An outcome analysis comparing the thoracodorsal and internal mammary vessels as recipient sites for microvascular breast reconstruction: a prospective study of 100 patients

The thoracodorsal vessels have been the standard recipient vessels for the majority of surgeons performing free transverse rectus abdominis musculocutaneous (TRAM) flap reconstructions. Recently, the internal mammary vessels have been recommended as the first-choice recipient vessels for microvascular breast reconstruction. This approach requires a shorter pedicle length, allows for central placement of flap tissue, and avoids axillary scarring. The use of the internal mammary vessels may provide for a shorter operative time and a higher-quality aesthetic reconstruction. The authors performed a prospective trial examining the differences in operative and aesthetic outcomes between each recipient site. A prospective trial of 108 consecutive free-tissue transfers was conducted in 100 patients. The first 60 TRAM flap patients were randomized so that 30 flaps were anastomosed to the internal mammary vessels and 30 were anastomosed to the thoracodorsal vessels, whereas the recipient vessels for the remaining 40 patients were left to the discretion of the surgeon. Of the 40 nonrandomized patients, 10 patients underwent reconstruction using the internal mammary vessels and 30 patients underwent reconstruction using the thoracodorsal vessels. The patients' medical history and hospital course were noted. To evaluate aesthetic outcome, a group of five blinded nonmedical observers and three blinded plastic surgeons graded the reconstructions in the 60 TRAM flap patients for symmetry and overall aesthetic result on a scale of 1 to 5. Blinded practitioners administered postoperative questionnaires to patients regarding recovery time and satisfaction with the aesthetic result. Forty-three flaps were transferred to the internal mammary vessels and 65 were transferred to the thoracodorsal vessels. No significant differences existed between groups with regard to age of preoperative risk factors. Average operative time was 6 hours in each group. Average hospital stay was 5.8 days in each group. Conversion from initial recipient vessel to a secondary recipient site occurred in 12.5 percent of internal mammary reconstructions and 7 percent of thoracodorsal reconstructions. All converted internal mammary cases occurred in left-sided reconstructions and were attributable to problems with the veins. Overall, 20 percent of left-sided internal mammary reconstructions were found to have an inadequate recipient vein. Unusable thoracodorsal vessels were found only in delayed reconstructions, at a rate of 15 percent in the delayed setting. All flaps from converted procedures survived without complications. Average follow-up was 20 months, during which time there was one flap loss in the thoracodorsal group. There were no significant differences in complication rates between groups. Average aesthetic grade was 3.6 in each group. Postoperative recovery time and overall patient satisfaction were not significantly different between groups. Either recipient site can provide for a safe and acceptable result; however, surgeons should be aware of conversion rates and plan appropriately if recipient vessels appear unusable for free-tissue transfer.     

Internal mammary vessels as a model for power Doppler imaging of recipient vessels in microsurgery.

The aim of this interdisciplinary study was to evaluate power Doppler imaging as a method of collecting reliable preoperative data concerning the diameters and topography of exemplary internal mammary vessels as recipient vessels in reconstructive microsurgery. Thirteen female patients (range, 37 to 58 years; mean, 45.6 years) were examined preoperatively with power Doppler imaging from the first to the fifth intercostal space parasternally and bilaterally. These data are compared with measurements obtained intraoperatively in each individual. Mean velocity in the artery in the second intercostal space on the right side is 47.11 cm/sec (range, 15 to 90 cm/sec) and on the left side is 42.25 cm/sec (range, 18 to 95 cm/sec). Mean velocity in the vein in the second intercostal space on the right side is 17.80 cm/sec (range, 10 to 30 cm/sec) and on the left side is 13.06 cm/sec (range, 5.3 to 32 cm/sec). The topographic results are in close agreement with intraoperative measurements and previous anatomical studies. Sonographic preoperative data of arteries (mean, 1.88 mm) show slightly smaller diameters than intraoperative measurements (mean, 2.08 mm), whereas veins show slightly larger diameters in sonography (mean, 2.33 mm) than intraoperatively (mean, 2.12 mm). Mean sonographic diameter of artery ranges from 2.14 mm (second intercostal space) to 1.46 mm (fifth intercostal space), of the vein from 2.76 (second intercostal space) to 1.25 mm (fifth intercostal space). In one case, a vein was not detectable. This noninvasive method leads to confirmation of the preoperative choice of the optimal recipient vessels for free tissue transfer and does not harm the patient.     

Thoracodorsal vessels as recipient vessels for the free TRAM flap in delayed breast reconstruction.

The internal mammary vessels have been recommended as the first choice recipient vessels for delayed breast reconstruction with the free TRAM flap. This approach has avoided surgery in the previously operated axilla, has required a shorter pedicle length, and has allowed for more medial placement of the TRAM tissue. Frequency of nonusable axillary vessels has been reported at 11 percent, with a 6 percent incidence of flap loss in the delayed reconstructive setting. We reviewed our experience with the thoracodorsal vessels as recipient vessels in delayed free TRAM breast reconstruction to assess more accurately the adequacy of these potential recipient vessels. All patients undergoing delayed TRAM reconstruction were reviewed. Forty-seven of 300 consecutive TRAM procedures were for planned delayed free reconstruction. In seven of the patients (15 percent), the thoracodorsal vessels were found to be inadequate for free reconstruction. A supercharged pedicled TRAM was used for reconstruction in each of these seven patients. Average operating room time was 7 hours. Mean follow-up time was 38 months. Nineteen percent of all patients developed at least one complication. Twelve percent of free TRAM patients developed a complication, whereas 57 percent of supercharged patients developed a postoperative complication. The difference in complication rates was statistically significant. The thoracodorsal vessels have provided an adequate recipient vessel in 85 percent of delayed free TRAM reconstructions, comparable to previous reports. Pedicling and supercharging the flap, in those situations in which the thoracodorsal vessels were inadequate, were associated with an increased incidence of postoperative complications. This suggests that in the delayed reconstructive setting, higher-risk patients benefit from free reconstruction over supercharged reconstructions. A second recipient vessel should be used when the thoracodorsal vessels are inadequate for planned free TRAM reconstruction. In these circumstances, we would recommend the use of the internal mammary vessels followed by the thoracoacromial vessels as reliable alternative recipient sites for delayed free TRAM reconstruction.     

Perforator flap breast reconstruction using internal mammary perforator branches as a recipient site: an anatomical and clinical analysis

A variety of useful recipient sites exist for breast reconstruction with free flaps, and correct selection remains a significant decision for the surgeon. Among the main pedicles, the disadvantages of the internal mammary vessels are the necessity of costal cartilage resection and the impairment of future cardiac bypass. This study was designed to reduce morbidity and to seek alternative recipient vessels. In the anatomical part of the study, 32 parasternal regions from 16 fresh cadavers were used. The locations and components of internal mammary perforator branches were measured and a histomorphometric analysis was performed. In the clinical part of the study, 36 patients underwent 38 deep inferior epigastric perforator (DIEP) flap and two superior gluteal artery perforator flap breast reconstructions (31 immediate and four bilateral). The recipient vessels were evaluated. In the anatomical study, there were 22 perforating vessels, with 14 (63.6 percent) on the second intercostal space and 11 (50 percent) with one artery and vein. The average (+/-SD) internal and external perforator artery diameters were 598.48 +/- 176.68 microm and 848.97 +/- 276.68 microm, respectively. In the clinical study, 13 successful anastomoses (32.5 percent) were performed at the internal mammary perforator branches (second and third intercostal spaces) with 12 DIEP flaps and one superior gluteal artery perforator flap (all performed as immediate reconstructions). One case of intraoperative vein thrombosis and one case of pedicle avulsion during flap molding were observed. The anatomic and clinical studies demonstrated that the internal mammary perforator branch as a recipient site is a further refinement to free flap breast reconstruction. However, it is neither a reproducible technique nor potentially applicable in all patients. Preoperative planning between the general surgeon and the plastic surgeon is crucial to preserve the main perforator branches during mastectomy. The procedure was not demonstrable in late reconstructions. The main advantages of internal mammary perforator branches used as recipient sites are sparing of the internal mammary vessels for a possible future cardiac surgery, prevention of thoracic deformities, and reduction of the operative time by limited dissection. Despite this, limited surgical exposure, caliber incompatibility, and technical difficulties have to be considered as the main restrictions.     

Choice of recipient vessels in delayed TRAM flap breast reconstruction after radiotherapy

This study compared the use of the internal mammary and thoracodorsal recipient vessels in a uniform group of patients who underwent delayed TRAM flap reconstruction after radiotherapy, focusing on usability rates and outcomes. The authors identified 123 delayed TRAM flap patients who had undergone postmastectomy radiotherapy from a prospective database (1990 to 2001). Recipient vessel unusability rates were calculated on the basis of reports of inspection of a vessel, either by direct intraoperative dissection or by findings from color Doppler examination (internal mammary vessels only). Charts were reviewed for outcomes including flap loss, vascular complications, fat necrosis, and lymphedema; t-test and chi-square analyses were performed to compare outcomes and unusability rates, and multiple regression analysis was performed to determine factors influencing outcome. Of the 123 planned free TRAM flaps, 106 were completed as free flaps and 17 were performed as pedicled flaps because of unusable recipient vessels. Of the free flaps, 45 were anastomosed to the internal mammary vessels, 55 to the thoracodorsal vessels, and six to other vessels. The internal mammary and thoracodorsal groups did not differ significantly in body mass index, abdominal scars, smoking history, time delay between irradiation and TRAM flap reconstruction, or flap ischemia time. Radiation doses to the axilla (thoracodorsal), internal mammary chain, and supraclavicular fossa were similar between the groups. The internal mammary vessels were rejected in 11 (20 percent) of 56 cases, and the thoracodorsal vessels were rejected in 19 (26 percent) of 74 cases (p = 0.42). In cases with unusable internal mammary vessels, 46 percent (n = 5) had inadequate veins, 27 percent (n = 3) had inadequate arteries, and in 27 percent (n = 3) both vessels were inadequate. In the 19 cases with unusable thoracodorsal vessels, 84 percent (n = 16) were excessively scarred, 11 percent (n = 2) had inadequate vessels, and 5 percent (n = 1) were absent. Outcomes were similar regardless of recipient vessels used (internal mammary versus thoracodorsal): total flap loss, 0 percent versus 4 percent (p = 0.20); vascular complications, 6.7 percent versus 11 percent (p = 0.46); arm lymphedema, 4.4 percent versus 9 percent (p = 0.37); partial flap loss, 9 percent versus 6 percent (p = 0.54); and fat necrosis, 18 percent versus 15 percent (p = 0.69). Multivariate analysis revealed a trend for higher complication rates in smokers and with the use of the thoracodorsal vessels as the recipients. Overall, no discernible unusability or outcome differences were detected between the internal mammary and thoracodorsal groups.     

Use of circumflex scapular vessels as a recipient pedicle for autologous breast reconstruction: a report of 40 consecutive cases

The use of the circumflex scapular pedicle as a recipient vessel for breast reconstruction in a series of 40 consecutive cases in 37 patients is reported. There were 3 bilateral reconstructions and 34 unilateral reconstructions. Twenty-one cases were immediate reconstructions, and 19 cases were secondary reconstructions. The diameter of the artery varied from 1.5 mm to 3 mm and systematically matched with the diameter of the epigastric artery. The artery was a branch of the subscapular system in 82.5 percent of cases (33 of 40). In 17.5 percent of cases (7 of 40), the artery was a direct branch of the axillary artery. The length of available pedicle between the axillary vessel and the distal part where it can be divided (on its division between scapular and parascapular artery) was of 76 +/- 13 mm for the artery and 72 +/- 12 mm for the vein. The vein was unique in 77.5 percent of cases. The diameter was similar to the artery diameter when unique. There was a dual venous system in 21 of 40 cases (52.5 percent) but in 15 cases (37.5 percent), one of the two veins was dominant. In the seven cases for which the veins were dual and of equivalent diameter, the epigastric veins were also dual and allowed a second anastomosis. Clinically, the anastomosis was always possible on the artery. In one case of reconstruction after Halstedt mastectomy, no vein could be found, because all the veins had been ligated previously. One venous thrombosis (2.5 percent) and one arterial thrombosis were experienced. Both were treated by revised anastomoses and did not compromise late results. The circumflex scapular pedicle is a reliable and simple recipient site for breast reconstruction. It allows a unique site of dissection in immediate reconstruction and avoids division of the thoracodorsal pedicle. The technique is now used exclusively at this institution.