Allan C. Roth

Morphologic analysis of the microcirculation during reperfusion of ischemic skeletal muscle and the effect of hyperbaric oxygen.

The morphologic events in the microcirculation that lead to reperfusion injury of ischemic skeletal muscle remain incompletely understood. The purpose of this experiment was to evaluate leukocyte endothelial adherence characteristics and dynamic changes in microvessel caliber during reperfusion of an in vivo skeletal muscle ischemia preparation. In addition, the effect of hyperbaric oxygen treatment on these microcirculatory changes also was studied. An intravital microscopy preparation of a transilluminated gracilis muscle in 27 rats was used to observe a total of 101 arterioles and 63 venules (13 to 73 μm diameter). Baseline hemodynamics were videotaped for 30 minutes following muscle isolation. The animals were divided into six groups: (1) sham, no ischemia, (2) 4 hours of global ischemia only, (3) no ischemia plus hyperbaric oxygen (one 2.5 ATA/1 hour of treatment with 100% oxygen), (4) 4 hours of ischemia plus hyperbaric oxygen during ischemia, (5) 4 hours of ischemia plus hyperbaric oxygen immediately on reperfusion, and (6) 4 hours of ischemia plus hyperbaric oxygen 1 hour after reperfusion. Changes in arteriolar and venular diameters at specific times during 3 hours of reperfusion were recorded, and the number of adherent and slow-rolling leukocytes in 100-μm venular segments were counted and compared with baseline measurements. The proximity of arterioles to venules was classified as adjacent (<15 μm) or distant (>15 μm). No significant changes in leukocyte endothelial adherence or arteriolar diameter were noted in group 1 sham or group 3 nonischemic hyperbaric oxygen-treated rats when compared with baseline measurements. A significant increase in adherent leukocytes was observed in group 2 ischemic venules (+14.9 ± 2.5) within 5 minutes of reperfusion, which was maintained for 3 hours. Reperfusion measurements of arteriolar diameter in group 2 ischemic muscle preparations demonstrated an initial vasodilation that was followed at 1 hour by a progressive and severe vasoconstriction (−46.9 ±11.3 percent at 3 hours) in arterioles adjacent to venules that was not seen in distant arterioles. The increase in adherent leukocytes seen in group 2 ischemic venules was significantly reduced by hyperbaric oxygen treatment given during ischemia (group 4) or up to 1 hour during reperfusion (groups 5 and 6). In addition, the progressive ischemic arteriolar vasoconstriction was inhibited in all groups (4, 5, and 6) treated with hyperbaric oxygen. These results suggest that (1) leukocyte venular endothelial adherence and microarteriolar vasoconstriction are important morphologic events leading to reperfusion injury of skeletal muscle, (2) this vasoconstriction is seen primarily in arterioles that are in close proximity to venules, and (3) hyperbaric oxygen treatment does not exacerbate reperfusion injury, but rather appears to protect the microcirculation by reducing venular leukocyte adherence and inhibiting progressive adjacent arteriolar vasoconstriction.

The effect of hyperbaric oxygen on reperfusion of ischemic axial skin flaps: a laser Doppler analysis.

This study evaluates the microvascular reperfusion of ischemic skin flaps with and without acute hyperbolic oxygen (HBO) treatment. Thirty-two axial pattern epigastric skin flaps (3 x 6 cm) in male Wistar rats were subjected to 8 hours of global ischemia by pedicle clamp occlusion. The rats were divided into the following control and two experimental groups: Control (n = 12) with ischemia, no HBO; Group 1 (n = 11) with HBO treatment (three 1.75-hour dives, 2.5 absolute atm, 100% O2) during ischemia; and Group 2(n = 9) with HBO treatment (two 1.75-hour dives) immediately after ischemia. Laser Doppler flows were recorded in two distal standardized flap locations at 0.5, 2, 4, and 18 hours after reperfusion in control rats and Group 1 rats and at 18 hours only in Group 2 rats, using a Med-Pacific 6000 laser Doppler unit. Mean distal flap laser Doppler flows (mV) were Control: 0.5 hours=23.2 ± 11.9, 2 hours=52.8 ± 27.3,4hours=53.6 ± 32.1,18hours=40.2 ± 36.2; Group 1: 0.5 hours=71.8 ± 30.9 (p<0.05 vs. control), 2 hours=74.3 ± 27.3, 4 hours=67.4 ± 20.6, 18 hours=79.1 ± 40.3 (p<0.05 vs. control); and Group 2:18 hours=90.3 ± 47.9 (p<0.05 vs. control). It is concluded that acute HBO treatment of ischemic rat skin flaps improves distal microvascular perfusion as measured by laser Doppler flowmetry. This effect is observed for HBO treatment given either during or immediately after prolonged global ischemia. Although the mechanism of this beneficial effect on the microcirculation remains unknown, these results correlate with earlier findings that HBO treatment improves survival of rat axial skin flaps subjected to prolonged global ischemia.

Stimulation of angiogenesis to improve the viability of prefabricated flaps

&NA; The cutaneous area in a prefabricated myocutaneous flap surviving after elevation is dependent on the rate and amount of vascular ingrowth that occurs from the underlying muscle. Two modalities, basic fibroblast growth factor and hyperbaric oxygen, were used separately and together in a prefabricated myocutaneous flap animal model to improve flap survival. The semimembranous muscle, based on the saphenous vessels of 40 female Wistar rats weighing between 250 and 325 grams, was tunneled under the ipsilateral abdominal skin and sutured in place. A 3 × 5‐cm silicone sheet was placed beneath the muscle flap, and the ipsilateral epigastric vessels were ligated. Four groups of 10 animals each received one of the following treatment regimes: a 1‐ml normal saline infusion into the saphenous arterial pedicle, a 1‐ml infusion of basic fibroblast growth factor (1.0 &mgr;g/gm of muscle), a 1‐ml normal saline infusion and 14 hyperbaric oxygen treatments, or a 1‐ml basic fibroblast growth factor infusion and 14 hyperbaric oxygen treatments. After 1 week, the muscle, still based on the saphenous vessels, was elevated with a 3 × 5‐cm abdominal skin paddle. The flap was sutured back in place, leaving the silicone sheet intact. The surviving area of each flap was measured 1 week later after it had demarcated into viable and necrotic regions. Laser Doppler skin perfusion measurements were taken before and after flap elevation and before animal euthanasia. Sixteen flaps, 4 in each group, were examined histologically for vascularity by means of hematoxylin and eosin staining. There was a statistically significant increase in flap survival area when either basic fibroblast growth factor or hyperbaric oxygen was used alone. Further improvement was noted with combination therapy. Histology confirmed improved vascularity in the basic fibroblast growth factor and hyperbaric oxygentreated flaps. This study shows a significant and reliable increase in the area of prefabricated myocutaneous flap survival using either basic fibroblast growth factor or hyperbaric oxygen. There is a further complementary effect when these two modalities are combined, leading to near complete flap survival through improved vascularity. (Plast. Reconstr. Surg. 101: 1290, 1998.)

IN VITRO TOXICITY TESTING FOR ANTIBACTERIALS AGAINST HUMAN KERATINOCYTES

The use of cultured human keratinocytes in an in vitro comparison of topical antibacterial toxicity for epithelial cells was examined. The complement of three assessments allows testing of epithelial migration, growth, and survival. The three assessments included (1) flow cytometry for determination of cell survival, (2) a comparison of confluent cell culture growth after antibacterial exposures, and (3) an evaluation of cell migration using a technique of dermal explains to study radial migration. A comparative ranking of the toxicities of the various topical antibacterials was determined with the three assessments. This has confirmed anecdotal reports that many of the topical antibacterials are cell-toxic and may inhibit wound healing. This information can be directly extrapolated to the clinical setting, unlike many of the animal data for wound healing that currently exist.

Ischemia-reperfusion injury in skeletal muscle: CD 18-dependent neutrophil-endothelial adhesion and arteriolar vasoconstriction.

&NA; The purpose of this study was to evaluate if the venular neutrophil‐endothelial adhesion associated with ischemia‐reperfusion of skeletal muscle is dependent on leukocyte adhesion glycoprotein CD18 function and to determine if this interaction influences the vasoactive response in nearby arterioles. An in vivo microscopy preparation of transilluminated gracilis muscle in 13 male Wistar rats was used for this experiment. Observations of nonischemic muscle (sham) demonstrated this preparation to be stable for 8 hours with negligible change in neutrophil adherence or arteriole diameter. Three groups were evaluated in this study: (1) sham, no ischemia, no treatment (n = 5, 20 arterioles, 20 venules), (2) 4 hours of global ischemia only (n = 4, 19 venules, 22 arterioles), and (3) 4 hours of ischemia plus monoclonal antibody against CD18 (n = 4, 12 venules, 9 arterioles). The murine monoclonal antibody (WT.3, Seikagaku America, Inc.), which binds the rat leukocyte function antigen 1 CD18 chain, was infused into the contralateral femoral vein 30 minutes prior to reperfusion. The number of leukocytes rolling and adherent to endothelium (15 seconds of observation) was counted in 100‐&mgr;m venular segments, and arteriole diameters were measured at various times during reperfusion. All counts and measurements were normalized to baseline preischemic readings for each animal. Mean changes from baseline were compared between groups. The increase in ischemia‐reperfusion—induced neutrophil‐endothelial adherence in venules was blocked by monoclonal antibody, but rolling behavior was not changed. The ischemia‐reperfusion—induced progressive vasoconstriction in arterioles was blocked by monoclonal antibody. These results suggest that (1) neutrophil‐endothelial adherence function associated with ischemia‐reperfusion in this model is CD18‐dependent, (2) neutrophil rolling function does not appear to be dependent on CD18, and (3) neutrophil CD18 function is a prerequisite for ischemia‐reperfusion‐induced arteriolar vasoconstriction. These findings provide important mechanistic information that may help explain the deleterious microcirculatory events associated with ischemia‐reperfusion injury of skeletal muscle. (Plast. Reconstr. Surg. 99: 2002, 1997.)

Digital images in the diagnosis of wound healing problems

The use of digital wound images could allow remote consultation among patients, physicians, or other care-givers located at quite distant sites by means of the Internet. To evaluate the efficacy and validity of digital images for the evaluation of wounds, the ability and reliability of surgeons to diagnose and make treatment suggestions using digital images of several types of wounds were compared. Twenty-four wound images on 35-mm slides were selected for use in this study. Each slide image was digitized at 24-bit color with a resolution of 640 pixels horizontal by 425 pixels vertical and stored as a JPEG file. These images were then presented as a slide show on a video monitor, with resolution set at 640 x 480. Six physicians examined the images, first in digital format and later in the original slide form. Each observer assessed each wound and possible treatment options by filling out a questionnaire using a series of yes/no questions. For all observers, there was an 87 percent agreement between digital and slide images (p = 0.004). The agreement between the digital and slide images was measured for each individual observer using a kappa coefficient. The agreement level corresponded to the experience of the observer, with the kappa values ranging from greater than 0.8 (almost perfect agreement) for the attending plastic surgeon to just greater than 0.5 (moderate agreement) for the intern. With this study, the feasibility of distance wound consultation using digital images of a quality consistent with consumer-grade digital photography was demonstrated.

The effects of chronic ketorolac tromethamine (toradol) on wound healing.

Intramuscular ketorolac is a commonly used nonsteroidal anti-inflammatory (NSAI) agent for analgesia in surgical patients. Increasing numbers of surgical patients are chronically taking some form of an NSAI drug. We examined the effects of “chronic” intramuscular ketorolac on the healing of a closed linear surgical wound in the rat. Wistar rats were pretreated with 4 mg per kilogram per day ketorolac intramuscularly prior to receiving dorsal incisional wounds. The ketorolac treatment was continued and after 2 weeks the wounds were excised and separated with a tensiometer to measure mechanical properties. Breaking strength was directly measured, tensile strength was calculated, and collagen concentrations at the wound site were determined. A significant decrease in the mean breaking strength was seen in the ketorolac-treated animals when compared to controls. The ketorolac-treated animals had a mean tensile strength less than the controls, although this difference did not reach statistical significance. The mean collagen concentration of the ketorolac-treated wounds was significantly less than the untreated wounds. Use of ketorolac for just 1 week prior to surgery in rats produced a significant decrease in the breaking strength of their wounds. With the increasing use of ketorolac in surgical patients as well as the increasing use of oral NSAI drugs, more study of this effect is warranted.

Vitamin A-soaked gelfoam sponges and wound healing in steroid-treated animals.

Previous work has shown improved wound healing after the administration of systemic vitamin A in patients on chronic steroids. In contrast there have been mixed reports on the effect of topical vitamin A on wound healing in steroid-treated patients. Previous laboratory work has suggested that the topical application of vitamin A may be beneficial to wound healing in a sutured wound in a steroid-treated rat. Due to some inconsistencies in previous studies and steroid animal models, we sought to develop a better wound-healing model in a steroid-treated rat and to assess the effect of topical vitamin A as part of the wound closure. With preliminary studies, we developed a consistent and reliable wound-healing model in a steroid-treated rat using dexamethasone in contrast to cortisone acetate, which had been used in previous studies. Next, rats were randomized into 8 groups, some of which received steroids. Wounds were treated with saline or vitamin A topically or via a soaked gelfoam sponge. Rats were wounded 1 week after the commencement of steroid administration. Wounds were repaired and allowed to heal for 2 weeks. Strips of the wounds were then harvested and tested for tensile strength and breaking strength using a tensiometer. Wound edges were then fixed and wound surface area was measured using digital planimetry. Steroid treatment resulted in consistent weight loss and failure to gain weight as well as decreased breaking strength. Tensile strength was not decreased. Vitamin A applied for 10 minutes before wound closure and a gelfoam sponge alone placed before wound closure both resulted in an increased breaking strength and tensile strength. Vitamin A applied via a gelfoam sponge did not statistically increase breaking or tensile strength.

Nipple-Areolar Perfusion and Reduction Mammaplasty: Correlation of Laser Doppler Readings with Surgical Complications

&NA; Clinical assessment of nipple‐arcolar perfusion by color, capillary refill, and temperature during and after reduction mammaplasty is generally satisfactory. However, the estimation of vascular perfusion in patients with very large breasts or dark‐skinned women is difficult. If marginal perfusion of the nipple‐arcolar areas is undetected, necrosis is likely. To this point, no studies have tested the ability of the laser Doppler perfusion monitor to give absolute alarm values that would suggest marginal perfusion in the nipple‐areola following reduction mammaplasty. We therefore completed a prospective study of areolar perfusion during surgery and for up to 24 hours following reduction mammaplasty. Fifty‐four patients were studied and data collected from 104 breasts. Laser Doppler perfusion was measured with a LASERFLO BPM2 Blood Perfusion Monitor (Vasamedics, St. Paul, Minn.). Perfusion values were recorded for each breast following anesthesia but prior to the incisions, at the end of surgery, and every 2 hours for 24 hours. Patients were divided into three groups according to their follow‐up results; no complications (92 breasts), minor complications (9 breasts), and patients with tissue necrosis (3 breasts). The no complications group had a perfusion of 4.8 ml/min/100 gm following the reduction procedure, while the minor complications and tissue necrosis groups had average perfusions of 1.4 and 0.8, respectively, immediately after incision closure. The average tissue removed from each group was 811, 1171, and 2118 gm for the no complications, minor complications, and tissue necrosis groups, respectively. The results from this study suggest that a laser Doppler perfusion monitor could prove useful for monitoring areolar perfusion following reduction mammaplasty, especially in patients with extremely large breasts and or dark skin. Our studies have shown that laser Doppler perfusion values that consistently are in the range of 1.0 to 2.0 ml/min/100 gm indicate marginal perfusion, and the recovery of these patients should be followed closely. Furthermore, patients with consistent perfusion values equal to or less than 1.0 coupled with other clinical signs of low perfusion should be considered for suture removal and or free nipple graft. (Plast. Reconstr. Surg. 97: 381, 1996.)

Role of the thromboxane A2 receptor in the vasoactive response to ischemia-reperfusion injury.

Neutrophil-endothelial adhesion in venules and progressive vasoconstriction in arterioles seem to be important microcirculatory events contributing to the low flow state associated with ischemia-reperfusion injury of skeletal muscle. Although the neutrophil CD-18 adherence function has been shown to be a prerequisite to the vasoconstrictive response, the vasoactive substances involved remain unknown. The purpose of this study was to evaluate the role of thromboxane A2 receptor in the arteriole vasoactive response to ischemia-reperfusion injury. An in vivo microscopy preparation of transilluminated gracilis muscle in male Wistar rats (175 +/- 9 g) (n = 12) was used for this experiment. Three experimental groups were evaluated in this study: (1) sham, flap raised, no ischemia (20 venules, 20 arterioles), (2) 4 hours of global ischemia only (19 venules, 22 arterioles), and (3) 4 hours of global ischemia + thromboxane A2 receptor antagonist (ONO-3708) (17 venules, 20 arterioles). ONO-3708 (5 mg/kg), a specific competitive antagonist of thromboxane A2 receptor, was infused at a rate of 0.04 ml/minute into the contralateral femoral vein 30 minutes before reperfusion. Mean arterial blood pressure was not changed at this dose of ONO-3708 (88 +/- 6 mmHg before infusion, 81 +/- 4 mmHg after infusion, n = 3). The number of leukocytes rolling and adherent to endothelium (15-sec observation) were counted in 100-microm venular segments, and arteriole diameters were measured at 5, 15, 30, 60, and 120 minutes of reperfusion. Leukocyte counts and arteriole diameters were analyzed with two-way factorial analysis of variance for repeated measures and Duncans post hoc mean comparison. Statistical significance was indicated by a p < or = 0.05. The ischemia-reperfusion-induced vasoconstriction was significantly reduced by the thromboxane A2 receptor antagonist (ONO-3708). The mean arteriole diameters at 30, 60, and 120 minutes reperfusion were significantly greater in the treated animals than in the ischemia-reperfusion controls. Despite a significant increase in treated mean arteriole diameters, 30 percent of arterioles still demonstrated vasoconstriction. Neutrophil-endothelial adherence was not reduced by ONO-3708. Thromboxane A2 receptor blockade significantly reduces but does not eliminate ischemia-reperfusion-induced vasoconstriction in this model. This finding suggests that additional and perhaps more important vasoactive mediators contribute to vasoconstriction. Furthermore, thromboxane A2 receptor blockade has no effect on polymorphonuclear endothelial adherence.