Journal of Plastic, Reconstructive & Aesthetic Surgery (2012) 65, 634e639

Treatment of complex ischial pressure soreswith free partial lateral latissimus dorsimusculocutaneous flaps in paraplegic patients

Jinguang He, Hua Xu, Tao Wang, Sunxiang Ma, Jiasheng Dong*

Department of Plastic and Reconstructive Surgery, Shanghai 9th People’s Hospital, Shanghai Jiao Tong University School ofMedicine, 639 Zhi Zao Ju Rd., Shanghai 200011, PR China

Received 14 August 2011; accepted 3 October 2011

KEYWORDSPressure sores;Ischial;Free partial latissimusdorsi flap;Donor-site morbidity

* Corresponding author. Tel.: þ86 2163136856.

E-mail address: jsdong2011@hotm

1748-6815/$-seefrontmatterª2011Bridoi:10.1016/j.bjps.2011.10.001

Summary In paraplegic patients dependent on their upper body for mobility, the latissimusdorsi muscle is generally unacceptable for microsurgical reconstruction of complex ischialdefect. To avoid total muscle function loss, a portion of the lateral latissimus dorsi musculo-cutaneous flap can instead be harvested. From February 1999 to March 2009, 11 paraplegicpatients with complex ischial pressure sores were prospectively recruited. The reconstructionwas performed using a free partial lateral latissimus dorsi musculocutaneous flap. The follow-up period ranged from 18 to 114 months (mean, 60 months). All flaps survived postoperatively.No recurrence occurred in our series. All patients experienced various degrees of back tight-ness, shoulder weakness and limited shoulder motion since surgery, which were relieved within9 months. The free partial lateral latissimus dorsi musculocutaneous flap can be a good alter-native for covering severe infected ischial defect. Shoulder functional deficits will lessen overtime and normal function will be regained gradually.ª 2011 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published byElsevier Ltd. All rights reserved.

Despite a structured multidisciplinary strategy that hasbeen advocated in recent years, the incidence and recur-rence rate of pressure sores are still disproportionatelyhigh.1e3 Ischial pressure sores are most common and oftenoccur in wheelchair-bound paraplegic patients.1,4 The

23271699x5120; fax: þ86 21

ail.com (J. Dong).

tishAssociationofPlastic,Reconstruc

aetiology and pathogenic progress are multifactorial. Tobegin with, the sustained unrelieved pressure causes irre-versible ischaemial damage to soft tissues over the ischialtuberosity. Because of underlying muscles being moresusceptible to the ischaemia, the injury under the skin isoften more severe and measured larger at its base than atthe skin surface4 (Figure 1(A)). The moist wound environ-ment is very suitable for bacterial growth, which canfurther aggravate the degree of tissue necrosis. In chroniccircumstances, the duration of infection and extensive

tiveandAestheticSurgeons.PublishedbyElsevierLtd.All rightsreserved.

Figure 1 (A) The left ischial pressure sore before the debridement. The external appearance frequently underestimates theextent of the wound. (B) The view after the debridement. The size of ischial skin defects was 7 � 10 cm. A free musculocutaneousflap may be optimal for providing larger healthy tissue bulk and avoiding interference with local pedicled flaps if needed for thesacrum or trochanter. (C) A partial lateral latissimus flap had been harvested based on the lateral perforators of descendingvascular branch. Innervations of the remaining muscle were preserved well. The donor site had been primarily closed. (D) Therewas no recurrence at 63 months after the operation. (E) The donor site scar at 63 months after the operation. No limited shouldermotion was observed by the physical examination.

Treatment of complex ischial pressure sores 635

inflammatory response can be observed in healthy-appearing skin as well as underlying subcutaneous tissue,fascia and bone.5 Furthermore, ulcers may also extend anderode into nearby deep structures, such as the hip joint,urethra or rectum.6

In general, the treatment of ischial pressure ulcersincludes relief of pressure, nutritional support, antibiotictreatment, thorough debridement and coverage of the

tissue defects. The surgical reconstruction of tissuedefects will be of little value if other issues are ignored.7

Patients with small- to medium-sized ulcers in the ischialregion may be candidates for the treatment with localpedicled flaps. Clinical options include the adipofascialturnover flap,8 posterior medial thigh fasciocutaneousflap,9 pedicled anterolateral thigh flap10 and glutealperforator flap.11

636 J. He et al.

However, the treatment of complex ischial ulcers is stilla great challenge to date. The wound often has a deep deadspace, complicated with severe infection and tissuenecrosis. Besides, there are multiple previous scars withpoor blood supply around the wound. Consequently, therehave been few appropriate local flaps left for coverage. Afree musculocutaneous flap may be an alternative. Thelatissimus dorsi musculocutaneous flap is an extremelyreliable and versatile flap and is commonly used in recon-structive surgery.12 It can provide adequate tissue bulk andenhance the vascularity of the wound. To avoid thepotential loss of the total muscle function, a partial laterallatissimus dorsi flap could instead be harvested witha portion of the lateral latissimus muscle while the bloodsupply and innervations to the remaining muscle arepreserved well.13,14 In this article, we present our surgicalprocedure for coverage of a complex ischial ulcer usinga free partial lateral latissimus dorsi musculocutaneousflap. Functional outcomes at the donor site following thelatissimus dorsi muscle transfer had also been thoroughlyexplored.

Patients and methods

Patients

From February 1999 to March 2009, 11 spinal cord injuredpatients (nine males, two females) diagnosed as grade IIIor IV ischial pressure sores with no osteomyelitis wereprospectively recruited in our department. An informedconsent was obtained from all patients for the use oftheir data. A free partial lateral latissimus dorsi muscu-locutaneous flap transfer was performed for the recon-struction of the ischial tissue defect. The mean age was47 years (range 26e72). The body mass index (BMI)ranged from 18.4 to 25.3 kg me2 (mean 21.4). Fourpatients had a history of smoking. There were threepatients with diabetes and two patients with malnutrition(albumin < 3.5 g dle1) before the surgery (Table 1). Inaddition, all the cases had previous scars around thewound (Table 2).

The follow-up was done every month in the first yearand then two times every year. The donor-site morbiditiesfollowing latissimus dorsi musculocutaneous flap transferwere also assessed by a questionnaire and physical exam-ination at 1, 3, 6 and 9 months after the operation,respectively. The record included back scar, back tight-ness, shoulder weakness and interference with dailyactivities, which were subjectively scaled by patients. The

Table 1 Demographics of patients before the surgery.

Number of subjects 11Age range (year), (mean) 26e72, (47)Female 2Male 9BMI, (range) 18.4e25.3, (21.4)Smoking 4Diabetes 3Malnutrition (albumin<3.5 g/dl) 2

range of shoulder motion was scaled as four categorieswhich were mainly based on the extent of shoulderabduction: ‘not affected’, indicating no limitation ofshoulder mobility; ‘severe’, indicating the extent ofabduction was less than 90�; ‘moderate’, indicating theextent of abduction was more than 90� but less than 135�;and ‘mild’, indicating the extent of abduction was morethan 135 but less than 180�.

Surgical technique

All the operations were performed by our senior author(J.S. Dong). A series of bedside debridements were usuallydone before the reconstruction (Figure 1(A)). After theremoval of all non-viable tissues using a scalpel, a spec-imen was regularly sent for diagnosis of bacterial growth.When there were no obvious infection and necrosis in thewound, consideration can be given to surgical closure. Theoperation was performed under general anaesthesia.Patients were placed first in the prone position and thesize of ulcer cavity was measured after a thoroughdebridement (Figure 1(B)). Superior gluteal vessels weredissected by splitting the gluteus maximus muscle and usedas recipient vessels. Patients were then transferred toa lateral position for latissimus dorsi musculocutaneousflap harvest. The transverse and the descending branch ofthe thoracodorsal neurovascular bundle were identifiedand dissected under the deep surface of the latissimus asthey entered the muscle. When the location of lateralperforators penetrating the muscle was carefully identi-fied, a portion of the lateral latissimus muscle with a skinpaddle was then harvested. The thoracodorsal nervebranch to the remaining latissimus was left intact(Figure 1(C)). The extent of muscle was dependent on theavailability of eliminating the dead space and carryingperforators that supplied the overlying skin paddle.Generally, the muscle was less than 3 cm in width and 9 cmin length. The pedicle dissection could extend as far as thesubscapular artery when a long vascular pedicle wasneeded. A skin graft was used to cover the donor site thatcould not be closed directly. At the recipient site, an end-to-end anastomosis was created between the thor-acodorsal vessels and the superior gluteal vessels. No veingraft was used for pedicle lengthening. A suction drain wasregularly placed under the flap and removed when thedrainage nearly ceased. Patients were kept in prone orlateral decubitus position for at least 2 weeks after theoperation.

Results

The mean size of the skin defect after debridement was7 � 13 cm. All flaps survived postoperatively. One patientdeveloped a seroma at the donor site 8 days after thesurgery. The wound healed conservatively within 2 weeks.The extent of the follow-up period ranged from 18 to 114months, with an average of 60 months. There was norecurrence during our follow-up period (Table 2),(Figure 1(D)).

To observe the potential donor-site morbidity, allpatients were interviewed and examined at 1 month

Table 2 Preoperative data and postoperative outcomes of 11 ischial ulcers treated with free partial lateral latissimus dorsimusculocutaneous flaps.

Age (year) Malnutrition (albumin<3.5 g/dl) Previous pedicled flaps Sizea (cm) Complication Outcome Follow-up (mo)

34 No 1 6 � 10 None Success 11447 No 2 6 � 12 None Success 10626 No 3 7 � 10 None Success 4763 Yes 2 10 � 15 Donor site seroma Success 3548 No 1 6 � 16 None Success 9272 No 3 5 � 12 None Success 2251 No 4 7 � 11 None Success 8430 No 1 9 � 15 None Success 6542 No 3 7 � 10 None Success 6359 Yes 1 7 � 12 None Success 1845 No 2 6 � 14 None Success 19a It indicates the size of ulcer cavity after debridement.

Treatment of complex ischial pressure sores 637

postoperatively. Two patients were dissatisfied with theirback scar due to continuous pruritus of the scar. Allpatients experienced various degrees of back tightness,shoulder weakness and limited range of shoulder motion.The daily activities were impaired subsequently during theperiod. However, the condition gradually improved aftersystemic practice taught by experts. When questionedabout shoulder weakness, limited range of shouldermotion and daily activity impairment at 9 months post-operatively, patients complained of none. Only threepatients still felt a minimal back tightness (Figure 1(E)),(Table 3).

Table 3 The donor site morbidities following the transferof partial lateral latissimus dorsi musculocutaneous flaps.

Variables 1 month 3 months 6 months 9 months

Back scarsAcceptable 0 6 11 11Average 9 5 0 0Unacceptable 2 0 0 0

Back tightnessNo 0 3 8 8Minimal 1 3 2 3Moderate 7 3 1 0Severe 3 2 0 0

Shoulder weaknessNo 0 3 6 11Minimal 2 5 5 0Moderate 6 2 0 0Severe 3 1 0 0

Shoulder motionNot affected 0 7 11 11Minimal 1 2 0 0Moderate 5 2 0 0Severe 5 0 0 0

Daily activityNot affected 0 2 11 11Minimal 1 6 0 0Moderate 4 1 0 0Severe 6 2 0 0

Discussion

Ischial pressure sores are the most common types on thepelvis in spinal cord-injured patients. The management ofcomplex ischial ulcers still represents a challenge forplastic surgeons due to high local recurrence. When infec-ted and deep ulcer cavities are located over the ischium,large healthy muscular tissues with sufficient bulk areneeded. Reconstruction with local flaps may be notoptimal, because it is difficult to cover the defectscompletely with no tension closure after the extensivedebridement.7 Moreover, for paraplegic patients treatedfor pressure sores, the muscle fibres in local flaps used forischial defect coverage are often atrophied due to muscledenervation, and the colour of muscle is hard to bedistinguished from the surrounding subcutaneous tissues. Inaddition, the vascularity of surrounding tissue is often verypoor due to the existence of severe infection and chronicinflammation.

Under these circumstances, reconstruction with freeflaps is an option.15 Our systemic review found that therewere 19 cases of free flaps for pressure sores in the ischialor scaroischial region. Free flaps comprised two cases of thefillet leg flap, five cases of the plantar flap, four cases ofthe gastrocnemius flap and eight cases of the latissimusdorsi flap. One flap necrosis occurred after surgery due tovenous congestion and two cases had recurrences duringthe follow-up period15e24 (Table 4). Although free flaps aretime-consuming and resources demanding, the risk ofrecurrence after reconstruction with free flaps is signifi-cantly lower than that with local flaps.3 For severelyinfected and deep ulcers in the ischial region, the plantarflap and the atrophied gastrocnemius flap are not sufficientfor filling dead space. Meanwhile, the fillet leg flap isusually the last resort due to the severe body imagedestruction. By contrast, our clinical study demonstratedthat the partial lateral latissimus dorsi musculocutaneousflap can provide sufficient tissues to cover complex defectsin the ischial region with no recurrence during our long-term follow-up period. The key point of using free flapsfor reconstruction was the search for suitable recipientvessels in this region. Park Sanghoon previously reportedthat superior and inferior gluteal vessels were constantly

Table 4 Reconstruction with free flaps in ischial/sac-roischiatic region in previous literatures (14, 18e26).

Variable No.

LocationIschion 13Sacroischiatic 6Free flapFillet leg 2Plantar 5Gastrocnemius 4Latissimus dorsi 8Recipient vesselsa

Superior gluteal vessels 4Inferior gluteal vessels 6Deep femoral vessels 3Intercostal vessels 2Superficial femoral vessels 2Inferior epigastric vessels 1ComplicationBursitis 1Anastomotic revision 3Venous congestion 1OutcomeSuccess 16Flap loss 1Recurrence 2a One anastomosis was not clear in the reference andexcluded.

638 J. He et al.

located at the centre of the region, which can beapproached by splitting the gluteus maximus muscle.21 Inour clinical practice, they were our first choice for the freeflap transfer in the region. Although the short vascularlength was their major drawback, the sufficient pediclelength of the latissimus dorsi musculocutaneous flap couldcompensate it well.

In the microsurgical reconstruction of pressure sores,the advantages of using the latissimus dorsi muscle transferinclude adequate tissue bulk for eliminating dead space;improving the vascularity of the wound and superiorinfection control; a consistent, long and large-calibrevascular pedicle; and a vertical for the perforators thatsupply the overlying skin paddle.

However, for patients with spinal cord injury, it isgenerally accepted that latissimus dorsi muscle transfershould be avoided due to potential shoulder functionimpairment. To maintain the muscle function and theaesthetic appearance of the donor site, several modifica-tions of the flap have recently been established. Thethoracodorsal artery perforator flap needs no sacrifice ofthe amount of latissimus dorsi muscle and preserves themuscle function completely.25 However, dissecting theperforators from the muscle is a delicate process andrequires advanced technical skills. Besides, it lacks suffi-cient tissue bulk for eliminating the deep cavity of ischialdefects. Schwabegger et al. further simplified the proce-dure and contained only a tiny lateral muscle segment inthe flap, which was used as a carrier for skin flap vascu-larity. This type of ‘muscle-sparing’ latissimus dorsi

myocutaneous flap saves the main nerve supply to theremaining muscle, and thus maintains the muscle functionand body contour.26 Nevertheless, a minor muscle bulk isstill its limitation. Buntic et al. also proposed a partialsuperior latissimus dorsi musculocutaneous flap forcoverage of a medium-sized defect, leaving intact inner-vation and function of the lateral and inferior muscle.12 Inour clinical experience, we found that the design of a skinpaddle in the vertical direction allowed an easier directskin closure or fewer skin grafts under moderate tension.Therefore, a similar partial lateral latissimus dorsi muscu-locutaneous flap could be harvested based on the lateralbranch of the thoracodorsal system. The segment of muscleserves the main function of eliminating the dead space,enhancing infection control and carrying perforators thatsupply the overlying skin paddle.

Although the adverse effect had been considerednegligible, we also recorded the shoulder functionalchanges in detail following the transfer of partial laterallatissimus dorsi musculocutaneous flaps. All patientsexperienced various degrees of back tightness, shoulderweakness and limited shoulder motion since surgery.However, these functional deficits were mainly caused byscar retraction after the removal of a large skin paddle.They were only temporary and disappeared gradually whenthe scar softened. In our cases, there were no patientscomplaining of the limits of shoulder function at 9 monthsafter surgery.

Finally, it should be noted that, although the free partiallateral latissimus dorsi musculocutaneous flap is a goodchoice for coverage of severe infected ischial defect withno permanent shoulder functional impairment, it is notsuitable for the ischial defect requiring a large amount ofmuscle. This is because the donor-site muscle function andform would irreversibly be impaired when a large latissimusmuscle is harvested. The maximum amount of the musclethat can be harvested with minimal donor site morbiditywill be investigated in our future studies.

Conflict of interest/Funding

The authors report no conflicts of interest or funding.

References

1. Bauer J, Phillips LG. MOC-PSSM CME article: pressure sores.Plast Reconstr Surg 2008;121:1e10.

2. Middleton JW, Lim K, Taylor L, et al. Patterns of morbidity andrehospitalisation following spinal cord injury. Spinal Cord 2004;42:359e67.

3. Keys KA, Daniali LN, Warner KJ, Mathes DW. Multivariatepredictors of failure after flap coverage of pressure ulcers.Plast Reconstr Surg 2010;125:1725e34.

4. Daniel RK, Wheatley D, Priest D. Pressure sores and paraplegia:an experimental model. Ann Plast Surg 1985;15:41e9.

5. Schiffman J, Golinko MS, Yan A, et al. Operative debridementof pressure ulcers. World J Surg 2009;33:1396e402.

6. Acarturk TO. Treatment of large ischial ulcers communicatingwith the hip joint with proximal femoral resection and recon-struction with a combined vastus lateralis, vastus intermediusand rectus femoris musculocutaneous flap. J Plast ReconstrAesthet Surg 2009;62:1497e502.

Treatment of complex ischial pressure sores 639

7. Sorensen JL, Jorgensen B, Gottrup F. Surgical treatment ofpressure ulcers. Am J Surg 2004;188:42e51.

8. Lin H, Hou C, Xu Z, Chen A. Treatment of ischial pressure soreswith double adipofascial turnover flaps. Ann Plast Surg 2010;64:59e61.

9. Lee SS, Huang SH, Chen MC, et al. Management of recurrentischial pressure sore with gracilis muscle flap and V-Y profundafemoris artery perforator-based flap. J Plast Reconstr AesthetSurg 2009;62:1339e46.

10. Kua EH, Wong CH, Ng SW, Tan KC. The island pedicled ante-rolateral thigh (pALT) flap via the lateral subcutaneous tunnelfor recurrent ischial ulcers. J Plast Reconstr Aesthet Surg 2011;64:e21e23.

11. Kim YS, Lew DH, Roh TS, et al. Inferior gluteal artery perfo-rator flap: a viable alternative for ischial pressure sores. J PlastReconstr Aesthet Surg 2009;62:1347e54.

12. Buntic RF, Horton KM, Brooks D, Lee CK. The free partialsuperior latissimus muscle flap: preservation of donor-site formand function. Plast Reconstr Surg 2008;121:1659e63.

13. Theeuwes HP, Gosselink MP, Bruynzeel H, et al. An anatomicalstudy of the length of the neural pedicle after the bifurcationof the thoracodorsal nerve: implications for innervated freepartial latissimus dorsi flaps. Plast Reconstr Surg 2011;127:210e4.

14. Watanabe K, Kiyokawa K, Rikimaru H, et al. Anatomical studyof latissimus dorsi musculocutaneous flap vascular distribution.J Plast Reconstr Aesthet Surg 2010;63:1091e8.

15. Lemaire V, Boulanger K, Heymans O. Free flaps for pressuresore coverage. Ann Plast Surg 2008;60:631e4.

16. Chen HC, Weng CJ, Noordhoff MS. Coverage of multipleextensive pressure sores with a single filleted lower leg myo-cutaneous free flap. Plast Reconstr Surg 1986;78:396e8.

17. Yamamoto Y, Nohira K, Shintomi Y, et al. Reconstruction ofrecurrent pressure sores using free flaps. J Reconstr Microsurg1992;8:433e6.

18. Sekiguchi J, Kobayashi S, Ohmori K. Free sensory and non-sensory plantar flap transfers in the treatment of ischialdecubitus ulcers. Plast Reconstr Surg 1995;95:156e65.

19. Hallock GG. Closure of an ischial pressure sore using a freegastrocnemius musculocutaneous flap with a long venouspedicle. Br J Plast Surg 1995;48:504e6.

20. Hung SJ, Chen HC, Wei FC. Free flaps for reconstruction of thelower back and sacral area. Microsurgery 2000;20:72e6.

21. Park S. Muscle-splitting approach to superior and inferiorgluteal vessels: versatile source of recipient vessels for free-tissue transfer to sacral, gluteal, and ischial regions. PlastReconstr Surg 2000;106:81e6.

22. Jones JW. Reinnervated medial gastrocnemius free flap forclosure of a recurrent ischial pressure sore: case report. JReconstr Microsurg 2002;18:397e400.

23. Feliciano B, Paige KT, Beshlian KM. Latissimus dorsi free flapsfor complex ischiosacral defects. Am J Surg 2007;193:648e50.

24. de la Fuente TP, Gonzalez I, Calderon-Munoz F. The role ofmedial gastrocnemius free flap in coverage of ischial pres-sure sore in paraplegic patients. Int J Surg 2008;6:72e77e76.

25. Spinelli HM, Fink JA, Muzaffar AR. The latissimus dorsiperforator-based fasciocutaneous flap. Ann Plast Surg 1996;37:500e6.

26. 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.