Autologous breast reconstruction with a free lumbar artery perforator flap

A 51-year-old woman had previously been treated for cancer in her right breast with a radical mastectomy and an axillary lymphadenectomy. No adjuvant therapy was given. A secondary breast reconstruction was performed by subpectoral placement of an adjustable saline implant. A reoperation was necessary because of implant rupture. Not satisfied with the aesthetic result and feeling uncomfortable with the prosthesis, she asked for an autologous reconstruction. Preoperative examination revealed a median Caesarean-section scar (Fig. 1) . As a result of congenital hip dysplasia, the patient had developed body asymmetry, with excessive subcutaneous tissue in her right flank. The abdominal scar made the use of a DIEP flap unattractive. The option of a free LAP flap was discussed with the patient. Preoperatively, the LAPs were localised with a hand-held ultrasound Doppler (Multi Dopplex II, 8 MHz) and marked on the skin (Fig. 2) . Colour Doppler flowmetry (General Electric Logiq 9) showed a third LAP, with a diameter of 2.5 mm and a flow of 25 cm s⁻¹. The flap was outlined on the skin, with the axis passing obliquely from the third LAP to the anterior superior iliac spine (Fig. 3a) . Flap harvest was performed with the patient in the lateral decubitus position. After incising the skin and subcutaneous tissue, we opened the thoracolumbar fascia medially over the erector spinae muscle. The fascia was elevated with a light retractor to identify the sensory nerves and the perforators. Under direct visual protection of the perforators, the flap was elevated from anterolateral to medial in a subfascial plane. To obtain sufficient pedicle length, the selected perforator with its comitant vein had to be dissected down between the erector spinae and quadratus lumborum muscles. The recipient vessels were the internal mammary vessels, which were dissected simultaneously in a two-team approach. The donor site was closed, the patient was turned to the supine position, the microsurgical anastomosis was carried out, and the breast was shaped (Fig. 3b). Owing to the earlier reconstruction in this patient, the flap could be almost completely buried beneath the mastectomy flaps. Therefore, no neurorrhaphy between the sensory nerve and the fourth intercostals nerve was performed. After microvascular anastomosis, the well-perfused flap, weighing 1150 g, was trimmed to 750 g. To compensate for postoperative volume reduction and ptosis, the breast was reconstructed approximately 20% larger and placed slightly higher on the chest wall than the opposite breast (Fig. 4) . The postoperative course was uneventful, and the patient left hospital 8 days postoperatively.

Autologous breast reconstruction has evolved enormously in the last decade, especially since the introduction of the DIEP flap. Breast reconstruction with the DIEP flap can give excellent aesthetic results with little donor-site morbidity, making it popular with patients and reconstructive surgeons. An abdominal scar can interfere with the use of a DIEP flap, and a previous abdominoplasty can make it impossible to use this technique. In those patients where a DIEP flap is contraindicated, the surgeon has to look for alternatives. The LAP flap may be such an alternative. The flap is easy to harvest, no muscle is sacrificed and donor-site morbidity is minimal. The resulting scar is easily covered by underwear (Fig. 5) . Kato et al are credited for their excellent cadaveric and clinical study of the pedicled LAP flap. Their fluorescein study showed that the skin territory supplied by the second lumbar artery alone extends from the posterior midline to the lateral border of the rectus sheath and up to 10 cm above the anterior superior iliac spine. The pedicled flap has proven its reliability in several publications.^{,  ,  ,  ,}  Dissection of six LAP flaps on fresh cadavers showed us that a pedicle length of 4 cm is easily achieved. The perforators arise between the erector spinae and quadratus lumborum muscles, and no intramuscular dissection is necessary. Preoperative colour Doppler flowmetry aids the selection of the most suitable perforator. By anastomosing the perforator artery and its comitant vein to the internal mammary vessels, we were able to position the flap favourably on the thorax. Blondeel et al showed that nerve repair in free DIEP flaps leads to earlier postoperative restoration of sensation, increases the quality and quantity of sensation in the flap and has a higher chance of providing erogenous sensation. Isenberg made a similar observation in breast reconstruction with innervated TRAM flaps. The sensory nerve supplying the LAP flap is easily identified intraoperatively. Anastomosing this nerve to the fourth intercostal nerve will probably give a sensate flap. The need to reposition the patient for breast moulding increases the operating time slightly. Postoperatively, patients may notice a slight hypoaesthesia in the L1 and L2 dermatomes. For cosmetic reasons, operative correction of the opposite flank may be necessary at a later date.

In conclusion, the free LAP flap can be used in autologous breast reconstruction as an alternative to other perforator flaps. The LAP flap is a reliable fasciocutaneous perforator flap. No intramuscular dissection is necessary, which makes the flap easy to harvest. Donor-site morbidity is minimal, and the resulting scar is easily hidden by underwear. Anastomosing the sensory nerve of the flap to the fourth intercostal nerve is an option.

We thank Rod Wolstenholme, graphic designer, for his valuable support in providing the illustrations.