RELIABILITY OF FREE-FLAP COVERAGE IN DIABETICFOOT ULCERS

OMER OZKAN, M.D.,* O. KORAY COSKUNFIRAT, M.D., and H. EGE OZGENTAS, M.D.

As microsurgery advances, microsurgical free-tissue transfers have become the reconstructive method of choice over staged or primaryamputation, and enabling independent ambulation in difficult lower-extremity wounds. In this report, we present our experiences with free-tissue transfer for the reconstruction of soft-tissue defects in 13 diabetic foot ulcers. Following radical debridement, soft-tissue reconstructionwas achieved in the following ways: anterolateral thigh fasciocutaneous flap in 5 patients, radial forearm fasciocutaneous flap in 3 patients,lateral arm fasciocutaneous flap in 1 patient, gracilis musculocutaneous flap in 1 patient, tensor fascia latae flap in 1 patient, deep inferiorepigastric perforator flap in 1 patient, and a parascapular flap in the remaining patient. In 8 cases, diabetic wounds were in the foot, whilewounds were at the level of the lower leg in the remaining patients. In all patients, vascular patency was confirmed by the Doppler technique.In suspicious cases, arteriography was then performed. While all flaps survived well in the postoperative period, one patient died fromcardiopulmonary problems on postoperative day 16 in an intensive care unit. Amputation was necessary in the early postoperative periodbecause of healing problems. In the remaining 10 cases, all flaps survived intact. In one case, arterial revision was performed successfully.The ultimate limb salvage rate was 83% for the 12 patients. Independent ambulation was achieved in these cases. During the follow-up periodof 8 months to 2 years, no ulcer recurrence was noted, and no revascularization or vascular bypass surgery was needed before or after thefree-tissue transfers. The authors conclude that free-tissue transfer for diabetic foot ulcers is a reliable procedure, despite pessimisticopinions regarding the flap survival and low limb salvage rates. It should be considered a useful reconstructive option for serious defects inwell-selected cases. ª 2005 Wiley-Liss, Inc. Microsurgery 25:107�112, 2005.

Improvements in the diagnosis and treatment of dia-betes mellitus and its complications such as retinopathy,nephropathy, neuropathy, and foot ulcers have allowedpatients to live much longer and expect a better qualityof life. Diabetic gangrene is one of the most devastatingcomplications of this serious disease. It is well-estab-lished that some 20% of all diabetic patients who enterhospital are admitted for foot problems.1 Although theetiology of a diabetic wound is generally attributed todecreased sensation and occasionally to reduced bloodsupply,2�4 the pathogenetic mechanisms of neuropa-thies, ischemia, and microangiopathies may contributein varying degrees to the adverse healing that results inserious and complex tissue defects with exposed bonesand tendons. In diabetic patients, wounds are suscepti-ble to serious and recalcitrant infections, and in manycases, this problem may become of major importance.5,6

Local wound care, including serial and sometimes vig-orous debridement of infected and nonviable tissueswith an appropriate antibiotic regimen, may be needed.Under favorable circumstances, conservative treatmentmay overcome the wound problem. Nevertheless,should there be insufficient or inappropriate local tissueswith exposed vital tissues, then even skin grafting and

local flaps may not be able to solve the problem. Incertain cases, the transfer of well-vascularized distanttissues can improve wound healing and lead to inde-pendent ambulation without the need to resort to aprosthesis or walking devices.

In this report, we present our experiences with 13such patients treated with free-tissue transfer for limbsalvage who could not be treated by simple wound-closure techniques. The difficult nature of diabeticwounds and the reliability and advantages of distanttissue transfers in these challenging situations are dis-cussed.

MATERIALS AND METHODS

From March 2002�September 2003, 13 various freeflaps were used to reconstruct soft-tissue defects of thelower extremities that existed as a complication of dia-betes mellitus. There were 12 males and 1 female, andtheir ages were between 39�78 years. All patients weretreated conservatively with minor interventions such aslocal debridement and skin-grafting procedures beforethe free flap transfer. The sources of free flaps were asfollows: in 5 patients, an anterolateral thigh fasciocu-taneous flap; in 3 patients, a radial forearm fasciocuta-neous flap; in one patient, a lateral arm fasciocutaneousflap; in one patient, a gracilis musculocutaneous flap; inone patient, a tensor fascia latae flap; in one patient, adeep inferior epigastric perforator flap; and in theremaining patient, a parascapular flap. While in 8 casesthe diabetic wounds were in the foot, in the others, thewounds were at the level of the lower leg. In all patients,the vascular structure and its suitability for microvas-

Department of Plastic and Reconstructive Surgery, Akdeniz University Schoolof Medicine, Antalya, Turkey

Grant sponsor: Akdeniz University Scientific Research Projects Unit.

*Correspondence to: Dr. Omer Ozkan, Plastik ve Rekonstruktif CerrahiAnabilim Dali, Akdeniz Universitesi Hastanesi, B Blok kat 5, Antalya, Turkey07059. E-mail: omozkan@hotmail.com

Received 18 May 2004; Accepted 20 September 2004

Published online 4 January 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/micr.20094

ª 2005 Wiley-Liss, Inc.

cular anastomosis were confirmed by the Dopplertechnique. In still-suspicious cases, conventional arterialangiography was performed. Before free-tissue transfer,necrotic and infected tissues were excised and removed.In all cases plain radiograms were obtained to examinethe bony structure, In selected cases, scintigraphy wasperformed to diagnose the existence of osteomyelitis. Anappropriate antibiotic treatment was administered ininfected cases and continued postoperatively in accor-dance with advice from our Department of InfectiousDiseases. The major medical problems that occurred inour series included chronic obstructive pulmonary dis-ease in 4 patients, coronary artery disease in 3 patients,and chronic renal failure in 1 patient.

PATIENT REPORTS

Patient 1

A 61-year-old male patient presented with a necroticwound in the medial aspect of the plantar surface of the

right foot (Fig. 1A). He had been operated on and skingrafted twice: 2 months and 1 month prior. The necrotictissue was radically excised. The medial plantar vesselswere exposed and prepared as recipient vessels. A radialforearm fasciocutaneous flap (5 · 8 cm) was used tocover the defect. The donor area was covered with asplit-thickness skin graft. The patient’s recovery wasuneventful, and he was discharged 15 days postopera-tively (Fig. 1B).

Patient 2

A 48-year-old male patient had had a compositetissue defect of the lateral aspect of the distal lowerextremity for 3 months (Fig. 2A). After surgicaldebridement, the bony structure and tendons were ex-posed. A parascapular flap (8 · 21 cm) was used tocover the defect. The pedicle of the flap was anasto-mosed to the posterior tibial artery and vein. The donor

Figure. 1. A: Necrotic wound in medial aspect of plantar surface of

right foot. B: Postoperative view at 15 months, after reconstruction of

defect with radial forearm free flap.

Figure. 2. A: Preoperative view of composite tissue defect of lateral

aspect of distal lower extremity. B: Reconstruction with parascapular

flap. Postoperative view at 3 months.

108 Ozkan et al.

area was closed primarily. The patient healed unevent-fully, and the flap survived (Fig. 2B).

Patient 3

A 43-year-old male patient presented with a necroticcomposite tissue defect with purulent drainage (Fig.3A). Direct radiogram and scintigraphic imaging of thefoot revealed an osteomyelitic image (Fig. 3B). Appro-priate antibiotic treatment was administered, and radi-cal debridement was performed. The medial plantarartery was prepared as the recipient artery. An antero-lateral thigh flap (10 · 18 cm) was used to cover thedefect. The donor area was closed primarily. Postoper-ative recovery was uneventful, and the flap survived(Fig. 3C).

Patient 4

A 59-year-old male patient presented with a necrosisformation over the anterior surface of the lower leg anddorsum of the foot (Fig. 4A). After radical debridementwas performed (Fig. 4B), the defect was covered with an

anterolateral thigh flap (12 · 21 cm). The pedicle of theflap was anastomosed to the anterior tibial vessels. Thedonor area was covered with a split-thickness skin graft.The patient’s recovery was uneventful, and he was dis-charged 13 days postoperatively (Fig. 4C).

RESULTS

No any major anesthetic complications occurred inany of the operations. While all flaps survived wellpostoperatively, one patient, reconstructed with an an-terolateral thigh flap, died from cardiopulmonaryproblems on day 16 postoperatively in an intensive careunit. Although the viability of flaps was unquestionable,a below-the-knee amputation in one patient with agracilis musculocutaneous flap was performed on post-operative day 15, and an above-the-knee amputationwas performed on another patient with a tensor fascialatae flap on day 17 postoperatively because ofrecalcitrant osteomyelitis and problematic healingbetween the flap the and wound base, as well as ischemicproblems around the original wound. In the remaining

Figure. 3. A: A 43-year-old male with necrotic composite tissue

defect complicated by purulent drainage. B: Scintigraphic confir-

mation of ostemyelitis at level of distal part of first metatarsal bone

and phalanges (arrow). C: Wound was debrided, and amputation of

first toe was performed. Defect was covered with free anterolateral

thigh flap. Postoperative view at 4 months.

Diabetic Foot Ulcers 109

10 cases, all flaps survived intact. In one case, an arterialanastomosis was revised successfully. The overall limbsalvage rate was about 83% for the 12 patients. In thesepatients, the average hospitalization period wasapproximately 15 days after free-flap procedure. In allcases, vascular anastomoses between the flap and re-cipient vessels were performed directly without usingany vein grafts. Independent ambulation was achievedin these cases. During the follow-up period of 8�26months, no ulcer recurrence was noted. No revascular-ization or vascular bypass surgery has been needed be-fore or after free-tissue transfers to date. All operationswere carried out by the same surgical team, and free-tissue transfer was performed after debridement of thewound and preparation of the recipient vessels. Nohematoma or seroma was observed, and all donor siteshealed uneventfully.

DISCUSSION

The basic etiology of diabetic wounds is attributed toneuropathy, which has anatomic, metabolic, and ische-

mic components.2,3,7 Chronic high blood sugar levelsalter the intracellular myoinositol sorbitol pathways inthe neurons, resulting in swelling and finally the dys-function of nerves.8 This nerve dysfunction causes theloss of protective sensation in the feet against externalagents such as pressure or trauma.

Vascular structures are also disturbed in diabeticpatients, leading to gangrenous ulcers.4 In some cases,vascular occlusions may require surgical bypass torevascularize or augment the vascularization of distalsegments.

In diabetic patients, the depressed immune systemexposes wounds to resistant and recalcitrant infections,especially in bony structures.5,6 Osteomyelitis in thesepatients requires more aggressive treatment than innondiabetic patients. In addition, whatever the etio-pathogenesis, adequate therapy requires completeresection of necrotic and infected tissues, decompres-sion of all suspected compartments, and filling in of allpotential dead spaces with appropriate tissues as wellas the simultaneous administration of appropriateantibiotic therapy. Based on this important concept,

Figure. 4. A: Necrotic ulcer over anterior surface of lower leg and

dorsum of foot in a 59-year-old male. B: Intraoperative view after

radical debridement. C: Postoperative view at 14 months, after

reconstruction with an anterolateral thigh flap.

110 Ozkan et al.

the transfer of well-vascularized tissue may be ofconsiderable benefit in overcoming wound-healingproblems that result from poorly vascularized and in-fected neighboring tissues. Furthermore, after ade-quate debridement of such complicated wounds, theexposure of bones and tendons will be inevitable inmost cases.

Although local muscle and other types of flapsshould remain as reconstructive options when over-coming small foot and lower leg wounds that have ex-posed bones, joints, and tendons in diabetic patients,local tissues are insufficient as a source of well-vascu-larized, good-quality material in most cases. In certainwell-selected cases, the transfer of well-vascularized tis-sues using microsurgical techniques can improve woundhealing and may decrease hospitalization time, and mayincrease a patient’s chance of achieving independentambulation.

The advent of free-tissue transfer has providedmultiple options for difficult defects of the lowerextremity, to preserve and maintain the functional andaesthetic status of the limb.9�12 In the literature, thereare several reports on diabetic wounds that discuss theeffect of free flaps on salvaging the extremity.13�18 Mostof these reports suggest that microsurgically transferredtissues enhance the revascularization of the criticallyischemic extremity through the development of vascularconnections at the free-flap-surrounding environmentaltissue interface. Using this advanced surgical technique,large diabetic wound ulcers that are normally unre-sponsive to conventional therapies and often lead toamputation can be treated successfully.

In the literature, various authors used free-tissuetransfer both experimentally and clinically.15�20 Mostseries reported successful results and acceptable ambu-lation rates with free-tissue transfer, even after vascularbypass surgery.

Revascularization of the surrounding environmentby well-vascularized tissue transfer is one of the mostfrequently mentioned issues in diabetic wounds.14�21

Van Landuyt et al. studied this subject in three clinicalcases.21 They demonstrated vascular connections at thefree-flap-surrounding interface in an ischemic diabeticlimb by angiography. Vermassen and van Landuyt de-scribed their experience of combined vascular recon-struction and free-flap transfer in 45 patients.14 Theircombined survival rate was 65% at the end of 3 years.They concluded that their techniques provided advan-tages of immediate soft-tissue coverage limiting ampu-tation level and healing time and early independentambulation. They also mentioned the enhancement ofneovascularization of the free-flap-surrounding tissueinterface, and the limiting effect of well-vascularizedtissue transfer on the infection rate.

When dealing with a diabetic wound of even smalldimensions, one should keep in mind that the evidentulcer may not express the actual size of the problem.Chronic osteomyelitis in these patients is frequentlyassociated with skin ulceration with continuation of thesinus tracts. Wound care should be focused on the in-fected and necrotic structures by exposing all involvedsinus tracts and compartments. After the exposure ofthese tracts and drainage of the purulent material,wound care will result in favorable outcomes, and thesuccess rate of reconstruction will be improved.

In the literature in selected cases, a temporary arte-riovenous fistula was performed with a long vein graftloop before microsurgical free tissue transfer.22�24 Thenthe flap vessels were anastomosed to the arterial andvenous limbs of the divided loop in an end-to-endmanner. In our series, in all cases, vascular anastomosesbetween the flap and the recipient vessels were per-formed directly, without using vein grafts. Although thisarteriovenous preliminary fistula may be safer than thesimultaneous use of a long interpositional vein graft, inour opinion, providing direct anastomosis (if possible),such as choosing long pedicled flaps, is the safest andmost reliable procedure. This long pedicle will providethe advantages of tension-free anastomoses away fromthe infected wound, without increasing the number ofanastomoses along the course of the pedicle (Fig. 5).

Although several types of free flaps have been usedfor diabetic foot wounds, free muscle tissues remain themost commonly used flaps in the current litera-ture.13,14,16 Except when needing additional mass to filla large defect, a thin fasciocutaneous flap may be anappropriate free-flap choice to minimize bulk. In ourseries we mostly used fasciocutaneous flaps to mini-malize bulkiness and to provide an aesthetically moreacceptable limb. In our opinion, fasciocutaneous flaps

Figure. 5. Long pedicle of flap provides tension-free anastomoses

away from possibly infected diabetic wound.

Diabetic Foot Ulcers 111

also provide the ability to use the surface area of theflap as a monitor to check the vitality of the flap in theearly period of transfer. This may not be possible whenusing muscle flaps, because in these flaps the vascu-larity of the tissue may not be objectively observed inits entirety.

In conclusion, the present series demonstrates thatfree-tissue transfer can be safely performed in properlyselected patients. All patients were selected as candidatesfor free-tissue transfer after a complete evaluation ofpossible medical problems and suitable recipient vesselconfirmation by Doppler ultrasonography, and in somecases by conventional lower-extremity arterial angiog-raphy. Free-tissue transfer is a possible alternative tostaged or primary amputation, especially in extensivesoft-tissue loss with vital-tissue exposures when localtissues are insufficient or are of poor quality. It shouldbe remembered that the key point for successful free-tissue transfer and an improved limb salvage rate is ameticulous for evaluation of the patients and theirwounds, to ascertain whether they are suitable for such aprocedure. In this manner, unsuccessful interventionsand low limb salvage rates will be avoided.

ACKNOWLEDGMENT

We thank Akdeniz University Scientific ResearchProjects Unit for their support.

REFERENCES

1. Colen LB. The diabetic foot. In: Cohen M, editor. Mastery ofplastic and reconstructive surgery. Boston: Little, Brown and Co;1994.

2. Dolman CL. The morbid anatomy of diabetic neuropathy. Neu-rology 1963; 13:135�142.

3. Eames RA, Lange LS. Clinical and pathological study of ischaemicneuropathy. J Neurol Neurosurg Psychiatry 1967; 30:215�226.

4. LoGerfo FW, Coffman JD. Current concepts. Vascular andmicrovascular disease of the foot in diabetes. Implications for footcare. N Engl J Med 1984;311:1615�1619.

5. Caputo GM, Joshi N, Weitekamp MR. Foot infections in patientswith diabetes. Am Fam Physician 1997;56:195�202.

6. Balsells M, Viade J, Millan M, Garcia JR, Garcia-Pascual L, delPozo C, Anglada J. Prevalence of osteomyelitis in non-healing

diabetic foot ulcers: usefulness of radiologic and scintigraphicfindings. Diabetes Res Clin Pract 1997;38:123�127.

7. Clements RS. Diabetic neuropathy: new concepts of its etiology.Diabetes 1979:28:604�611.

8. Gabbay KH. The sorbitol pathway and the complications of dia-betes. N Engl J Med 1973;288:831�836.

9. Gonzalez MH, Tarandy Dl, Troy D, Phillips D, Weinzweig N.Free tissue coverage of chronic traumatic wounds of the lower leg.Plast Reconstr Surg 2002;109:592�600.

10. Moran SL, Illig KA, Green RM, Serletti JM. Free-tissue transferin patients with peripheral vascular disease: a 10 year experience.Plast Reconstr Surg 2002;109:999�1006.

11. Kaplan I, Ada S, Ozerkan F, Bora A, Ademoglu Y. Reconstruc-tion of soft tissue and bone defects in lower extremity with freeflaps. Microsurgery 1998;18:176�181.

12. Haas F, Seibert FJ, Koch H, Hubmer M, Moshammer HE, PiererG, Scharnagl E. Reconstruction of combined defects Achillestendon and the overlying soft tissue with a fascia lata graft and afree fasciocutaneous lateral arm flap. Ann Plast Surg 2003;51:376�382.

13. Musharrafieh R, Saghieh S, Macari G, Atiyeh B. Diabetic footsalvage with microsurgical free-tissue transfer. Microsurgery2003;23:257�261.

14. Vermassen FE, van Landuyt K. Combined vascular reconstructionand free flap transfer in diabetic arterial disease. Diabetes MetabRes Rev 2000;16:33�36.

15. Shenaq SM, Dinh TA. Foot salvage in arteriolosclerotic and dia-betic patients by free flaps after vascular bypass: report of twocases. Microsurgery 1989;10:310�314.

16. Lai CS, Lin SD, Yang CC, Chou CK, Wu SF, Chang CH. Limbsalvage of infected diabetic foot ulcers with microsurgical free-muscle transfer. Ann Plast Surg 1991;26:212�220.

17. Karp NS, Kasabian AK, Siebert JW, Eidelman Y, Colen S.Microvascular free-flap salvage of the diabetic foot: a 5-yearexperience. Plast Reconstr Surg 1994;94:834�840.

18. Armstrong MB, Villalobos RE, Leppink DM. Free-tissue transferfor lower-extremity reconstruction in the immunosuppressed dia-betic transplant recipient. J Reconstr Microsurg 1997;13:1�5.

19. Cooley BC, Hanel DP, Anderson RB, Foster MD, Gould JS. Theinfluence of diabetes on free flap transfer: I. Flap survival andmicrovascular healing. Ann Plast Surg 1992:29:58�64.

20. Banis JC Jr, Richardson JD, Derr JW Jr, Acland RD. Microsur-gical adjuncts in salvage of the ischemic and diabetic lowerextremity. Clin Plast Surg 1992;19:881�893.

21. Van Landuyt K, Monstrey S, Blondeel P, Tonnard P, VermassenF. Revascularization by ingrowth of a free flap: fact or fiction?Microsurgery 1996;17:417�422.

22. Atiyeh BS, Sfeir RE, Hussein MM, Husami T. Preliminary arte-riovenous fistula for free-flap reconstruction in the diabetic foot.Plast Reconstr Surg 1995;95:1062�1069.

23. Atiyeh BS, Musharafieh RS. Staged arteriovenous fistula and freeflap transfer. Ann Plast Surg 1997;38:193�194.

24. Atiyeh BS, Khalil IM, Hussein MK, Al Amm CA, MusharafiehRS. Temporary arteriovenous fistula and microsurgical free tissuetransfer for reconstruction of complex defects. Plast Reconstr Surg2001;108:485�488.

112 Ozkan et al.