Search

We have shown previously that when a flap is delayed, the maximal anatomic effect on the arterial side of the circulation is focused at the level of the reduced-caliber choke vessels that link adjacent vascular territories. These anastomotic vessels were noted to increase in size to the dimension of true anastomoses. However, we did not define when this occurred. The present experiment therefore was designed to elucidate the chronologic sequence of events that occur in the "choke" vessels using a rabbit flank skin flap as the experimental model. A long two-territory osteocutaneous flank flap was designed on one side of each rabbit (n = 30), with the opposite unoperated side serving as a control. The flap was elevated and sutured back in place. At various times postoperatively, namely, 1 (n = 2), 2 (n = 2), 3 (n = 2), 4 (n = 2), 6 (n = 2), 8 (n = 2), 12 (n = 2), 24 (n = 2), 48 (n = 2), and 72 (n = 2) hours and 7 days (n = 10), the animals were sacrificed, and total-body arteriograms were obtained using a lead oxide mixture. The density and size of the choke arteries between the territories in the flap and their counterparts on the control side were assessed by histologic analysis (n = 3). We observed a sequential dilation of choke vessels during the delay period. In particular, we found that the vessels increased rapidly in size between the 48- and 72-hour studies.(ABSTRACT TRUNCATED AT 250 WORDS)

In 1987, the results of a series of total-body investigations of the arterial system of the skin and underlying deep tissues were published. This resulted in the angiosome concept. In 1990, a similar series of studies of the venous network was published. In both investigations, it was noted that "vessels hitchhike with nerves." This anatomic study analyzes these neurovascular relationships in the skin and in the underlying muscles. Seven fresh human cadavers and nine animals were studied over a 2-year period. The entire integument of each and a total of 538 human and 72 animal muscles were removed and analyzed. Either the arterial or the venous system was injected with a radiopaque lead oxide mixture, and the dissected nerves were labeled with fine wires, being segregated later by a subtraction radiography technique. The results of these investigations are presented, with special emphasis placed on the design of long axial skin flaps placed along neurovascular systems and their relationship with the current design of skin flaps. The muscles are classified according to their extrinsic and intrinsic neurovascular supplies, and suggestions are made as to how they may or may not be subdivided into functional units for local and distant transfer. The cutaneous nerves, as well as the motor nerves of the muscles, were invariably accompanied by a longitudinal system of arteries and veins that often was the dominant supply to the region. Whether the nerves appeared together with the vessels, whether the nerves crossed them at an angle, or whether they approached the vessels from opposite directions, in each case the main trunk of the vessel or some of its branches soon "peeled off" to course parallel to the nerve. This information provides the basis for the design of long skin flaps placed along neurovascular systems. Indeed, it reveals that many of the current "axial" or "fasciocutaneous" skin flaps used in clinical practice are in fact neurovascular flaps. The muscles are classified into four types according to their extrinsic and intrinsic neurovascular supplies. Type I muscles are supplied by a single unbranched nerve. In type II muscles, the nerve branches before entering the muscle. Type III muscles receive multiple motor nerves from the same nerve trunk, and type IV muscles are supplied from multiple nerve trunks. Suggestions are made as to how muscles of each type may or may not be subdivided into functional neurovascular units for local and distant transfer.(ABSTRACT TRUNCATED AT 400 WORDS)

The deep venous system of the upper and lower extremities was injected with a lead oxide mixture in 2 fresh human cadavers, dissected, radiographed and the sites of the venous valves located. These studies confirmed that the macrovenous connections between the venae comitantes of the distributing arteries were insufficient in number to bypass the venous valves in conventional, distally based reverse flow flaps (e.g. radial, ulnar, peroneal) but revealed an alternative microvenous interconnecting pathway which surrounds the artery as the venae arteriosa. This pathway was investigated in a series of distally based reverse flow saphenous flaps in dogs, comparing flaps where the microvenous connections were left intact (non-skeletonised) with those where these vessels were disconnected with the operating microscope (skeletonised). All non-skeletonised flaps survived subtotally or totally whereas total necrosis was observed in 70% of the skeletonised flaps. Finally a series of haemodynamic studies was performed to test valve competency including extrinsic pressure on the valves. It is concluded that the macrovenous and microvenous pathways, coupled with the variable anatomy of the venous valves, are major factors in determining the survival of reverse flow flaps.

Experimental skin flaps have been utilized by researchers for almost a century to investigate many of the perplexing questions in plastic and reconstructive surgery, yet the underlying vascular anatomy of these flaps is addressed infrequently. The purpose of this study was to predict the survival of experimental skin flaps prior to their elevation in guinea pigs and rabbits, planned on a knowledge of the underlying vascular anatomy. On the basis of our previous anatomic studies, 17 guinea pigs and 15 rabbits were used in separate experiments. In experiment 1, two parallel flank flaps of identical dimensions were compared on one side of each guinea pig. The dorsal flap encompassed the vascular territories of multiple perforators, while the ventral flap embraced two perforators (two-territory flap). Viability was assessed on days 3 and 7 by inspection and fluorescein dye injection. All ventral flaps survived to a greater extent than the dorsal flaps. Whole-body fresh cadaver lead oxide injections were performed to provide cutaneous angiograms. It was found in each flap that the area of skin viability corresponded to the capture of one to two adjacent vascular territories on the artery at its base. In the second experiment, a multiple-territory osteocutaneous flap was designed on one side of the torso of the rabbit using the Doppler probe. It was based on the thoracodorsal artery and embraced the skin and a 1 x 2 cm segment of iliac crest bone in the adjacent deep circumflex iliac artery angiosome. Using the same criteria as in experiment 1, we found that one to two adjacent viable skin territories were captured on the thoracodorsal artery. In addition, viability of the iliac bone was confirmed in every case by angiography, fluorochrome labeling, and india ink injection studies indicating the capture of deep structures of the deep circumflex iliac artery angiosome. This study reinforces the angiosome concept and indicates that one adjacent vascular territory may be captured reliably in experimental guinea pig and rabbit skin flaps. In addition, we have described a reliable osteocutaneous flap model in the rabbit.

A successful clinical case of scalp replantation is presented where a suitable artery could not be identified in the amputated part. Reperfusion of the flap was achieved by an arteriovenous inflow anastomosis and a conventional venous anastomosis for drainage. The aesthetic and physiological success of this case suggests that arteriovenous scalp replantation is a potential avenue for treatment where arterio-arterial repair is not possible. The possible explanation for the success in this patient is the paucity of valves that exist in the venous network of the scalp and forehead, noted in our previous anatomical studies. Knowledge of the replanted tissue venous architecture provides an anatomical basis to understand the potential limitations of arteriovenous replantation.

A number of experiments were conducted to study the anatomic changes in a flap following a surgical delay using the Doppler probe to add precision to the technique. After scanning the integument of a series of anesthetized animals with the probe, each was sacrificed; a total-body arterial injection was performed with a lead oxide mixture, the integument and deep tissues were radiographed separately, and the results were correlated and compared with our previous human studies. The dog was selected from the range of animals examined, and the arterial networks of a number of skin and muscle flaps were studied with and without a surgical delay. The study included the use of a tissue expander. Results revealed that an adjacent cutaneous perforator could be captured with safety on the artery at the base of an undelayed flap; that the survival length of that flap was related to the distance between perforators; that the necrosis line of the flap usually appeared in the zone of choke vessels connecting adjacent territories; that a surgical delay results in a dilatation of existing vessels with maximal effect in the zone of choke arteries; that the most effective delay was obtained by elevating the flap in stages from the base, leaving detachment of the tip until last; that tissue expansion is a form of surgical delay, with particular emphasis on vessel hypertrophy; and that similar changes occur when a muscle is delayed. The clinical applications of this investigation are presented in Part II of this anatomic review of the delay phenomenon.

This paper applies the anatomic concepts and data obtained from our animal experimental studies of the delay phenomenon to a series of clinical cases. Similar clinical results were obtained to those seen in Part I of our study when skin flaps were raised with and without a delay, when a tissue expander was used, and when the delay technique was extended to musculocutaneous flaps. In each instance, the cutaneous perforators were identified with the Doppler probe to facilitate the delay of specific vessels rather than dividing those at random. Intraoperative arteriograms and venograms reveal that the choke arteries dilate and the anatomically unfavorable valved vein segments become regurgitant. The end result is the observation that at least one additional anatomic vascular territory can be added to the length of a flap with safety following a surgical delay.

This is a comparative study of the vasculature of the integument and underlying deep tissues of a range of mammals and other vertebrates. The investigation was conducted in the pig, monkey, dog, cat, possum, guinea pig, rat, rabbit, duck, and toad. The results from each are compared not only to each other, but also to previously performed human studies. The arterial network of the fresh animal cadaver was injected with a mixture of lead oxide and gelatin. The vascular anatomy of the skin, deep tissues, and individual muscles was defined by dissection, cutaneous perforator counts, photography, and radiography. A similar pilot study of the venous framework was performed in the pig, dog, and rabbit that included maps of the sites and orientations of the valves. The vasculature of the integument and deep tissues was correlated, and we found that we were able to define angiosomes (composite blocks of tissue supplied by the same source vessel) in each animal. Results revealed a marked dissimilarity of the overlying cutaneous vessels in many cases, yet a striking resemblance of the vascular architecture of the deep tissues. The size and density of the cutaneous perforators bore a close relation to the degree of the skin mobility, being large and sparse where the skin was mobile and smaller and more densely grouped where the integument was tethered or fixed. The cutaneous vasculature of the human resembled that of the monkey closely, was similar to that of the dog, cat, and possum, and was dissimilar to that of the pig, rat, guinea pig, and rabbit. Studies of the amphibian and bird bore many resemblances to those of the mammals. They provided basic concepts regarding modification of the animals' vascular anatomy in response to the functional demands of the species. In each animal, the arteries formed an unbroken network throughout the body. This consisted of anatomic territories linked by anastomotic vessels that were usually of reduced caliber. The pattern of the venous system was almost identical. Valved venous territories were linked by avalvular (oscillating) veins. The common denominator in the vascular system is the capillary bed. Conceptually, the anatomic arrangement of the arteries and veins, reproduced in each species, appears to be a sophisticated mechanism to allow equilibration of flow and pressure arriving at and departing from the capillary bed. The angiosome concept is reinforced by the animal studies. Although this investigation is essentially a detailed pilot study, it embraces many animals commonly used for experimentation and provides a reference atlas of their vasculature.(ABSTRACT TRUNCATED AT 400 WORDS)

The venous drainage of the peripheral nerves was studied in the upper and lower limbs of two human fresh cadaver subjects after total body perfusion with a radio-opaque lead oxide mixture. Four patterns of extraneural drainage were identified in which the venae nervosa drained: directly to the venae comitantes of the neurovascular bundle; indirectly via nearby veins, derived usually from muscles; to the periarterial venous plexus; or, in the case of the cutaneous nerves, to the perivenous plexus. The various patterns of the drainage along the length of the radial, median, ulnar, sciatic, anterior and posterior tibial nerves were identified. A rich longitudinal plexus of veins exists on and within the nerve which appears to be mainly free of valves. The large venae nervosa usually contained valves, whereas the tiny veins draining the nerves were valveless or exhibited a sentinel valve at their entry point into a larger venous channel. The clinical implications of these results are discussed in relation to the mobilisation of nerves, the use of island nerve flaps, possible donor sites for free arteriolised neurovenous flaps and the compressive nerve syndromes.

The venous architecture of the integument and the underlying deep tissues was studied in six total-body human fresh cadavers and a series of isolated regional studies of the limbs and torso. A radiopaque lead oxide mixture was injected, and the integument and deep tissues were dissected and radiographed. The sites of the venous perforators were plotted and traced to their underlying parent veins that accompany the source (segmental) arteries. A series of cross-sectional studies were made in one subject to illustrate the course of the perforators between the integument and the deep tissues. The veins were dissected under magnification to identify the site and orientation of the valves. Results revealed a large number of valveless (oscillating) veins within the integument and deep tissues that link adjacent valved venous territories and allow equilibration of flow and pressure throughout the tissue. Where choke arteries define the arterial territories, they are matched by boundaries of oscillating veins in the venous studies. The venous architecture is a continuous network of arcades that follow the connective-tissue framework of the body. The veins converge from mobile to fixed areas, and they "hitchhike" with nerves. The venous drainage mirrors the arterial supply in the deep tissues and in most areas of the integument in the head, neck, and torso. In the limbs, the stellate pattern of the venous perforators is modified by longitudinal channels in the subdermal network. However, when an island flap is raised, these longitudinal channels are disconnected, and once again the arterial and venous patterns match. Our venous studies add strength to the angiosome concept. Where source arteries supply a composite block of tissue, we have demonstrated radiologically and by microdissection that the branches of these arteries are accompanied by veins that drain in the opposite direction and return to the same locus. Hence each angiosome consists of matching arteriosomes and venosomes. The clinical implications of these results are discussed with particular reference to the design of flaps, the delay phenomenon, venous free flaps, the pathogenesis of flap necrosis, the "muscle pump," varicose veins, and venous ulceration.

The Doppler probe was used to identify the dominant cutaneous perforating arteries in a series of 10 patients. The results were compared with our previous total body fresh cadaver anatomical studies and a close correlation was found. The instrument was used in a series of patients to plan the base, the axis and the dimensions of skin flaps for local, distant and free transfer. A dominant perforator was located at the base of the flap, the surrounding skin was scanned to identify the next dominant perforator in each direction and the appropriate axis was chosen by drawing a line between two nominated perforators. Often the flap was based distally or its axis departed from the main course of the supplying vessel. The technique proved to be simple and reliable and in many cases flaps of unusual dimensions and directions were transferred successfully. The instrument provides a useful link between the anatomical dissecting room and the operating theatre.

A series of 31 fresh cadaver injection, dissection and radiographic studies were undertaken to define the vascular architecture of the anterior chest wall and to correlate the findings of previous writers in this area. The findings confirmed that the dominant supply to this region is from the internal thoracic artery which interconnects in all layers with the posterior intercostals, the lateral thoracic, the acromio-thoracic and the transverse cervical arteries. The dominant cutaneous perforators of these vessels are concentrated around the perimeter of the pectoralis major muscle, the costal margin and over the interdigitations of the serratus anterior muscle in the mid-axillary line. The clinical significance of these findings is discussed with particular reference to the design of various breast reduction procedures and the planning of flaps in this area. Of special note is the observation that the concentration of large perforators which exists along the inframammary crease arises from the anterior intercostal vessels in the fifth and sixth intercostal spaces.

In the last decade, free vascularized bone transfer has become an established technique, with success rates above 90 per cent reported in most series. These operations have special application for those situations in which the bone defect is large, the recipient bed is poorly vascularized, and there is an associated soft tissue deficiency. The donor sites now include the fibula, iliac crest, rib, metatarsal, radius, and scapula. Because this is a rapidly expanding field of surgical endeavor, much of the work is unpublished and information in some instances has been obtained necessarily from recent clinical meetings and by personal communication with surgeons. Our experience of 41 bone transfers, using the fibula and the iliac crest as the donor grafts, over a 9-year period has been presented. In general, the fibula is recommended for the reconstruction of a long bone. In the lower extremity it should be placed within the medullary cavity of the recipient bone ends so as to be sited in the line of the weight-bearing stress. The iliac osteocutaneous graft is ideal for reconstructing the mandible and other curved bones but can be used also for shorter defects of the tibia. With appropriate osteotomies the curvature of this bone can be increased or straightened to satisfy the requirements of the recipient site. Stress fracture of a long graft is not uncommon in the lower extremity, especially in adults. Nevertheless, this is followed usually by rapid formation of callus at the fracture site and hypertrophy of the bone. The use of angiography, replica bone models, and trial operations on cadavers is recommended in planning. Operative refinements include a simpler dissection of the fibula with a thinner sleeve of muscle; shaping the iliac bone graft while it is still attached at the donor site; and the suture of vein grafts to the bone graft on a side table when the pedicle is too short. Finally, it must be remembered that these are still relatively sophisticated procedures that place demands on personnel and hospital routine. They do not replace conventional bone grafting techniques but offer an alternative solution for difficult situations (Fig. 19). John Hunter stated, "A surgeon must approach the victim of his operation with a sacred dread and reluctance." Although a century has passed this is still a sage message.

For the common problem of lower limb injury with extensive. Loss of skin and bone, a new method of free vascularized bone grafting has been used and integrated with an appropriate soft tissue flap repair. The technique was developed to salvage two legs which would otherwise have been amputated. The preliminary result in Case 2 is encouraging. This case would appear to be the first successful distant transfer of a composite fibular graft by microvascular anastomoses to be reported in man.

The venous architecture of the muscles of the body was studied in 425 specimens obtained from four fresh cadavers after total body injection with a radio-opaque lead oxide mixture. In 40 muscles the site and orientation of the valves was identified with the operating microscope. The venous network of each muscle was compared with similar arterial studies. The venous territories in each muscle matched the arterial territories. Where arterial territories were linked with choke arteries the venous territories were linked by veins devoid of valves which allowed bidirectional flow--the "oscillating veins". Elsewhere the valves of adjacent territories were directed away from each other and towards their respective pedicles. These anatomical observations reinforce our angiosome concept. The muscles were classified into Type A, B and C where there were one, two or more territories respectively. When a skin paddle is designed distally on a Type B or C muscle its venous return at first must negotiate the anatomical obstruction of the valves of the distal muscle territory before reaching the safety of the oscillating veins and the venous territory of the drainage pedicle. Finally, afferent veins were noted entering many muscles from the superficial and deep tissues. This highlights the importance of the muscles in aiding venous return by their muscle pump action.

The results of 100 cadaver dissections of the subscapular-thoracodorsal arterial system are presented. These results confirm the anatomical reliability of the thoracodorsal artery and report the consistent presence of a branch of the thoracodorsal artery to the serratus anterior muscle. The branches to serratus anterior were of sufficient dimensions to support either a latissimus dorsi flap or a serratus anterior flap. A direct cutaneous branch from the thoracodorsal artery to the skin of the axilla was identified in only 47% of the dissections.

A new free flap is described based on the deltoid vessels of the acromiothoracic axis. The flap is comprised of the clavicular head of pectoralis major muscle with overlying skin. It is also possible to harvest vascularised clavicular bone with the flap. The vascular anatomy is reviewed and the technique of raising the flap described. Its clinical application is illustrated with five cases. Four of these were intra-oral reconstructions and the fifth a composite osteo-musculocutaneous flap to a lower limb following trauma.

A new method of nerve grafting is described that introduces the one stage distant transfer of a living nerve graft by microvascular techniques. The development of this operation is outlined and the first clinical application of this technique is presented together with the preliminary reports of 2 new cases. The blood supply of peripheral nerves is examined, and consequently a neurovascular bundle is defined as the best graft for free transfer. Donor sites are suggested for both normal and unusual circumstances, and a sequence of preoperative investigations and intraoperative precautionary steps is recommended, which can also be used as a basis for microvascular free transfer of other tissues. This technique of nerve grafting is designed primarily to ensure viability of the nerve graft. Because the procedure is exacting and time consuming, it is recommended only when conventional techniques carry high risk of graft necrosis, situations in which the nerve gap is large, the recipient bed is scarred, or the free transfer of a thick nerve is desired. There are disadvantages as well as advantages of the free vascularized nerve grafts. Important new developments in the field may depend upon advances in immunology.

Three successful cases of distal finger replantation are described where suitable veins were unavailable for anastomosis after arterial flow had been re-established. To prevent infarction, calcium heparin was injected subcutaneously into the replants at intervals over 9 days. This allowed the venous blood to escape into the dressings while an adequate microvenous circulation was re-established, thus simulating the effect of a leech. Complete survival of all three distal replantations was observed. No attempted case has been unsuccessful. The procedure is simple and can be administered by the nursing staff. It avoids some of the problems associated with the use of medicinal leeches and systemic heparin. Applications of this technique in other areas of replantation and flap surgery are suggested.

An anatomical and hemodynamic study of the deltopectoral flap in cadaver dissections and clinical cases is described. Anatomically, the deltopectoral flap represents a fusion of 3 vascular regions supplied by 3 different sets of vessels. Hemodynamically, the flap consists of a lateral cutaneous flap nourished by a medial arterial pedicle. The method of choice for delay of a deltopectoral flap is discussed.

This paper reports the findings of an anatomical study of the primary and secondary muscular branches of the thoracodorsal artery within the latissimus dorsi muscle. Twenty latissimus dorsi muscles were studied, harvested from 13 fresh cadavers and the findings support the hypothesis that it is possible to thin safely a latissimus dorsi muscle flap at the time of flap transfer.

Radiographic studies of the deep superior epigastric artery (DSEA) and its connections within the soft tissues of the abdominal wall were performed in 64 fresh cadavers. The patterns of anastomosis between the deep superior epigastric artery and the deep inferior epigastric artery (DIEA) were noted. Type I (29 percent) revealed a single deep superior epigastric artery and deep inferior epigastric artery, type II (57 percent) revealed a double-branched system of each vessel, and type III (14 percent) revealed a system of three or more major branches. In each case, the two systems were united by choke vessels in the segment of muscle above the umbilicus. The supply to the various transverse and vertical skin flaps from the deep superior epigastric artery was defined as a series of captured anatomic territories bounded by choke vessels. The upper transverse and vertical flaps had the best supply, and the TRAM flap had the most tenuous supply. Midline crossover occurs predominantly in the subdermal plexus and on the surface of the rectus sheath. Modifications of the design of the TRAM flap, the case for a delay procedure, the wisdom of including a strip of anterior rectus sheath, and the risks of splitting the muscle with respect to its nerve supply and vascular patterns are discussed on an anatomic basis.

The precise vascular territory and the variations of the acromio-thoracic axis were investigated in a series of 60 fresh cadavers and 50 formalin fixed specimens using dissection, ink injection and barium radiographic studies. The sternocostal portion and the clavicular head of the pectoralis major were found to have virtually independent vascular and nerve supplies. The pectoral artery supplied the former, whereas the deltoid artery nourished the latter. The dominant supply from the pectoral artery to the rib cage was found to enter around the fourth rib in the mid clavicular line. This supply is associated with a previously undescribed origin of the pectoralis major muscle in this region. The supply to the sternum was determined as indirect via the captured territory of the internal mammary system. The dominant supply to the skin from the pectoral artery arose laterally along the free lower border of the muscle as fasciocutaneous branches. The deltoid artery supplies the skin over the shoulder by numerous small branches which emerge from the intramuscular septa of the deltoid muscle. In addition a large axial artery was noted. In most cases this arose from the deltoid artery or its acromial branch and coursed laterally. It is noteworthy that the majority of skin paddles of the pectoralis major myocutaneous flap currently used in clinical practice are designed medially and inferiorly around the perimeter of the muscle and onto the rectus sheath. In these situations such flaps are not supplied directly by the pectoral artery. In fact, they are supplied indirectly by cutaneous branches belonging to the internal mammary/superior epigastric system which are captured by arterial connections with the pectoral artery. These occur predominantly in the pectoralis major muscle. Suggestions, based on these anatomical studies, are offered to improve the versatility and safety of flaps designed in this area.

The blood supply to the skin and underlying tissues was investigated by ink injection studies, dissection, perforator mapping and radiographic analysis of fresh cadavers and isolated limbs. The results were correlated with previous regional studies done in this department. The blood supply is shown to be a continuous three-dimensional network of vessels not only in the skin but in all tissue layers. The anatomical territory of a source artery in the skin and deep tissues was found to correspond in most cases, giving rise to the angiosome concept. Arteries follow closely the connective tissue framework of the body. The primary supply to the skin is by direct cutaneous arteries which vary in calibre, length and density in different regions. This primary supply is reinforced by numerous small indirect vessels, which are "spent" terminal branches of arteries supplying the deep tissues. An average of 374 major perforators was plotted in each subject, revealing that there are still many more potential skin flaps. Our arterial roadmap of the body provides the basis for the logical planning of incisions and flaps. The angiosomes defined the tissues available for composite transfer.

The successful revascularized free transfer of a composite flap of groin skin with attached iliac bone is presented in two patients. The findings in the preliminary dissections and injection studies in cadavers are discussed, and the superficial circumflex iliac system is recommended as a stem for free flaps of groin skin and iliac bone. If iliac bone alone is required for a revascularized graft, then the deep circumflex iliac system would seem suitable for the stem. Further clinical cases are necessary to determine the indications for this procedure, and patients should be carefully selected. The need for preoperative preparation and planning on a sound anatomical background is emphasized, and preoperative angiography seems advisable. The ability to sew small vessels together reliably is one thing: the when and where is another.

The deep inferior epigastric vascular system nourishes a vast area of the anterior abdominal wall and offers a versatile variety of tissue combinations to the reconstructive surgeon for local or free flap transfer. Its supply to and beyond the rectus abdominis muscle provides the basis for flap design. Eighteen cases are presented to illustrate various muscle, myocutaneous and myosubcutaneous flaps that have been used to date to repair defects in the extremities and in the head and neck. The ease of dissection, coupled with the large calibre and length of the vascular stem, are key factors which combine to offer a speedy and a reliable technique. The donor site scar has been improved by the use of a suprapubic transverse incision to dissect the pedicle and this exposure provides access to the iliac crest for bone grafts where indicated. Time will evaluate this technique and further designs and modifications will undoubtedly be forthcoming. Meanwhile the initial results have been most promising.

The rationale for the use of revascularised nerve grafts is well established and has been proved both experimentally and clinically. It is a technique that is particularly useful if one wishes to transfer large thick grafts to repair major nerves or insert vascularised cable grafts to bridge long defects in a poorly vascularised fibrous bed of tissue. An experimental technique has been devised to raise a composite neuro-venous graft and by reversing the vein component provide an arterialised composite nerve graft that can be safely transferred to different parts of the body. Experimental work in cadavers and dogs has confirmed these observations and details of seven clinical cases are given. The results are encouraging and in patients with nerve injuries where there are associated vascular and nerve defects, as in the forearm and hand, the technique of a composite reversed neuro-venous graft which is arterialised, should be seriously considered.

The vascular anatomy of the supplying vessels to the iliofemoral, the deltopectoral, the axillary, and the thoracoabdominal regions has been investigated in detail in fresh cadavers. Other donor sites were also examined. The anatomical relationship of the supplying vessels is discussed-and the design, conversion, and utilization of these skin areas as donor sites for free flap transfer are detailed. The iliofemoral region is suggested as the best donor site for most free flap procedures.

The vascular territories of the superior and the deep inferior epigastric arteries were investigated by dye injection, dissection, and barium radiographic studies. By these means it was established that the deep inferior epigastric artery was more significant than the superior epigastric artery in supplying the skin of the anterior abdominal wall. Segmental branches of the deep epigastric system pass upward and outward into the neurovascular plane of the lateral abdominal wall, where they anastomose with the terminal branches of the lower six intercostal arteries and the ascending branch of the deep circumflex iliac artery. The anastomoses consist of multiple narrow "choke" vessels. Similar connections are seen between the superior and the deep inferior epigastric arteries within the rectus abdominis muscle well above the level of the umbilicus. Many perforating arteries emerge through the anterior rectus sheath, but the highest concentration of major perforators is in the paraumbilical area. These vessels are terminal branches of the deep inferior epigastric artery. They feed into a subcutaneous vascular network that radiates from the umbilicus like the spokes of a wheel. Once again, choke connections exist with adjacent territories: inferiorly with the superficial inferior epigastric artery, inferolaterally with the superficial circumflex iliac artery, and superiorly with the superficial superior epigastric artery. The dominant connections, however, are superolaterally with the lateral cutaneous branches of the intercostal arteries. For breast reconstruction, it would appear that prior ligation of the deep inferior epigastric artery would be of advantage when elevating the lower abdominal skin on a superiorly based rectus abdominis musculocutaneous flap. The vascularity of this flap would be further increased by positioning some part of the skin paddle over the dense pack of large paraumbilical perforators. Based on these anatomic studies, the relative merits of the superior and deep inferior epigastric arteries with respect to local and distant tissue transfer using various elements of the abdominal wall are discussed in detail.

The blood supply to the fibula from the anterior tibial and peroneal arteries was investigated in 66 fresh cadaver limbs by India ink injection and radiographic studies. These studies revealed a reliable blood supply to the proximal epiphysis and the proximal two-thirds of the diaphysis of the fibula from the anterior tibial artery. Although the peroneal artery provided the dominant supply to the shaft of the bone, its contribution to the head of the fibula was scanty. The cutaneous supply of each vessel was defined and correlated with the territories of the other vessels supplying the leg. The vascular anatomy of the leg was examined in 100 clinical angiograms, revealing 21 limbs with anomalies. This reinforces the case for preoperative angiogram studies where transfer of the fibula is contemplated. The technique of harvesting the proximal growth plate and the shaft of the fibula on the anterior tibial vessels alone is described and illustrated with two successful clinical cases, each followed for 4 years. The growth rate of the transferred epiphysis averaged 1 cm per year.

The use of free vascularized tissue transfers in reconstructive surgery is an exciting new development. Such procedures require careful planning and meticulous attention to detail. Illustrative cases are described to indicate the value and scope of this concept, which has been made possible by the development of microsurgery.

This study has investigated the action of two commonly available preservatives, chlorbutol and chlorocresol, on fresh cadaver tissue. If the arterial system of the subject is perfused first with either preservative then the progression of tissue necrosis can be delayed for up to 1 month. The preservative can be combined with the lead oxide-gelatin mixture described by Rees and Taylor without compromising the results of radiographic studies. Both chlorocresol and chlorbutol have the advantage over formalin of retaining the normal colour and texture of the tissues. A regimen is suggested which has all of the advantages of the mixture originally described by Salmon, but is much simpler and cheaper to produce. It has proved successful when used in whole cadavers but the visceral contents must be removed at an early stage.

A versatile flap based on the deep inferior epigastric artery is presented. Its application for local and free-flap transfer is discussed and presented in two clinical cases. The possible inclusion of bone as an osteocutaneous flap is outlined. The flap can be raised with ease and speed, and the donor site is repaired as a linear scar. Since the latissimus dorsi flap with its pedicle based in the axilla has become the workhorse for the upper torso, this extended flap may become its distal counterpart with its pedicle based on the groin.

A simple, inexpensive injectable substance is reported for analysis of the arterial circulation in fresh human or animal cadavers. The technique is a modification of that reported by Salmon in 1936 and utilizes lead oxide and gelatin. This combination is highly radiopaque; it perfuses the small radicles of the vascular tree and sets to a firm rubbery consistency to fascilitate dissection.