Journal of Plastic, Reconstructive & Aesthetic Surgery (2013) 66, 193e200

Clinical application of the thoracodorsal arteryperforator flaps

Li-chang Yang a, Xian-cheng Wang a,*, Michael L. Bentz b, Bai-rong Fang a,Xiang Xiong a, Xiao-fang Li a, Qin Lu a, Qian Wu a, Rui-feng Wang a, Wei Feng a,Ji-yong He a, Yu-ying Wang a

aDepartment of Plastic Surgery and Aesthetic Surgery and Burns, Second XiangYa Hospital, Xiangya Medical School,Central South University, Changsha, Hunan 410011, ChinabDivision of Plastic and Reconstructive Surgery, University of Wisconsin, Madison, WI 53792-3284, USA

Received 15 February 2012; accepted 10 September 2012

KEYWORDSPerforator flaps;Thoracodorsal artery;Latissimus dorsi

* Corresponding author. Tel.: þ86 73E-mail address: wangxiancheng64@

1748-6815/$-seefrontmatterª2012Brihttp://dx.doi.org/10.1016/j.bjps.2012.0

Summary Background: Thoracodorsal artery perforator (TAP) flaps without latissimus dorsimuscle e have been used for reconstruction of the extremities, head and neck as free styleflaps, and as pedicled flaps for reconstruction of the chest wall and axillary wounds. This retro-spective study aimed to analyse the clinical applications and relevant anatomic findings of TAPflaps.Methods: From April 2007 to August 2011, 67 free or pedicled TAP flaps were transferred in 67patients for reconstruction of wounds of the extremities, chest wall and axilla. Eight were usedas free flaps for reconstruction of extremities, and 59 used as pedicled flaps for reconstructionof axillary or chest wounds. Patient ages ranged from 7 to 55 years (26.04 � 12.83). Perforatorarteries were detected and identified with a hand-held Doppler. The size of flaps ranged from 6by9 to14by18cm((8.66�2.05)by (12.62�2.03)). Flapsweredesignedwith theperforatorarteryincluded, with all flaps based on one or two perforator arteries.Results: All of theflaps survived.Therewere noproblemswith vascular spasmorocclusion. Signif-icant venous congestion was not observed in any of the cases. Two cases developed minor wounddehiscence but healed with conservative therapy. None of the donor sites developed seromas.Conclusions: The thoracodorsal artery flapwithout associated latissimus dorsimuscle canprovidea thin, large and reliable flapwith robust blood supply. TheTAPflap can significantly reduce donorsite morbidity.ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published byElsevier Ltd. All rights reserved.

15295139.hotmail.com (X.-c. Wang).

tishAssociationofPlastic,ReconstructiveandAestheticSurgeons.PublishedbyElsevierLtd.All rightsreserved.9.005

Table 1 The information of the 67 patients.

Patient Sex Age (yr) Flapsize (cm)

No. ofperforators

ReconstructedArea

Types offlaps

Lengthof vesselpedicles (cm)

Time ofdissectionperforator (h)

Treatmentof donorregion

Healing timeof donorregion (days)

1 M 35 10 � 13 1 Axillary Pedicle flap 12.5 1.5 Graft skin 92 F 27 7 � 10 1 Chest Pedicle flap 12 1.5 Direct closure 143 F 10 8 � 14 1 Axillary Pedicle flap 8 2 Graft skin 94 M 42 10 � 15 2 Lower extremity Free flap 7 2 Graft skin 95 F 11 6 � 10 1 Axillary Pedicle flap 11 1.5 Direct closure 126 M 7 7 � 12 1 Chest Pedicle flap 9 1.5 Direct closure 127 M 15 7 � 11.5 1 Axillary Pedicle flap 11 1.5 Direct closure 128 F 27 6 � 9 1 Axillary Pedicle flap 13 1.5 Direct closure 129 F 21 9 � 13.5 1 Chest Pedicle flap 12 1.5 Graft skin 910 F 29 11 � 14 2 Upper extremity Free flap 15 2 Graft skin 911 M 31 14 � 18 2 Lower extremity Free flap 13 2 Graft skin 912 M 19 11 � 15 2 Axillary Pedicle flap 14 2 Graft skin 913 F 36 7 � 12 1 Chest Pedicle flap 13.5 1.5 Direct closure 1214 M 37 8 � 12 1 Chest Pedicle flap 14.5 1 Direct closure 1215 M 13 10 � 13 1 Axillary Pedicle flap 10 1 Graft skin 916 F 15 7 � 13 1 Axillary Pedicle flap 12 1 Direct closure 1217 M 55 13 � 15 2 Chest Pedicle flap 15 1.5 Graft skin 918 M 42 10 � 13.5 1 Lower extremity Free flap 9 1 Graft skin 919 M 12 6 � 10 1 Axillary Pedicle flap 13 1.5 Direct closure 1220 M 9 7 � 10 1 Axillary Pedicle flap 9 1.5 Direct closure 1221 F 15 11 � 13.5 1 Chest Pedicle flap 15 1.5 Graft skin 922 M 23 9 � 14 2 Chest Pedicle flap 16 2 Graft skin 923 M 22 10 � 16 2 Upper extremity Free flap 8 2 Graft skin 924 F 31 7 � 13.5 1 Chest Pedicle flap 15 1.5 Direct closure 1225 F 54 8 � 10 1 Chest Pedicle flap 15 1.5 Direct closure 1226 M 12 10 � 13 2 Axillary Pedicle flap 12 1.5 Graft skin 927 F 35 11 � 15 2 Lower extremity Free flap 9 1.5 Graft skin 928 F 38 8 � 13 2 Axillary Pedicle flap 13 1.5 Direct closure 1229 M 25 8 � 12 1 Axillary Pedicle flap 11 1 Direct closure 1230 M 29 7 � 12 1 Chest Pedicle flap 15 1.5 Direct closure 1231 M 16 14 � 18 2 Upper extremity Free flap 10.5 1.5 Graft skin 932 M 42 11 � 14 2 Axillary Pedicle flap 11.5 1.5 Graft skin 933 M 22 7 � 11 1 Chest Pedicle flap 15.5 1.5 Direct closure 1234 F 16 7.5 � 13 1 Axillary Pedicle flap 13 1.5 Direct closure 1235 M 23 8 � 12 1 Axillary Pedicle flap 14 1.5 Direct closure 1236 M 14 7 � 12 1 Axillary Pedicle flap 12 1.5 Direct closure 1237 M 39 10 � 13 1 Chest Pedicle flap 18 1.5 Graft skin 938 F 45 13 � 17 2 Lower extremity Free flap 11 1.5 Graft skin 939 M 51 8 � 13 1 Chest Pedicle flap 16 1.5 Direct closure 1240 M 46 7.5 � 10 1 Chest Pedicle flap 16 1.5 Direct closure 12

194L.-ch

angYangetal.

41 M 15 7 � 9 1 Axillary Pedicle flap 13 1.5 Direct closure 1242 M 23 7 � 11 1 Chest Pedicle flap 17 1.5 Direct closure 1243 F 13 8 � 13 1 Axillary Pedicle flap 11.5 1.5 Direct closure 1244 F 15 9 � 13 1 Axillary Pedicle flap 13 1.5 Graft skin 945 M 16 10 � 13 1 Axillary Pedicle flap 14 1.5 Graft skin 946 M 36 7 � 10 1 Chest Pedicle flap 15 1.5 Direct closure 1247 M 17 8 � 12.5 1 Axillary Pedicle flap 12 1.5 Direct closure 1248 F 11 7.5 � 12 1 Axillary Pedicle flap 10 1.5 Direct closure 1249 M 13 7 � 11 1 Axillary Pedicle flap 12 1.5 Direct closure 1250 M 17 6 � 10 1 Axillary Pedicle flap 13 1.5 Direct closure 1251 F 32 10 � 14 1 Chest Pedicle flap 15 1.5 Graft skin 952 M 48 7 � 12 1 Chest Pedicle flap 16 1.5 Direct closure 1253 M 53 6 � 9.5 1 Axillary Pedicle flap 11 1.5 Direct closure 1254 M 22 7 � 11 1 Axillary Pedicle flap 13 1.5 Direct closure 1255 M 14 8 � 14 1 Axillary Pedicle flap 10.5 1.5 Direct closure 1256 F 34 10 � 13 1 Chest Pedicle flap 17 1.5 Graft skin 957 M 27 11 � 14 2 Axillary Pedicle flap 12.5 2 Graft skin 958 F 28 8 � 12 1 Axillary Pedicle flap 11.5 1.5 Direct closure 1259 F 19 7 � 11 1 Chest Pedicle flap 16 1.5 Direct closure 1260 M 23 10 � 14 1 Chest Pedicle flap 17 1.5 Graft skin 961 M 16 11 � 14 2 Axillary Pedicle flap 12 1.5 Graft skin 962 M 12 7 � 10 1 Axillary Pedicle flap 8 1.5 Direct closure 1263 F 38 6 � 10 1 Chest Pedicle flap 16.5 1.5 Direct closure 1264 M 13 10 � 15 2 Axillary Pedicle flap 11 1.5 Graft skin 965 M 18 10 � 14 1 Axillary Pedicle flap 12.5 1.5 Graft skin 966 F 42 13 � 15 2 Chest Pedicle flap 15 2 Graft skin 967 F 39 7 � 10.5 1 Axillary Pedicle flap 13 1 Direct closure 12

Clin

icalapplica

tionofTAPflaps

195

Figure 1 A 42-year-old male presented with a defect of thelower extremity after a motor vehicle accident.

Figure 2 A 10 � 15-cm TAP flap was designed.

196 L.-chang Yang et al.

With progress understanding the anatomy of perforatorflaps, the thoracodorsal artery perforator (TAP) flapswithout latissimus dorsi muscle have been used progres-sively in reconstructive surgery.1e4 Since 1995, TAP flapshave been used for reconstruction of the extremities,5,6

head and neck7,8 and chest wall1 and also transferred aspedicled flaps to cover axillary wounds,9e11 pectus exca-vatum12 and the back.13 TAP flaps can also be transferred ascomposite flaps with latissimus muscle flaps or scapula flapsto cover multiple anatomically complicated wounds.14

From August 2006 to August 2011, we performed 67 freeor pedicled TAP flaps on 67 patients (Table 1).

Surgical technique

The lateral border of the latissimus dorsi muscle wasidentified in the lateral decubitus position. The perforatorarteries were identified by the hand-held Doppler, and flapswere designed to include one or two perforators. Thelocation of the perforator in adults is 8e12 cm caudal to theposterior axillary fold and 2e4 cm medial to the lateralmargin of the latissimus dorsi muscle. For children, theperforators were also mapped by Doppler. The flap isorientated in the same direction as the latissimus dorsimuscle fibres. Dissection is started from the anterior borderof the flaps towards the perforator points. The dissectionwas completed between the subfascial plane and the deepplane of latissimus dorsi muscle. All of our cases hadperforators that were pulsatile and larger than 0.5 mmintra-operatively. After identifying the pulsating perfo-rator, dissection was continued through the muscle untilthe thoracodorsal vessels were identified. The remainder ofthe flap was then incised. The vascular pedicle wasdissected to achieve adequate length of the pedicle. Whenincluding two perforator arteries, we separated the musclebetween the perforators, connecting them. The thor-acodorsal nerve should be preserved. In this study, if theflap width was not greater than 8 cm, the donor site couldbe closed primarily, otherwise skin grafting is necessary.

Case reports

Case 4

A 42-year-old male presented with a defect of the lowerextremity after a motor vehicle accident. After debride-ment of the wound, the tibia was exposed. A 10 � 15-cmfree TAP flap was performed with an end-to-end anasto-mosis to the anterior tibial artery and vein. The flapsurvived completely. A minor wound dehiscence healed byconservative therapy (Figures 1e4).

Case 8

A 27-year-old female presented with scar contracture ofthe axilla after tumour resection. She could not fullyabduct her arm. After releasing the scar, a 6 by 9-cmpedicled TAP flap including one perforator artery wastransferred to reconstruct the axillary defect. The flapsurvived completely. The donor area was directly closed. At

6 months’ follow up, there was no limitation of armadduction and abduction (Figures 5e8).

Case 25

A 54-year-old female presented with a chest mass of 12months’ duration. The tumour was extensively resectedwith the pathologic evaluation identifying a fibrosarcoma.An 8 by 10-cm pedicle TAP flap was transferred to recon-struct the chest defect (Figures 9e12).

Results

From August 2006 to August 2011, TAP flaps were performedin 67 patients (42 male and 25 female). Their ages rangedfrom 7 to 55 years (26.04 � 12.83). Thirty-five pedicled TAPflaps were performed for axillary reconstruction, and 24cases for coverage of the chest defects after injury ortumour resection. Eight cases were performed for recon-struction of the extremity using free TAP flaps. The sizes offlaps ranged from 6 by 9 to 14 by 18 cm ((8.66 � 2.05) by(12.62 � 2.03)). Dissection time was 1.52 � 0.23 h. Thelength of the pedicle varied from 7 to 17 (12.81 � 2.5). Allflaps survived with no problems relating to vascular spasm

Figure 3 The TAP flap was ready to be transferred.

Figure 4 The flaps survived with a minor wound dehiscencehealed with conservative therapy.

Figure 6 After releasing the scar, a 7 � 11 cm pedicle TAPflap was designed.

Figure 7 The TAP flap was elevated with one perforatorartery.

Clinical application of TAP flaps 197

or occlusion. Significant venous congestion was notobserved in any of the cases. Only two cases developedwith minor wound dehiscence, and healed with conserva-tive therapy. None of the donor sites developed seromas(Table 2).

Figure 5 A 27-year-old female suffered from scar contrac-ture of axilla after tumour resection, she could not abduct herarm completely.

Figure 8 The TAP flap survived and motion of upper arm isnot limited six months after operation.

Figure 9 A 54-year-old female presented with a chest mass.

Figure 11 The TAP pedicle flap was transferred to recon-struct the chest defect.

198 L.-chang Yang et al.

Discussion

Heitmann et al.15 dissected 16 fresh cadavers and foundthat the thoracodorsal artery bifurcated into transverseand descending branches at points that occurred at about4 cm below the bottom edge of the scapula, and roughly2.5 cm posterior to the anterior edge of the latissimus dorsimuscle. There were 64 perforators greater than 0.5 mm indiameter in all of the 20 fresh cadavers, and 36 perforators(about 56%) were found originating from the lateral branch,and 28 perforators from the horizontal branch. Thomaset al.16 found that the most proximal perforator waslocated at a point that was 3 cm posterior to the anterioredge of the latissimus dorsi muscle, and was parallel to thesubscapular angle. Angrigiani et al.17 reported that there

Figure 10 A 8 � 10-cm TAP pedicle flap was dissected.

were two to three perforators originating from thedescending branch of the thoracodorsal artery in all spec-imens, and that these perforators were consistentlypresent. The first perforator was located approximately8 cm below the posterior axillary fold. Schaverien et al.18

found that at least one perforator originated from thedescending branch. He also found that 58% of all perfora-tors originated from the descending branch and 39% fromthe thoracodorsal artery were found in the region locatedbetween 9.5 and 15.4 cm from the posterior axillary fold,and within 4.3 cm of the lateral border of the latissimusmuscle. Kim et al.3 described that the perforator is usuallylocated at approximately 8e12 cm below the posterioraxillary fold and at approximately 2e4 cm from the anterior

Figure 12 Twoweeks after surgery, the TAP flap survivedwell.

Table 2 The mean � SD information of the 67 patients.

Patient age (yr) Flap size (cm) Length of vesselpedicles (cm)

Time of dissectionperforator (h)

Mean � SD 26.04 � 12.83 (8.66 � 2.05) � (12.62 � 2.03) 12.80 � 2.52 1.52 � 0.23

Clinical application of TAP flaps 199

edge of the latissimus dorsi muscle in adults. For children,Schwabegger et al.19 located that the perforator atapproximately 6 cm below the posterior axillary fold inchildren. Although the reports seem different, variouslandmarks are still helpful to locate these perforators.

In our cases, we did not solely rely on the anatomiclandmarks. It is very important to identify the location ofthe perforators by Doppler examination before surgery,especially for children. We found that the first perforator isusually located at approximately 6e8 cm below theposterior axillary fold and 2e3 cm from the anterior edge ofthe latissimus dorsi muscle. Because the first perforator islarger than others and was presented in all patients, weprefer to utilise the first perforator.

One perforator was adequate in most flaps, but fora large flap we would prefer to include two perforators inthe flap. There are different opinions about whether a stripof muscle should be retained to protect the perforatorartery while dissecting. Schwabegger et al.19 emphasisedthat a 2 by 8-cm muscle bundle should be retained whendissecting the perforator. Xu et al.4 reported that a 3e5 by6e8-cm muscle cuff should be preserved to avoid injury tothe perforators. We are concerned that harvesting a strip oflatissimus dorsi muscle will affect its function, and makethe pedicle bulky possibly requiring debulking at a secondstage. We did not preserve any surrounding muscle cuff,and all flaps survived well. Children’s latissimus dorsimuscle is usually thin (less than 1 cm), so the distance thatthe perforator travels in the muscle is short and relativelyeasier to dissect. It usually took us 1e2 h for dissection ofthe perforator. If the size of the flap was large, we usuallyincluded two perforator arteries. We detached the musclebetween the two arteries, and dissected to the trunk of thethoracodorsal artery to obtain enough pedicle length. Toget enough length of the arterial pedicle for local transfer,we usually dissected the perforator to the origin of thor-acodorsal artery. Occasionally, we divided the perforator atthe level of the circumflex scapular artery to extend theperforator to the subscapular artery if needed. Afterdissection, we delivered the flaps through the latissimusdorsi muscle, or less likely divided the muscle and trans-ferred the flap directly, which could damage a small part ofthe muscle. After the dissection is finished, the dividedmuscle should be repaired before the closure of the donorsite. Although the process of division of muscle may affectits function, at that time of follow-up no patients lostfunction or changed their job, or gave up sports because ofmuscle dysfunction. We made no formal assessment ofpostoperative muscle function.

Compared to latissimus dorsi myocutaneous flaps andscapular flaps,20 thoracodorsal artery flap has severaladvantages: First, it offers a larger and thinner flap withoutlatissimus dorsi muscle, with thickness of the flap approx-imately 0.4e0.8 cm after partial subcutaneous adipose

tissue resection. It is suitable for reconstruction of headand neck, anterior tibial and axillary wounds that requirefunctional and aesthetic recovery in a single surgicalprocedure. Second, due to the retention of latissimus dorsimuscle, the morbidity of the donor area is reduced. It alsodecreases the complications of the seroma and haematomathat often occur with latissimus dorsi myocutaneous flapharvest. Third, if the width of the flap was less than 8 cm,the donor area would be closed directly without skingrafting, with the scar direction and the axillary fold con-cealing the scar itself. Fourth, it can provide a long pedicleof up to 18 cm. Fifth, it is possible to design a sensate TAPflap with attached intercostal nerves. Sixth, based on theanatomic features of the subscapular artery, the TAP flapwith associated scapular skin flap, para-scapular flap orlatissimus dorsi flap can be harvested so as a chimaeric flapto reconstruct different functional units. The scapular flapor para-scapular flap can be another good option forreconstruction of axillary defects, in order to obtain longpedicles to the level of the bifurcation of the subscapularartery. Extending the flap triangular space allowed the flapto cover the anterior part of the axillary contractures, butthe pedicle may not be reliable long enough for chestreconstruction.21,22 The main disadvantage is that itrequires patient repositioning in surgery, which increasessurgery time.

Funding

None.

Conflict of interest

None.

Acknowledgements

None.

References

1. Ortiz CL, Mendoza MM, Sempere LN, Sanz JS, Torres AN,Barraquer EL. Versatility of thepedicled thoracodorsal arteryperforator (TDAP) flap in soft tissue reconstruction. Ann PlastSurg 2007;58:315e20.

2. Lin CT, Huang JS, Yang KC, Hsu KC, Chen JS, Chen LW. Reli-ability of anatomical landmarks for skin perforators of thethoracodorsal artery perforator flap. Plast Reconstr Surg 2006;118:1376e86.

3. Kim SE, Rhyou IH, Suh BG, Chung KC. Use of thoracodorsalartery perforator flap for soft tissue reconstruction in children.Ann Plast Surg 2006;56:45le4.

200 L.-chang Yang et al.

4. Xu YB, Xiang JP, Liu XL. Design and application of the latissimusdorsi perforator based muscle sparing flap. Chin J Microsurg2006;29:335e7.

5. Hamdi M, Van Landuyt K, Monstrey S, Blondeel P. A clinicalexperience with perforator flaps in the coverage of extensivedefects of the upper extremity. Plast Reconstr Surg 2004;113:1175e83.

6. Jeon B, Lim SY, Pyon JK, Bang SI, Oh KS, Mun GH. Secondaryextremity reconstructionwith free perforator flaps for aestheticpurposes. J Plast Reconstr Aesthet Surg 2011;64:1483e9.

7. Mun GH, Lim SY, Hyun WS, Bang SI, Oh KS. Correction of tem-porommasseteric contour defomity using the dual paddlethoracodorsal artery perforator adiposal flap. J ReconstrMicrosurg 2006;22:335e42.

8. Guerra AB, Lyons GD, Dupin CL, Metzinger SE. Advantages ofperforator flaps in reconstruction of complex defects of thehead and neck. Ear Nose Throat J 2005;84:44le47.

9. Rehman N, Kannan RY, Hassan S, Hart NB. Thoracodorsal arteryperforator (TAP) type I V-Y advancement flap in axillaryhidradenitis suppurativa. Br J Plast Surg 2005;58:441e4.

10. Kulahci Y, Sever C, Uygur F, Oksuz S, Sahin C, Duman H. Pre-expanded pedicled thoracodorsal artery perforator flap forpostburn axillary contracture reconstruction. Microsurgery2011;31:26e31.

11. Busnardo FF, Coltro PS, Olivan MV, Busnardo AP, Ferreira MC.The thoracodorsal artery perforator flap in the treatment ofaxillary hidradenitis suppurativa effect on preservation of armabduction. Plast Reconstr Surg 2011;128(4):949e53.

12. Sinna R, Perignon D, Qassemyar Q, et al. Case report. A doublethoracodorsal artery perforator flap technique for the treat-ment of pectus excavatum. Eplasty 2010;10:281e6.

13. Narushima M, Yamamoto T, Yamamoto Y, Hirai R, Mihara M,Koshima I. Lateral intercostal artery perforator-based reversed

thoracodorsal artery flap for reconstruction of a chronic radi-ation ulcer of the lower back wall. Ann Plast Surg 2011;67(4):352e6.

14. Momeni A, Krischak S, Bannasch H. The thoracodorsal arteryperforator flap with a vascularized scapular segment forreconstruction of a composite lower extremity defect. Micro-surgery 2006;26:515e8.

15. Heitmann C, Guerra A, Metzinger SW, Levin LS, Allen RJ. Thethoracodorsal artery perforator flap: anatomic basis and clin-ical application. Ann Plast Surg 2003;51:23e9.

16. Thomas BP, Geddes CR, Tang M, Williams J, Morris SF. Thevascular basis of the thoracodorsal artery perforator flap. PlastReconstr Surg 2005;116:818e22.

17. Angrigiani C, Grilli D, Siebert J. Latissimus dorsi musculocuta-neous flap without muscle. Plast Reconstr Surg 1995;96:1608e14.

18. Schaverien M, Wong C, Bailey S, Saint-Cyr M. Thoracodorsalartery perforator flap and latissimus dorsi myocutaneous flap e

anatomical study of the constant skin paddle perforator loca-tions. J Plast Reconstr Aesthet Surg 2010;63:2123e7.

19. Schwabegger AH, Harpf C, Rainer C. Muscle-sparing latissimusdorsi myocutaneous flap with maintenance of muscle inner-vation, function, and aesthetic appearance of the donor site.Plast Reconstr Surg 2003;111:1407e11.

20. Zevallos JP, Urken ML. Reverse-flow scapular osteocutaneousflap for head and neck reconstruction. Head Neck 2012 Apr 5.http://dx.doi.org/10.1002/hed.22914. [Epub ahead of print].

21. Turkaslan T, Turan A, Dayicioglu D, Ozsoy Z. Uses of scapularisland flap in pediatric axillary burn contractures [correction ofconractures]. Burns 2006;32:885e90.

22. Ogawa R, Hyakusoku H, Murakami M, Koike S. Reconstruction ofaxillary scar contractures e retrospective study of 124 casesover 25 years. Br J Plast Surg 2003;56:100e5.