Experimental cooling-induced preconditioning attenuates skin flap failure.

BACKGROUND: Microvascular perfusion failure is a leading cause of tissue necrosis in reconstructive surgery. In the present experimental study the effect of local hypothermia was investigated as a possible preconditioning procedure that could induce stress proteins such as heat-shock protein (HSP) 70 and HSP-32 (haem oxygenase (HO) 1). The effect on flap microcirculation and survival was also studied. METHODS: Ears of hairless mice were subjected to local hypothermia (30 min, 4 degrees C) 24 h before flap creation. A pedicled flap was elevated by incision of four-fifths of the base of the ear. Microcirculatory dysfunction and tissue necrosis were analysed quantitatively over 5 days by means of intravital fluorescence microscopy. HO-1 and HSP-70 protein expression were determined by western blot analysis. HO-1 distribution within the flap tissue was also analysed by immunohistochemistry. Animals with unconditioned flaps served as controls. RESULTS: Cooling induced a marked expression of HO-1 without induction of HSP-70 protein. This was paralleled by a significant improvement in microvascular perfusion (P < 0.050) that was predominantly regulated by the dilatation of nutritive capillaries. The cooling-mediated improvement in microcirculation resulted in a significant reduction in final flap necrosis (P < 0.050). CONCLUSION: In this experimental study preoperative cooling was associated with the expression of HO-1 and was an effective conditioning procedure.

Microvascular transfer-related abrogation of capillary flow motion in critically reperfused composite flaps.

Capillary flow motion is defined as rhythmic fluctuations of blood flow in the capillaries. Although critical perfusion has been demonstrated to induce capillary flow motion, little is known about the role of capillary flow motion in microvascular free flaps. The aim of this study was to elucidate the tissue-confined incidence and consequence of capillary flow motion in microvascularly transferred composite flaps, using intravital fluorescence microscopy. In Wistar rats, transferred osteomyocutaneous flaps (n = 7), which were exposed to 1 h of ischaemia during the anastomotic procedure followed by 1 h of reperfusion, were subjected to critical perfusion by stepwise reduction of the femoral-artery blood flow to 0.15 ml min(-1), 0.10 ml min(-1) and 0.05 ml min(-1). Pedicled osteomyocutaneous flaps that were not subjected to ischaemia (n=8) served as controls. In pedicled flaps critical perfusion induced capillary flow motion in the muscle, but not in the skin, subcutis and periosteum. In these flaps, the functional capillary density was preserved in all tissues analysed, including the skeletal muscle. Additional sympathetic denervation of the pedicled flaps did not change the incidence or pattern of capillary flow motion. In contrast, after flap transfer capillary flow motion in muscle tissue did not occur during critical perfusion. As a consequence, a shutdown of perfusion of individual capillaries was observed, resulting in a significant reduction (P<0.05) in functional capillary density, not only in the subcutis, skin and periosteum but also in the muscle itself. Thus, our data suggest that the microcirculatory control of pedicled osteomyocutaneous flaps is preserved during critical perfusion by skeletal-muscle capillary flow motion, whereas this protective regulatory mechanism is lost during the initial reperfusion period after flap transfer, probably not because of denervation but because of surgery- and/or ischaemia-reperfusion-associated injury.

Reduction of inflammatory response in composite flap transfer by local stress conditioning-induced heat-shock protein 32.

BACKGROUND: The failure of composite flaps despite anastomotic patency is thought to be mediated by the inflammatory response within the microvasculature, which results from unavoidable surgical trauma and transfer-related ischemia-reperfusion. Evidence suggests that stress conditioning may improve flap survival; however, the molecular mechanisms of protection are far from being clear. Therefore, we analyzed whether stress conditioning-induced heat-shock protein 32 is effective to prevent the inflammatory response in transferred osteomyocutaneous flaps. METHODS: In a rat model, leukocyte-endothelial cell interaction and endothelial integrity disruption as early indicators of the inflammatory response were quantitatively analyzed in muscle, subcuticular tissue, and periosteum of microvascularly transferred osteomyocutaneous flaps by using intravital fluorescence microscopy. Twenty-four hours before flap transfer, stress conditioning was induced by local heating of the left hindlimb up to 42.5 degrees C for 30 minutes. In additional animals, stress conditioning-induced activity of heat-shock protein 32 was inhibited by tin protoporphyrin-IX. Unconditioned flaps served as controls. RESULTS: In all tissues analyzed, control flaps showed significant leukocyte adherence in postcapillary venules, increased intercellular adhesion molecule-1 (ICAM-1) expression, and endothelial integrity disruption, but a lack of heat-shock protein 32. In contrast, stress conditioning induced marked heat-shock protein 32 expression, which was associated with a significant reduction (P <.05) of leukocyte adherence, ICAM-1 expression, and endothelial hyperpermeability. The inhibition of heat-shock protein 32 by tin protoporphyrin-IX completely abolished the stress conditioning-induced amelioration of the inflammatory response in all tissues analyzed. CONCLUSIONS: Stress conditioning by local heat-shock priming reduces the inflammatory response in osteomyocutaneous flaps. The protective effect is predominantly mediated by the induction of heat-shock protein 32.

Local heat-shock priming-induced improvement in microvascular perfusion in osteomyocutaneous flaps is mediated by heat-shock protein 32.

BACKGROUND: Stress conditioning is thought to improve microvascular free flap survival but the mechanisms of protection are not clear. The aim of this study was to determine whether local induction of heat-shock protein (HSP) 32 improves microvascular perfusion in transferred osteomyocutaneous flaps. METHODS: The hindlimb harvest region of osteomyocutaneous flaps in Wistar rats was subjected to stress conditioning by local heating (30 min, 42.5 degrees C) 24 h before microvascular flap transfer. In a second group of animals, after heat-shock priming, the action of HSP-32 was inhibited by tin protoporphyrin IX. Animals with unconditioned flaps served as controls. After transfer, the microcirculation of the muscle, cutaneous, subcutaneous and periosteal tissue of the flap was analysed quantitatively for 6 h using intravital fluorescence microscopy. RESULTS: Immunohistochemistry revealed that HSP-32 was detectable only after priming and not in unconditioned flaps. Priming did not alter functional capillary density or capillary red blood cell velocity compared with that in unconditioned flaps. However, heat-shock priming induced significant capillary dilatation (P < 0.05) and thus a substantial increase in capillary blood flow volume (P < 0.05) in all tissues of the transferred flaps. Inhibition of HSP-32 by tin protoporphyrin IX completely abolished the priming-induced improvement in capillary perfusion, as indicated by the lack of increased capillary diameters and volumetric blood flow. CONCLUSION: The present study demonstrated that stress conditioning by local heat-shock priming improves nutritive perfusion in osteomyocutaneous flaps by capillary dilatation, probably mediated through the vasoactive action of HSP-32.

Vasomotion in critically perfused muscle protects adjacent tissues from capillary perfusion failure.

We analyzed the incidence and interaction of arteriolar vasomotion and capillary flow motion during critical perfusion conditions in neighboring peripheral tissues using intravital fluorescence microscopy. The gracilis and semitendinosus muscles and adjacent periosteum, subcutis, and skin of the left hindlimb of Sprague-Dawley rats were isolated at the femoral vessels. Critical perfusion conditions, achieved by stepwise reduction of femoral artery blood flow, induced capillary flow motion in muscle, but not in the periosteum, subcutis, and skin. Strikingly, blood flow within individual capillaries was decreased (P < 0.05) in muscle but was not affected in the periosteum, subcutis, and skin. However, despite the flow motion-induced reduction of muscle capillary blood flow during the critical perfusion conditions, functional capillary density remained preserved in all tissues analyzed, including the skeletal muscle. Abrogation of vasomotion in the muscle arterioles by the calcium channel blocker felodipine resulted in a redistribution of blood flow within individual capillaries from cutaneous, subcutaneous, and periosteal tissues toward skeletal muscle. As a consequence, shutdown of perfusion of individual capillaries was observed that resulted in a significant reduction (P < 0.05) of capillary density not only in the neighboring tissues but also in the muscle itself. We conclude that during critical perfusion conditions, vasomotion and flow motion in skeletal muscle preserve nutritive perfusion (functional capillary density) not only in the muscle itself but also in the neighboring tissues, which are not capable of developing this protective regulatory mechanism by themselves.

Acceleration of wound healing by topical drug delivery via liposomes.

BACKGROUND: Despite intensive research, impaired wound healing remains a considerable complication. Therefore, topically applied liposome-encapsulated buflomedil hydrochloride was investigated for its ability to improve wound repair in normal (n=16) and ischemic (n=16) skin tissue. METHODS: Experiments were performed using the wound healing model of the ear of hairless mice. Standardized skin wounds (4.25 mm(2)) were created by circular excision of the epidermal layer and the subcutaneous tissue. Liposomes were applied daily until complete neovascularization of the wound occurred. Tissue regeneration by complete epithelialization and neovascularization of the wound area, microcirculatory parameters, and leukocyte-endothelium interaction were investigated by means of intravital microscopy. Microvascular perfusion was assessed by laser-Doppler flowmetry. RESULTS: Topical application of buflomedil liposomes led to a significantly (P<0.05) accelerated wound closure in both normal (9.6+/-0.7 days) and ischemic (13.4+/-0.1 days) skin tissue compared with animals that were treated with unloaded liposomes (controls; 13.1+/-0.8 days; 15.3+/-0.6 days). Complete neovascularization of the wound was also enhanced (P<0.05) in buflomedil-treated animals (normal tissue 18.8+/-0.4 days; ischemic tissue 19.6+/-0.7 days) compared with controls (20.6+/-0.6 days; 22. 6+/-1.2 days). CONCLUSION: These data suggest that buflomedil-loaded liposomes might be of beneficial use for clinical wound care.

In vivo analysis of the microcirculation of osteomyocutaneous flaps using fluorescence microscopy.

Previous studies have indicated that freely transferred osteomyocutaneous flaps may fail despite anastomotic patency. While microvascular dysfunction is thought to be one of the major causes for this type of flap failure, little is known of its underlying mechanisms, probably due to the lack of adequate experimental models allowing detailed intravital microcirculatory analysis. Herein we report quantitative analysis of the microcirculation of periosteum, muscle, subcutis and skin by intravital fluorescence microscopy using an osteomyocutaneous free flap model in the hindlimb of rats. The microcirculation of the different tissues was studied after microanastomotic transfer (free flap), and was compared to that after solely elevating the tissue, mimicking a pedicled osteomyocutaneous flap. Transferred flaps, which were exposed to 1 h of ischaemia during the anastomotic procedure, showed a slight but significant decrease (P< 0.05) of functional capillary density in muscle, subcutis and skin when compared with the microcirculation of pedicled flaps, while capillary diameters, red blood cell velocity and blood flow of perfused capillaries remained almost unaffected. The decrease of functional capillary density was associated by a significant (P< 0.05) inflammatory response, as indicated by the increased number of leukocytes adherent to the endothelial lining of postcapillary venules. While the functional capillary density of periosteum was not affected by the free transfer procedure, the inflammatory response was found similar when compared with that observed in muscle and subcutis. Thus, our study indicates that even after a short 1-h ischaemic time period, capillary perfusion failure and leukocyte-endothelial cell interaction are the main events, characterising microvascular dysfunction after free transfer of osteomyocutaneous flaps. Using the model described herein, intravital microscopic analysis of the microcirculation proved an appropriate tool to study the individual microvascular response after free tissue transfer, and may thus be used to evaluate the effectiveness of novel therapeutic regimens which aim at counteracting microcirculatory dysfunction in free osteomyocutaneous flaps.

In vivo analysis of antithrombotic effectiveness of recombinant hirudin on microvascular thrombus formation and recanalization.

PURPOSE: This study was undertaken to evaluate in vivo the effect of recombinant hirudin (r-hirudin [HBW 023]), a potent thrombin inhibitor, on the process of microvascular thrombus formation and recanalization. METHODS: Thrombosis was induced photochemically in distinct arterioles (n = 25) and venules (n = 30) of the ear of 16 hairless hr/hr mice (8 to 10 weeks old, 25 to 30 g of body weight). r-Hirudin (1 mg/kg of body weight) was administered intravenously directly before thrombus induction; saline-treated animals served as controls. Thrombus formation (i.e., first platelet deposition at the endothelial lining [FPD]; inner luminal diameter reduction to 50% [D/2]; complete vessel occlusion [CVO]), vessel recanalization, microcirculatory parameters, and leukocyte-endothelial cell interaction were analyzed by means of intravital fluorescence microscopy. RESULTS: Hirudin significantly delayed the process of thrombus formation compared with saline-treated controls in both arterioles (FPD: 381 +/- 80 vs 137 +/- 25 seconds, P < 0.05; D/2: 627 +/- 49 vs 501 +/- 71 seconds; CVO: 925 +/- 78 vs 854 +/- 60 seconds) and venules (FPD: 173 +/- 11 vs 59 +/- 4 seconds; D/2: 342 +/- 54 vs 228 +/- 27 seconds; CVO: 541 +/- 85 vs 344 +/- 43 seconds; P < 0.05). In addition, r-hirudin-treated animals showed an increased rate of vessel recanalization at 24 hours after thrombus induction (arterioles: 54% [7 of 13] vs 0% [0 of 12], P < 0.05; venules: 77% [10 of 13] vs 53% [9 of 17]), whereas microcirculatory parameters and leukocyte-endothelial cell interaction were not affected. CONCLUSION: Our data indicate that r-hirudin not only counteracts the process of thrombus formation but also promotes vessel recanalization, thus supporting its use in clinical microvascular surgery.

A novel approach for comparative study of periosteum, muscle, subcutis, and skin microcirculation by intravital fluorescence microscopy.

Herein, we report a new model, which allows comparative study of the microcirculation of different peripheral tissues, i.e., periosteum, skeletal muscle, subcutis, and skin. Using dextran-insensitive Wistar rats gracilis and semitendinosus muscles of the left hindlimb were prepared in association with their appertaining tibial fragments, subcutis, and skin. Blood supply was guaranteed by the femoral artery via the saphenous vessels. High-resolution intravital epi-illumination microscopy of the two muscles displayed the typical microvascular architecture with the capillaries running in parallel to each other (capillary density (CD) 128.4 +/- 4.5 cm-1). In subcutis and skin, capillaries were found arranged as interconnecting mesh-like networks with a density, which was significantly higher (P < 0.05) in subcutis (191.0 +/- 5.5 cm-1) compared with skin (108.9 +/- 3.3 cm-1). Analysis of periosteal tissue revealed two distinct types of arrangements of microvascular architecture. Adjacent to the major feeding and draining vessels of the periosteum, capillaries were organized in densely meshed shunt-like networks, revealing the highest capillary density (242.7 +/- 13.2 cm-1; P < 0.05) of all tissues studied. Periosteal capillaries distant from the major feeding and draining vessels were arranged in parallel to the longitudinal axis of the tibial bone and presented with a density similar to that of the skeletal muscle (128. 6 +/- 9.4 cm-1). Topical application of acetylcholine for analysis of physiological reactivity of the microvasculature showed dose-dependent arteriolar dilation. Moreover, a 3-min upstream femoral artery occlusion demonstrated an appropriate hyperemic response in all tissues studied, indicating intact myogenic control. A prolonged period of ischemia (120 min) followed by reperfusion (60 min) caused massive (P < 0.05) leukocyte-endothelial cell interaction in postcapillary venules, similarly as reported in other microvascular tissue preparations. We propose that the model presented provides a good approach to all peripheral tissues for both the analysis of the physiology of tissue-confined microvascular control and the development of novel therapeutic strategies to counteract manifestation of nutritional dysfunction and inflammatory response in disease.

A new model for quantitative in vivo microscopic analysis of thrombus formation and vascular recanalisation: the ear of the hairless (hr/hr) mouse.

The alteration of rheological blood properties as well as deterioration of vascular perfusion conditions and cell-cell interactions are major determinants of thrombus formation. Herein, we present an experimental model which allows for quantitative in vivo microscopic analysis of these determinants during both thrombus formation and vascular recanalisation. The model does not require surgical preparation procedures, and enables for repeated analysis of identical microvessels over time periods of days or months, respectively. After i.v. administration of FITC-dextran thrombus formation was induced photochemically by light exposure to individual arterioles and venules of the ear of ten anaesthetised hairless mice. In venules, epi-illumination induced rapid thrombus formation with first platelet deposition after 0.59 +/- 0.04 min and complete vessel occlusion within 7.48 +/- 1.31 min. After a 24-h time period, 75% of the thrombosed venules were found recanalised. Marked leukocyte-endothelial cell interaction in those venules indicated persistent endothelial cell activation and/or injury, even after an observation period of 7 days. In arterioles, epi-illumination provoked vasomotion, while thrombus formation was significantly (p <0.05) delayed with first platelet deposition after 2.32 +/- 0.22 min and complete vessel occlusion within 20.07 +/- 3.84 min. Strikingly, only one of the investigated arterioles was found recanalised after 24 h, which, however, did not show leukocyte-endothelial cell interaction. Heparin (300 U/kg, i.v.) effectively counteracted the process of thrombus formation in this model, including both first platelet deposition and vessel occlusion. We conclude that the model of the ear of the hairless mouse allows for distinct in vivo analysis of arteriolar and venular thrombus formation/recanalisation, and, thus, represents an interesting tool for the study of novel antithrombotic and thrombolytic strategies, respectively.

Leukocyte rolling in venules of striated muscle and skin is mediated by P-selectin, not by L-selectin.

Leukocyte rolling in post-capillary venules is mediated by adhesion molecules of the selectin family expressed on both leukocytes (L-selectin) and endothelial cells (E- and P-selectin). With the use of intravital fluorescence microscopy, the effects of antibodies against these selectins were analyzed in the skinfold chamber model of BALB/c mice and the ear model of homozygous hairless mice (hr/hr) that permit chronic observation of striated muscle and skin microcirculation in awake animals, respectively. Mice were injected intravenously with monoclonal antibodies (MAb) to murine L-selectin and E-selectin and affinity-purified polyclonal antibodies to P-selectin. The antibodies, which are known to block cell adhesion, were tested by immunoprecipitation to selectively bind to L-, E-, or P-selectin. Leukocyte rolling was a constant finding in both microcirculation models in the absence of inflammatory stimuli. In both models, injection of anti-P-selectin antibodies completely prevented baseline leukocyte rolling over an observation period of 2 h (P < 0.01 vs. baseline), while no effects were seen after administration of either anti-L-selectin or anti-E-selectin MAb. Treatment with the isotype-matched control antibodies did not affect leukocyte rolling in either model. We conclude that leukocyte rolling in postcapillary venules of murine striated muscle and skin is a physiological process mediated via P-selectin, whereas L- and E-selectin appear not to play a significant role under these circumstances.