J Plast Surg Hand Surg, 2014; 48: 191–196 © 2014 Informa Healthcare ISSN: 2000-656X print / 2000-6764 online DOI: 10.3109/2000656X.2013.852100
ORIGINAL ARTICLE
The free groin flap in the rat: a model for improving microsurgical skills and for microvascular perfusion studies Jens Wallmichrath1, R. G. H. Baumeister2, O. Gottschalk3, R. E. Giunta1 & A. Frick1 1
Hand Surgery, Plastic Surgery and Aesthetic Surgery, University Hospital of the Ludwig-Maximilians-University-Grosshadern, Munich, Germany, 2Surgical Clinic Munich Bogenhausen, Munich, Germany and 3Department of Trauma Surgery, University Hospital of the LudwigMaximilians-University-Grosshadern, Munich, Germany Abstract The goal of this study was to evaluate the free groin flap in the rat transplanted to the neck as a tool for extending microsurgical skills and to assess its suitability as a model for microvascular perfusion studies following secondary venous ischaemia. An analysis of 60 consecutive groin flap transplantations was performed in male Sprague Dawley rats with special regard to anatomy and operation times (Part I, animals No. 1–60). Following flap transplantation, the animals No. 10–30 (n = 21) were used for the determination of the critical time period of a complete venous stasis of the free groin flap resulting in a total flap loss (Part II). The flaps of animals No. 31–41 (n = 11) were used for assessing the feasibility and reproducibility of intra-vital video microscopy (IVM) of the flaps (Part III). The mean total operation time decreased from 166 (± 26) minutes ins the first 10 animals to 126 (± 21) minutes and 130 (± 12) minutes in the latter two groups of 10 animals, respectively. After a critical period of 35 minutes of a complete artificial venous stasis a complete flap necrosis occurred. IVM detected a higher functional capillary density of the skin of the transplanted groin flaps in the animals in which the flaps were rinsed with 1 ml of Ringer’s lactated solution prior to I/R. In conclusion, this model is simple and reliable. The model may be a useful tool for evaluating and comparing the effects of various anticoagulants or vasomotor drugss on microvascular perfusion in critically compromised free flaps. Key Words: Groin flap, microsurgery, microsurgical training, intra-vital microscopy, ischaemia, reperfusions
Introduction Free flap reconstructions and microsurgical replantations require extensive microsurgical skills to optimize the success rate. Short operating times with minimal ischaemic periods are beneficial to proper wound healing, short rehabilitation periods, reduced financial expenses, and a good functional result. Thus, an authentic microsurgical training is a profitable investment for prospective microsurgeons. Various flap models exist in laboratory rats with a long history of use both as training models and for microsurgical experimental studies [1–4]. The thickness of the free groin flap based on the superficial iliac vessels is the feature of this flap since this makes flap survival solely dependent on the vascular pedicle. The goal of this study was to assess the learning curve of the free groin flap in the rat transplanted to the neck and its suitability as a model for microvascular perfusion studies following secondary venous ischaemia. This study consists of three parts: . .
Part I: Evaluation of the learning curve of the flap as a tool for extending microvascular and experimental skills. Part II: Identification of the critical time period of a complete secondary venous stasis of the flap resulting in a complete flap loss. This should serve as a basis for the development of an animal model for the examination of positive effects of vasoactive and/or anticoagulant agents on compromised free flaps following fatal secondary venous stasis.
.
Part III: Finally, the applicability of this model for experimental in vivo perfusion studies in free skin flaps involving intra-vital video microscopy (IVM) was performed. Here, the special focus was on the influence of disturbing artefacts due to limited flap size, cardiopulmonary motions of the region of the neck and the fluorescence of hair follicles.
Materials and methods Experimental design An analysis of 60 consecutive free groin flap transplantations was performed with special regard to anatomy and operation times (Part I, animals No. 1–60). Following flap transplantation, the animals No. 10–30 (n = 21) were used for the determination of the critical time period of a complete venous stasis of the free groin flap resulting in a flap loss (Part II). The flaps of animals No. 31–41 (n = 11) were used for assessing the feasibility and reproducibility of intra-vital video microscopy (IVM) of the skin of these compromised free groin flaps (Part III, exact experimental design is given below). The remaining animals were included in various perfusion studies (data not shown). Animal model and flap preparation Experiments were performed according to the guiding principles for research involving animals and the German legislation on protection of animals. Approval was obtained by the local
Correspondence: Jens Wallmichrath, MD, Hand Surgery, Plastic Surgery and Aesthetic Surgery, Surgical Clinic and Policlinic, University Hospital of the Ludwig-Maximilians University, Marchioninistrasse 15, D-81377 Munich, Germany. Tel: +49 89 7095 3502. Fax: +49 89 7095 6505. E-mail: [email protected] (Accepted 2 October 2013)
192 J. Wallmichrath et al. governmental animal care committee. Male Sprague-Dawley rats, weighing between 200–250 g, were purchased from Charles River (Sulzfeld, Germany), and housed at the Institute for Surgical Research (Walter Brendel Zentrum), LudwigMaximilians-University, Munich, Germany. Animals were allowed to accommodate to the standard conditions of the animal facility with free access to rat chow and water for at least 1 week. Solid food was withheld from the rats 12 hours prior to surgery. The rats were anaesthetised by intraperitoneal injection of 0.15 mg/kg Medetomidin (Pfizer GmbH, Karlsruhe, Germany), 0.01 mg/kg Fentanyl (CuraMED Pharma GmbH, Karlsruhe, Germany), and 2 mg/kg Midazolam (Hoffmann-La-Roche AG, Grenzach-Wyhlen, Germany) and supplemented as required. The animals were placed on a heated operating table in a supine position with the extended hind leg. The region of the right groin was shaved using an electrical shaver without harming the skin. The flap borders were outlined on the skin using a round template with 3 cm in diameter centred in the right groin (Figure 1). After the dissection of the skin and the subcutaneous tissue the flap can bluntly be separated from the abdominal muscles as well as the inguinal ligament, exposing the superficial epigastric vessels on the deep surface of the adipocutaneous groin flap. Then the comparatively thick and sturdy subcutaneous tissue of the lateral end of the groin was dissected and the superficial iliac vessels were ligated using Vicryl 5-0 (Polyglactin 910; Ethicon, Germany). The microsurgical preparation was started distally at the femoral vessels and preparation was continued proximally towards the groin (Figure 2). First, the femoral artery and vein were dissected and the femoral nerve was microsurgically isolated from the vessels and excluded from the flap. Now, the femoral vessels still intact, the muscular branches can easily be dissected and ligated. Microvascular preparation was continued with exposure and isolation of the proximal femoral artery and vein serving as the pedicle of the flap. Here, the profunda system was left intact not to compromise the blood supply of the hind leg. Finally, the distal end of the femoral artery was ligated 6 mm distally to the branch point of the flap
Flap
Pedicle diev fn
il
fa
fv / p
fv mb
Figure 2. The anatomy of the inguinal vessels in the right groin after elevation of the adipocutaneous groin flap. The hook is inserted in the proximal medial wound edge (fn = femoral nerve; fa = femoral artery; fv = femoral vein; mb = muscular branch; fv / p = femoral vessels and branch point of the profundal vessels; il = inguinal ligament; diev = deep inferior epigastric vessels).
vessels using Vicryl 5-0 (Polyglactin 910; Ethicon, Germany). The femoral vein was ligated separately. Following preparation of the recipient site, the vascular pedicle was transected and the stumps of the femoral vessels were ligated distally to the branch point of the profunda system. The flap size was outlined on the skin in the neck using the round template 3 cm in diameter arranged centrally in the neck (Figure 1). The left carotid artery and the internal jugular vein were exposed mainly by blunt dissection. The proximal ends were irrigated using Heparin solution (100 IU/ml) and clamped temporarily with single Acland vessel clamps Type B-3 (Art. No. 00400V and 00400A, respectively; S&T AG, Neuhausen, Switzerland). The distal vessel ends were ligated and the flap was transferred to the neck (Figure 3). The arteries and then the veins were anastomosed in an end-to-end fashion using a single stitch technique requiring up to 10 stitches (Ethilon 9-0 for the arterial and 8-0 for the venous anastomosis; Ethicon, Germany).
3cm
3cm
cm
Figure 1. Localisation of the donor site of the groin flap in the right inguinal region and the recipient site in the center of the cervical region.
1
2
3
Figure 3. The free adipocutaneous groin flap.
The free groin flap in the rat Then the Acland clamps were removed to allow reperfusion of the flap. Finally, the flap was sutured in place using single stitches and the donor site was closed with a buried single suture (Ethilon 3-0, Ethicon). Anaesthesia was terminated by intraperitoneal injection of 0.75 mg/kg atipamezolhydrochloride (Pfizer GmbH), 0.20 mg/kg Flumazenil (Hoffmann-La-Roche AG) and 0.12 mg/kg Naloxon (Ratiopharm GmbH, Ulm, Germany). The animals were monitored until recovery from anaethesia and then returned to their cages. After 20 hours, the animals were re-anaesthetised as described above and the flap and the contralateral groin were depilated using keratolytic creme (Elca Med Enthaarungscreme, Asid Bonz, Herrenberg, Germany). The wound was partially re-opened and the vascular pedicle was exposed. The patency of the anastomoses and the uncompromised blood flow was assessed by the method of Acland [5] using two microsurgical tweezers. Part II (determination of a fatal secondary venous stasis) For determining the shortest time period necessary to produce a complete flap necrosis, the venous clamping was performed for 20, 25, 30, or 35 minutes, respectively, in four groups with n = 5 animals each. Flap viability was assessed after 14 days. Part III (applicability of intra-vital video microscopy) Pre-operatively, the animals were randomized into group A (n = 5; no infusion) or group B (n = 5; infusion of 1 ml of Ringer’s solution). Following a positive Acland test 20 hours after the transplantation a sterile polyethylene tube was introduced into the distal stump of the femoral artery (Figure 4). In group A (control group) no infusion was administered. In group B, 1 ml of sterile Ringer’s solution was infused over 10 minutes using an infusion pump (Perfusor compact, B. Braun Melsungen AG, Melsungen, Germany). Afterwards, the catheter was removed and the arterial stump was ligated. One minute after the time period of the infusion a venous Acland vessel clamp (Type B-3, Art.No. 00400V, S&T AG, Neuhausen, Switzerland) was applied to the flap vein in all groups to produce a fatal venous stasis for 35 minutes (Figure 5). Then the clamp was removed and 1 minute later the Acland test was repeated on both artery and vein. The wound was closed as described above and the animal was transferred to a heated
Flap
193
a
v
Figure 5. The catheter is removed and the flap vein is clamped for 35 minutes using a microvascular clamp generating a temporary venous stasis.
microscope table in a supine position. IVM was carried out on the flap (Periplan, Leitz Wetzlar, with a CCD camera; Epiillumination with an adjustable100 W Xenon lamp (Zeiss, Oberkochen, BW, Germany)). The video analysis was performed offline (Cap-Image, Dr. Zeintl, Heidelberg, Germany). Blood was gained by cardiac puncture of a donor rat using a heparinised syringe and the thrombocytes were labelled with CFDA-SE (Carboxyfluorescein diacetate succinimidyl ester, Molecular Probes, Eugene, OR), as described elsewhere [6]. The fluorescent thrombocytes were infused intravenously and IVM was performed. The parameters measured were counts of floating and adherent (‘sticking’) thrombocytes (i.e. adhesion >30 seconds). Afterwards FITC-Dextran (Fluorescein isothiocyanate–dextran, Mw 50.000, St. Louis, MO) was used as a plasma marker. The amount of the injected fluorescence marker and time of exposure were minimised strictly to avoid illumination artefacts. The parameters measured were functional capillary density (FCD) and capillary diameters. The contralateral groin served as control. Measurements were performed 5-times in the central region (i.e. within a central circus of 1 cm in diameter) and another 5-times within the peripheral region. After a healing period of 14 days all animals were sacrificed and the vitality of the flaps was assessed. Statistical analysis The program PASW Statistics 17.0 (IBM Corporation, New York, NY) was used for statistical analysis. The Wilcoxon Rank-Sum Test was performed for statistical evaluations. Values of p < 0.05 were considered to be significant.
a Catheter v
Figure 4. The catheter for the application of the test substance is introduced into the distal stump of the femoral artery allowing a direct rinsing of the flap with the test solution.
Results Part I: The free groin flap transplantation The results of the analysis of the transplantation procedures are given in Table I. The body weight of the animals at time of surgery was 383 (± 86) g. The length of the vascular pedicle was 30 (± 2.4) mm, the vessel diameters ranged from 0.6–2.0 mm. The flap wet weight was 2.7 (± 0.7) g. The mean total operation time was 145 (± 28) minutes. The mean total operation time decreased from 166 (± 26) minutes in the first 10 animals to
194 J. Wallmichrath et al. Table I. The total operating time and time periods for flap preparation, the arterial anastomosis, and the venous anastomosis, respectively.
Total operating time (M [min]) Flap preparation (M [min]) Duration of ischaemia (M [min]) Arterial anastomosis (M [min]) Venous anastomosis (M [min]) Failure rate*, n (%)
All animals (No. 1–60)
Animal No. 1–10
Animal No. 11–20
Animal No. 21–30
Animal No. 31–40
Animal No. 41–50
Animal No. 51–60
145 (± 28) 46 (± 10) 59 (± 12) 22 (± 6) 19 (± 6) n = 8/60 (13%)
166 (± 26) 50 (± 8) 73 (± 18) 25 (± 8) 20 (± 6) n = 3/10 (30%)
141 (± 15) 46 (± 12) 60 (± 7) 21 (± 4) 21 (± 6) n = 2/10 (20%)
153 (± 28) 51 (± 7) 62 (± 11) 23 (± 7) 20 (± 2) n = 1/10 (10%)
140 (± 22) 43 (± 12) 52 (± 6) 21 (± 4) 18 (± 4) n = 1/10 (10%)
126 (± 21) 42 (± 6) 55 (± 10) 24 (± 5) 19 (± 9) n = 0/10 (10%)
130 (± 12) 45 (± 6) 55 (± 9) 20 (± 5) 15 (± 4) n = 1/10 (0%)
The failure rate comprises animals with flap perfusion problems (n = 5/60) as well as loss of an animal due to bad medical conditions (n = 3/60).
*
126 (± 21) minutes and 130 (± 12) minutes in the last two groups of 10 animals, respectively. This means a significant reduction in the total operating time by 24% (i.e. 40 minutes). The time period needed for flap preparation was comparatively constant with 50 (± 8) minutes in the beginning and ~ 42–45 minutes in the last animals (reduction by 10–16%). Also, the ischaemic period showed an improvement from 73 (±18) minutes in the first 10 animals to ~ 55 minutes in the last 20 animals (i.e. a reduction by 25%). The mean time needed for the arterial anastomoses was 22 (± 6) minutes, with 25 (± 8) minutes in the first 10 animals and 20–24 minutes in the last 30 animals. The mean time needed for the venous anastomoses was 19 (± 6) minutes, with 20 (± 6) minutes in the first 10 animals and 15 (±4) minutes in the last 10 animals. The post-operative weight loss during the first 24 hours was M = 4% (0–7%) of the body weight at the time of flap transplantation. The overall flap failure rate was 8/60 flaps (i.e. 13%) in the first consecutive series. The flap failure rates of clusters of 10 animals are given in Table I. The failure rates decrease with experience and were quite stable from animal No. 20 on. Part II: Determination of the critical time period of a complete venous stasis The four groups of five animals each sustained a complete venous stasis of the free flaps for 20, 25, 30, and 35 minutes,
respectively. This resulted in a complete necrosis of the flaps in the four groups in n = 0/5, n = 1/5, n = 4/5, and n = 5/5 cases, respectively. Part III: Intra-vital video microscopy (IVM) The results of the measurement of the functional capillary density (FCD) in the skin of the contralateral native groin in both groups are given in Figure 6. The native groins of the animals without infusion (Group A, n = 5) and with infusion of lactated Ringer’s solution (Group B, n = 5) prior to I/R show no statistically significant difference with regard to the FCD (Figure 6). On the other hand, the FCD of the skin of the transplanted groin flaps of group B show statistically significant higher values (p £ 0.01), as illustrated in Figure 7. The capillary diameters (CD) in the native groin were 4.9 ± 0.4 mm (Group A) and 4.7 ± 0.4 mm (Group B). The CD in the transplanted groin flap were 7. 4 ± 0.8 mm (Group A) and 5.8 ± 0.6 mm (Group B). The differences in the CD were statistically not significant between the groups. The mean number of thrombocytes sticking to the endothelium of the capillaries in the native groin was 0.2 per field of view in both groups. The mean number of thrombocytes sticking to the endothelium of the venules in the transplanted groin flap was 2.5 ± 1 per field of view in group A and 4.3 ± 2 per field of view in group B. The differences were statistically not significant between the two groups.
250 200
150
100 50
0 n1
n2 n3 n4 Group A
n5
n1 Groin
n2 n3 n4 Group B
n5
Figure 6. Functional capillary density [mm per field] in the skin of the contralateral native groin (Intravital video microscopy, FITC-Dextran). The animals without infusion (Group A, n = 5) and with infusion of lactated Ringer’s solution (Group B, n = 5) prior to I/R show no statistical significant difference with regard to the functional capillary density.
Functional capillary density [mm]
Functional capillary density [mm]
250
200 150 100 50
0
n1
n2
n3 n4 n5 n1 Group A Flap
n2 n3 n4 Group B
n5
Figure 7. The functional capillary density (FCD) of the skin of the transplanted groin flap in group A (n = 5; no infusion) and group B (n = 5; with infusion of 1ml Ringer’s lactate solution prior to I/R), respectively. The flaps of group B show a statistical significant higher FCD (p > 0.01).
The free groin flap in the rat Discussion Part I: Microsurgical training model: Clinical background and applicability The first clinical experience in microsurgical preparation of blood vessels and nerves is often gained in the treatment of acute traumatic lesions or replantation surgery. The variety of the injuries, the post-traumatic alterations of the anatomic structures, and the irregularity of their incidence make the training difficult. Furthermore, the outcome depends on many variables besides the technical skills of the microsurgeon. Standardised and authentic conditions as well as regular training periods can raise the training curve. These pre-requisites can solely be found in animal models. Although many different microsurgical training models exist in a variety of animal species, the rat model is the most spread. Just in the rat there is a vast number of microsurgical flaps described such as the superficial epigastric or the pectoralis flap, the latissimus dorsi flap, the gastrocnemius muscle flap, or the omental flap [4,7,8]. The distinctive features of the free adipocutaneous groin flap as a microsurgical training model are the large diameter vessels and the comparatively easy harvest. The flap can be raised safely in a very short time period (M = 46 minutes), and the single and long microvascular pedicle of this flap can be anastomosed in an endto-end or end-to-side fashion to the recipient vessels. The vessel diameters usually fit the diameters of the carotid artery and the jugular vein, respectively. Its vascular pedicle is easily dissected from the surrounding connective tissue, on the contrary to e.g. flaps based on the axillary vessels [9]. The latter additionally have the significant disadvantage of severe complications such as bleeding and air embolism as a result of the proximity to the thoracic cavity [9,10]. An important advantage of the groin flap as compared with the thinner superficial epigastric flap is that the groin flap contains enough subcutaneous tissue to prevent the skin of the flap to survive solely by diffusion like a free skin graft. This is proven by the complete necrosis of all flaps after the generation of a venous stasis over 35 minutes (‘all-ornothing’-rule). Further, this free flap model qualifies for perfusion studies with identical standard conditions already after completion of no more than 20 flap transplantations, as it can be deduced from the flap failure rate (c.f. Table I). Part II: Determination of the critical time period of a complete venous stasis As with clinical experience, free flaps are much more susceptible to venous stasis than local flaps. It can be postulated that the critical period of a complete venous stasis normally leading to a flap failure is dependent on the particular type of flap and the microsurgical technique applied. The exact determination of this period in the particular model applied is essential with regard to further experiments in which effects of vasoactive or anticoagulant drugs on flap survival could be investigated. Both too short and too long periods of venous stasis could lead to a higher number of animals needed for a statistically significant drug effect. A too weak I/R damage may result in a spontaneously increased flap survival and, thus, protective effects of the test substance are harder to detect, especially since partial necrosis seems to be uncommon [11]. On the other hand, too severe I/R damage makes a successful flap protection more unlikely. Frick et al. [11] designated a period of 20 minutes as a fatal
195
event in transplanted free groin flaps in the rat. Lacking other comparable experimental studies we started our experiments to determine the fatal period of complete secondary venous stasis with 20 minutes and increased it stepwise according to the flap failure rate. The comparatively long result of 35 minutes as a fatal period is probably due to differences in the microsurgical procedure such as e.g. the use of other microsurgical clamps for the venous clamping. Part III: Experimental model for intra-vital video microscopy of comprised free flaps Microsurgical free transfer of the flap can be performed to the cervical or the contralateral groin region. For perfusion studies we recommend the cervical region as a recipient site because of the prevention of autocannibalization and to minimize mechanical skin alterations. The irritated flap surface can affect skin perfusion studies due to hyperaemia and fibrin appositions. Additionally, the preservation of the contralateral groin gives an optimal reference, e.g. for IVM. The location of the groin distant to the thorax and abdomen reduces surface movements due to cardiopulmonary action. The connection of the flap to the carotid artery as a central blood vessel assures a good blood supply and a sufficient blood pressure. This reduces the risk of a flap loss or regional hypoperfusion due to problems such as arterial hypotension. The results of the measurement of the FCD in the skin of the contralateral native groin show a minor variance of the values, with a mean value of 183 mm (SD = 32) in group A and 169 mm (SD = 30) in group B (c.f. Figure 7). This could be taken as qualification of the test method. On the contrary, the IVM of the transplanted flap in the region of the neck is more susceptible to interfering movements due to cardiopulmonary action. The FCD in the transplanted flaps was significantly lower than in the native groin (c.f. Figures 6 and 7). This is probably the case because of the severe ischaemia/reperfusion damage. However, an improvement in the FCD was detected when Ringer’s solution was administered before the venous stasis (Figure 7, group B). This might be ascribed to a haemodilution and a dilution of coagulation factors and toxic metabolic products, which accumulate during the ischaemic period. The consecutive improvement in microvascular perfusion (FCD) was measured 5 minutes after the onset of reperfusion. Interestingly, it did not increase flap survival in group B. The increased capillary diameters and the increased sticking of labelled thrombocytes in the flap skin are due to the complex I/R damage. The metabolic products obviously lead to a vascular hyperemia (i.e. increased FCD and capillary distension) as well as early activation of thrombocytes increasing rolling and sticking to the endothelial surface. However, the capillary diameters and the counts of sticking thrombocytes within the flaps were not significantly different between the two groups. Thus, rinsing the flap with Ringer’s solution did not significantly affect these parameters.
Conclusion The free groin flap transplanted to the neck is a simple and reliable model that may be used for refining microsurgical skills. The surgical procedure closely resembles a clinical
196 J. Wallmichrath et al. free flap transfer. Absolvation of a microsurgical course involving synthetic bench models and ~ 20 free flap transplantations with this animal model may provide a reliable and steady level of microsurgical experience for the application of this model for microvascular perfusion studies. The shortest time period of a complete venous stasis normally leading to a complete flap necrosis in this particular model was determined to be 35 minutes. This enables us to look for drugs increasing the survival of compromised free flaps with a minimum number of experimental animals and without sacrificing human safety. Further studies (work in progress) show that the fate of these ‘lost’ flaps can be manipulated by administration of certain drugs. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. References [1] Cooley BC, Werker PM. Review article of rat muscle and myocutaneous flap models. J Reconstr Microsurg 1995;11: 83–5. [2] Petry JJ, Wortham KA. The anatomy of the epigastric flap in the experimental rat. Plast Reconstr Surg 1984;74:410–13.
[3] Tilgner A, Herrberger U, Oswald P. Myocutaneous flap models in the rat. Anatomy, histology and operative technique of the latissimus dorsi myocutaneous flap. Z Versuchstierkd 1988;31: 225–32. [4] Zhang F, Sones WD, Lineaweaver WC. Microsurgical flap models in the rat. J Reconstr Microsurg 2001;17:211–21. [5] Acland R. Signs of patency in small vessel anastomosis. Surgery 1972;72:744–8. [6] Ishikawa M, Cooper D, Arumugam TV, et al. Plateletleukocyte-endothelial cell interactions after middle cerebral artery occlusion and reperfusion. J Cereb Blood Flow Metab 2004;24:907–15. [7] Zhang F, Kao SD, Walker R, Lineaweaver WC. Pectoralis major muscle free flap in rat model. Microsurgery 1994;15:853–6. [8] Zhang F, Lineaweaver WC, Kao S, et al. The greater omentum transplantation model in the rat. Microsurgery 1994;15: 269–73. [9] Allen RJ Jr, Chen CM, Warren SM. Free pectoral skin flap in the rat based on the long thoracic vessels: a new flap model for experimental study and microsurgical training. Ann Plast Surg 2008;61:479–80. [10] Miyamoto S, Takushima A, Okazaki M, et al. Free pectoral skin flap in the rat based on the long thoracic vessels: a new flap model for experimental study and microsurgical training. Ann Plast Surg 2008;61:209–14. [11] Frick A, Baumeister RG, Menger MD, et al. Secondary ischaemia in experimental free flaps–treatment by long acting prostacyclin analogues. Br J Plast Surg 1999;52:392–8.
Copyright of Journal of Plastic Surgery & Hand Surgery is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
The free groin flap in the rat: a model for improving microsurgical skills and for microvascular perfusion studies Jens Wallmichrath1, R. G. H. Baumeister2, O. Gottschalk3, R. E. Giunta1 & A. Frick1 1
Hand Surgery, Plastic Surgery and Aesthetic Surgery, University Hospital of the Ludwig-Maximilians-University-Grosshadern, Munich, Germany, 2Surgical Clinic Munich Bogenhausen, Munich, Germany and 3Department of Trauma Surgery, University Hospital of the LudwigMaximilians-University-Grosshadern, Munich, Germany Abstract The goal of this study was to evaluate the free groin flap in the rat transplanted to the neck as a tool for extending microsurgical skills and to assess its suitability as a model for microvascular perfusion studies following secondary venous ischaemia. An analysis of 60 consecutive groin flap transplantations was performed in male Sprague Dawley rats with special regard to anatomy and operation times (Part I, animals No. 1–60). Following flap transplantation, the animals No. 10–30 (n = 21) were used for the determination of the critical time period of a complete venous stasis of the free groin flap resulting in a total flap loss (Part II). The flaps of animals No. 31–41 (n = 11) were used for assessing the feasibility and reproducibility of intra-vital video microscopy (IVM) of the flaps (Part III). The mean total operation time decreased from 166 (± 26) minutes ins the first 10 animals to 126 (± 21) minutes and 130 (± 12) minutes in the latter two groups of 10 animals, respectively. After a critical period of 35 minutes of a complete artificial venous stasis a complete flap necrosis occurred. IVM detected a higher functional capillary density of the skin of the transplanted groin flaps in the animals in which the flaps were rinsed with 1 ml of Ringer’s lactated solution prior to I/R. In conclusion, this model is simple and reliable. The model may be a useful tool for evaluating and comparing the effects of various anticoagulants or vasomotor drugss on microvascular perfusion in critically compromised free flaps. Key Words: Groin flap, microsurgery, microsurgical training, intra-vital microscopy, ischaemia, reperfusions
Introduction Free flap reconstructions and microsurgical replantations require extensive microsurgical skills to optimize the success rate. Short operating times with minimal ischaemic periods are beneficial to proper wound healing, short rehabilitation periods, reduced financial expenses, and a good functional result. Thus, an authentic microsurgical training is a profitable investment for prospective microsurgeons. Various flap models exist in laboratory rats with a long history of use both as training models and for microsurgical experimental studies [1–4]. The thickness of the free groin flap based on the superficial iliac vessels is the feature of this flap since this makes flap survival solely dependent on the vascular pedicle. The goal of this study was to assess the learning curve of the free groin flap in the rat transplanted to the neck and its suitability as a model for microvascular perfusion studies following secondary venous ischaemia. This study consists of three parts: . .
Part I: Evaluation of the learning curve of the flap as a tool for extending microvascular and experimental skills. Part II: Identification of the critical time period of a complete secondary venous stasis of the flap resulting in a complete flap loss. This should serve as a basis for the development of an animal model for the examination of positive effects of vasoactive and/or anticoagulant agents on compromised free flaps following fatal secondary venous stasis.
.
Part III: Finally, the applicability of this model for experimental in vivo perfusion studies in free skin flaps involving intra-vital video microscopy (IVM) was performed. Here, the special focus was on the influence of disturbing artefacts due to limited flap size, cardiopulmonary motions of the region of the neck and the fluorescence of hair follicles.
Materials and methods Experimental design An analysis of 60 consecutive free groin flap transplantations was performed with special regard to anatomy and operation times (Part I, animals No. 1–60). Following flap transplantation, the animals No. 10–30 (n = 21) were used for the determination of the critical time period of a complete venous stasis of the free groin flap resulting in a flap loss (Part II). The flaps of animals No. 31–41 (n = 11) were used for assessing the feasibility and reproducibility of intra-vital video microscopy (IVM) of the skin of these compromised free groin flaps (Part III, exact experimental design is given below). The remaining animals were included in various perfusion studies (data not shown). Animal model and flap preparation Experiments were performed according to the guiding principles for research involving animals and the German legislation on protection of animals. Approval was obtained by the local
Correspondence: Jens Wallmichrath, MD, Hand Surgery, Plastic Surgery and Aesthetic Surgery, Surgical Clinic and Policlinic, University Hospital of the Ludwig-Maximilians University, Marchioninistrasse 15, D-81377 Munich, Germany. Tel: +49 89 7095 3502. Fax: +49 89 7095 6505. E-mail: [email protected] (Accepted 2 October 2013)
192 J. Wallmichrath et al. governmental animal care committee. Male Sprague-Dawley rats, weighing between 200–250 g, were purchased from Charles River (Sulzfeld, Germany), and housed at the Institute for Surgical Research (Walter Brendel Zentrum), LudwigMaximilians-University, Munich, Germany. Animals were allowed to accommodate to the standard conditions of the animal facility with free access to rat chow and water for at least 1 week. Solid food was withheld from the rats 12 hours prior to surgery. The rats were anaesthetised by intraperitoneal injection of 0.15 mg/kg Medetomidin (Pfizer GmbH, Karlsruhe, Germany), 0.01 mg/kg Fentanyl (CuraMED Pharma GmbH, Karlsruhe, Germany), and 2 mg/kg Midazolam (Hoffmann-La-Roche AG, Grenzach-Wyhlen, Germany) and supplemented as required. The animals were placed on a heated operating table in a supine position with the extended hind leg. The region of the right groin was shaved using an electrical shaver without harming the skin. The flap borders were outlined on the skin using a round template with 3 cm in diameter centred in the right groin (Figure 1). After the dissection of the skin and the subcutaneous tissue the flap can bluntly be separated from the abdominal muscles as well as the inguinal ligament, exposing the superficial epigastric vessels on the deep surface of the adipocutaneous groin flap. Then the comparatively thick and sturdy subcutaneous tissue of the lateral end of the groin was dissected and the superficial iliac vessels were ligated using Vicryl 5-0 (Polyglactin 910; Ethicon, Germany). The microsurgical preparation was started distally at the femoral vessels and preparation was continued proximally towards the groin (Figure 2). First, the femoral artery and vein were dissected and the femoral nerve was microsurgically isolated from the vessels and excluded from the flap. Now, the femoral vessels still intact, the muscular branches can easily be dissected and ligated. Microvascular preparation was continued with exposure and isolation of the proximal femoral artery and vein serving as the pedicle of the flap. Here, the profunda system was left intact not to compromise the blood supply of the hind leg. Finally, the distal end of the femoral artery was ligated 6 mm distally to the branch point of the flap
Flap
Pedicle diev fn
il
fa
fv / p
fv mb
Figure 2. The anatomy of the inguinal vessels in the right groin after elevation of the adipocutaneous groin flap. The hook is inserted in the proximal medial wound edge (fn = femoral nerve; fa = femoral artery; fv = femoral vein; mb = muscular branch; fv / p = femoral vessels and branch point of the profundal vessels; il = inguinal ligament; diev = deep inferior epigastric vessels).
vessels using Vicryl 5-0 (Polyglactin 910; Ethicon, Germany). The femoral vein was ligated separately. Following preparation of the recipient site, the vascular pedicle was transected and the stumps of the femoral vessels were ligated distally to the branch point of the profunda system. The flap size was outlined on the skin in the neck using the round template 3 cm in diameter arranged centrally in the neck (Figure 1). The left carotid artery and the internal jugular vein were exposed mainly by blunt dissection. The proximal ends were irrigated using Heparin solution (100 IU/ml) and clamped temporarily with single Acland vessel clamps Type B-3 (Art. No. 00400V and 00400A, respectively; S&T AG, Neuhausen, Switzerland). The distal vessel ends were ligated and the flap was transferred to the neck (Figure 3). The arteries and then the veins were anastomosed in an end-to-end fashion using a single stitch technique requiring up to 10 stitches (Ethilon 9-0 for the arterial and 8-0 for the venous anastomosis; Ethicon, Germany).
3cm
3cm
cm
Figure 1. Localisation of the donor site of the groin flap in the right inguinal region and the recipient site in the center of the cervical region.
1
2
3
Figure 3. The free adipocutaneous groin flap.
The free groin flap in the rat Then the Acland clamps were removed to allow reperfusion of the flap. Finally, the flap was sutured in place using single stitches and the donor site was closed with a buried single suture (Ethilon 3-0, Ethicon). Anaesthesia was terminated by intraperitoneal injection of 0.75 mg/kg atipamezolhydrochloride (Pfizer GmbH), 0.20 mg/kg Flumazenil (Hoffmann-La-Roche AG) and 0.12 mg/kg Naloxon (Ratiopharm GmbH, Ulm, Germany). The animals were monitored until recovery from anaethesia and then returned to their cages. After 20 hours, the animals were re-anaesthetised as described above and the flap and the contralateral groin were depilated using keratolytic creme (Elca Med Enthaarungscreme, Asid Bonz, Herrenberg, Germany). The wound was partially re-opened and the vascular pedicle was exposed. The patency of the anastomoses and the uncompromised blood flow was assessed by the method of Acland [5] using two microsurgical tweezers. Part II (determination of a fatal secondary venous stasis) For determining the shortest time period necessary to produce a complete flap necrosis, the venous clamping was performed for 20, 25, 30, or 35 minutes, respectively, in four groups with n = 5 animals each. Flap viability was assessed after 14 days. Part III (applicability of intra-vital video microscopy) Pre-operatively, the animals were randomized into group A (n = 5; no infusion) or group B (n = 5; infusion of 1 ml of Ringer’s solution). Following a positive Acland test 20 hours after the transplantation a sterile polyethylene tube was introduced into the distal stump of the femoral artery (Figure 4). In group A (control group) no infusion was administered. In group B, 1 ml of sterile Ringer’s solution was infused over 10 minutes using an infusion pump (Perfusor compact, B. Braun Melsungen AG, Melsungen, Germany). Afterwards, the catheter was removed and the arterial stump was ligated. One minute after the time period of the infusion a venous Acland vessel clamp (Type B-3, Art.No. 00400V, S&T AG, Neuhausen, Switzerland) was applied to the flap vein in all groups to produce a fatal venous stasis for 35 minutes (Figure 5). Then the clamp was removed and 1 minute later the Acland test was repeated on both artery and vein. The wound was closed as described above and the animal was transferred to a heated
Flap
193
a
v
Figure 5. The catheter is removed and the flap vein is clamped for 35 minutes using a microvascular clamp generating a temporary venous stasis.
microscope table in a supine position. IVM was carried out on the flap (Periplan, Leitz Wetzlar, with a CCD camera; Epiillumination with an adjustable100 W Xenon lamp (Zeiss, Oberkochen, BW, Germany)). The video analysis was performed offline (Cap-Image, Dr. Zeintl, Heidelberg, Germany). Blood was gained by cardiac puncture of a donor rat using a heparinised syringe and the thrombocytes were labelled with CFDA-SE (Carboxyfluorescein diacetate succinimidyl ester, Molecular Probes, Eugene, OR), as described elsewhere [6]. The fluorescent thrombocytes were infused intravenously and IVM was performed. The parameters measured were counts of floating and adherent (‘sticking’) thrombocytes (i.e. adhesion >30 seconds). Afterwards FITC-Dextran (Fluorescein isothiocyanate–dextran, Mw 50.000, St. Louis, MO) was used as a plasma marker. The amount of the injected fluorescence marker and time of exposure were minimised strictly to avoid illumination artefacts. The parameters measured were functional capillary density (FCD) and capillary diameters. The contralateral groin served as control. Measurements were performed 5-times in the central region (i.e. within a central circus of 1 cm in diameter) and another 5-times within the peripheral region. After a healing period of 14 days all animals were sacrificed and the vitality of the flaps was assessed. Statistical analysis The program PASW Statistics 17.0 (IBM Corporation, New York, NY) was used for statistical analysis. The Wilcoxon Rank-Sum Test was performed for statistical evaluations. Values of p < 0.05 were considered to be significant.
a Catheter v
Figure 4. The catheter for the application of the test substance is introduced into the distal stump of the femoral artery allowing a direct rinsing of the flap with the test solution.
Results Part I: The free groin flap transplantation The results of the analysis of the transplantation procedures are given in Table I. The body weight of the animals at time of surgery was 383 (± 86) g. The length of the vascular pedicle was 30 (± 2.4) mm, the vessel diameters ranged from 0.6–2.0 mm. The flap wet weight was 2.7 (± 0.7) g. The mean total operation time was 145 (± 28) minutes. The mean total operation time decreased from 166 (± 26) minutes in the first 10 animals to
194 J. Wallmichrath et al. Table I. The total operating time and time periods for flap preparation, the arterial anastomosis, and the venous anastomosis, respectively.
Total operating time (M [min]) Flap preparation (M [min]) Duration of ischaemia (M [min]) Arterial anastomosis (M [min]) Venous anastomosis (M [min]) Failure rate*, n (%)
All animals (No. 1–60)
Animal No. 1–10
Animal No. 11–20
Animal No. 21–30
Animal No. 31–40
Animal No. 41–50
Animal No. 51–60
145 (± 28) 46 (± 10) 59 (± 12) 22 (± 6) 19 (± 6) n = 8/60 (13%)
166 (± 26) 50 (± 8) 73 (± 18) 25 (± 8) 20 (± 6) n = 3/10 (30%)
141 (± 15) 46 (± 12) 60 (± 7) 21 (± 4) 21 (± 6) n = 2/10 (20%)
153 (± 28) 51 (± 7) 62 (± 11) 23 (± 7) 20 (± 2) n = 1/10 (10%)
140 (± 22) 43 (± 12) 52 (± 6) 21 (± 4) 18 (± 4) n = 1/10 (10%)
126 (± 21) 42 (± 6) 55 (± 10) 24 (± 5) 19 (± 9) n = 0/10 (10%)
130 (± 12) 45 (± 6) 55 (± 9) 20 (± 5) 15 (± 4) n = 1/10 (0%)
The failure rate comprises animals with flap perfusion problems (n = 5/60) as well as loss of an animal due to bad medical conditions (n = 3/60).
*
126 (± 21) minutes and 130 (± 12) minutes in the last two groups of 10 animals, respectively. This means a significant reduction in the total operating time by 24% (i.e. 40 minutes). The time period needed for flap preparation was comparatively constant with 50 (± 8) minutes in the beginning and ~ 42–45 minutes in the last animals (reduction by 10–16%). Also, the ischaemic period showed an improvement from 73 (±18) minutes in the first 10 animals to ~ 55 minutes in the last 20 animals (i.e. a reduction by 25%). The mean time needed for the arterial anastomoses was 22 (± 6) minutes, with 25 (± 8) minutes in the first 10 animals and 20–24 minutes in the last 30 animals. The mean time needed for the venous anastomoses was 19 (± 6) minutes, with 20 (± 6) minutes in the first 10 animals and 15 (±4) minutes in the last 10 animals. The post-operative weight loss during the first 24 hours was M = 4% (0–7%) of the body weight at the time of flap transplantation. The overall flap failure rate was 8/60 flaps (i.e. 13%) in the first consecutive series. The flap failure rates of clusters of 10 animals are given in Table I. The failure rates decrease with experience and were quite stable from animal No. 20 on. Part II: Determination of the critical time period of a complete venous stasis The four groups of five animals each sustained a complete venous stasis of the free flaps for 20, 25, 30, and 35 minutes,
respectively. This resulted in a complete necrosis of the flaps in the four groups in n = 0/5, n = 1/5, n = 4/5, and n = 5/5 cases, respectively. Part III: Intra-vital video microscopy (IVM) The results of the measurement of the functional capillary density (FCD) in the skin of the contralateral native groin in both groups are given in Figure 6. The native groins of the animals without infusion (Group A, n = 5) and with infusion of lactated Ringer’s solution (Group B, n = 5) prior to I/R show no statistically significant difference with regard to the FCD (Figure 6). On the other hand, the FCD of the skin of the transplanted groin flaps of group B show statistically significant higher values (p £ 0.01), as illustrated in Figure 7. The capillary diameters (CD) in the native groin were 4.9 ± 0.4 mm (Group A) and 4.7 ± 0.4 mm (Group B). The CD in the transplanted groin flap were 7. 4 ± 0.8 mm (Group A) and 5.8 ± 0.6 mm (Group B). The differences in the CD were statistically not significant between the groups. The mean number of thrombocytes sticking to the endothelium of the capillaries in the native groin was 0.2 per field of view in both groups. The mean number of thrombocytes sticking to the endothelium of the venules in the transplanted groin flap was 2.5 ± 1 per field of view in group A and 4.3 ± 2 per field of view in group B. The differences were statistically not significant between the two groups.
250 200
150
100 50
0 n1
n2 n3 n4 Group A
n5
n1 Groin
n2 n3 n4 Group B
n5
Figure 6. Functional capillary density [mm per field] in the skin of the contralateral native groin (Intravital video microscopy, FITC-Dextran). The animals without infusion (Group A, n = 5) and with infusion of lactated Ringer’s solution (Group B, n = 5) prior to I/R show no statistical significant difference with regard to the functional capillary density.
Functional capillary density [mm]
Functional capillary density [mm]
250
200 150 100 50
0
n1
n2
n3 n4 n5 n1 Group A Flap
n2 n3 n4 Group B
n5
Figure 7. The functional capillary density (FCD) of the skin of the transplanted groin flap in group A (n = 5; no infusion) and group B (n = 5; with infusion of 1ml Ringer’s lactate solution prior to I/R), respectively. The flaps of group B show a statistical significant higher FCD (p > 0.01).
The free groin flap in the rat Discussion Part I: Microsurgical training model: Clinical background and applicability The first clinical experience in microsurgical preparation of blood vessels and nerves is often gained in the treatment of acute traumatic lesions or replantation surgery. The variety of the injuries, the post-traumatic alterations of the anatomic structures, and the irregularity of their incidence make the training difficult. Furthermore, the outcome depends on many variables besides the technical skills of the microsurgeon. Standardised and authentic conditions as well as regular training periods can raise the training curve. These pre-requisites can solely be found in animal models. Although many different microsurgical training models exist in a variety of animal species, the rat model is the most spread. Just in the rat there is a vast number of microsurgical flaps described such as the superficial epigastric or the pectoralis flap, the latissimus dorsi flap, the gastrocnemius muscle flap, or the omental flap [4,7,8]. The distinctive features of the free adipocutaneous groin flap as a microsurgical training model are the large diameter vessels and the comparatively easy harvest. The flap can be raised safely in a very short time period (M = 46 minutes), and the single and long microvascular pedicle of this flap can be anastomosed in an endto-end or end-to-side fashion to the recipient vessels. The vessel diameters usually fit the diameters of the carotid artery and the jugular vein, respectively. Its vascular pedicle is easily dissected from the surrounding connective tissue, on the contrary to e.g. flaps based on the axillary vessels [9]. The latter additionally have the significant disadvantage of severe complications such as bleeding and air embolism as a result of the proximity to the thoracic cavity [9,10]. An important advantage of the groin flap as compared with the thinner superficial epigastric flap is that the groin flap contains enough subcutaneous tissue to prevent the skin of the flap to survive solely by diffusion like a free skin graft. This is proven by the complete necrosis of all flaps after the generation of a venous stasis over 35 minutes (‘all-ornothing’-rule). Further, this free flap model qualifies for perfusion studies with identical standard conditions already after completion of no more than 20 flap transplantations, as it can be deduced from the flap failure rate (c.f. Table I). Part II: Determination of the critical time period of a complete venous stasis As with clinical experience, free flaps are much more susceptible to venous stasis than local flaps. It can be postulated that the critical period of a complete venous stasis normally leading to a flap failure is dependent on the particular type of flap and the microsurgical technique applied. The exact determination of this period in the particular model applied is essential with regard to further experiments in which effects of vasoactive or anticoagulant drugs on flap survival could be investigated. Both too short and too long periods of venous stasis could lead to a higher number of animals needed for a statistically significant drug effect. A too weak I/R damage may result in a spontaneously increased flap survival and, thus, protective effects of the test substance are harder to detect, especially since partial necrosis seems to be uncommon [11]. On the other hand, too severe I/R damage makes a successful flap protection more unlikely. Frick et al. [11] designated a period of 20 minutes as a fatal
195
event in transplanted free groin flaps in the rat. Lacking other comparable experimental studies we started our experiments to determine the fatal period of complete secondary venous stasis with 20 minutes and increased it stepwise according to the flap failure rate. The comparatively long result of 35 minutes as a fatal period is probably due to differences in the microsurgical procedure such as e.g. the use of other microsurgical clamps for the venous clamping. Part III: Experimental model for intra-vital video microscopy of comprised free flaps Microsurgical free transfer of the flap can be performed to the cervical or the contralateral groin region. For perfusion studies we recommend the cervical region as a recipient site because of the prevention of autocannibalization and to minimize mechanical skin alterations. The irritated flap surface can affect skin perfusion studies due to hyperaemia and fibrin appositions. Additionally, the preservation of the contralateral groin gives an optimal reference, e.g. for IVM. The location of the groin distant to the thorax and abdomen reduces surface movements due to cardiopulmonary action. The connection of the flap to the carotid artery as a central blood vessel assures a good blood supply and a sufficient blood pressure. This reduces the risk of a flap loss or regional hypoperfusion due to problems such as arterial hypotension. The results of the measurement of the FCD in the skin of the contralateral native groin show a minor variance of the values, with a mean value of 183 mm (SD = 32) in group A and 169 mm (SD = 30) in group B (c.f. Figure 7). This could be taken as qualification of the test method. On the contrary, the IVM of the transplanted flap in the region of the neck is more susceptible to interfering movements due to cardiopulmonary action. The FCD in the transplanted flaps was significantly lower than in the native groin (c.f. Figures 6 and 7). This is probably the case because of the severe ischaemia/reperfusion damage. However, an improvement in the FCD was detected when Ringer’s solution was administered before the venous stasis (Figure 7, group B). This might be ascribed to a haemodilution and a dilution of coagulation factors and toxic metabolic products, which accumulate during the ischaemic period. The consecutive improvement in microvascular perfusion (FCD) was measured 5 minutes after the onset of reperfusion. Interestingly, it did not increase flap survival in group B. The increased capillary diameters and the increased sticking of labelled thrombocytes in the flap skin are due to the complex I/R damage. The metabolic products obviously lead to a vascular hyperemia (i.e. increased FCD and capillary distension) as well as early activation of thrombocytes increasing rolling and sticking to the endothelial surface. However, the capillary diameters and the counts of sticking thrombocytes within the flaps were not significantly different between the two groups. Thus, rinsing the flap with Ringer’s solution did not significantly affect these parameters.
Conclusion The free groin flap transplanted to the neck is a simple and reliable model that may be used for refining microsurgical skills. The surgical procedure closely resembles a clinical
196 J. Wallmichrath et al. free flap transfer. Absolvation of a microsurgical course involving synthetic bench models and ~ 20 free flap transplantations with this animal model may provide a reliable and steady level of microsurgical experience for the application of this model for microvascular perfusion studies. The shortest time period of a complete venous stasis normally leading to a complete flap necrosis in this particular model was determined to be 35 minutes. This enables us to look for drugs increasing the survival of compromised free flaps with a minimum number of experimental animals and without sacrificing human safety. Further studies (work in progress) show that the fate of these ‘lost’ flaps can be manipulated by administration of certain drugs. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. References [1] Cooley BC, Werker PM. Review article of rat muscle and myocutaneous flap models. J Reconstr Microsurg 1995;11: 83–5. [2] Petry JJ, Wortham KA. The anatomy of the epigastric flap in the experimental rat. Plast Reconstr Surg 1984;74:410–13.
[3] Tilgner A, Herrberger U, Oswald P. Myocutaneous flap models in the rat. Anatomy, histology and operative technique of the latissimus dorsi myocutaneous flap. Z Versuchstierkd 1988;31: 225–32. [4] Zhang F, Sones WD, Lineaweaver WC. Microsurgical flap models in the rat. J Reconstr Microsurg 2001;17:211–21. [5] Acland R. Signs of patency in small vessel anastomosis. Surgery 1972;72:744–8. [6] Ishikawa M, Cooper D, Arumugam TV, et al. Plateletleukocyte-endothelial cell interactions after middle cerebral artery occlusion and reperfusion. J Cereb Blood Flow Metab 2004;24:907–15. [7] Zhang F, Kao SD, Walker R, Lineaweaver WC. Pectoralis major muscle free flap in rat model. Microsurgery 1994;15:853–6. [8] Zhang F, Lineaweaver WC, Kao S, et al. The greater omentum transplantation model in the rat. Microsurgery 1994;15: 269–73. [9] Allen RJ Jr, Chen CM, Warren SM. Free pectoral skin flap in the rat based on the long thoracic vessels: a new flap model for experimental study and microsurgical training. Ann Plast Surg 2008;61:479–80. [10] Miyamoto S, Takushima A, Okazaki M, et al. Free pectoral skin flap in the rat based on the long thoracic vessels: a new flap model for experimental study and microsurgical training. Ann Plast Surg 2008;61:209–14. [11] Frick A, Baumeister RG, Menger MD, et al. Secondary ischaemia in experimental free flaps–treatment by long acting prostacyclin analogues. Br J Plast Surg 1999;52:392–8.
Copyright of Journal of Plastic Surgery & Hand Surgery is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
