The superior gluteal artery perforator flap: an additional tool in the treatment of sacral pressure sores

ritish Journal of Plastic Surgery (1999), 52, 385 391 1999 The ritish ssociation of Plastic Surgeons The superior gluteal artery perforator flap: an additional tool in the treatment of sacral pressure sores. M. Verpaele, P. N. londeel, K. Van Landuyt, P. L. Tonnard,. Decordier, S. J. Monstrey and G. Matton Department of Plastic and Reconstructive Surgery, University Hospital Gent, Gent, elgium SUMMRY. We describe the use of a large skin subcutaneous tissue flap based on one perforator of the superior gluteal artery (SG) to reconstruct large midline posterior defects in one stage. The integrity of the gluteus muscles is preserved and we feel this is particularly important in non-paralysed patients. Donor sites were always closed primarily. Use of the superior gluteal artery perforator (SGP) flap preserves the entire contralateral side as a future donor site. On the ipsilateral side, the gluteal muscle itself is preserved and all flaps based on the inferior gluteal artery are still possible. We recommend this flap in an area where reconstructive possibilities are limited, as it preserves other reconstructive flap options, both on the ipsilateral and contralateral sides. Keywords: pressure sore, decubitus, sacral, flaps, perforator, superior gluteal artery. Paralysed patients often develop large sacral pressure sores, which are subject to recurrence and which need repeated reconstruction. When large musculocutaneous flaps (which may be bilateral) are committed primarily, reconstructive possibilities for the future are diminished. Sacral pressure sores may also present in non-paralysed patients. It is preferable that these defects are not closed with musculocutaneous flaps in order to spare functional muscle. However, the defects may be too large to be reconstructed with fasciocutaneous flaps. Our experience with the DIEP (deep inferior epigastric perforator) flap and the SGP (superior gluteal artery perforator) flap in breast reconstruction stimulated us to develop the SGP flap, which provides an ample amount of tissue with good vascularity, to cover large sacral pressure sores in one stage. This flap is reliably pedicled on a lateral perforator and elevation of the flap does not sacrifice the vascularity or innervation of the underlying gluteus maximus muscle. In our hands, the SGP flap has now become an important additional tool for the treatment of large sacral pressure sores. The flap has a good range of movement and is easily moved, whilst the donor defect is always closed primarily. We feel that the anatomy is sufficiently straightforward for this flap to be dissected by surgical trainees. the medial and middle thirds of a line drawn between the posterior superior iliac spine (PSIS) and the apex of the greater trochanter (T) of the femur 1 (Fig. 1). s the SG supplies the suprapiriformis portion of the gluteus maximus muscle, only perforators located above the piriformis muscle (PM) are used. They are found superiorly to a line drawn between the greater natomy and operative technique The superior and inferior gluteal arteries (SG and IG) are the terminal branches of the internal iliac artery. The SG gives off a deep branch to the gluteus medius muscle and then runs through the gluteus maximus muscle, 1 ending in cutaneous arteries mainly located in the superolateral gluteal region. 2 The SG emerges at the border of the sacrum at the junction of Figure 1 natomical landmarks: the SG emerges at the junction of the middle and medial thirds of a line drawn between the posterior superior iliac spine (PSIS) and the lateral border of the greater trochanter; the perforators are found superiorly to the piriformis muscle (PM), which is above a line between the top of the greater trochanter (T) and a point halfway between the PSIS and coccyx (C). 385

386 ritish Journal of Plastic Surgery C Figure 2 Principle of the procedure. () The flap is dissected free from the underlying gluteus maximus muscle and still perfused through two perforators of the SG, which emerge from between two separate muscle bundles. () The most proximal perforator has been divided to allow maximal medial translation of the flap. (C) The flap is sutured into the defect and the donor site is closed primarily. trochanter of the femur and a point halfway between the PSIS and the coccyx (C) (Fig. 1). With the patient in the prone position, the anatomical landmarks are drawn and the SG and its perforators are indicated with the help of unidirectional Doppler ultrasound. s the arc of movement of the flap is determined by the length of the pedicle, it is important to design the flap around the most lateral perforator to create the longest pedicle possible (Fig. 2). The range of the flap is then easily estimated, usually more than 1.5 times the distance between the SG and the perforator. With this knowledge, one can plan how long the flap should be designed medial to the perforator to reach the furthest end of the defect. We used a conservative approach to the flap, designing the upper border of the flap as a sector of an imaginary rotation flap (Fig. 3), into which the flap could be converted should no suitable perforator be encountered during the dissection. This provided the technique with an additional safety factor. The superior border of the flap is incised first, without bevelling, through skin, subcutaneous tissue and fascia to the muscle. From there the flap is detached from the muscle, until the chosen perforator

SGP flap in the treatment of sacral pressure sores 387 Figure 3 Postoperative view of a patient in whom no suitable perforator could be found during SGP flap dissection. The conservative approach consists of designing the cranial border of the flap as a sector of an imaginary rotation flap (arrows), which can be used for salvage of the procedure, as demonstrated in this case. is encountered. Using loupe magnification, the vessel is then dissected from between the muscle fibres, carefully dividing its small muscular side branches. Usually the vessels lie in a fibrous perimysium, cleaving between the muscle fibres 3 (Fig. 4). Once the vessel is found suitable and dissected, the inferior margin of the flap is incised and the flap is further detached from the muscle. ll other branches emerging from the SG are kept intact, safeguarding the blood supply to the muscle (Fig. 5). No motor nerves are cut in this process. The flap can then be transposed into the defect, as demonstrated in Figure 8, and the donor defect closed primarily after minimal undermining. Suction drainage is applied under the flap and in the donor area until less than 10 cc is collected per 24 h. The patient is kept in a supine position in an air-fluidised bed for approximately 3 weeks, after which gradual mobilisation of the patient is begun. Figure 4 Intraoperative view. () Shows the freed gluteal muscle fibres below, the flap above, well demonstrating the pennate organisation of the muscle, which is attached to the flap by fibrous septa of perimysium (arrows). The vessel loop is placed around a medial perforator, the scissors are under a lateral perforator, still encased in its fibrous septum. () Pedicle length of 9 cm. Results etween pril 1996 and February 1998 we used this flap in 15 patients, of which nine were para- or tetraplegic, and six non-paralysed (Table 1). The age ranged from 20 to 76 years, with a mean age of 49.9 years. The flap size ranged between 10 7 cm and 20 9 cm. Pedicle lengths up to 10 cm were achieved. Twelve flaps survived completely. One patient developed a large haematoma under the flap 2 days postoperatively due to a detached vascular clip. The flap was ultimately lost. Three flaps suffered marginal necrosis, which was debrided with ultimate closure of the defect. Five patients developed a seroma under an intact flap. fter bursectomy, these flaps healed primarily, with the exception of one patient with a liver failure due to alcoholism. Figure 5 Intraioperative view showing the important mobility potential of the flap through the length of the pedicle. Note that the flap is harvested away from the zone of injury, which is located outside of the picture, at the left lower corner.

388 ritish Journal of Plastic Surgery Table 1 List of patients; mean age is 49.9 years; six patients were non-paralysed, nine patients were paralysed Number ge P = paralysed Flap size (cm) Complications NP = not paralysed 1 68 P 12 15 None 2 20 P 11 14.5 None 3 45 P 20 9 Seroma 4 76 NP 12 10 Haematoma partial flap necrosis 5 76 NP 7 10 Seroma 6 61 P 14 10 None 7 59 P 15 6.5 Seroma 8 20 NP 14 12 None 9 40 P 13 9 Fistula outside flap 10 71 NP 13 10 Partial flap necrosis seroma 11 60 NP 18 11 Seroma 12 24 P 15 9 None 13 30 P 16 8 Marginal necrosis 14 27 P 18 11 None 15 72 NP 18 8.5 Marginal necrosis deceased 1 month postoperatively C Figure 6 Case 1. () large grade III sacral pressure sore in a 69- year-old paraplegic female. The extent of the undermining is indicated by the small arrows. () Transposition, without traction of a SGP flap (14 10 cm) into the defect. The ulcerated area visible at the lower medial border of the flap was resected before closure. (C) Stable result 18 months postoperatively. Case reports Case 1 69-year-old female with a history of spastic tetraplegia (patient 1, see Table 1) was referred with a large sacral (Fig. 6) and a smaller trochanteric pressure sore. She was febrile upon admission, although there were no signs of osteomyelitis on bone scan. The bursa of the ulcer was completely excised and a flap of 12 15 cm based on a single SG perforator was moved into the defect (Fig. 6), and inset without any tension. The ulcerated area visible at the lower medial border of the flap was resected before closure. The donor defect was closed primarily. fter a standard 3 weeks of nursing in an air-fluidised bed, all wounds were closed and a stable coverage of the sacrum was provided. The patient was moved into a normal bed with a watermattress and mobilised gradually. She was discharged to a nursing home a few days thereafter. The patient remains recurrence-free after 18 months of follow-up (Fig. 6C). Case 2 Consultation was sought for a 19-year-old male patient (patient 8, see Table 1) who developed a large midline sacral grade III pressure sore during a 2-month stay in the intensive therapy unit after polytrauma. edside necrectomy, enhanced

SGP flap in the treatment of sacral pressure sores 389 D C Figure 7 Case 2. () Defect after debridement of a sacral decubitus ulcer in a 19-year-old non-paralysed male. The flap (arrow) is dissected from a zone not immediately adjacent to the defect and is ready for transposition. () Early postoperative photograph showing primary healing of all wounds. (C) One year postoperatively, the patient is able to stand on the leg of the affected side, showing no signs of gluteal muscle dysfunction. (D) MRI image, transverse section, of the patient s gluteal region, 19 months postoperatively, showing no significant difference in thickness between the gluteal muscles (GM), nor signs of atrophy or myositis. The flap (FL) is visible in position on the sacrum. The donor site scar can be noted as a depression in the subcutis (thick arrow). (E) Coronal section, showing a linear scar in the gluteus maximus muscle (small arrows). E nutrition and local treatment for 2 weeks preceded surgical reconstruction. fter debridement, the defect measured 14 12 cm (Fig. 7). It was closed primarily with a unilateral SGP flap, without any tension on the suture lines. The patient could not be positioned in an air-fluidised bed because of the presence of external fixation devices and was nursed on a low-airloss bed for 3 weeks. The sutures were then removed and gradual mobilisation was commenced. Healing was uneventful (Fig. 7). Eight weeks postoperatively the patient was walking. fter 1 year of follow-up the result remains stable and there is no functional donor site morbidity (Fig. 7C). Nineteen months postoperatively, MRI scan analysis of the gluteal muscles revealed no significant difference in muscle thickness or quality. narrow scar was seen in the donor site muscle (Fig. 7D, E). Discussion Since Koshima and Soeda reported the feasibility of dissecting a musculocutaneous perforator from between the muscle fibres to produce the first DIEP flap, 4 this concept has considerably gained in popularity. 3,7,8 The DIEP and SGP free flaps being the standard procedure for breast reconstruction in our unit, we considered it safe to apply the SGP flap for coverage of sacral pressure sores. In 1993, Koshima et al published their early results with gluteal perforator-based flaps for repair of sacral pressure sores. 2 However, the majority of Koshima s patients received a flap based on several perforators, which needed to be rotated over 60 180 to

390 ritish Journal of Plastic Surgery Figure 8 Transposition (interrupted arrows) of a SGP flap into the defect (long arrow). Note how the thickness of the flap (opposing arrows) provides ample bulk to fill the defect, which is the case even in thin patients. cover the defect. lso, the flaps were taken from the parasacral area, being the zone immediately adjacent to the injured zone. We found that it is possible to raise large skin subcutaneous flaps, based on one single muscle perforator, at a distance from the injured area. The sliding gluteus maximus musculocutaneous flap, as described by Ramirez et al in 1984, has become a widely accepted procedure for coverage of sacral pressure sores in non-paraplegic patients. 5 ilateral musculocutaneous flaps have become the standard procedure for coverage of large sacral defects. The flaps are transposed into the defect, taking advantage of the laterally oblique course of the pedicle. Defects up to 17 cm in width were reported to be covered using bilateral sliding flaps. 5 few years later, the same author commented that the flap is contraindicated when the defect reaches the vicinity of the gluteal pedicles. 6 The gluteal pedicles have a length not exceeding 4 cm, 1,5 so that a medial translation of muscle mass of about 4 cm is the maximum to be expected. s the point of emergence of the superior gluteal artery lies 5 cm from the midline, 9 defects wider than 5 cm are likely to need a bilateral transfer. defect width of 10 cm is the theoretical limit for coverage of sacral decubitus ulcers with the sliding gluteus musculocutaneous flap, a limitation which is confirmed by Ramirez himself. 6 The sacrifice of gluteus maximus muscles and thus also future reconstructive possibilities, the increased blood loss, the increased operating time and the tension on the edges of the flaps are other disadvantages of this technique. y dissecting the perforator vessel from the muscle, a pedicle length of 8.5 10 cm is easily obtained (Fig. 4), giving the flap an impressive mobility and the possibility of covering large defects with a unilateral flap. s has been determined before, the perforators of the SG mainly perfuse the superolateral gluteal region. 2 Choosing a lateral perforator creates the longest possible pedicle, thereby giving the flap a very large arc of movement. This allows the surgeon to replace the ulcer in most cases with undamaged tissue from a distant, untraumatised zone. However, in some very large defects the margins of the defect need to be included in the flap, thereby utilising the skin of the whole gluteal area. The main movement of the flap is a translation. Limited rotation of the flap (up to 90 ) applies minimal torsion to the vascular pedicle, which is, however, of no concern. The flap can be inset without tension on any of the suture lines and the donor defect can be closed primarily in all cases. The flaps from this region have also been shown to provide more than ample bulk to obliterate dead space, as demonstrated in Figure 8. The preservation of muscle integrity and muscle function is one of the greatest assets of the perforator flap principle. Especially in non-paralysed patients who will need full function of the gluteal muscles for recovery of ambulation, the knowledge that function is kept intact may significantly lower the threshold towards decubitus reconstruction with good-quality tissue. Muscle integrity can be judged intraoperatively by the colour of the muscle, 10 and postoperatively by the absence of functional problems in the non-paralysed patients (Fig. 7C). MRI imaging of the gluteal region shows no significant difference in gluteus maximus muscle thickness. There are no signs of muscle atrophy or myositis (Fig. 7D). coronal cross-section reveals a narrow scar in the gluteus maximus muscle without distortion (Fig. 7E). The dissection of the pedicle takes some time, but is straightforward as it lies in an avascular plane. This gives the additional advantage that the blood loss is kept to a minimum, compared to any gluteus flap of which the dissection of the sacral origin can be quite bloody. Subcutaneous tunnelling of a de-epithelialised portion of the flap should be avoided, as the tunnelled portion may suffer fat necrosis and hence induce wound infection and dehiscence. However, this does not appear to compromise the viability of the flap. Seroma formation appeared initially to be a problem. We could detect no clear correlation with age, sex, paralysis, infection or nutritional status. We suspected the explanation might lie in the smoothness of the flap s undersurface, preventing prompt flap adherence. In the last four cases the fascia was partly removed from the flap, in an attempt to create a better adherence. ll these flaps healed without seroma formation. The conservative approach to the flap makes it a safe procedure, even at the beginning of the learning curve. If no perforators had been found at the expected sites, salvage would have been possible with bilateral rotation flaps. The rotation flap suture lines, however, always show some tension and the tip of the flap is less reliable and less bulky than the SGP flap would be. The superior gluteal artery perforator flap provides us with a large, bulky and safe skin subcutaneous flap to cover sacral pressure sores. There is no significant donor site morbidity, no bridges are burned and neither muscle nor muscle function is sacrificed. References 1. Strauch, Yu HL. tlas of Microvascular Surgery. New York: Thieme Medical Publishers, 1993; 104. 2. Koshima I, Moriguchi T, Soeda S, Kawata S, Ohta S, Ikeda. The gluteal perforator-based flap for repair of sacral pressure sores. Plast Reconstr Surg 1993; 91: 678 83.

SGP flap in the treatment of sacral pressure sores 391 3. llen RJ, Tucker C Jr. Superior gluteal artery perforator free flap for breast reconstruction. Plast Reconstr Surg 1995; 95: 1207 12. 4. Koshima I, Soeda S. Inferior epigastric artery skin flaps without rectus abdominis muscle. r J Plast Surg 1989; 42: 645 8. 5. Ramirez OM, Orlando JC, Hurwitz DJ. The sliding gluteus maximus myocutaneous flap: its relevance in ambulatory patients. Plast Reconstr Surg 1984; 74: 68 75. 6. Ramirez OM. The sliding gluteus maximus flap. In: Strauch, Vasconez LO, Hall-Findlay EJ (eds), Grabb s Encyclopedia of Flaps, Vol. 3. New York: Little, rown and Company, 1990; 1544. 7. llen RJ, Treece P. Deep inferior epigastric perforator flap for breast reconstruction. nn Plast Surg 1994; 32: 32 8. 8. londeel PN, oeckx WD. Refinements in free flap breast reconstruction: the free bilateral deep inferior epigastric perforator flap anastomosed to the internal mammary artery. r J Plast Surg 1994; 47: 495 501. 9. Minami RT, Mills R, Pardoe R. Gluteus maximus myocutaneous flaps for repair of pressure sores. Plast Reconstr Surg 1977; 60: 242 9. 10. londeel PN, Vanderstraeten GG, Monstrey SJ, et al. The donor site morbidity of the free DIEP flaps and free TRM flaps for breast reconstruction. r J Plast Surg 1997; 50: 322 30. The uthors lexis M. Verpaele MD, FCCP, Consultant Plastic Surgeon, Phillip N. londeel MD, PhD, FCCP, ssociate Professor, Koenraad Van Landuyt MD, FCCP, ssociate Professor, Patrick L. Tonnard MD, FCCP, Consultant Plastic Surgeon, enoit Decordier MD, FCCP, Consultant Plastic Surgeon, Stanislas J. Monstrey MD, PhD, FCCP, Professor and Chief, Guido Matton MD, FCS, Emeritus Professor and former Chief, Department of Plastic and Reconstructive Surgery, University Hospital Gent, De Pintelaan 185, -9000 Gent, elgium. Correspondence to Dr lexis M. Verpaele. Paper received 31 October 1997. ccepted 11 January 1999, after revision.