Forefoot reconstruction by reversed island flaps in diabetic patients

Soft-tissue coverage of the foot is often difficult, especially when the distal third of the foot (dorsal or plantar aspects) is involved. The clinical situation can be further complicated when diabetic patients are affected by painful and unstable wounds of this kind because of the familiar phenomenon of vasculopathy. The purpose of this study was to evaluate the possibility of using distally based foot flaps to cover forefoot defects in diabetic patients. Preoperative selection of patients was the key to this study; those who had other major disease, chronic infection, bone involvement, and/or insufficient foot vascularization were excluded from the study. The authors report a series of 12 diabetic patients in whom the reconstruction of medium-sized defects (ranging from 1.5 x 2.0 cm to 3.0 x 7.0 cm) of the forefoot was performed using distally based dorsalis pedis flaps or medial plantar flaps. The transferred flaps survived and adapted well to the defects, except for one flap in a patient who had a slight venous insufficiency at outset. Wearing their own footwear, patients could walk after 20 to 30 days. After the follow-up period (3 months to 3 years), no skin breakdown in the treated areas was observed. Temporary donor-site pain was reported by medial plantar flap patients, and partial skin graft loss at the donor site occurred in some of the dorsalis pedis patients. The authors suggest that in selected cases, medium-sized soft-tissue defects involving the dorsal aspects or the weight-bearing areas of the diabetic foot can be successfully covered with distally based island flaps.     

The concept of fillet flaps: classification, indications, and analysis of their clinical value

Tissue of amputated or nonsalvageable limbs may be used for reconstruction of complex defects resulting from tumor and trauma. This is the "spare parts" concept.By definition, fillet flaps are axial-pattern flaps that can function as composite-tissue transfers. They can be used as pedicled or free flaps and are a beneficial reconstruction strategy for major defects, provided there is tissue available adjacent to these defects.From 1988 to 1999, 104 fillet flap procedures were performed on 94 patients (50 pedicled finger and toe fillets, 36 pedicled limb fillets, and 18 free microsurgical fillet flaps).Nineteen pedicled finger fillets were used for defects of the dorsum or volar aspect of the hand, and 14 digital defects and 11 defects of the forefoot were covered with pedicled fillets from adjacent toes and fingers. The average size of the defects was 23 cm2. Fourteen fingers were salvaged. Eleven ray amputations, two extended procedures for coverage of the hand, and nine forefoot amputations were prevented. In four cases, a partial or total necrosis of a fillet flap occurred (one patient with diabetic vascular disease, one with Dupuytren's contracture, and two with high-voltage electrical injuries).Thirty-six pedicled limb fillet flaps were used in 35 cases. In 12 cases, salvage of above-knee or below-knee amputated stumps was achieved with a plantar neurovascular island pedicled flap. In seven other cases, sacral, pelvic, groin, hip, abdominal wall, or lumbar defects were reconstructed with fillet-of-thigh or entire-limb fillet flaps. In five cases, defects of shoulder, head, neck, and thoracic wall were covered with upper-arm fillet flaps. In nine cases, defects of the forefoot were covered by adjacent dorsal or plantar fillet flaps. In two other cases, defects of the upper arm or the proximal forearm were reconstructed with a forearm fillet. The average size of these defects was 512 cm2. Thirteen major joints were salvaged, three stumps were lengthened, and nine foot or forefoot amputations were prevented. One partial flap necrosis occurred in a patient with a fillet-of-sole flap. In another case, wound infection required revision and above-knee amputation with removal of the flap.Nine free plantar fillet flaps were performed-five for coverage of amputation stumps and four for sacral pressure sores. Seven free forearm fillet flaps, one free flap of forearm and hand, and one forearm and distal upper-arm fillet flap were performed for defect coverage of the shoulder and neck area. The average size of these defects was 432 cm2. Four knee joints were salvaged and one above-knee stump was lengthened. No flap necrosis was observed. One patient died of acute respiratory distress syndrome 6 days after surgery.Major complications were predominantly encountered in small finger and toe fillet flaps. Overall complication rate, including wound dehiscence and secondary grafting, was 18 percent. This complication rate seems acceptable. Major complications such as flap loss, flap revision, or severe infection occurred in only 7.5 percent of cases. The majority of our cases resulted from severe trauma with infected and necrotic soft tissues, disseminated tumor disease, or ulcers in elderly, multimorbid patients.On the basis of these data, a classification was developed that facilitates multicenter comparison of procedures and their clinical success. Fillet flaps facilitate reconstruction in difficult and complex cases. The spare part concept should be integrated into each trauma algorithm to avoid additional donor-site morbidity and facilitate stump-length preservation or limb salvage.     

Posterior interosseous free flap with extended pedicle for hand reconstruction

The anatomy of the posterior interosseous vessels makes them suitable as a donor area of free flap. The skin island can be designed on the perforating vessels of the distal third of the forearm, up to the dorsal wrist crease, to increase the pedicle length (7 to 9 cm). A series of nine flaps transferred to reconstruct hand defects is presented. All flaps were designed over the dorsal distal forearm, and dimensions permitted direct closure of the donor site (up to 4 to 5 cm wide). Apart from a linear scar, donor morbidity was negligible. All transfers were successful. Although its dissection is somewhat tedious, the anatomy of the vascular pedicle is suitable for microanastomosis and the skin island is thin, although hairy. The posterior interosseous free flap with extended pedicle may be a good choice when limited amounts of thin skin and a long vascular pedicle are needed.     

Primary closure of radial forearm flap donor defects with a bilobed flap based on the fasciocutaneous perforator of the ulnar artery

To primarily repair a series of radial forearm flap donor defects, a total of 10 bilobed flaps based on the fasciocutaneous perforator of the ulnar artery were designed at the Chang Gung Memorial Hospital in Kaohsiung in the period from January of 2002 to January of 2003. All patients were male, with ages ranging from 36 to 67 years. The forearm donor defects ranged in size from 5 x 6 cm to 8 x 8 cm, with the average defect being 47 cm. One to three sizable perforators from the ulnar artery were consistently observed in the distal forearm and were most frequently located 8 cm proximal to the pisiform, which could be used as a pivot point for the bilobed flap. The bilobed flap consisted of two lobes, one large lobe and one small lobe. With elevation and rotation of the bilobed flap, the large lobe of the flap was used to repair the radial forearm donor defect and the small lobe was used to close the resultant defect from the large lobe. All bilobed flaps survived completely, without major complications, and no skin grafting was necessary. Compared with conventional methods for reconstruction of radial forearm donor defects, such as split-thickness skin grafting, the major advantage of this technique is its ability to reconstruct the donor defect with adjacent tissue in a one-stage operation. Forearm donor-site morbidity can be minimized with earlier hand motion, and better cosmetic results can be obtained. Furthermore, because a skin graft is not used, no additional donor area is necessary. However, this flap is suitable for closure of only small or medium-size donor defects. A lengthy postoperative scar is its major disadvantage.     

Bipedicle strap flaps in reconstruction of longitudinal dorsal skin defects of the digits

Dorsal skin defects in which the loss of integument is longitudinal in shape are not uncommon after injury by rotating machinery and by glass shearing along the length of the digit. This shape of defect is difficult to reconstruct with commonly used flaps but lends itself to reconstruction by the use of longitudinal bipedicle strap flaps moved across the dorsum of the finger from lateral to medial. A variant of this traditional technique was used in the reconstruction of 28 dorsal digital defects. The incidence of these defects and the need for this reconstructive technique were analyzed by a review of 1077 patients with dorsal digital injuries treated in a 6-year period between 1989 and 1995. Approximately 20 percent of all dorsal digital injuries requiring flap reconstruction were suitable for reconstruction with bipedicle strap flaps.     

Osteocutaneous posterior interosseous flap for reconstruction of the metacarpal bone and soft-tissue defects in the hand

A vascularized bone segment of the ulna together with a posterior interosseous fasciocutaneous flap is harvested, including a cuff of the extensor pollicis longus muscle. The authors treated five male patients with metacarpal bone and soft-tissue defects of the hand using a distally based island osteocutaneous posterior interosseous flap. Their ages at the time of surgery ranged from 15 to 37 years (mean, 24 years). The bone defects were in the first metacarpal in three cases, the fourth metacarpal in one, and the fifth metacarpal in one. The length of the donated ulna ranged from 3 to 7 cm (mean, 5 cm). The follow-up period ranged from 5 to 92 months (mean, 39 months). All flaps survived completely. The posterior interosseous flap provides thin skin of good texture, together with vascularized bone, for a one-stage reconstruction of the metacarpal bone and soft-tissue defects in the hand.     

Clinical application of the free thin anterolateral thigh flap in 31 consecutive patients

In this retrospective study, 31 reconstructions using thin anterolateral thigh flaps and six cadaveric dissections of the thigh were investigated in consideration of the anatomic variations of the perforator vessels in the adipose layer, the safe area of flap circulation, and the clinical indications.Three variations of the perforator vessel course in the adipose layer were predicted correctly. The safe radius of a thin anterolateral thigh flap with a thickness of 3 to 4 mm was determined to be approximately 9 cm from the point where the perforator met the skin. The use of a thin anterolateral thigh flap for reconstruction of the neck, axilla, anterior tibial area, dorsum of the foot, circumference on the ankle, forearm, and dorsum of the hand was therefore recommended.     

The sandwich temporoparietal free fascial flap for tendon gliding.

Microsurgical transfer of the superficial and deep temporal fascia based on the superficial temporal vessels has been documented. This article analyzes the functional recovery when each layer of this facial flap is placed on either side of reconstructed or repaired tendons, to recreate a gliding environment. This fascial flap also provided a thin, pliable vascular cover in selected defects of the extremities.Six patients (four male and two female) with tendon loss and skin scarring of the hand (three dorsum, one palmar, and one distal forearm) and posttraumatic scarring of the ankle with tendoachilles shortening (one patient) underwent this procedure. No flap loss was witnessed. Good overall functional recovery and tendon excursion were observed. Complication of partial graft loss was observed in two patients.     

The radial forearm flap: reconstructive applications and donor-site defects in 35 consecutive patients

Thirty-five consecutive patients treated with the radial forearm flap were reviewed. This flap was used in head and neck reconstruction in 25 patients, soft-tissue cover of an extremity in 9 patients, and as a new technique for penile reconstruction in 1 patient. Osteocutaneous flaps were used for mandibular reconstruction in 13 patients. In 6 patients innervated flaps were used to provide sensation on the dorsum of the hand or on the weight-bearing surface of the foot. There was only one total flap failure and no partial failures. Recipient-site complications were few, with prompt healing and very acceptable appearance. Donor-site complications included partial loss of the skin graft with tendon exposure in 10 patients (33 percent), an unsatisfactory appearance in 5 patients (17 percent), and one case of radial fracture (8 percent). On functional testing, there was no significant loss of strength or joint mobility in the donor extremity in 19/20 patients. The authors recommend measures to reduce donor-site morbidity and conclude that, with an acceptable donor site, this flap is valuable in a variety of reconstructive applications.     

Clinical applications of the posterior rectus sheath-peritoneal free flap

Soft-tissue injuries involving the dorsum of the hand and foot continue to pose complex reconstructive challenges in terms of function and contour. Requirements for coverage include thin, vascularized tissue that supports skin grafts and at the same time provides a gliding surface for tendon excursion. This article reports the authors' clinical experience with the free posterior rectus sheath-peritoneal flap foil dorsal coverage in three patients. Two patients required dorsal hand coverage; one following acute trauma and another for delayed reconstruction 1 year after near hand replantation. A third patient required dorsal foot coverage for exposed tendons resulting from skin loss secondary to vasculitis. In all three patients, the flap was harvested through a paramedian incision at the lateral border of the anterior rectus sheath. After opening the anterior rectus sheath, the rectus muscle was elevated off of the posterior rectus sheath and peritoneum. When elevating the muscle, the attachments of the inferior epigastric vessels to the posterior rectus sheath and peritoneum were preserved while ligating any branches of these vessels to the muscle. Segmental intercostal innervation to the muscle was preserved. The deep inferior epigastric vessels were then dissected to their origin to maximize pedicle length and diameter. The maximum dimension of the flaps harvested for the selected cases was 16 X 8 cm. The anterior rectus sheath was closed primarily with non-absorbable suture. Mean follow-up was 1 year, and all flaps survived with excellent contour and good function in all three patients. Complications included a postoperative ileus in one patient, which resolved after 5 days with nasogastric tube decompression.     

Distally based lateral supramalleolar adipofascial flap for reconstruction of the dorsum of the foot and ankle

Soft-tissue reconstruction of the dorsum of the foot and ankle has long been a challenge for reconstructive surgeons. Limitations in the available local tissue and donor-site morbidity restrict the options. In an effort to solve these difficult problems, the authors have begun to use a distally based lateral supramalleolar adipofascial flap. This report presents the authors' early experience with seven patients treated with this flap. The patients' ages ranged from 5 to 26 years; four of the patients were male and three were female. The cause of the soft-tissue defects involved acute trauma and chronic scar contracture. The flap and the adjoining raw area were covered with a full-thickness skin graft, and the donor site at the lateral aspect of the leg was closed primarily without grafting. A skin graft was taken from the groin area, which was closed primarily. Compared with the other flaps, this adipofascial flap was thinner and produced less bulkiness to the recipient site and minor aesthetic sequelae to the donor site. It is believed that this flap is versatile and effective and is a good addition to the available techniques used by reconstructive surgeons for coverage of the dorsum of the foot and ankle.     

Anterolateral thigh flap in the treatment of postburn flexion contractures of the knee

The use of the anterolateral thigh fasciocutaneous flap in the reconstruction of soft-tissue defects around the knee among burn patients is described. The anterolateral thigh fasciocutaneous flap was used for eight patients (all male; mean age, 45 years; age range, 32 to 60 years). Flexion contracture was observed for seven patients with unhealed wounds and one patient with a healed burn wound. The anterolateral thigh flap was used as a free flap for six patients and as a distally based island flap for two patients. The flaps ranged from 8 to 17 cm in width and from 12 to 30 cm in length. Seven flaps were based on a musculocutaneous perforator, and two of them were thinned before transfer to the defect. A true septocutaneous perforator was observed in only one case. The mean follow-up period was 12.5 months (range, 3 to 23 months). Only one flap exhibited distal superficial necrosis, which did not compromise the final result. All patients returned to ambulatory status in 15 to 22 days. Extensor splints were applied to prevent mobilization of the skin graft at the flap donor site for only 7 days. The anterolateral thigh flap has many advantages for the reconstruction of postburn flexion contracture of the knee, as follows: (1) very large thin flaps can be elevated, (2) the two-team approach is possible, (3) color and texture matches are good, (4) the donor-site scar can be easily hidden, and (5) the technique allows early mobilization and patients can return to normal daily activity in a short time. Free or distally based anterolateral thigh flaps are a good choice, both aesthetically and functionally, for the reconstruction of soft-tissue defects of the knee region.     

The boomerang flap in managing injuries of the dorsum of the distal phalanx

Finding an appropriate soft-tissue grafting material to close a wound located over the dorsum of a finger, especially the distal phalanx, can be a difficult task. The boomerang flap mobilized from the dorsum of the proximal phalanx of an adjacent digit can be useful when applied as an island pedicle skin flap. The vascular supply to the skin flap is derived from the retrograde perfusion of the dorsal digital artery. Mobilization and lengthening of the vascular pedicle are achieved by dividing the distal end of the dorsal metacarpal artery at the bifurcation and incorporating two adjacent dorsal digital arteries into one. The boomerang flap was used in seven individuals with injuries involving the dorsal aspect of the distal phalanx over the past year. Skin defects in all patients were combined with bone,joint, or tendon exposure. The authors found that the flap was reliable and technically simple to design and execute. This one-step procedure preserves the proper palmar digital artery to the fingertip and has proven valuable for the coverage of wide and distal defects because it has the advantages of an extended skin paddle and a lengthened vascular pedicle. When conventional local flaps are inadequate, the boomerang flap should be considered for its reliability and low associated morbidity.     

Prelaminating the fascial radial forearm flap by using tissue-engineered mucosa: improvement of donor and recipient sites.

In reconstructive surgery, prelamination of free flaps using split-thickness skin is an established technique to avoid the creation of a considerable defect at the donor site, for example, in the case of a radial forearm flap. For oral and maxillofacial surgery, this technique is less than optimal for the recipient site because the transferred skin is inadequate to form a lining in the oral cavity. To create mucosa-lined free flaps, prelamination using pieces of split-thickness mucosa has been performed. However, the availability of donor sites for harvesting mucosa is limited. The present study combines a tissue-engineering technique with free flap surgery to create mucosa-lined flaps with the intention of improving the tissue quality at the recipient site and decreasing donor-site morbidity. On five patients undergoing resection of squamous cell carcinoma of the oral cavity, the radial forearm flap was prelaminated with a tissue-engineered mucosa graft to reconstruct intraoral defects. Using 10 x 5 mm biopsies of healthy mucosa, keratinocytes were cultured for 12 days and seeded onto collagen membranes (4.5 x 9 cm). After 3 days, the mucosal keratinocyte collagen membrane was implanted subcutaneously at the left or right lower forearm to prelaminate the fascial radial forearm flap. One week later, resection of the squamous cell carcinoma was performed, and the free fascial radial forearm flap pre- laminated with tissue-engineered mucosa was transplanted into the defect and was microvascularly anastomosed. Resection defects up to a size of 5 x 8 cm were covered. In four patients, the graft healed without complications. In one patient, an abscess developed in the resection cavity without jeopardizing the flap. During the postoperative healing period, the membrane detached and a vulnerable pale-pink, glassy hyperproliferative wound surface was observed. This surface developed into normal-appearing healthy mucosa after 3 to 4 weeks. In the postoperative follow-up period, such functions as mouth opening and closing and speech attested to the success of the tissue-engineering technique for flap prelamination.     

Clinical applications of free soleus and peroneal perforator flaps

Clinical applications of two free lateral leg perforator flaps are described: a free soleus perforator flap that is based on the musculocutaneous perforator vessels from one of the three main arteries in the proximal lateral lower leg, and a free peroneal perforator flap that is based on the septocutaneous or direct skin perforator vessels from the peroneal artery in the distal and middle thirds of the lateral lower leg. The authors applied free soleus perforator flaps to 18 patients and free peroneal perforator flaps to five patients with soft-tissue defects. The recipient site was the great toe in 14 patients, the hand and fingers in five patients, the leg in two patients, and the upper arm and the jaw in one patient each. The largest soleus perforator flap was 15 x 9 cm, and the largest peroneal perforator flap was 9 x 4 cm. Vascular pedicle lengths ranged from 6.5 to 10 cm in soleus perforator flaps and from 4 to 6 cm in peroneal perforator flaps. All flaps, except for the flap in one patient in the peroneal perforator flap series, survived completely. Advantages of these flaps are that there is no need to sacrifice any main artery in the lower leg, and there is minimal morbidity at the donor site. For patients with a small to medium soft-tissue defect, these free perforator flaps are useful.     

Reconstruction of composite through-and-through mandibular defects with a double-skin paddle fibular osteocutaneous flap

Microsurgical reconstruction of composite through-and-through defects of the oral cavity involving mucosa, bone, and external skin has often required two free flaps or double-skin paddle scapular or radial forearm flaps for successful functional and aesthetic outcomes. A safe, reliable technique using a double-skin paddle fibular osteocutaneous flap to restore the intraoral lining, mandibular bone, and external skin is described. A large elliptical or rectangular skin paddle is designed 90 degrees to the longitudinal axis of the fibula, over the junction of the middle and distal thirds of the lower leg, based only on the posterolateral septocutaneous perforators. This skin flap can be draped anteriorly and posteriorly over the fibular bone to reconstruct both the intraoral defect and the external skin defect. The area between the two skin islands of the intraoral flap and the external flap is deepithelialized and left as a dermal bridge between the two skin islands, as opposed to the creation of two separate vertical skin paddles, each based on a septocutaneous perforator. The transverse dimension of the flap can be as great as 14 cm, extending to within 1 to 2 cm of the tibial crest anteriorly and as far as the midline posteriorly, and with a length of up to 26 cm, this flap should be more than sufficient for reconstruction of most through-and-through defects. This technique has allowed the successful reconstruction of large composite defects, with missing intraoral lining, mandibular bone, and external skin, for 16 patients, with 100 percent survival of both skin islands in all cases and without the development of any orocutaneous fistulae.     

Comparative study of two series of distally based fasciocutaneous flaps for coverage of the lower one-fourth of the leg, the ankle, and the foot

Skin defects over the lower one-fourth of the leg and over the foot are difficult to cover. Two types of pedicled fasciocutaneous flaps used to cover such defects were studied: the lateral supramalleolar flap and the distally based sural neurocutaneous flap. The series consisted of 27 and 36 cases, respectively. The lateral supramalleolar flap was used 27 times: for skin defects over the ankle (4), foot (16), and leg (7). The distally based sural neurocutaneous flap was used 42 times: over the foot (24), ankle (13), and leg (5). Fourteen of these patients were 65 years of age or older, and local vascularity was diminished in 16 cases. The flaps were evaluated clinically twice: in the immediate postoperative period for survival or for partial or total flap necrosis, and again to determine the presence of pain at the donor or recipient sites and the cosmetic appearance. Thirty-nine patients (62 percent) were reviewed subsequently, with a mean follow-up of 5 years for the supramalleolar flap and 2 years for the sural neurocutaneous flap. The results were evaluated for the presence or absence of pain, the appearance of the flap, the disability due to the insensate nature of the flap, and the presence or absence of secondary ulceration. Painful neuromata were noted in three cases with the sural neurocutaneous flap, whereas complete necrosis of the supramalleolar artery flap occurred in three patients. The distally based sural neurocutaneous island flap is very reliable, even in debilitated patients. Though the lateral supramalleolar artery flap offers the possibility of covering the same areas as the sural neurocutaneous flap, it is much less reliable in the presence of diminished local vascularity (18.5 percent failure rate as compared with 4.8 percent for the sural neurocutaneous flap). Because the procedure can cover extensive defects and is easy to perform, the distally based sural neurocutaneous flap was the method of choice for covering skin defects over the foot, heel, ankle, and the lower one-fourth of the leg. The lateral supramalleolar artery flap is indicated only when the sural neurocutaneous flap is contraindicated.     

A reverse ulnar hypothenar flap for finger reconstruction

A reverse-flow island flap from the hypothenar eminence of the hand was applied in 11 patients to treat palmar skin defects, amputation injuries, or flexion contractures of the little finger. There were three female and eight male patients, and their ages at the time of surgery averaged 46 years. A 3 x 1.5 to 5 X 2 cm fasciocutaneous flap from the ulnar aspect of the hypothenar eminence, which was located over the abductor digiti minimi muscle, was designed and transferred in a retrograde fashion to cover the skin and soft-tissue defects of the little finger. The flap was based on the ulnar palmar digital artery of the little finger and in three patients was sensated by the dorsal branch of the ulnar nerve or by branches of the ulnar palmar digital nerve of the little finger. Follow-up periods averaged 42 months. The postoperative course was uneventful for all patients, and all of the flaps survived without complications. The donor site was closed primarily in all cases, and no patient complained of significant donor-site problems. Satisfactory sensory reinnervation was achieved in patients who underwent sensory flap transfer, as indicated by 5 mm of moving two-point discrimination. A reverse island flap from the hypothenar eminence is easily elevated, contains durable fasciocutaneous structures, and has a good color and texture match to the finger pulp. This flap is a good alternative for reconstruction of palmar skin and soft-tissue defects of the little finger.     

Prelaminated free flap reconstruction of complex central facial defects

This article is a review of five patients who underwent reconstruction of nasal and paranasal facial defects with prelaminated forearm free flaps. The defects resulted from thermal injury, gunshot wound, excision of tumor, and arteriovenous malformation (n = 2). The forearm flaps were based on the radial artery (n = 4) and ulnar artery (n = 1) and were prelaminated with grafts of skin and cartilage. All flaps were successfully transferred to the face, but revisions were needed to separate the subunits and improve appearance. A prelaminated free flap should be considered for a patient requiring reconstruction of a complex central facial defect.     

Enhanced neovascularization of rat tubed pedicle flaps with low perfusion of the wound margin

To study the role of ischemia due to low perfusion as the inciter of neovascularization, caudally based 3 X 9 cm skin flaps were created on the dorsum of 50 Sprague-Dawley rats. After injection of 0.2 ml 10% fluorescein, the animals were divided into two groups. In group I (n = 25), the distal margin of the flap tip was 1 cm proximal to the border of the fluorescence (good perfusion). In group II (n = 25), the flap was cut 1 cm distally in the nonfluorescent part (poor perfusion). The tips of the tubed flaps were transferred to a wound bed on the right flank. After 10 days, the pedicles were ligated, so that flap survival depended totally on the new vascular supply from the inset area of the flap. The flaps in group I showed a significantly higher rate of necrosis of 52.4 +/- 15.1 percent versus 1.7 +/- 1.4 percent in group II (p less than 0.0001), although the flap length in group I (5.85 +/- 1.16 cm) was less than in group II (7.15 +/- 0.95 cm; p = 0.0001). A nearly three times larger amount of tissue based on the new blood supply survived in group II compared to group I. Xerograms after injection of PbO2-gelatine on day 10 showed an increased ingrowth of blood vessels in group II. After excluding the delay phenomenon as the cause for the difference in necrosis rate, it is concluded that the only possible explanation is an enhancement of neovascularization by a perfusion gradient between the wound margins.