW, Gardner MJ, Cannon P,Richardson PC: Observer variability inrecording the peripheral pulses. Br Heart]30:661-665, 1968

134. Lawson IR, Ingman SR, Masih Y, FreemanB: Reliability of palpation of pedal pulsesas ascertained by the kappa statistic. J AmGeriatrics Soc 28:300-303,1980

135. Brearley S, Shearman CP, Simms MH:Peripheral pulse palpation: an unreliablephysical sign. Ann R Coll Surg fim>l74:169-171, 1992

136. Christensen JH, Freundilish M, JacobsenBA, Falstie-Jensen N: Clinical relevance ofpedal pulse palpation in patients sus-pected of peripheral arterial insufficiency. JIntern Med 226:95-99, 1989

137. Osmundson PJ, O'Fallon WM, Clements


Technical Review

IP, Kazmier FJ, Zimmerman Br, Palumbo:Reproducibility of noninvasive tests ofperipheral occlusive arterial disease. ] VaseSurg 2:678-683,1985

138. Lazarus HM, Albo D Jr, Welling D, HuttoW: Doppler ankle pressures and stiff arter-ies. In Noninvasive Cardiovascular DiagnosisCurrent Concepts. Diethrich EB, Ed. Balti-more, University Park Press, 1978, p.127-135

139. Everhart JE, Pettitt DJ, Knowler WC, RoseFA, Bennett PH: Medial arterial calcifica-tion and its association with mortality andcomplications of diabetes. Diabetologia 31,1988

140. Young MJ, Adams JE, Boulton AJM,Cavanagh PR: Medial arterial calcificationin the feet of diabetic patients and matchednon-diabetic control subjects. Diabetologia36:615-621, 1993

141. Goebel FD, Fuessl HS: Monckeberg's scle-rosis after sympathetic denervation in dia-betic and non-diabetic subjects.Diabetologia 24:347-350, 1983

142. Emanuele MA, Buchanan BJ, Abraira C:Elevated leg systolic pressures and arterialcalcification in diabetic occlusive vasculardisease. Diabetes Care 4:289-292, 1981

143. Duprez D, Missault LWA, Clement DL:Comparison between ankle and toe indexin patient with peripehral arterial disease.MAngiol 6:295-297, 1987

144. Bone GE, Pomajzl MJ: Toe blood pressureby photoplethysmography: an index ofhealing in forefoot amputation. Surgery89:569-574, 1981

145. Carter SA, Strandness DE Jr: Measurementof distal arterial blood pressure: a simpleand valuable guide to the diagnosis andtherapy of arterial occlusive disease. ScandJ CM Lab Invest 128:357-361,1973

146. Uccioli L, Monticone G, Durola L, Russo E.Mormile F, Mennuni G, Menzinger G:Autonomic neuropathy influences greattoe blood pressure. Diabetes Care17:284-287, 1994

147. Orchard TJ, Strandness DE Jr: Assessmentof peripheral vascular disease in diabetes:report and recommendations of an inter-national workshop sponsored by theAmerican Diabetes Association and theAmerican Heart Association September18-20, 1992 New Orleans, Circulation88:819-28, 1993

148. Moneta GL, Strandness GL: Peripheralarterial duplex scanning. ] Card Ultrasound115:645-651,1987

149. Chantelau E, Haage P: An audit of cush-ioned diabetic footwear: relation to patientcompliance. Diabet Med 11:114-116,1994

150. Edmonds ME, Blundell MP, Morris ME,Thomas EM, Cotton LT, Watkins PJ:Improved survival of the diabetic foot: therole of a specialized foot clinic. Q J Med60:763-771, 1986

151. Sims DS Jr, Cavanagh PR, Ulbrecht JS:Risk factors in the diabetic foot. ] Am PhysTherAssoc 68:1887-1902, 1988

152. Duffy JC, Patout CA Jr: Management of theinsensitive foot in diabetes: lessons learnedfrom Hansens disease. Mil Med 155:575-579,1990

153. Brill LR, Cavanaugh PR, Doucette, Mar-garet M, Ulbrecht JS: Prevention, Protectionand Recurrence Reduction of Diabetic Neu-ropathyic Foot Ulcers. East Setauket, NY,Curative Technologies, 1994

154. Mazze R, Deeb L, Palumbo PJ: AlteringPhysicians' Practice Patterns: A Nation-wide Educational Experiment: evaluationof the Clinical Education Program of theAmerican Diabetes Association. DiabetesCare 9:420-425, 1986

155. Deeb LC, Pettijohn FP, ShirahJK, FreemanG: Interventions among primary-carepractitioners to improve care for pre-ventable complications of diabetes. Dia-betes Care 11:275-280, 1988

156. Payne TH, Gabella BA, Michael SL, YoungWF, Pickard J, Hofeldt FD, Fan F,StombergJS, Hamman RF: Preventive carein diabetes mellitus: current practice inurban health-care system. Diabetes Care12:745-747,1989

157. Mayfield JA, Rith-Najarian SJ, Acton KJ,Schraer CD, Stahn RM, Johnson MH,Gohdes D: Assessment of diabetes care bymedical record review. Diabetes Care17:918-923, 1994

158. Cohen SJ: Potential barriers to diabetescare. Diabetes Care 6:499-500, 1983

159. del Aguila MA, Reiber GE, Koepsell TD:How does provider and patient awarenessof high-risk status for lower-extremityamputation influence foot-care practice?Diabetes Care 17:1050-1054, 1994

160. Tillo TH, Giurini JM, Habershaw GM,Chrzan JS, Rowbotham JL: Review ofmetatarsal osteotomies for the treatment ofneuropathic ulcerations. ] Am Podiatr MedAssoc 80:211-217, 1990

161. Patel VG, Wieman TJ: Effect of metatarsalhead resection for diabetic foot ulcers onthe dynamic plantar pressure distribution.Am J Surg 167:297-300, 1994

162. Brand FN, Abbott RD, Kannel WB: Dia-betes, intermittent claudication, and riskof cardiovascular events. Diabetes 38:504-509, 1989

163. Widmer LK, Biland L: Incidence andcourse of occlusive peripheral artery dis-ease in geriatric patients. Int Angiol4:289-294,1985

164. Creasy TS, McMillan PJ, Fletcher EWL,Collin J, Morris PJ: Is percutaneous trans-luminal angioplasty better than exercisefor claudication? Preliminary results froma prospective randomized trial. Eur] VaseSurg 4:135-140, 1990

165. Williams LR, Ekers MA, Collins PS, Lee JF:Vascular rehabilitation: benefits of a struc-

tured exercise/risk modification program.J Vase Surg 14:320-326,1991

166. Currie ID, Wilson YG, Baird RN, LamontPM: Treatment of intermittent claudica-tion: the impact on quality of life. Eur JVase Surg 10:356-361, 1995

167. Conte MS, Belkin M, Donaldson MC,Baum P, Mannick JA, Whittemore AD:Femortibial bypass for claudication: doresults justify an aggressive approach? JVase Surg 21:873-881, 1995

168. Wolf GL, Wilson SE, Cross AP, DeupreeRH, Stason WB: Surgery and balloonangioplasty for peripheral vascular disease:a randomized clinical trial. J Vase Interven-tional Radiol 4:63-648, 1993

169. Luther M: The influence of arterial recon-structive surgery on the outcome of criticalleg. Eur J Vase Surg 8:682-689, 1994

170. LoGerfo FW, Gibbons GW, Pomposelli FBJr, Campbell DR, Miller A, Freeman DV,Quist WC: Trends in the care of the dia-betic foot: expanded role of arterial recon-struction. Arch Surg 127:617-621, 1992

171. Tunis SR, Bass EB, Steinberg EP: The use ofangioplasty, bypass surgery, and amputa-tion in the management of peripheral vas-cular disease. N EnglJ Med 325:556-562,1991

172. Plummer ES, Albert SG: Foot care assess-ment in patients with diabetes: a screeningalgorithm for patient education and refer-ral. Diabetes Educ 21:47-51, 1995

173. De Backer M, De Vroey C, Lesaffre E,Scheys I, De Keyser P: Twelve weeks ofcontinuous oral therapy for toenail ony-chomycosis caused by dermatophytes: adouble-blind comparative trial ofterbinafine 250 mg/day versus itracona-zole 200 mg/day. J Am Acad Dermatol38:S57-S63,1998

174. GoodfieldM: Short-duration therapy withterbinafine for dermatophyte onychomy-cosis: a multicenter trial. Br J Dermatol126 (Suppl. 39):33-35,1992

175. De Doncker P, Gupta AK, Maynissen G,Stoffels P, Heremans A: Itraconazole pulsetherapy for onychomycosis and dermato-mycoses: an overview. J Am Acad Dermatol37:968-974, 1997

176. Sarnow MR, Veves A, Giurini JM, Rosen-blum BI, Chrzan JS, Habershaw GM: In-shoe foot pressure measurements indiabetic patients with at-risk feet and inhealthy subjects. Diabetes Care17:1002-1006,1994

177. Rose NE, Feiwell LA, Cracchiolo A, III: Amethod for measuring foot pressures usinga high resolution, computerized insolesensor: the effect of heel wedges on plan-tar pressure distribution and center offorce. Foot Ankle 13:263-270, 1992

178. Perry JE, Ulbrecht JS, Derr JA, CavanaghPR: The use of running shoes to reduceplantar pressures in patients who have dia-betes. J Bone Jt Surg 77-A:1819-1828,1995


Technical Review

179. Nawoczenski DA, BirkeJA, Coleman WC:Effect of rocker sole design on plantar fore-foot pressures. J Am Podiatr Med Assoc78:455-460, 1988

180. Soulier SM: The use of running shoes inthe prevention of plantar diabetic ulcers. JAm Podiatr Assoc 76:395-400, 1986

181. Boulton A, Franks C, Betts R, DuckworthT, Ward J: Reduction of abnormal footpressures in diabetic neuropathy using anew polymer insole material. Diabetes Care7:42-46, 1984

182. Lord M, Hosein R: Pressure redistributionby molded inserts in diabetic footwear: apilot study. J Rehabil Res Dev 31:214-221,1994

183. Mueller MJ, Strube MJ, Allen BT: Thera-peutic footwear can reduce plantar pres-sures in patients with diabetes andtransmetatarsal amputation. Diabetes Care20:637-641, 1997

184. Brodsky JW, Kourosh S, Stills M, MooneyV: Objective evaluation of insert materialsfor diabetic and athletic footwear. FootAnkle 9:111-116,1988

185. Veves A, Masson EA, Fernando DJS, Boul-ton AJM: Use of experimental paddedhosiery to reduce abnormal foot pressuresin diabetic neuropathy. Diabetes Care12:653-655, 1989

186. Murray HJ, Veves A, Young MJ, Richie

DH, Boulton AJM: Role of experimentalsocks in the care of the high-risk diabeticfoot, a multicenter patient evaluationstudy: American Group for the Study ofExperimental Hosiery in the Diabetic Foot.Diabetes Care 16:1190-1192, 1993

187. Tovey FI: The manufacture of diabeticfootwear. DiabetMed 1:69-71, 1984

188. Uccioli L, Faglia E, Monticone G, FavalesF, Durola L, Aldeghi A, Quarantiello A,Calia P, Menzinger G: Manufactured shoesin the prevention of diabetic foot ulcers.Diabetes Care 18:1376-1378,1995

189. Breuer U: Diabetic patients' compliancewith bespoke footwear after healing ofneuropathic foot ulcers. Diabete Metab20:415-419, 1994

190. Bloomgarden ZT, Karmally W, Metzger MJ,Brothers M, Nechemias C, Bookman J,Faierman D, Ginsberg-Fellner F, Rayfield E,Brown WV: Randomized, controlled trialof diabetic patient education: improvedknowledge without improved metabolicstatus. Diabetes Care 10:263-272,1987

191. Barth R, Campbell LV, Allen S, Jupp JJ,Chisholm DJ: Intensive educationimproves knowledge, compliance, andfoot problems in type 2 diabetes. DiabetMed 8:111-117, 1991

192. Kruger S, Guthrie D: Foot care: knowledgeretention and self-care practices. Diabetes

Educ 18:487-490,1992193. Pieber TR, Holler A, Siebenhofer A, Brun-

ner GA, Semlitsch B, Schattenberg S: Eval-uation of a structured teaching andtreatment programme for type 2 diabetesin general practice in a rural area of Aus-tria. DiabetMed 12:349-354, 1995

194. Malone JM, Snyder M, Anderson G, Bem-hard VM, Hollo way GA, Bunt TJ: Preven-tion of amputation by diabetic education.AmJSurg 158:520-524, 1989

195. Crausaz FM, Clavel S, Liniger C, AlbeanuA, Assal J-P: Additional factors associatedwith plantar ulcers in diabetic neuropathy.DiabetMed 5:771-775, 1988

196. Thomson FJ, Masson EA: Can elderly dia-betic patients co-operate with routine footcare? Age Ageing 21:333-337, 1992

197. Raviola CA, Nichter LS, Baker JD, BusuttilRW: Cost of treating advanced legischemia-bypass graft vs primary amputa-tion. Arch Surg 123:495-496, 1988

198. Gibbons GW, Marcaccio EJ Jr, Burgess AM,Pomposellil FB Jr, Freeman DV, CampbellDR, Miller A, LoGerfo FW: Improvedquality of diabetic foot care, 1984 vs 1990.Arch Surg 128:576-581, 1993

199. Wooldridge J, Moreno L: Evaluation of thecosts to Medicare of covering therapeuticshoes for diabetic patients. Diabetes Care17:541-547, 1994


Documents

Preventive Foot Care in People With Diabetes foot care … · of Family Medicine, 1110 W Michigan St., Long Hospital, Room 200, Indiana University, Indianapolis, IN 46202. E-mail: