DIABETES/METABOLISM RESEARCH AND REVIEWS R E V I E W A R T I C L EDiabetes Metab Res Rev 2008; 24(Suppl 1): S81–S83.Published online 20 March 2008 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/dmrr.858

Diabetic foot surgery: classifying patientsto predict complications

Nicholas J. BevilacquaLee C. RogersDavid G. Armstrong*

Center for Lower ExtremityAmbulatory Research (CLEAR), Dr.William M. Scholl College of PodiatricMedicine, Rosalind FranklinUniversity of Medicine and Science,Chicago, IL, USA

*Correspondence to:David G. Armstrong, Center forLower Extremity AmbulatoryResearch (CLEAR), Dr. William M.Scholl College of Podiatric Medicine,Rosalind Franklin University ofMedicine and Science, Chicago, IL,USA.E-mail: David.armstrong@rosalind-franklin.edu

Received: 9 October 2007Revised: 23 January 2008Accepted: 23 January 2008

Summary

The purpose of this article is to describe a classification of diabetic footsurgery performed in the absence of critical limb ischaemia. The basis of thisclassification is centred on three fundamental variables that are present inthe assessment of risk and indication: (1) presence or absence of neuropathy(the loss of protective sensation); (2) presence or absence of an open wound;(3) presence or absence of acute limb-threatening infection. The conceptualframework for this classification is to define distinct classes of surgery inan order of theoretically increasing risk for high-level amputation. Theseinclude: Class I: elective diabetic foot surgery (procedures performed to treata painful deformity in a patient without the loss of protective sensation);Class II: prophylactic (procedure performed to reduce the risk of ulcerationor reulceration in a person with the loss of protective sensation but withoutan open wound); Class III: curative (procedure performed to assist in healingan open wound); and Class IV: emergency (procedure performed to limitthe progression of acute infection). The presence of critical ischaemia inany of these classes of surgery should prompt a vascular evaluation toconsider (1) the urgency of the procedure being considered and (2) possiblerevascularization prior to or temporally concomitant with the procedure. It isour hope that this system begins a dialogue amongst physicians and surgeonswhich can ultimately facilitate communication, enhance perspective, andimprove care. Copyright 2008 John Wiley & Sons, Ltd.

Keywords diabetes; foot; limb salvage; reconstruction; surgery; ulcer

The lifetime risk of a person with diabetes developing a foot ulceration hasbeen estimated at 25% [1]. The causal pathways leading to the developmentof a diabetic foot ulcer consist of the peripheral neuropathy, deformity,and minor repetitive trauma [2]. Increased plantar pressure is a factor, butby itself does not cause ulceration. However, when combined with sensoryneuropathy, the conditions can favour ulcer formation [3]. Masson et al.evaluated two groups of patients with similar peak plantar pressures, thefirst group had rheumatoid arthritis and normal sensation, and the secondgroup had diabetes and sensory neuropathy. There were no ulcerations inthe rheumatoid group whereas 31% of the diabetic neuropathic group hadan ulcer history. Foot deformity must be addressed and pressure reductionmay be accomplished extrinsically through padding and accommodation withshoes, or intrinsically through surgery.

Over the past decade, there has been an increasing interest and numer-ous descriptive studies detailing various surgical techniques in treatmentof the high-risk diabetic foot. While there has been a relative dearth ofstudies evaluating specific procedures, recent studies have suggested somepotential benefit from judicious intervention in this population. In responseto the increasing attention to this area of intervention and the lack of a concise

Copyright 2008 John Wiley & Sons, Ltd.

S82 N. J. Bevilacqua et al.

nomenclature of indications, a diabetic foot surgeryclassification system was proposed in 2003 [4]. Thebasis of this classification system centres on historicalvariables such as presence or absence of neuropathy,an open wound, or limb threatening infection [4]. Thissystem divides non-vascular diabetic foot surgery into fourclasses: elective, prophylactic, curative, and emergency.

Elective diabetic foot surgery is performed on a patientwith diabetes for the alleviation of pain or restoration offunction [4]. These patients have the protective sensationintact and an absence of critical limb ischaemia. Thesepatients are at no increased risk for ulceration over a non-diabetic patient, and there is a low risk of morbidityassociated with elective foot surgery in persons withdiabetes without complications [5].

Prophylactic surgery is performed on patients withneuropathy and deformity, but without an open wound.The goal of prophylactic surgery is to restore function,reduce plantar pressure, and lower the risk of futureulceration. Foot deformities are common in those withdiabetes and are not always visible to superficialinspection. Motor neuropathy causes intrinsic muscleatrophy and can lead to a deformity causing abnormalloading and increased plantar pressures resulting inulceration. Lavery et al. found that patients withneuropathy and deformity had a greater than 12-timeslikelihood for ulceration when compared with patientswithout neuropathy [6].

Armstrong et al. compared postoperative complica-tions, in persons with and without diabetes, of a singlehammertoe correction. After a follow-up period of 3 years,no complications were noted in the persons with neuropa-thy and diabetes but no history of ulceration, as well as thesubjects without diabetes. However, persons with diabetesand a history of ulceration were more likely to experiencea postoperative infection (14.3%). The long-term out-comes after prophylactic surgery were good, with 96.3%of patients remaining ulcer-free for a mean of 3 years [7].In a more recent manuscript, Armstrong and colleaguesreported a postoperative infection rate of 6.7% followingClass II surgery [5].

At least 40% of patients with diabetic Charcot neuro-osteoarthropathy (CN) will develop an ulcer and theyhave a much increased risk ratio for ulceration [6].Patients presenting with an unstable foot secondary to CNcan achieve bony fusions and limb salvage with surgicalreconstruction [8].

Limited joint mobility is as important a risk factor forulceration as frank deformity [9]. A common exampleis hallux limitus, where the limited metatarsophalangealjoint dorsiflexion increases pressure at the distal halluxduring ambulation, specifically ‘toe-off’. This can lead tothe formation of calluses under the hallux and subsequentulceration. This is a clear indication for prophylacticsurgery to increase motion, reducing distal pressure.

Curative diabetic foot surgery is performed on patientswith open wounds [4]. These procedures are designedto heal the wound and reduce the risk of reulceration.Rosenblum and colleagues [10] reviewed their experience

with a hallux interphalangeal joint arthroplasty in45 feet for chronic neuropathic ulcers of the great toe.Overall, 41 feet (91%) healed and had no evidence ofrecurrence in the follow-up period. The authors reportedan 11% complication and 4% postoperative infection rate.The authors concluded that the hallux interphalangealjoint arthroplasty is a valuable procedure for chroniculcerations of the hallux.

Ankle equinus (less than 10◦ of dorsiflexion at theankle joint) increases plantar forefoot pressures and mayincrease the risk for ulceration [11]. An Achilles tendonlengthening has been shown to reduce plantar forefootpressures [12] and is beneficial to prevent ulceration. Linet al. reported rapid healing of previously recalcitrantplantar wounds with persons undergoing an Achillestendon lengthening procedure. The authors also reporteda significantly lower rate of recurrence [13]. Muelleret al. subsequently reported a similar trend toward lowerulcer recurrence when performing an Achilles tendonlengthening [14].

Although the primary goal of curative surgery isto heal an ulcer, the procedure should also focus onreducing the risk for future ulcerations. This becomeschallenging when treating patients with CN presentingwith a plantar ulceration. At times, a simple ostectomymay heal the wound, but more often these patients requirecomplex reconstruction. Catanzariti et al [15]. performeda retrospective review of patients who underwent anostectomy for chronic or recurrent ulceration in themidfoot secondary to diabetic CN. The authors reviewed20 patients (18 medial ulcers and 9 lateral ulcers) andfound a 74% healing rate with the majority of failedprocedures involving lateral plantar foot wounds (six ofseven). The authors concluded that an ostectomy forulcers involving the lateral plantar foot often fails andrequires complex reconstruction [15].

Emergency surgery is, as the name implies, a surgicalemergency and may have to be performed in thepresence of critical limb ischaemia. Patients requiringrevascularization must be evaluated by a vascularsurgeon immediately postoperatively. Emergency surgeryis performed to limit the progression of an acute limb-or life-threatening infection [4]. These procedures ofteninvolve partial amputations and can be a staged procedureto return to the operating theatre for multiple surgicaldebridements and eventually delayed primary closure.

Armstrong and colleagues validated the aforemen-tioned risk-based classification system, which wasdesigned to assist the surgeon in assessing risk when deter-mining a rational for foot surgery in persons with diabetes[5]. The investigators designed a retrospective cohortmodel and abstracted medical records from 180 patientswith diabetes, 76.1% male, aged 57.8 ± 11.2 years whounderwent a surgical procedure. All surgeries were clas-sified using the previously presented diabetic foot surgeryclassification scheme and patients were equally assignedinto the four classes. Patients with significant vasculardisease were excluded from the study. The investigatorsfound a significant trend toward increasing postoperative

Copyright 2008 John Wiley & Sons, Ltd. Diabetes Metab Res Rev 2008; 24(Suppl 1): S81–S83.DOI: 10.1002/dmrr

Classification of Diabetic Foot Surgery S83

Table 1. Prevalence of complications by foot surgery classi-fication

Class I Class II Class III Class IV

Ulceration/reulceration 0 2.2 11.1 24.4Postoperative Infection 2.2 6.7 20.0 100Amputation 0 2.2 6.7 48.9

complications, including the risk of peri- and postop-erative infection, ulceration, and reulceration, as wellas amputation with increasing class of foot surgery. Asexpected, there was no morbidity reported with electivesurgery, and a very high rate of ulceration, infection, andamputation in emergency surgery [5] (Table 1).

The results of their study suggest that a non-vascular diabetic foot surgery classification system maybe predictive of peri- and postoperative complicationsand may assist the surgeon in assessing risk better whendetermining a rationale for and the type of surgery inpersons with diabetes. This system, however, did notinclude persons with severe peripheral arterial diseaseand should not be generalized to the entire diabeticpopulation. Variables such as presence or absence ofneuropathy, an open wound, and acute infection whencollated, may better help the clinician identify andcommunicate risk to patients and their families and setreasonable goals in the surgical treatment of patients withdiabetic feet [5].

Conflict of interestThe authors have no conflicts of interest.

References

1. Singh N, Armstrong DG, Lipsky BA. Preventing foot ulcers inpatients with diabetes. JAMA 2005; 293: 217–228.

2. Reiber GE, Vileikyte L, Boyko EJ, et al. Causal pathways forincident lower-extremity ulcers in patients with diabetes fromtwo settings. Diabetes Care 1999; 22(1): 157–162.

3. Masson EA, Hay EM, Stockley I, Veves A, Betts RP, Boulton AJ.Abnormal foot pressures alone may not cause ulceration. DiabetMed 1989; 6: 426–428.

4. Armstrong DG, Frykberg RG. Classification of diabetic footsurgery: toward a rational definition. Diabet Med 2003; 20:329–331.

5. Armstrong DG, Lavery LA, Frykberg RG, Wu SC, Boulton AJ.Validation of a diabetic foot surgery classification. Int Wound J2006; 3: 240–246.

6. Lavery LA, Armstrong DG, Vela SA, Quebedeaux TL, Fleis-chli JG. Practical criteria for screening patients at high risk fordiabetic foot ulceration. Arch Intern Med 1998; 158: 158–162.

7. Armstrong DG, Lavery LA, Stern S, Harkless LB. Is prophylacticdiabetic foot surgery dangerous? J Foot Ankle Surg 1996; 35:585–589.

8. Baravarian B, Van Gils CC. Arthrodesis of the charcot foot andankle. Clin Podiatr Med Surg North Am 2004; 21: 271–289.

9. Fernando DJS, Masson EA, Veves A, Boulton AJM. Relationshipof limited joint mobility to abnormal foot pressures and diabeticfoot ulceration. Diabetes Care 1991; 14: 8–11.

10. Rosenblum BI, Giurini JM, Chrzan JS, Habershaw GM. Prevent-ing loss of the great toe with the hallux interphalangeal jointarthroplasty. J Foot Ankle Surg 1994; 33: 557–560.

11. Lavery LA, Armstrong DG, Boulton AJM. Ankle equinusdeformity and its relationship to high plantar pressure in alarge population with diabetes mellitus. J Am Podiatr Med Assoc2002; 92: 479–482.

12. Armstrong DG, Stacpoole-Shea S, Nguyen H, Harkless LB.Lengthening of the Achilles tendon in diabetic patients whoare at high risk for ulceration of the foot. J Bone Joint Surg Am1999; 81: 535–538.

13. Lin SS, Lee TH, Wapner KL. Plantar forefoot ulceration withequinus deformity of the ankle in diabetic patients: theeffect of tendo-achilles lengthening and total contact casting.Orthopaedics 1996; 19: 465–475.

14. Mueller MJ, Sinacore DR, Hastings MK, Strube MJ, Johnson JE.Effect of achilles tendon lengthening on neuropathic plantarulcers. A randomized clinical trial. J Bone Joint Surg 2003; 85A:1436–1445.

15. Catanzariti AR, Mendicino R, Haverstock B. Ostectomy fordiabetic neuroarthropathy involving the midfoot. J Foot AnkleSurg 2000; 39: 291–300.

Copyright 2008 John Wiley & Sons, Ltd. Diabetes Metab Res Rev 2008; 24(Suppl 1): S81–S83.DOI: 10.1002/dmrr