EXPERIMENTAL Hydrogen Peroxide Priming of the Venous Architecture: A New Technique That Reveals the Underlying Anatomical Basis for Venous Complications of DIEP, TRAM, and Other Abdominal Flaps Kwok Hao Lie, M.A., M.B.B.Chir. G. Ian Taylor, A.O., M.B.B.S., M.D. Mark W. Ashton, M.B.B.S., M.D. Parkville, Victoria, Australia
Background: Previous studies of venous anatomy lack the detail of their arterial counterparts because of (1) the technical challenge of retrograde perfusion against competent valves and (2) anterograde venous perfusion failing to adequately delineate the area of interest. We introduced a novel technique: retrograde hydrogen peroxide priming that dilates veins and renders valves incompetent, thereby facilitating complete cadaveric venous perfusion. Methods: The superficial and deep venous systems of 41 hemiabdomens and 20 hemichests of unembalmed human cadavers were primed by retrograde injection with 6% hydrogen peroxide. Specimens were then injected with lead oxide contrast, radiographed, and dissected. In five hemiabdomens, the valves were mapped by dissection. Results were compared with archival venous studies of six total body injections, six abdominal lipectomy specimens, and two intraoperative venograms of delayed transverse rectus abdominis musculocutaneous flaps. Results: Unprecedented venous filling of the anterior torso was demonstrated. Two types of superficial-to-deep venous connections were defined: large venae communicantes and small venae comitantes. Venae communicantes (>2 mm) formed major connections between large superficial and deep veins, mostly within 5 cm of the umbilicus in the abdomen, the axilla and fifth or sixth intercostal space parasternally. Seventy-four percent of venae communicantes coursed with arteries greater than 1.0 mm. Four major longitudinal valved subcutaneous pathways of the superficial inferior epigastric vein and superficial circumflex iliac vein were defined bilaterally with large avalvular transverse connections in the midline and small-caliber connections laterally that explain venous complications seen sometimes in transverse abdominal flaps. Conclusion: Retrograde hydrogen peroxide priming of veins in cadavers renders valves incompetent and facilitates detailed venous studies that help refine flap design and explain venous complications. (Plast. Reconstr. Surg. 133: 790e, 2014.)
I
n a systematic review of 17,096 deep inferior epigastric artery perforator (DIEP) flaps comprising all articles with data on DIEP flaps published between 1989 and August of From the Taylor Laboratory, Department of Anatomy and Neuroscience, University of Melbourne. Received for publication August 23, 2013; accepted November 15, 2013. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000228
790e
2011, there were 152 reported cases of total flap failure.1 Of these, 67 had reported causes, 19 (28 percent) of which were arterial in nature and 27 (40 percent) of which were venous problems. Of these venous problems, a significant subset were cases of inadequate venous outflow despite a patent venous microanastomosis. There are Disclosure: The authors have no financial interest to declare in relation to the content of this article.
www.PRSJournal.com
Volume 133, Number 6 • Venous Complications of Abdominal Flaps also reports of DIEP flaps exhibiting venous congestion that underwent reexploration, were found to have patent venous outflows, and were salvaged by venous decompression with a second venous anastomosis.2–6 However, there is no consensus on the best method for this to be achieved. There is the need, therefore, for a better understanding of the superficial and deep venous architecture of the anterior abdominal wall and the venous pathways of drainage of DIEP and other abdominal flaps. Previous studies of the head and neck using retrograde perfusion have delineated the venous anatomy in great detail because of the paucity of valves.7 In the extremities, anterograde perfusion has been successful, as sequential tourniquets to the limbs directed the flow of contrast between the superficial and deep veins.8 In the torso, however, valves have compromised previous studies. Retrograde injection often produced “blowouts” at valve sites, and anterograde injection filled only one branch of the venous tree. Perforator skin flaps are designed on a cutaneous artery and its venae comitantes that pierce the deep fascia and drain to the deep veins. However, there are two venous systems draining the integument that are important to flap success: (1) the large (>2 mm) valved subcutaneous veins (e.g., superficial inferior epigastric vein) that develop first in the embryo, drain large areas, and connect with the deep system at various fixed skin sites by means of large venae communicantes; and (2) the smaller stellate venae comitantes (e.g., of deep inferior epigastric veins) that develop later in the embryo,9,10 and whose tributaries generally match the territory of their cutaneous artery angiosome. These secondary “cutaneous venosomes” connect radially with neighboring venosomes and with tributaries of the primary venous system by means of small-caliber avalvular “oscillating” veins that allow bidirectional flow7 (Fig. 1). These connections between the primary subcutaneous and the secondary perforating veins are essential to the successful drainage of flaps that span multiple cutaneous angiosome territories11 such as the DIEP flap. They also explain why, for example, a radial forearm free flap can be drained on either the cephalic vein or the radial venae comitantes, or a free groin flap can be drained on either the superficial inferior epigastric vein or the venae comitantes of either the superficial circumflex iliac or superficial inferior epigastric artery. This article focuses on the interconnections between the various venous pathways in the abdomen.
Fig. 1. Schematic diagram of the superficial venous network. The superficial venous network comprises valved segments (blue) linked by a network of avalvular, bidirectional, small-caliber “oscillating” veins (yellow). There are two types of connections to the deep venous network: large-caliber venae communicantes (a), and small-caliber venae comitantes (b) that accompany the perforating arteries. (Reproduced from Taylor GI, Caddy CM, Watterson PA, Crock JG. The venous territories (venosomes) of the human body: Experimental study and clinical implications. Plast Reconstr Surg. 1990;86:185–213.)
MATERIALS AND METHODS Twenty anterior torsos (40 sides) and one hemitorso were harvested from 21 fresh unembalmed cadavers (aged 53 to 88 years), donated to the University of Melbourne Body Donor Programme, approved for research by the university’s research ethics committee (HEAG 0829896). The first 21 specimens spanned the subcostal margins to the groin crease, subsequently expanded to include the chest and saphenofemoral junction to complete the anterior torso. In the latter 20, the clavicles and proximal third of the arm were kept superiorly. Lateral margins were the posterior axillary line and, inferiorly, the proximal third of the thigh was included. All tissues deep to pleura and peritoneum were discarded. The internal mammary, deep inferior epigastric veins, and axillary vein branches to the breast were cannulated at their origin. The superficial inferior epigastric vein, superficial circumflex iliac vein, superficial external pudendal veins, and venae comitantes of the superficial inferior epigastric artery were cannulated at the saphenous bulb. All other tributaries of the saphenous bulb were ligated. Paired deep inferior epigastric veins and venae comitantes of the superficial inferior epigastric artery were cannulated separately.
791e
Plastic and Reconstructive Surgery • June 2014 Five to 6 ml of 6% hydrogen peroxide was pulsed gently into each cannula. Leaks were ligated, cauterized, or clamped. Oxygen released by intravascular decomposition of hydrogen peroxide is the key factor that renders venous valves incompetent, thereby facilitating good retrograde filling (Fig. 2). A pilot study using air alone filled only the main venous trunks and did not reach the peripheral venules. Contrast consisting of 100 g of lead oxide and 15 g of milk powder in 500 ml of water was injected.12 Serial radiographs were obtained before and during injection. Injection was ceased when the venous network was filled to the specimen margins or if uncontrollable extravasation obscured detail. Next, the deep inferior epigastric arteries were injected with radiolucent India ink– stained gelatin to compare arterial versus venous perforator size. In 30 specimens, tissues superficial to the deep fascia were removed and radiographed separately. In six specimens, the length of the superficial
Fig. 2. The effects of hydrogen peroxide priming on the venous network. Radiograph of an anterior torso wall specimen with the umbilicus, nipple-areola complex, and two preexisting surgical scars marked with washers and lead wire sutured onto the skin. The veins have been primed by injection with hydrogen peroxide into the cannulae, visible on the radiograph, which has distended and rendered valves incompetent in the deep inferior epigastric, internal mammary, and left superficial circumflex iliac veins (yellow arrowheads).
792e
inferior epigastric vein and underlying deep inferior epigastric vein were harvested en bloc and the cross-sections radiographed. In five, veins were dissected open to locate valves. Results were compared with archival studies of six full-body venous injections, six venous injections of fresh abdominal lipectomy specimens, and two intraoperative venograms of delayed transverse rectus abdominis musculocutaneous (TRAM) flaps.
RESULTS Deep Venous Network Inferiorly, the deep inferior epigastric veins traveled within or deep to the rectus abdominis as paired venae comitantes (2 to 5 mm) on either side of the deep inferior epigastric artery and its branches. Above the linea semicircularis, they coursed vertically in the middle of rectus abdominis. Below this, they ran obliquely and joined the external iliac veins as a single vein. In 11 of 37 cases, this was very short (2 mm) are marked with blue dots and the umbilicus is colored yellow. (Above, right) Deep tissues separate from the skin and subcutaneous fat. Note the paired internal mammary and deep inferior epigastric veins bilaterally. (Below, left) Tracing of the radiograph showing valved (blue) and avalvular (yellow) segments of the primary superficial inferior epigastric (SIEV), superficial circumflex iliac (SCIV), superficial external pudendal (SEPV), internal mammary perforating (IMV), and lateral thoracic (LT) veins. Large-caliber venae communicantes are marked as blue dots and (below, right) the territories of these veins are defined by lines drawn across avalvular junctions between them.
793e
Plastic and Reconstructive Surgery • June 2014 between the suprapubic crease and the umbilicus at a 45-degree angle in 36 of 41 specimens. Medially, two branches formed avalvular midline crossovers (2 to 3 mm diameter): a semicircular supraumbilical branch that crossed the midline midway between the umbilicus and xiphoid process (37 of 41), and an infraumbilical branch that left the superficial inferior epigastric vein within 2 cm of the suprapubic crease and passed obliquely superomedially to meet its opposite vein 2 to 4 cm below the umbilicus, forming a chevron (32 of 41 specimens; four abdomens had midline scars). In every specimen, superficial venous arcades, found at the dermal-fat junction, linked these four main branches. Superficial Circumflex Iliac Vein The superficial circumflex iliac vein was filled in 33 of 41 specimens. Valved at first, all veins observed passed obliquely along the inguinal ligament. Two centimeters medial to the anterior superior iliac spine, they turned and continued vertically, joining the thoracoepigastric vein and other superficial veins draining the chest wall and breast, by means of avalvular channels. Only small (1.5 mm) tributaries within 5 cm of the saphenous bulb that formed a diamond-shaped avalvular plexus across the midline between the suprapubic crease and 2 cm inferior to the upper margin of the pubic hairline. The entire plexus lay within 5 mm of the dermal-fat junction. The superficial external pudendal vein plexus was noted in 41 of 41 specimens. In 13 of 41 specimens, the top branch of the superficial external pudendal vein plexus drained into the superficial inferior epigastric vein rather than directly into the saphenous bulb as the superficial external pudendal vein. Venae Comitantes of the Superficial Inferior Epigastric Artery The venae comitantes of the superficial inferior epigastric artery were paired, running alongside the superficial inferior epigastric artery and its branches. They were smaller (
Background: Previous studies of venous anatomy lack the detail of their arterial counterparts because of (1) the technical challenge of retrograde perfusion against competent valves and (2) anterograde venous perfusion failing to adequately delineate the area of interest. We introduced a novel technique: retrograde hydrogen peroxide priming that dilates veins and renders valves incompetent, thereby facilitating complete cadaveric venous perfusion. Methods: The superficial and deep venous systems of 41 hemiabdomens and 20 hemichests of unembalmed human cadavers were primed by retrograde injection with 6% hydrogen peroxide. Specimens were then injected with lead oxide contrast, radiographed, and dissected. In five hemiabdomens, the valves were mapped by dissection. Results were compared with archival venous studies of six total body injections, six abdominal lipectomy specimens, and two intraoperative venograms of delayed transverse rectus abdominis musculocutaneous flaps. Results: Unprecedented venous filling of the anterior torso was demonstrated. Two types of superficial-to-deep venous connections were defined: large venae communicantes and small venae comitantes. Venae communicantes (>2 mm) formed major connections between large superficial and deep veins, mostly within 5 cm of the umbilicus in the abdomen, the axilla and fifth or sixth intercostal space parasternally. Seventy-four percent of venae communicantes coursed with arteries greater than 1.0 mm. Four major longitudinal valved subcutaneous pathways of the superficial inferior epigastric vein and superficial circumflex iliac vein were defined bilaterally with large avalvular transverse connections in the midline and small-caliber connections laterally that explain venous complications seen sometimes in transverse abdominal flaps. Conclusion: Retrograde hydrogen peroxide priming of veins in cadavers renders valves incompetent and facilitates detailed venous studies that help refine flap design and explain venous complications. (Plast. Reconstr. Surg. 133: 790e, 2014.)
I
n a systematic review of 17,096 deep inferior epigastric artery perforator (DIEP) flaps comprising all articles with data on DIEP flaps published between 1989 and August of From the Taylor Laboratory, Department of Anatomy and Neuroscience, University of Melbourne. Received for publication August 23, 2013; accepted November 15, 2013. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000228
790e
2011, there were 152 reported cases of total flap failure.1 Of these, 67 had reported causes, 19 (28 percent) of which were arterial in nature and 27 (40 percent) of which were venous problems. Of these venous problems, a significant subset were cases of inadequate venous outflow despite a patent venous microanastomosis. There are Disclosure: The authors have no financial interest to declare in relation to the content of this article.
www.PRSJournal.com
Volume 133, Number 6 • Venous Complications of Abdominal Flaps also reports of DIEP flaps exhibiting venous congestion that underwent reexploration, were found to have patent venous outflows, and were salvaged by venous decompression with a second venous anastomosis.2–6 However, there is no consensus on the best method for this to be achieved. There is the need, therefore, for a better understanding of the superficial and deep venous architecture of the anterior abdominal wall and the venous pathways of drainage of DIEP and other abdominal flaps. Previous studies of the head and neck using retrograde perfusion have delineated the venous anatomy in great detail because of the paucity of valves.7 In the extremities, anterograde perfusion has been successful, as sequential tourniquets to the limbs directed the flow of contrast between the superficial and deep veins.8 In the torso, however, valves have compromised previous studies. Retrograde injection often produced “blowouts” at valve sites, and anterograde injection filled only one branch of the venous tree. Perforator skin flaps are designed on a cutaneous artery and its venae comitantes that pierce the deep fascia and drain to the deep veins. However, there are two venous systems draining the integument that are important to flap success: (1) the large (>2 mm) valved subcutaneous veins (e.g., superficial inferior epigastric vein) that develop first in the embryo, drain large areas, and connect with the deep system at various fixed skin sites by means of large venae communicantes; and (2) the smaller stellate venae comitantes (e.g., of deep inferior epigastric veins) that develop later in the embryo,9,10 and whose tributaries generally match the territory of their cutaneous artery angiosome. These secondary “cutaneous venosomes” connect radially with neighboring venosomes and with tributaries of the primary venous system by means of small-caliber avalvular “oscillating” veins that allow bidirectional flow7 (Fig. 1). These connections between the primary subcutaneous and the secondary perforating veins are essential to the successful drainage of flaps that span multiple cutaneous angiosome territories11 such as the DIEP flap. They also explain why, for example, a radial forearm free flap can be drained on either the cephalic vein or the radial venae comitantes, or a free groin flap can be drained on either the superficial inferior epigastric vein or the venae comitantes of either the superficial circumflex iliac or superficial inferior epigastric artery. This article focuses on the interconnections between the various venous pathways in the abdomen.
Fig. 1. Schematic diagram of the superficial venous network. The superficial venous network comprises valved segments (blue) linked by a network of avalvular, bidirectional, small-caliber “oscillating” veins (yellow). There are two types of connections to the deep venous network: large-caliber venae communicantes (a), and small-caliber venae comitantes (b) that accompany the perforating arteries. (Reproduced from Taylor GI, Caddy CM, Watterson PA, Crock JG. The venous territories (venosomes) of the human body: Experimental study and clinical implications. Plast Reconstr Surg. 1990;86:185–213.)
MATERIALS AND METHODS Twenty anterior torsos (40 sides) and one hemitorso were harvested from 21 fresh unembalmed cadavers (aged 53 to 88 years), donated to the University of Melbourne Body Donor Programme, approved for research by the university’s research ethics committee (HEAG 0829896). The first 21 specimens spanned the subcostal margins to the groin crease, subsequently expanded to include the chest and saphenofemoral junction to complete the anterior torso. In the latter 20, the clavicles and proximal third of the arm were kept superiorly. Lateral margins were the posterior axillary line and, inferiorly, the proximal third of the thigh was included. All tissues deep to pleura and peritoneum were discarded. The internal mammary, deep inferior epigastric veins, and axillary vein branches to the breast were cannulated at their origin. The superficial inferior epigastric vein, superficial circumflex iliac vein, superficial external pudendal veins, and venae comitantes of the superficial inferior epigastric artery were cannulated at the saphenous bulb. All other tributaries of the saphenous bulb were ligated. Paired deep inferior epigastric veins and venae comitantes of the superficial inferior epigastric artery were cannulated separately.
791e
Plastic and Reconstructive Surgery • June 2014 Five to 6 ml of 6% hydrogen peroxide was pulsed gently into each cannula. Leaks were ligated, cauterized, or clamped. Oxygen released by intravascular decomposition of hydrogen peroxide is the key factor that renders venous valves incompetent, thereby facilitating good retrograde filling (Fig. 2). A pilot study using air alone filled only the main venous trunks and did not reach the peripheral venules. Contrast consisting of 100 g of lead oxide and 15 g of milk powder in 500 ml of water was injected.12 Serial radiographs were obtained before and during injection. Injection was ceased when the venous network was filled to the specimen margins or if uncontrollable extravasation obscured detail. Next, the deep inferior epigastric arteries were injected with radiolucent India ink– stained gelatin to compare arterial versus venous perforator size. In 30 specimens, tissues superficial to the deep fascia were removed and radiographed separately. In six specimens, the length of the superficial
Fig. 2. The effects of hydrogen peroxide priming on the venous network. Radiograph of an anterior torso wall specimen with the umbilicus, nipple-areola complex, and two preexisting surgical scars marked with washers and lead wire sutured onto the skin. The veins have been primed by injection with hydrogen peroxide into the cannulae, visible on the radiograph, which has distended and rendered valves incompetent in the deep inferior epigastric, internal mammary, and left superficial circumflex iliac veins (yellow arrowheads).
792e
inferior epigastric vein and underlying deep inferior epigastric vein were harvested en bloc and the cross-sections radiographed. In five, veins were dissected open to locate valves. Results were compared with archival studies of six full-body venous injections, six venous injections of fresh abdominal lipectomy specimens, and two intraoperative venograms of delayed transverse rectus abdominis musculocutaneous (TRAM) flaps.
RESULTS Deep Venous Network Inferiorly, the deep inferior epigastric veins traveled within or deep to the rectus abdominis as paired venae comitantes (2 to 5 mm) on either side of the deep inferior epigastric artery and its branches. Above the linea semicircularis, they coursed vertically in the middle of rectus abdominis. Below this, they ran obliquely and joined the external iliac veins as a single vein. In 11 of 37 cases, this was very short (2 mm) are marked with blue dots and the umbilicus is colored yellow. (Above, right) Deep tissues separate from the skin and subcutaneous fat. Note the paired internal mammary and deep inferior epigastric veins bilaterally. (Below, left) Tracing of the radiograph showing valved (blue) and avalvular (yellow) segments of the primary superficial inferior epigastric (SIEV), superficial circumflex iliac (SCIV), superficial external pudendal (SEPV), internal mammary perforating (IMV), and lateral thoracic (LT) veins. Large-caliber venae communicantes are marked as blue dots and (below, right) the territories of these veins are defined by lines drawn across avalvular junctions between them.
793e
Plastic and Reconstructive Surgery • June 2014 between the suprapubic crease and the umbilicus at a 45-degree angle in 36 of 41 specimens. Medially, two branches formed avalvular midline crossovers (2 to 3 mm diameter): a semicircular supraumbilical branch that crossed the midline midway between the umbilicus and xiphoid process (37 of 41), and an infraumbilical branch that left the superficial inferior epigastric vein within 2 cm of the suprapubic crease and passed obliquely superomedially to meet its opposite vein 2 to 4 cm below the umbilicus, forming a chevron (32 of 41 specimens; four abdomens had midline scars). In every specimen, superficial venous arcades, found at the dermal-fat junction, linked these four main branches. Superficial Circumflex Iliac Vein The superficial circumflex iliac vein was filled in 33 of 41 specimens. Valved at first, all veins observed passed obliquely along the inguinal ligament. Two centimeters medial to the anterior superior iliac spine, they turned and continued vertically, joining the thoracoepigastric vein and other superficial veins draining the chest wall and breast, by means of avalvular channels. Only small (1.5 mm) tributaries within 5 cm of the saphenous bulb that formed a diamond-shaped avalvular plexus across the midline between the suprapubic crease and 2 cm inferior to the upper margin of the pubic hairline. The entire plexus lay within 5 mm of the dermal-fat junction. The superficial external pudendal vein plexus was noted in 41 of 41 specimens. In 13 of 41 specimens, the top branch of the superficial external pudendal vein plexus drained into the superficial inferior epigastric vein rather than directly into the saphenous bulb as the superficial external pudendal vein. Venae Comitantes of the Superficial Inferior Epigastric Artery The venae comitantes of the superficial inferior epigastric artery were paired, running alongside the superficial inferior epigastric artery and its branches. They were smaller (
