The influence of postoperative vessel occlusion on the viability of free microvascular skin-fat flaps and island flaps in rats

Using a method of vascular pedicle ligation without skin incision, some minor differences were found between the progress of neovascularization in microvascular free skin-fat flaps and island flaps, both with and without 1-hr ischemia in rats. Viable flap areas were established following vascular pedicle ligation (both vessels or only artery or vein), on the third day after flap replantation in both island and free flaps. Vascular pedicle occlusion on the fifth day after operation resulted in complete survival of free flaps only. Island flaps survived completely following vascular pedicle ligation on the seventh postoperative day. A clear correlation existed between the timing of vascular pedicle ligation and the necrosis and shrinkage rates. In free flaps, clinical signs of viability disturbances were seen earlier in single or total vascular pedicle occlusions, on the third or fourth day compared with island flaps. Insufficiency of the pedicle vein in island or free flaps was tolerated earlier than arterial or total pedicle occlusion. Venous occlusion three days after flap replantation caused significantly higher necrosis and shrinkage rates (p less than 0.05) in free flaps than in island flaps.     

Use of free interpositional vein grafts as pedicles for prefabrication of skin flaps

Pedicles created from a long vein graft increase the scope and applications of prefabricated skin flaps. This study reports the survival and pattern of neovascularization of lower abdominal skin flaps in rabbits based on a pedicle formed by interposition of a long vein graft between the divided ipsilateral femoral artery and vein. Flaps were elevated 2–5 weeks after pedicle implantation and the surviving area quantitated and vascular patterns examined 1 week later. Only 8 out of 35 flaps were greater than 50% alive, the most frequent cause of flap failure being pedicle non-patency. If the pedicle remained patent, complete flap survival was possible as early as 2 weeks after implantation. In non-patent pedicles, recanalization or formation of a new vascular network may, given at least 4 weeks, be sufficient to ensure partial flap survival. The findings indicate that implantation of a long, skeletonized vein graft is an unreliable method of prefabrication of abdominal skin flaps in this model. © 1994 Wiley-Liss, Inc.     

Neovascularized free fat flaps: an experimental model

Using the concepts of neovascularization and microsurgical free transfer, two types of free fat flaps were developed in a rat model. The technique for construction of the flap was to use either a free fat graft or a pedicle fat flap that were then wrapped around a predetermined vascular bundle, subsequently to neovascularize the enveloping fat tissue. Once adequate neovascularization occurred, these flat flaps could then be harvested on a neopedicle consisting of the enveloped artery and vein. These preparations were later examined for viability, and then freely transferred to the contralateral femoral vessels. After transfer, the flaps were again examined for viability and prepared for histologic examination. Both of these studies revealed that successful neovascularized free fat flaps could be created, employing techniques of neovascularization, and that those created using a pedicle fat flap in the initial stage reliably succeeded in neovascularization and free transfer, in contrast with the less successful free fat graft technique.     

Use of fascial grafts as an interface in flap prefabrication: an experimental study.

An experimental study was conducted to investigate whether a fascial graft can be used as an interface between a vascular pedicle and target tissue to augment tissue survival in a prefabricated flap. Thirty-six male Sprague-Dawley rats were divided into three experimental groups according to the type of the recipient bed prepared for the vascular implantation. The left saphenous vascular pedicle was used as the vascular source. A 9 x 9-cm inferiorly based peninsular abdominal flap was elevated in each animal. In group I, the pedicle was tacked beneath the abdominal flap, in which the epigastric fascial layer was untouched. In group II, a 3 x 5-cm graft of epigastric fascia was harvested from the abdominal flaps under loupe magnification. The graft was sutured back into its original position after a 180-deg rotation. The vascular pedicle was then implanted just beneath the center of the fascial graft. In group III, the same size of epigastric fascia was removed in the same manner as group II, exposing the subcutaneous layer for pedicle implantation. Four weeks later, abdominal flaps were raised as island flaps connected only to the saphenous pedicle and were sutured in place. Flap viability was assessed visually on day 7. Overall, the ultimate flap survival in group I was the largest, with some necrotic areas at the periphery of the flaps. In group II, flap survival was typically centralized over the fascial graft, and crescent-shaped necrosis was noted superiorly. In group III, an almost linear pattern of survival overlying the vascular pedicle was observed. The mean surviving flap area of group I (12.13 +/- 1.615 cm2) was statistically greater than that of group II (8.83 +/- 0.663 cm2, p < 0.001) and group III (6.3 +/- 0.815 cm2; p < 0.001). There was a statistically significant difference between the mean flap survival in groups II and III (p < 0.001). Vascular arborization was examined by microangiography, and specimens were processed for histological staining. In group II, vascularization was distributed in a larger area along the fascial graft in comparison with limited vascularization around the pedicle in group III. In this study it was revealed that the interposition of a fascial graft as an interface between the vascular source and the target tissue seems to increase the size of the prefabricated flap.     

Prefabrication of combined composite (chimeric) flaps in rats

The purpose of this study was to prefabricate a new combined composite (chimeric) flap that consists of four different tissues. The tissues were prefabricated around two independent pedicles that ultimately join as a single main pedicle. In the inguinal area of 36 rats, the saphenous and the superficial inferior epigastric (SIE) pedicles were dissected and prepared as vascular carriers. A fascial graft and a local muscle flap were wrapped around the saphenous pedicle. The SIE pedicle was then implanted under the abdominal skin to supply a future skin flap. An ear cartilage graft was also inserted under the abdominal skin and adjacent to the implanted SIE pedicle. After allowing 2-, 4-, 6-, and 12-week prefabrication periods in different groups of nine animals, the prefabricated tissues were raised around two pedicles nourished by the femoral pedicle and then transferred. Flap survival was assessed by observation, microangiography and histology. The skin flaps showed survival rates of 52 +/- 17% (mean +/- standard error of the mean), 64 +/- 16%, 86 +/- 11%, and 100 +/- 0% of the total areas in the 2-, 4-, 6-, and 12-week prefabricated flaps respectively. None of the control grafts that were prepared on the contralateral side survived totally. A significant difference was found between the 12- and 2-week (p < 0.008), 12- and 4-week (p < 0.02), and 6- and 2-week (p < 0.05) prefabrication groups. Histologically, fascial and cartilage grafts, and portions of muscle were viable in the 2- and 4-week groups. Also, noticeable necrosis was found in the skin flaps in these groups. The muscle showed mild (at 2, 4, and 6 weeks) and moderate (at 12 weeks) atrophy. After prefabrication for 6 weeks, all tissues demonstrated good survival. This study shows that a combined composite flap can be prefabricated successfully in rats after a 6-week period of prefabrication.     

Bone flap prefabrication: an experimental study in rabbits

The usual method to prefabricate a bone flap is to harvest a nonvascularized bone graft and to implant the artery and vein bundle between segments of bone graft. The basic problem of this method is sacrificing an artery for prefabrication. Another method for creating flap donor sites without using an artery is venous flap prefabrication. There are a few articles describing bone flap prefabrication, and these include implantation of both artery and vein as a vascular bundle. Also, there is no experimental study in the literature using a vein or an arterialized vein pedicle for bone flap prefabrication. As an experimental model for bone flap prefabrication, the rabbit ear vascular model was chosen. For the experiments 3 groups were formed. Each group contained 5 rabbits. In the first experimental group a vein was implanted between the halves of bone graft. In the second experimental group an arterialized vein was implanted between the halves of bone graft. To compare the viability of the bone graft of the 2 prefabrication groups, a bone graft was implanted into the subcutaneous pocket of the posterior auricular area in the third group. The authors examined 5 rabbits in each group by microangiography at the end of 6 weeks except for group 3. On microangiographic analysis, groups 1 and 2 showed patency of the vascular pedicle. There was no difference between these 2 groups from the point of view of vascular patency and bone appearance. Bone scintigraphy was performed for 5 rabbits in each group. On bone scintigraphic scans, the bone component of the flaps was visualized in groups 1 and 2, but not in group 3. A quantitative analysis of images was performed by drawing symmetric spherical regions of interest (ROIs) over both the implanted area and cranial bone. The uptake ratios were computed by dividing the mean counts in the implanted ROI by mean counts in the cranial bone ROI. The mean value was 0.86 +/- 0.02 in group 1 and 0.86 +/- 0.04 in group 2. A statistically significant uptake difference was not seen between venous and arterialized venous groups (P < 0.01). Histologic examination was performed all rabbits in each group, and demonstrated that the bony component was viable, showing osteocytes containing lacunae, osteoblasts along bony trabeculae, and vascular channels in groups 1 and 2. In group 3, the bony architecture of the graft was still apparent, but all bone within it was dead. There were no significant microangiographic, histologic, and scintigraphic differences between the 2 experimental methods.     

Prefabrication of skin flaps using vein grafts: an experimental study in rabbits.

Prefabrication provides a new method for creating donor sites which are not limited by natural vascular territories. There are several methods for prefabrication, and these include implantation of greater omentum, blood vessels or muscle flaps. Based on the concept that an arterio-venous (A-V) shunt results in sufficient neovascularisation to support a free flap, we used a rabbit model to investigate the characteristics of these flaps. Prefabrication of an abdominal wall donor site was performed using the left epigastric vein in 20 male New Zealand white rabbits. An 8 x 10 cm skin flap was elevated 10 days after prefabrication, either as an island or a free flap. Survival of the skin flaps exceeded 93% and was independent of position of the vascular pedicle, direction of blood flow, or nature of the flap (island or free flap). Angiograms showed a very rich neovascularisation within the prefabricated flap.     

Flap pedicle vena comitant as a vein graft donor source

A vena comitant segment harvested from a flap's pedicle can be used as an interpositional vein graft in selected microvascular cases. When a vascular pedicle includes paired venae comitantes, one of these can prove suitable for use as a vein graft while still allowing for venous outflow of the flap. An additional operative site and procedure to harvest a vein graft can be avoided if a vena comitant segment can be used. We present eight cases in which pedicle vena comitant segments were used as interpositional vein grafts. In six cases, vena comitant grafts were used to supercharge or augment venous outflow in transverse rectus abdominis myocutaneous (TRAM) flaps used for breast reconstruction. A vena comitant graft was used to revise the venous anastomosis in one deep inferior epigastric perforator (DIEP) flap. The arterial anastomosis was revised with a vena comitant graft in a gracilis muscle free flap. Our experience demonstrates the viability and utility of using the flap pedicle's vena comitant as a source of vein graft in selected cases.     

Prefabrication of large fasciocutaneous flaps using an isolated arterialised vein as implanted vascular pedicle.

Flap prefabrication represents a new trend in microsurgical tissue transfer. Based on the concept of neovascularisation, in Chinchilla Bastard rabbits (n=40), an isolated venous pedicle dissected from the femoral and saphena magna vein was arterialised by end-to-end anastomosis to the femoral artery at the inguinal ligament. This arterialised venous loop was implanted beneath a random-pattern vascularised abdominal fasciocutaneous flap as large as 8 x 15 cm(2) to investigate the development of neovascularisation at various evaluating times of 4, 8, 12, 16 and 20 days. To prevent neoangiogenesis from occurring between the underlying vascular bed and abdominal flap, a silicone sheet with the corresponding dimension of 8 cm x 15 cm x 0.25 mm was placed and fixed on the abdominal wall. The flap viability and the neovascularisation process in the prefabricated abdominal skin flaps were evaluated by macroscopic observation, blood analysis, selective microangiography and histology. The experimental results showed that newly formed vessels originating from the implanted isolated venous pedicle were evident on the angiograms 4 days after pedicle implantation. In the 8- and 12-day groups, newly formed vessels became larger and some were connected to the originally available vasculature in the abdominal fasciocutaneous flaps. In the 20-day group, entire flaps were perfused by the blood flow supplied from the newly implanted venous pedicles through newly formed vessels and their vascular connections. This study indicated that large flap prefabrication can be created by implantation of an isolated arterialised venous pedicle into a random-pattern vascularised fasciocutaneous flap. Twenty days appears to be the minimal length of time required after arterialised venous pedicle implantation for the maturation of neovascularisation in the prefabricated flap.     

Customized prefabricated neovascularized free flaps

Neovascularization of tissues into which a vascular pedicle has been implanted can result in the creation of a flap, or free flap, which is supported by those vessels as a neopedicle. This phenomenon allows the construction of customized prefabricated free flaps from tissue without restriction to naturally occurring vascular territories.     

Arterial flap (II): an experimental study in rabbits

Skin flaps are normally characterized by arterial inflow and venous outflow. However, there are several reports about experimental and clinical applications of venous and arterialized venous flaps. On the other hand, some studies evaluated the importance of different pedicle types in prefabricated flaps. It was discovered by chance that a prefabricated free flap, having arterial-only inflow, could be used successfully in eyelid reconstruction. An experimental study on arterial flaps in rabbits showed that an arterial-only inflow was adequate for flap viability until the establishment of venous neovascularization.Presented at the 14th Congress of Turkish Plastic and Reconstructive Surgery, Ankara, Turkey, October 1992     

Vascular resistance in human muscle flaps

Important differences in free muscle flap survival have been reported in the setting of long arterial and venous vein grafts. The authors provide insight into the etiology of flap failure by addressing the following question: Do differences in flap type result in clinically significant different vascular resistances and consequently anastomotic patency? A total of 15 human flaps were studied intraoperatively: 9 gracilis, 3 rectus abdominis, and 3 latissimus dorsi. The muscle was isolated on a single pedicle and hemodynamic stability was ensured. The venous pedicle was then divided. A timed collection of effluent was used to determine flow. Vascular resistance was calculated by dividing the change in pressure by the flow, and standardizing this for temperature and hematocrit. Average vascular resistance and standard deviation for the gracilis, rectus, and latissimus flaps was 10.34 +/- 7.77 mmHg per milliliter per minute, 2.79 +/- 1.50 mmHg per milliliter per minute, and 3.17 +/- 1.05 mmHg per milliliter per minute respectively. An inverse relationship between muscle vascular resistance and flap mass was found (p < 0.001). This indicates that larger muscles have less vascular resistance. The decreased resistance gives rise to higher flow rates and, as a result, potentially improved vein graft patency. The clinical implication is that a larger flap should be used when high flow-through is critical. The role of flap vascular territory makeup continues to be pursued.     

Four types of venous flaps for wound coverage: a clinical appraisal

Venous flaps were used for coverage of hand wounds over exposed bones, joints, and tendons in 28 patients. Four types of operations were performed. Among them type IV was the best. It included the advantages of perfusion in types I and III, but excluded their disadvantages. The viability of venous flaps was confirmed. Clinical observation showed that a venous flap is not merely a composite graft. The presence of a vascular network in the flap helps to ensure initial survival before the establishment of neovascularization between the venous flap and the recipient site. Partial loss of a flap was observed in two cases and complete failure was seen in one case. Viability versus flap size and modality of perfusion are analyzed. With total venous perfusion, small venous flaps usually survive better than large ones. For large venous flaps, arterialized venous perfusion is better than total venous perfusion. Venous flaps are useful for wound coverage of fingers and hand, but they do not replace cross-finger flaps or other conventional flaps when these simpler flaps are available.     

Experimental tracheal reconstruction with free prefabricated arterialized venous flap

Summary A technique of tracheal reconstruction in the rat model has been developed using the microvascular transfer of a prefabricated compound flap. This flap consists of a tubed cartilage graft incorporated within an arterialized venous flap created in the abdominal skin. Twenty-five Wistar male rats were operated on in three stages. First, the subcutaneous venous network was arterialized by doing an arteriovenous anastomosis. Seven days later, a tube of autologous cartilage was made from the xiphoid cartilage. This neotrachea was buried in a subcutaneous pocket within the arterialized abdominal area. Eight weeks later, the free flap containing the neotrachea, a skin island and a vascular pedicle was transferred to the neck, doing the afferent and efferent anastomosis to the carotid artery and the jugular vein. The neotrachea was interposed into a 1 cm tracheal defect. Microangiography and pathological studies were done in all cases. The pathologic study revealed a stable, viable cartilage tube with an internal lining of fibrous periprosthetic capsule. Angiography demonstrated enhanced vascularization in the abdominal flap. Flap viability was 85%. Postoperative survival ranged from 12 h to 5 days. In all cases, the animal died with a patent airway.     

兔股动脉和静脉预构轴型扩张皮瓣的实验观察

目的 通过兔股动脉、静脉预构轴型扩张皮瓣的微循环血流晕动态变化、光镜下结构的改变及其成活面积,为预构轴型扩张皮瓣的临床应用提供依据.方法 选择新西兰白兔40只,随机分为4组:预构轴型扩张皮瓣组、预构轴型不扩张皮瓣组、单纯预构轴型皮瓣组及无蒂游离皮瓣组,每组4只,前2组股动脉、静脉移位后,预构轴型扩张皮瓣组、预构轴型不扩张皮瓣组分别在肉膜深面置入容量为50 ml长方形皮肤软组织扩张器,预构轴型扩张皮瓣组7 d后开始注水;无蒂游离皮瓣组为对照组,未采取预构及扩张处理.定期对4组皮肤进行微循环血流量检测,并取样进行光镜观察.预构术后52 d,前3组形成以预构股动静脉血管束为蒂的岛状皮瓣,游离皮瓣组则形成无蒂游离皮瓣后均原位缝合,观察其成活面积.结果 预构轴型扩张皮瓣组较其他组微循环血流量增加,成活面积大[(97.54±2.73)%],光镜下改变显著(P<0.05).结论 扩张术能促进预构轴型皮瓣的血管化进程,明显增大预构轴型皮瓣成活面积,增加其移植的安全性.     

Gene therapy with adenovirus-mediated VEGF enhances skin flap prefabrication

We investigated the feasibility in rats of enhancing skin-flap prefabrication with subdermal injections of adenovirus-encoding vascular endothelial growth factor (Ad-VEGF). The left saphenous vascular pedicle was used as a source for vascular induction. A peninsular abdominal flap (8 x 8 cm) was elevated as distally based, keeping the epigastric vessels intact on both sides. After the vascular pedicle was tacked underneath the abdominal flap, 34 rats were randomly divided into three groups according to treatment protocol. The implantation site around the pedicle was injected with Ad-VEGF in group I (n = 10), with adenovirus-encoding green fluorescent protein (Ad-GFP) in control group I (n = 14), and with saline in control group II (n = 10). All injections were given subdermally at four points around the implanted vessel by an individual blinded to the treatment protocol. The peninsular flap was sutured in its place, and 4 weeks later, an abdominal island flap based solely on the implanted vessels was elevated. The prefabricated island flap was sutured back, and flap viability was evaluated on day 7. Skin specimens were stained with hematoxylin and eosin for histological evaluation. In two rats from each group, microangiography was performed to visualize the vascularity of the prefabricated flaps. There was a significant increase in survival of prefabricated flaps in the Ad-VEGF group compared to the control groups: Ad-VEGF, 88.9 +/- 6.1% vs. Ad-GFP, 65.6 +/- 9.4% (P < 0.05) and saline, 56.0 +/- 3.4% (P < 0.05). Sections from four prefabricated flaps treated with Ad-GFP revealed multiple sites of shiny deposits of green fluorescent protein around the area of local administration 1 day and 3 weeks after gene therapy. Histological examination done under high-power magnification (x400) with a light microscope revealed increased vascularity and mild inflammation surrounding the implanted vessel in all groups. However, we were unable to demonstrate any significant quantitative difference with respect to vascularity and inflammatory infiltrates in prefabricated flaps treated with Ad-VEGF compared with controls. Microangiographic studies showed increased vascularity around the implanted pedicle, which was similar in all groups. However, vascularization was distributed in a larger area in the prefabricated flaps treated with Ad-VEGF. In this study, the authors demonstrated that adenovirus-mediated VEGF gene therapy increased the survival of prefabricated flaps, suggesting that it may allow prefabrication of larger flaps and have the potential to reduce the time required for flap maturation.     

Some factors affecting the survival of venous flaps: an experimental study

The inferior epigastric venous flap of the rat was chosen for experimental studies of vascular flow alterations. The long saphenous vein was not selected for use; preliminary studies involving forced retrograde injection demonstrated that it drains blood primarily from the leg and foot, not from its overlying skin. Eight different flap designs were studies: group A, saphenous flap, simple isolation with pedicle; group B, inferior epigastric flap, simple isolation with pedicle; group C, inferior epigastric flap, simple isolation with interpositioned Silastic sheet; group D, cross-transferred venous pressure arterial flap; group E, cross-transferred, arterialised venous flap; group F, arterialized venous flap, in which arterial blood was shunted away via the main vein: group G, cross-transferred venous pressure venous flap; and group K. control, nonvascularized flap. Three more groups (H, I, J) were introduced by repeating the flap designs of groups E, F, and G but preceded by forced perfusion of the venous system using 5 cc of normal saline at 200 mm Hg. The best results were obtained with groups B, C, E, H, and I. The other flaps necrosed. The results are discussed based on the assumption that true retrograde venous blood flow does occur.     

Effects of hyperbaric oxygen treatment and heparin on the survival of unipedicled venous flaps: an experimental study in rats

The effects of hyperbaric oxygen (HBO) and heparin on the survival of the rat inferior epigastric venous flap were investigated. Preliminary transcutaneous oxygen measurements showed that partial oxygen pressure values of venous flaps increased at 2.5 ATA pressure while inhaling 100% oxygen. During the experiment, 128 venous flaps of two different sizes and 50 composite grafts were prepared bilaterally in 89 rats. Perivenous areolar tissue was removed from the pedicle vein in all flaps. Half of the venous flaps were isolated from the wound bed. Initial flap perfusion was tested by fluorescein injection during flap elevation. Four treatment groups were created: control, heparin, HBO, and HBO+heparin. After 6 days of treatment, the mean surviving flap area was calculated for each group. Surviving flaps were reelevated, final flap perfusion was tested by fluorescein injection, and flaps were harvested for histological examination. The mean survival rates of the HBO (26.56%) and the HBO+heparin (36.87%) groups were significantly higher than the control (0%) and the heparin (0%) groups (p<0.01). None of the composite grafts survived. Smaller flaps and nonisolated flaps survived better, although not significantly (p>0.05). Veins were enlarged both clinically and histologically. Fluorescein uptake was delayed during initial flap elevation but was normal during reelevation. These findings imply that the rat inferior epigastric venous flap may be an ischemic flap with capillary circulation through a single venous pedicle, but it needs HBO treatment to survive, especially during the acute period. Heparin treatment, reducing the flap size, and presence of a vascular wound bed also improve survival rates.     

Delayed prefabricated arterial composite venous flaps: an experimental study in rabbits

Prefabrication of composite arteriovenous flaps with implantation of an autologous graft (cartilage) or an alloplastic material (porous polyethylene) was studied in 40 rabbits. Abdominal flaps based on bilateral epigastric pedicles were elevated. An ear cartilage graft or a porous polyethylene implant was inserted under the flap. Two weeks after the operation, 10 flaps with cartilage graft and 10 flaps with porous polyethylene were raised, converted to arteriovenous flaps, and resutured in place in the experimental groups. In the other 20 rabbits of the control groups, the flaps (10 with cartilage graft and 10 with porous polyethylene) were raised and resutured in place as conventional axial flaps. At the end of the second and fourth week postoperatively, samples were obtained from the flap tissues (including a part of the graft or implantation material) and were prepared for histologic examination in all rabbits. The viable areas of all flaps were assessed at the end of fourth week after the second operation. The mean survival rates were 99.4%, 99.7%, 99.5%, and 99.8% in the arteriovenous and control flaps prefabricated with cartilage graft and the arteriovenous and control flaps prefabricated with porous polyethylene respectively. The features of wound healing in the experimental and control groups were similar. The study showed that arteriovenous perfusion can nourish a prefabricated flap containing an implanted material (autologous or alloplastic) and these 2-week delayed composite flaps have a similar survival rate to delayed prefabricated conventional axial flaps.     

Effect of vein grafting on the survival of microvascularly transplanted muscle flaps

Since vein grafting is often required during transplantation of free muscle flaps but is associated with a higher failure rate than those flaps transplanted with primary anastomoses, we sought to compare primary repair with the use of vein grafting in an experimental setting. We transplanted the gracilis muscle to the contralateral side in 98 rats using four different methods of vessel repair. In the Control group (n = 28), both femoral vessels were anastomosed primarily. In Experimental Group 1 (n = 25), the femoral artery was anastomosed with an epigastric vein graft and the femoral vein was anastomosed primarily. In Experimental Group 2 (n = 25), the femoral vein was anastomosed with a femoral vein graft and the femoral artery was anastomosed primarily. In Experimental Group 3 (n = 20), both femoral vessels were anastomosed with vein grafts. The Control and Experimental Groups 1–3 survival rates were 89.3, 76.0, 84.0, and 70.0%, respectively; none of the experimental group survival rates was significantly different from that of the control (P < 0.5). This study demonstrates that the use of size-matched, interpositional vein grafts in the arterial or venous pedicle of the rat gracilis muscle flap during transplantation did not significantly decrease flap survival as compared to primary arterial or venous anastomoses. If the observed failure rate persisted with expansion of the study groups to 60–100 animals each, the failure rate of flaps with vein grafts would be significantly lower and comparable to failure rates reported in some clinical series. The large numbers necessary to significantly show this decrease make this model impractical for further studies © 1997 Wiley-Liss, Inc. MICROSURGERY 17:512–516 1996