Search

Excision of head and neck melanoma is often limited by critical structures, which can lead to incomplete excision with positive pathologic margin and increased local recurrence rate. Complete excision with recommended margins and immediate reconstruction is possible when surgical oncologists and plastic surgeons work collaboratively. Our purpose was to evaluate local recurrence rate after excision and immediate reconstruction. We reviewed 98 consecutive patients treated for primary head and neck cutaneous melanoma at a single institution between 1999 and 2004. We assessed local recurrence rate. A total of 72 patients (73%) were followed for an average of 5.2 ± 1.7 years while 26 patients were excluded due to less than 1 year of follow-up. Adjacent tissue transfer was the most common reconstruction (87%). Local recurrence was reported in 2.8% and distant metastasis in 12.5% of patients. Immediate reconstruction after excision of head and neck melanoma can be safely performed with low local recurrence rate.

Although bone marrow-derived mesenchymal stem cells have been shown to promote repair when applied to cutaneous wounds, the mechanism for this response remains to be determined. The aim of this study was to determine the effects of paracrine signaling from mesenchymal stem cells on dermal fibroblast responses to injury including proliferation, migration and expression of genes important in wound repair. Dermal fibroblasts were co-cultured with bone marrow-derived mesenchymal stem cells grown in inserts, which allowed for paracrine interactions without direct cell contact. In this co-culture model, bone marrow-derived mesenchymal stem cells regulate dermal fibroblast proliferation, migration and gene expression. When co-cultured with mesenchymal stem cells, dermal fibroblasts show increased proliferation and accelerated migration in a scratch assay. A chemotaxis assay also demonstrated that dermal fibroblasts migrate towards bone marrow-derived mesenchymal stem cells. A PCR array was used to analyze the effect of mesenchymal stem cells on dermal fibroblast gene expression. In response to mesenchymal stem cells, dermal fibroblasts up-regulate integrin alpha 7 expression and down-regulate expression of ICAM1, VCAM1 and MMP11. These observations suggest that mesenchymal stem cells may provide an important early signal for dermal fibroblast responses to cutaneous injury.

Introduction: The neuropeptide, substance P (SP), up-regulates nitric oxide production (NO). The purpose of this study was to determine whether SP enhances response to cutaneous injury in nitric oxide synthase knockout (NOS null) mice., Methods: We studied mice with targeted deletions of the 3 NOS genes, neuronal NOS, inducible NOS, or endothelial NOS. Full thickness dorsal wounds were treated daily (d 0-6) with topical SP or normal saline (NaCl). Wounds were analyzed by flow cytometry for macrophage, leukocyte, endothelial, and dendritic cells. Healing time and wound epithelialization were compared using analysis of variance., Results: Wound closure in the 3 NOS null mice was slower than the control mice (P < 0.05). SP treatment enhanced wound closure in NOS null mice (P < 0.02). NOS null wounds exhibited reduced inflammation. SP increased macrophage, leukocyte, and dendritic cell densities at d 3 and d 7 (P < 0.05) in all NOS null mice. SP increased endothelial cell number in neuronal NOS and inducible NOS null mice, but not in endothelial NOS null mice (P > 0.05)., Conclusions: SP ameliorated the impaired wound healing response observed in NOS null mice by enhancing wound closure kinetics and epithelialization. SP increased inflammatory cell density in the wounds supporting the essential role of inflammatory cells, especially macrophages, in wound repair.

Body mass index (BMI) must be considered when selecting patients suitable for abdomen-based microsurgical breast reconstruction. It is unknown whether BMI or age affects quality or quantity of abdominal wall perforating blood vessels. The purpose of this study was to identify differences in abdominal wall perforating blood vessels among patients with different BMI and age. A retrospective review was conducted of 66 patients undergoing abdomen-based microsurgical breast reconstruction tissue transfer from 2000 to 2006. Median age was 48.6 years (+/- 8.2). Patients were divided into BMI < 25 (28 patients), BMI 25 to 30 (26 patients), or BMI > 30 (12 patients). Perforator size and location was determined by ultrasound data. There was a greater number of perforators in horizontal zone II compared with the remaining zones (P < 0.05, Bonferroni corrected). There were no differences between age or BMI and the number of perforators or average perforator size per patient. A significant positive linear association was found between the average perforator diameter and total number of abdominal wall perforators. We concluded there is no anatomical difference in perforator quality among patients with varying BMIs

Head and neck melanoma often approaches critical structures. Therefore, excision is often limited, leading to positive margins, and increased local recurrence. Immediate reconstruction carries concern for rearrangement or concealment of cancerous tissues. Therefore, reconstruction is often delayed until confirming negative margins on permanent pathology. Our purpose is to identify variables associated with a positive margin and establish criteria for reconstruction timing. We reviewed 117 consecutive patients who underwent wide local excision of head and neck melanoma. Reconstruction was immediate for 107 and delayed for 10. Six percent of patients had a positive margin after wide local excision with no difference in incidence between immediate and delayed reconstruction (P = 0.11). Tumor characteristics associated with a positive margin were locally recurrent, ulcerated, and T4 tumors (P < 0.05); and delayed reconstruction should be considered in these circumstances. Immediate reconstruction is safe for the majority of head and neck melanoma and should be based on knowledge of tumor characteristics.

The neuropeptide substance P (SP) is a known inflammatory mediator released from cutaneous peripheral nerve terminals. SP effects on cellular composition in the cutaneous response to injury remain unclear. Based on our previous observations about SP effects on wound repair, we hypothesized that topical SP increases inflammatory cell density infiltration early after injury. A full-thickness 1.5 x 1.5 cm(2) wound was created on the dorsum of 8-9-week-old C57BL/6J-m+Lepr(db) mice (db/db). Wounds were treated daily with 300 muL of either normal saline (0.9% NaCl) or 10(-9) M SP for 7 days. Three wounds from each group were harvested at 2, 3, 7, 14, and 28 days. Samples underwent enzymatic digestion and were incubated with fluorescent-labeled antibodies. Using flow cytometry, cellular content and density for each sample was derived. Masson Trichrome stained histology specimens were prepared to confirm results. Cell density in the SP-treated wounds (11.3 x 10(7) cells/g tissue, standard deviation [SD]+/-1.5 x 10(7)) was greater than in NaCl-treated wounds (7 x 10(7) cells/g tissue, SD+/-2.3 x 10(7), p<0.05) at day 7 postwounding. SP significantly increased the density of leukocytes (2.1 x 10(7), SD +/-3.6 x 10(6) vs. 1.8 x 10(7), SD+/-4.9 x 10(5), p<0.02) 3 days after wounding and the density of macrophages (2.9 x 10(7), SD+/-7.5 x 10(6) vs. 1.3 x 10(7), SD+/-1.4 x 10(6), p<0.05) 7 days after wounding. There were no significant differences in endothelial cell, leukocyte, or macrophage density at later time points. Topical SP treatment increases early inflammatory density in the healing wounds of db/db mice. These data support a role for nerve-mediated inflammation in cutaneous wound repair.

Background: Improved self-image and psychological well-being after breast reconstruction are well documented. To determine methods that optimized results with minimal morbidity, the authors examined their results and complications based on reconstruction method and timing., Methods: The authors reviewed all breast reconstructions after mastectomy for breast cancer performed under the supervision of a single surgeon over a 6-year period at a tertiary referral center. Reconstruction method and timing, patient characteristics, and complication rates were reviewed., Results: Reconstruction was performed on 240 consecutive women (94 bilateral and 146 unilateral; 334 total reconstructions). Reconstruction timing was evenly split between immediate (n = 167) and delayed (n = 167). Autologous tissue (n = 192) was more common than tissue expander/implant reconstruction (n = 142), and the free deep inferior epigastric perforator was the most common free flap (n = 124). The authors found no difference in the complication incidence with autologous reconstruction, whether performed immediately or delayed. However, there was a significantly higher complication rate following immediate placement of a tissue expander when compared with delayed reconstruction (p = 0.008). Capsular contracture was a significantly more common late complication following immediate (40.4 percent) versus delayed (17.0 percent) reconstruction (p < 0.001; odds ratio, 5.2; 95 percent confidence interval, 2.3 to 11.6)., Conclusions: Autologous reconstruction can be performed immediately or delayed, with optimal aesthetic outcome and low flap loss risk. However, the overall complication and capsular contracture incidence following immediate tissue expander/implant reconstruction was much higher than when performed delayed. Thus, tissue expander placement at the time of mastectomy may not necessarily save the patient an extra operation and may compromise the final aesthetic outcome.

Background: A subset of obese people develop a pannus hanging to the floor. This panniculus morbidus prevents weight loss, as the patient cannot exercise. It prevents hygiene, leading to a profound odor and ultimately results in intertrigo, cellulitis, and/or abdominal ulceration. The only two options are to live/die with it or resect it. Some of these people are otherwise ready for a weight loss program. For this group, resection of the panniculus morbidus may be indicated. The authors reviewed the literature and found the condition has not been addressed in this Journal since 1994 and was not considered in the recent supplement on body contouring. In 1998, the authors began resecting panniculus morbidus for this small group. The authors found the learning curve to be profoundly steep, with many wound complications, a finding that is quite in conflict with the literature on the subject, and decided to present their experience., Methods: The authors conducted a retrospective chart review of 23 patients and collected data on demographics, ambulation, hygiene, technique, complications, and outcome., Results: The technique of closure evolved as the authors struggled with complications. The current method of closure is three suture layers over four suction drains with a small wound vacuum-assisted closure device at each end of the incision. All patients ultimately healed and found it easier to ambulate and perform hygiene., Conclusion: Resection of panniculus morbidus is a beneficial salvage procedure for some morbidly obese people, but the learning curve is steep and the current literature is misleading.

Hypertrophic scarring after burns is an unsolved problem and remains as devastating today as it was in the 40s and it may be that the main reason for this is the lack of an accepted, useful animal model. The female, red Duroc pig was described as a model of hypertrophic scarring nearly 30 years ago but then vanished from the literature. This seemed strange since the authors reported that 12 of 12 pigs developed thick scar. In the mid 90s we explored the model and found that, indeed, the red Duroc pig does make thick scar. Other authors have established that the Yorkshire pig does not heal in this fashion so there is the possibility of a same species control. We have continued to explore the Duroc/Yorkshire model and herein describe our experiences. Is it a perfect model of hypertrophic scarring? No. Is it a useful model of hypertrophic scarring? Time will tell. We have now obtained gene expression data from the Duroc/Yorkshire model and analysis is underway.

Capsular contracture is the most common complication following breast augmentation or reconstruction with implants. We recently demonstrated that bone marrow-derived cells provide fibroblasts to murine skin during wound healing. To determine if bone marrow-derived cells were the cellular source of periprosthetic capsules, we created chimeric C57BL mice containing bone marrow cells from isogeneic enhanced green fluorescent protein (EGFP) mice and implanted with a textured silicone shell implant. We found that none of the mice developed infection or capsular contracture, but day 30 capsules were composed of 26.4 +/- 6.1% EGFP cells, and day 60 capsules had 21.8 +/- 10.3% EGFP cells. Immunohistochemistry revealed a small population of EGFP cells in the capsules that were myofibroblasts. Thus, breast implant capsules are partially composed of bone marrow-derived cells and, given the potential of these cells to become myofibroblasts, may explain the cellular source of capsular contracture when it develops.

Breast cancer chest wall recurrence is often treated with chemotherapy, radical surgery, and radiation. Extensive chest wall resection requires soft-tissue reconstruction with tissue that provides chest wall stability and durability for additional radiation. Local and regional muscle and musculocutaneous flaps are often used for reconstruction. Free flaps, such as the transverse rectus abdominis musculocutaneous flap, are used for large defects, although donor site morbidity can result. The free deep inferior epigastric perforator (DIEP) flap provides coverage for large defects and may have less donor site morbidity. We describe the use of the free DIEP flap to reconstruct large chest wall defects (mean, 501 cm2 defects) after the resection of recurrent breast cancer in two patients. One patient had 2% flap loss. No donor site morbidity occurred. The free DIEP flap is a durable and reliable flap that provided immediate and complete coverage of these large chest wall defects with no donor site morbidity and did not delay the administration of adjuvant therapy.

The etiology and treatment of hypertrophic scar remain puzzles even after decades of research. A significant reason is the lack of an accepted animal model of the process. The female, red Duroc pig model was described long ago. Since the skin of the pig is similar to that of humans, we are attempting to validate this model and found it to be encouraging. In this project we quantified myofibroblasts, mast cells and collagen nodules in the thick scar of the Duroc pig and compared these to the values for human hypertrophic scar. We found the results to be quite similar and so further validated the model. In addition, we observed that soon after wounding an inflammatory cell layer forms. The thickness of the inflammatory layer approaches the thickness of the skin removed as if the remaining dermis "knows" how much dermis is gone. In deep wounds this inflammatory layer thickens and this thickness is predictive of the thickness of the ultimate scar.

We hypothesized that the use of muscle flaps, known as tissue transfer (TT), at the time of abdominoperineal resection (APR) reduces perineal wound complications. A restrospective review of patients undergoing an APR at the University of Washington (1984-2003) was conducted. Perineal wound complications and eventual wound healing were compared in patients with and without TT. Ninety-two patients (mean age, 56.6 years) underwent APR; 23.9 per cent (n = 22) had concurrent TT. Patients undergoing TT were more likely to have cancer (91% vs. 77%, P = 0.05) and radiation therapy (86% vs. 52%, P < 0.01). Operative times were nearly 2 hours longer in patients having TT (7.4 hours +/- 2.5 hours vs. 5.6 hours +/- 1.8 hours, P = 0.03), but lengths of stay were similar (13 +/- 5.9 days vs. 12 +/- 7.6 days, P = 0.5). Patients undergoing TT had a higher rate of all wound-healing complications (59% vs. 40%, P = 0.1) and major wound-healing complications (32% vs. 26%, P = 0.6). However, these differences were not statistically significant. No differences in major complications were identified in patients with and without preoperative radiation therapy (26% vs. 28%, P = 0.8). Fifteen per cent (n = 14) of all patients failed to heal wounds at 6 months, but only 9 per cent (n = 2) of patients undergoing TT failed to heal their wounds at 6 months compared with 17 per cent (n = 12) in the non-TT group (P = 0.3). After controlling for important covariates, patients undergoing TT during an APR did not have a significantly lower rate of wound complications. The impact of TT on wound healing in patients with recurrent cancer and preoperative radiation therapy is suggestive of a benefit but requires prospective investigation.

The vertical reduction mammaplasty is an evolving technique. Its proponents report significantly decreased scarring, better breast shape, and more stable results compared with the standard inverted-T method, but the learning curve is long and cosmetic outcomes can be inconsistent. Many surgeons have experimented with the vertical closure before returning to methods more familiar to them. The authors present their modifications to the vertical reduction mammaplasty. Their changes simplify the preoperative markings and the intraoperative technique to shorten the learning curve while maintaining reliable aesthetic results. With the patient standing, only four preoperative marks are made: (1) the inframammary fold; (2) the breast axis; (3) the apex of the new nipple-areola complex; and (4) the medial and lateral limbs of the vertical incision. In the operating room, a medial or a superomedial pedicle is developed. Excess breast skin is resected with the inferior and lateral parenchyma as a C-shaped wedge. The lateral skin-adipose flap is redraped inferomedially and sutured to the chest wall. The inferior aspect of the breast is aggressively debulked and a gathering subcuticular stitch is started 2 cm below the nadir of the nipple-areola complex. Finally, a 38-mm to 42-mm nipple-areola complex marker is used to create a circular defect that is offset 0.5 cm medial to the vertical axis of the breast. In their series, 56 patients were treated and no major complications were noted. The median follow-up period was 17 months. The average reduction was 554.5 g per breast; however, the reduction was greater than 1000 g per breast in eight patients. The authors found that (1) chest wall anchoring improves lateral contour and minimizes axillary fullness; (2) aggressive debulking inferiorly avoids the persistent inferior bulge; and (3) starting the subcuticular gathering suture 2 cm below the nipple-areola complex followed by placement of a nipple-areola complex marker at the conclusion of the case prevents lateral deviation and corrects the nipple-areola complex teardrop deformity. These innovations accelerate the learning curve by simplifying the preoperative markings and lead to more consistent postoperative results and an improved cosmetic outcome. In conclusion, these modifications yield a simple, easily learned vertical reduction mammaplasty with aesthetically reliable results.

Hypertrophic scarring occurs after deep dermal wounds. Our understanding of the etiology is poor; one reason is the lack of an animal model. In 1972, Silverstein described scarring in the Duroc pig but the model was never confirmed nor disproved. Another reason, as we previously suggested, is that hypertrophic scarring only occurs within regions of human skin that contain cones and the cones have not been studied in relation to hypertrophic scarring. We, therefore (i) explored healing in the female, red Duroc model for similarities to human hypertrophic scarring, studying wound thickness, appearance, healing status at 3 weeks, histology, and immunocytochemical localization of decorin, versican, TGFbeta1 and IGF-1; and (ii) examined Duroc skin for cones. We found that healing after deep wounds in Duroc pigs is similar, but not identical, to human hypertrophic scarring. We also found that Duroc skin contains cones. Healing in the female, red Duroc pig is sufficiently similar to human hypertrophic scarring to warrant further study so that it can be accepted or rejected as a model of human hypertrophic scarring. In addition, the relationship of the cones to hypertrophic scarring needs further detail and can be studied in this model.

The vertical reduction mammaplasty can be challenging to learn. In addition, first attempts to perform the vertical reduction mammaplasty can lead to inconsistent aesthetic results. The authors describe their transition from a traditional inverted-T reduction mammaplasty to a modified vertical reduction mammaplasty based on a technique described by Elizabeth Hall-Findlay. In their early cases using the Hall-Findlay technique, they noted several aesthetic complications. These problems included a persistent vertical dog-ear deformity at the nadir of the incision, a teardrop deformity of the nipple-areola complex, lateral deviation of the nipple, and lateral axillary fullness. They developed several modifications to the Hall-Findlay technique to correct the aesthetic deficiencies and to simplify further the vertical reduction method. The authors think their innovations facilitate the transition from a traditional inverted-T breast reduction to a successful vertical reduction mammaplasty technique.

Following cutaneous injury, sensory nerves regenerate into the dermis and epidermis. Tissues that are innervated by sensory nerves synthesize neurotrophins such as nerve growth factor (NGF). The close anatomic proximity of nerves and capillaries throughout the skin suggests that mutual regulation may exist between nerve fibers and microvascular endothelial cells (MECs) during wound repair. Release of the neuropeptide substance P by sensory nerves induces endothelial cell rounding, capillary leak, and cytokine upregulation. We propose that dermal endothelial cells produce neurotrophins required for nerve fiber maintenance and regeneration. In this study, we demonstrate that substance P stimulates NGF messenger RNA expression by cultured human dermal MECs. Likewise, enzyme-linked immunosorbant assay demonstrated that conditioned medium from cultured dermal MECs contains NGF. NGF bioactivity in the supemates was verified by conditioned medium-induced clonal rat pheochromocytoma (PC-12) cell differentiation. This activity was inhibited by anti-NGF antibodies. Therefore, we have demonstrated that substance P, an inflammatory neuropeptide released by sensory nerve fibers, induces endothelial cells to produce NGF. Our data suggest that MECs may be unrecognized contributors to nerve regeneration after cutaneous injury.

Background. Patients with diabetic sensory neuropathy have significant risk of chronic ulcers. Insufficient nerve-derived mediators such as substance P (SP) may contribute to the impaired response to injury. Mutant diabetic mice (db/db), which develop neuropathy and have delayed healing, may provide a model to study the role of nerves in cutaneous injury.Methods. Skin from human chronic nonhealing ulcers and age-matched control skin was immunohistochemically evaluated for nerves. Nerve counts were also compared in murine diabetic (C57BL/KsJ-m+/+ Lepr(db); db/db) and nondiabetic (db/-) skin. Excisional wounds on the backs of db/db and db/- mice were grouped as: (a) untreated db/- mice; (b) untreated db/db mice; (c) db/db mice with polyethylene glycol (PEG); (d) db/db mice with PEG and SP 10(-9) M; or (e) db/db mice with PEG and SP 10(-6) M.Results. We demonstrated fewer nerves in the epidermis and papillary dermis of skin from human subjects with diabetes. Likewise, db/db murine skin had significantly fewer epidermal nerves than nondiabetic littermates. We confirmed increased healing times in db/db mice (51.7 days) compared to db/- littermates (19.8 days; P

Endothelial cell proliferation and migration is initiated by growth factors including FGF and VEGF that bind to specific transmembrane receptor tyrosine kinases. Mechanisms that regulate in vivo expression of fibroblast growth factor receptors (FGFR) and vascular endothelial growth factor receptors (VEGFR) are not well understood. Since it is well known that different matrices influence the proliferation and migration of endothelial cells in culture, we hypothesized that changes in the extracellular matrix environment can regulate growth factor receptors on endothelial cells. We cultured human microvascular endothelial cells on different matrices (vitronectin, laminin, fibronectin, fibrin, and collagen IV) and examined for the presence of growth factor receptors (FGFR-1, FGFR-2, VEGFR-1, and VEGFR-2). We show that vitronectin increased the presence of all four growth factor receptors and most notably, VEGFR-1. In contrast, fibrin decreased all four receptors, especially FGFR-1 and FGFR-2. Inhibiting phosphotyrosine signaling abolished immunostaining for all four receptors, regardless of the matrix, but was not dependent on activating the Fyn-Shc pathway. Cells plated on vitronectin in the presence of blocking antibodies to integrins alphavbeta3 and alphavbeta5 similarly decreased presence of these growth factor receptors. Our data suggests a possible mechanism of how matrix-integrin interactions regulate endothelial cell responsiveness to growth factors and anchorage-dependent cell growth.

Hypertrophic scarring is devastating for the patient, however the pathophysiology and treatment remain unknown after decades of research. The process follows deep dermal injury, occurs only on certain body parts, does not occur in the early fetus or in animals, and is a localized event. This suggests that an anatomic structure in human, deep dermis may be involved. The dermis is a matrix perforated by cones containing many structures including skin appendages and fat domes. We hypothesized that studying the cones might reveal a structure related to scarring. We examined tangential wounds from various body parts on human cadavers along with skin histology from various human body parts, the early fetus, partial thickness burns, hypertrophic scars, and two other species-rats and rabbits. We found that the cones may in fact be the structure. They exist where hypertrophic scar occurs-cheek, neck, chest, abdomen, back, buttock, arm, forearm, dorsal hand, thigh, leg, dorsal foot, helix and ear lobe. They do not exist where hypertrophic scar does not occur-scalp, forehead, concha, eyelid, palm, early fetus, and in rat, or rabbit. It also became apparent that the cones have been omitted from most considerations of skin histology. We suggest that the cones need to be studied in relation to hypertrophic scarring and restored to skin diagrams.

Hypertrophic scar is one form of abnormal wound healing. Previous studies have suggested that hypertrophic scar formation results from altered gene expression of extracellular matrix molecules. A broadscale evaluation of gene expression in hypertrophic scars has not been reported. To better understand abnormalities in hypertrophic scar gene expression, we compared messenger RNA expression in hypertrophic scars, normal scars, and uninjured skin with the use of complementary (c)DNA microarrays. Total RNA was extracted from freshly excised human hypertrophic scars, normal scars, or uninjured skin and reverse transcribed into cDNA with the incorporation of [33P] deoxycytidine triphosphate. The resulting radioactive cDNA probes were hybridized onto cDNA microarrays of 4000 genes. Hybridization signals were normalized and analyzed. In the comparison of tissue samples, mean intensities were calculated for each gene within each group (hypertrophic scars, normal scars, and uninjured skin). Ratios of the mean intensities of hypertrophic scars to normal scars, hypertrophic scars to uninjured skin, and normal scars to uninjured skin were generated. A ratio that was greater than 1 indicated upregulation of any particular gene and a ratio that was less than 1 indicated downregulation of any particular gene. Our data indicated that 142 genes were overexpressed and 50 genes were underexpressed in normal scars compared with uninjured skin, 107 genes were overexpressed and 71 were underexpressed in hypertrophic scars compared with uninjured skin, and 44 genes were overexpressed and 124 were underexpressed in hypertrophic scars compared with normal scars. Our analysis of collagen, growth factor, and metalloproteinase gene expression confirmed that our molecular data were consistent with published biochemical and clinical observations of normal scars and hypertrophic scars. cDNA microarray analysis provides a powerful tool for the investigation of differential gene expression in hypertrophic scar samples and either uninjured skin or normal scars. Our data validate the use of this technology for future studies on gene expression during repair processes of normal and abnormal wounds.

Hemangiomas appear at birth and undergo gradual regression within several years. Recent published studies have documented increased nerve numbers in port-wine stains and intramuscular vascular tumors. The aim of this study was to establish a relationship between angiogenesis and nerve growth in lesions that undergo neovascular proliferation followed by vessel involution. Twenty-two hemangiomas and arteriovenous malformations were studied using indirect immunocytochemistry with antibodies against the nerve markers protein gene product 9.5 (PGP 9.5) and calcitonin gene-related peptide (CGRP). Nerves and vessels were counted and compared. Our results indicate that PGP 9.5(+) and CGRP(+) nerves were most numerous in growing hemangiomas and numbers were reduced in involuting hemangiomas and vascular malformations. The percentage of CGRP(+) sensory nerves was markedly increased in growing hemangiomas (45.3%) compared with involuting hemangiomas (21.2). These data indicate that hemangiomas with increasing neovascularization have increased sensory nerve growth. Sensory nerve-derived neuropeptides are known to act as endothelial cell mitogens and may contribute to the angiogenesis in these vascular tumors. Conversely, angiogenic endothelial cells may secrete mediators that promote nerve fiber growth. These results suggest that endothelial cell proliferation and sensory nerve fiber growth may be closely related., (Copyright 2000 Academic Press.)

Background: Vitronectin has several putative functions including regulating hemostasis, cell adhesion, and cell migration. However, the targeted deletion of vitronectin in mice results in normal development and normal coagulation parameters. To determine whether vitronectin may be necessary for nondevelopmental processes, we examined the response to tissue injury in vitronectin-null mice., Methods: We examined wound healing in control and vitronectin-null mice by healing rate, zymography, reverse zymography, and Western blots., Results: We found that dermal wound healing was slightly delayed in mice lacking vitronectin. More importantly, we found extensive areas of delayed hemorrhage near the sprouting tips of microvessels between days 7 and 14, which temporally coincided with increased urokinase-type plasminogen activator and tissue-type plasminogen activator activity by zymography. Though Western blots confirmed the presence of plasminogen activator inhibitor-1 protein throughout wound repair and reverse zymograms showed decreased plasminogen activator inhibitor-1 activity between days 7 and 14., Conclusions: Loss of vitronectin in mice was associated with changes in the fibrinolytic balance, and this may have led to focal sites of delayed hemorrhage. The mechanism that resulted in decreased angiogenesis and the formation of larger blood vessels in response to tissue injury remains unknown. This study suggests that vitronectin may have several distinct functions that are not required for normal development but are manifested in response to tissue injury.

The growth and metastases of many solid tumors are dependent on the recruitment of new blood vessels. Tumor angiogenesis is most likely initiated by paracrine release of growth factors that bind to their corresponding endothelial cell surface receptors. To determine whether angiogenesis and growth factor receptor expression are consistent findings in malignant melanoma, primary human melanomas were examined for mRNA expression of receptors for fibroblast growth factors (FGFR-1, FGFR-2), vascular endothelial growth factor (VEGFR-1, VEGFR-2), and the receptors Tiel and Tie2. Charts were reviewed and archival formalin-fixed, paraffin-embedded primary tumors were obtained from patients with thin (<1 mm; n = 10), intermediate (1 to 4 mm; n = 10), or thick malignant melanoma (>4 mm; n = 8). Also examined was whether melanoma cell lines could induce endothelial growth factor receptor synthesis by metabolic labeling. It was found that tumor vascularity did not correlate with clinical stage, melanoma thickness, or clinical outcome. It was also found that melanoma cell lines were not capable of directly regulating endothelial cell synthesis of growth factor receptors. However, expression of Tiel and VEGFR-2 mRNA by the tumor vasculature in select stage IA-IIB patients, and FGFR-1 mRNA expression by the tumor cells in the same clinical stages was found. The expression of these growth factor receptors did not correlate with clinical outcome. These data suggest that angiogenesis is not a prominent characteristic of primary malignant melanoma lesions and that the endothelial cell expression of Tiel and VEGFR-2 in vivo is probably not directly induced by the tumor.

Angiogenesis, the formation of new blood vessels from pre-existing blood vessels, is thought to be critical for wound repair. Yet few studies have critically examined dermal wound repair in a system in which angiogenesis was impaired. Since alpha(v)-containing integrins are critical for angiogenesis, we administered either an alpha(v) integrin blocking antibody or cyclic Arg-Gly-Asp peptide into a murine excisional wound model to restrict wound angiogenesis. Although both methods markedly decreased wound angiogenesis, decreased angiogenesis had no significant effect on wound epithelization, contraction, or ultimate wound closure. These results suggest that if other cellular components of wound healing are intact, moderate impairment of angiogenesis alone does not necessarily retard normal wound healing.

Background: Response to tissue injury begins with the deposition of a fibrin-rich clot or the provisional matrix. The provisional matrix consists of plasma-borne matrix molecules that serve as scaffolding for the ensuing migration of cells. During wound repair multiple cell types must migrate through the clot-matrix scaffolding. The migration of these cells through the matrix is dependent on the activity of the fibrinolytic and proteolytic systems, which include the plasminogen activator (PA) system and matrix metalloproteinases (MMP). The aim of this study was to better understand the temporal activity of these enzymes during normal wound repair., Methods: We used the murine excisional wound model and extracted proteins under nonreducing conditions. With use of gelatin and casein zymography, we determined the activity of the MMPs during the course of wound repair. In addition, we quantified the activity of MMP-2 and MMP-9 by a standardized assay. Plasminogen zymograms were used to detect urokinase PA and tissue PA activity. Western blots were used to detect the natural inhibitor of PAs, plasminogen activator inhibitor type 1., Results: Our results demonstrate the temporal activity of MMP-2, MMP-3, MMP-7, and MMP-9 during the course of normal dermal repair. The activity of urokinase PA and tissue PA were also determined; it preceded the activity of the MMPs., Conclusions: We demonstrate the temporal activity of the 2 protease families, MMPs and PAs, in the normal process of cutaneous wound healing.

Introduction: Pulmonary arteriovenous malformations are a common cause of progressive cyanosis in children after cavopulmonary anastomoses. We analyzed the pulmonary histologic characteristics from children in whom pulmonary arteriovenous malformations developed after procedures that resulted in pulmonary arterial blood flow devoid of hepatic venous effluent., Methods: We performed routine histologic studies, immunohistochemical staining, and electron microscopic analysis of peripheral lung biopsy specimens from 2 children with angiographically proven pulmonary arteriovenous malformations. Microvessel density was determined with a computer-assisted, morphometric analysis system., Results: Histologic examination demonstrated large, dilated blood vessels ("lakes") and clustered, smaller vessels ("chains") in the pulmonary parenchyma. Microvessel density was significantly greater in these patients than in age-matched controls (P =.01). Immunohistochemistry demonstrated uniform staining for type IV collagen and alpha-smooth muscle actin, weak staining for the endothelial marker CD31 (cluster of differentiation, PECAM-1), and negative staining for proliferating cell nuclear antigen. Electron microscopy revealed endothelial irregularity, a disorganized basement membrane, and increased numbers of collagen and actin filaments beneath the endothelium., Conclusions: This study represents an attempt to characterize the histologic features of pulmonary arteriovenous malformations in children with congenital heart disease who have pulmonary arterial blood flow devoid of hepatic venous effluent. The histologic correlate of this condition appears to be greatly increased numbers of thin-walled vessels. Immunohistochemistry suggests that the rate of cellular proliferation is not increased in these lesions. The development of these techniques may provide a standardized histologic approach for this condition and aid in understanding its etiology.

Background: Hemangiomas offer an uncommon opportunity to study rapid vessel growth and spontaneous regression of a vascular human tumor. In contrast, venous malformations are another type of vascular tumor that grows slowly without spontaneous involution. Extracellular matrix (ECM) molecules modulate the responsiveness of endothelial cells to mitogenic stimuli such as basic fibroblast growth factor (bFGF), a well-recognized stimulant of angiogenesis. In this study we hypothesized that in hemangiomas, sites of angiogenesis may have a different ECM composition than sites of vascular regression., Materials and Methods: Using immunohistochemistry, we analyzed proliferating hemangiomas, regressing hemangiomas, venous malformations, and normal skin for the basement membrane ECM molecules collagen IV and laminin and plasma-borne ECM molecules fibronectin and vitronectin. We used metabolic labeling to determine whether primary human dermal microvascular endothelial cells regulated FGFR-1 or FGFR-2 when grown on these different matrices., Results: We found that proliferating hemangiomas showed extensive deposition of vitronectin in the subendothelial space. In contrast, regressing hemangiomas or venous malformations did not show vitronectin deposition. Venous malformations, which are composed of ectatic lakes of venous channels, also lacked laminin in their basement membranes. We also found that cultured microvascular endothelial cells grown on vitronectin increased synthesis of FGFR-1 and FGFR-2 protein., Conclusions: Changes in the ECM environment occur in conjunction with the angiogenic state of a vascular human tumor. Furthermore, changes in the ECM environment alone can directly regulate synthesis of angiogenic growth factor receptors., (Copyright 1998 Academic Press.)

Despite improvements in surgical repair of pressure sores, recurrence rates greater than 80 percent are reported, thus indicating that this difficult problem is not yet solved. Recurrence results in additional hospitalizations and increased medical expenses. Because associated general clinical and social issues are numerous for these patients, our physical medicine and rehabilitation colleagues are active participants in their perioperative medical care. In addition, the Department of Physical Medicine and Rehabilitation also directs a complete postreconstruction rehabilitation and education program for them. The results of surgically repaired pressure sores for patients managed in this collaborative fashion have not been previously reported. Pressure sore patients at the Harborview and University of Washington Medical Centers are evaluated by plastic surgery colleagues together with the Department of Physical Medicine and Rehabilitation. Patients believed to be candidates for complete postoperative rehabilitation are offered surgical repair and constitute this study cohort. Individuals who cannot cooperate with our protocol are treated nonoperatively and are not included in this study. A retrospective analysis of all 158 patients (mean age 34.5 years) operated on for 268 grade III and IV pressure sores between October of 1977 and December of 1989 was performed. Following surgical debridement and reconstruction, patients receive their principal medical care from the Department of Physical Medicine and Rehabilitation service while the Plastic Surgery Department manages the surgical wounds. Graduated patient mobilization is initiated in accord with a mutually agreed upon standardized protocol. New or primary sores numbered 174 (65 percent), and recurrent or secondary sores numbered 94 (35 percent). Mean patient follow-up was 3.7 years. The overall pressure sore recurrence rate (recurrence at the same site) was 19 percent, and the overall patient recurrence rate (previous patient developing a new sore) was 25 percent. Recurrence rates of 22 and 15 percent were noted for primary and secondary pressure sores, respectively. On most recent examination, 131 patients (83 percent) had intact pelvic and perineal skin. These results support a collaborative approach to the management of high-grade pressure sore patients. Our protocol of mutual patient evaluation followed by surgical reconstruction and postoperative rehabilitation yields notably low recurrence rates of both primary and secondary sores. In addition, the high percentage of patients who manifest long-term maintenance of skin integrity demonstrates the excellent education in personal skin and self-care that this approach provides. Not only do patients enjoy successful and durable reconstructive results, but additional hospitalizations and health care expenses implicit to pressure sore recurrence are consequently diminished. This collaborative clinical effort remains our standard of care.

Angiogenesis after tissue injury occurs in a matrix environment consisting of fibrin, fibronectin, and vitronectin as the major extracellular matrix (ECM) constituents. ECM-integrin interactions is critical for angiogenesis and failure to bind a ligand to certain integrin receptors (alpha[v]beta3 or alpha[v]beta5) inhibits angiogenesis. The ligand that binds to alpha(v)beta3 or alpha(v)beta5 integrin receptors during microvascular angiogenesis has not been identified. Our hypothesis is that provisional matrix molecules provide the environmental context cues to microvascular endothelial cells and promote angiogenesis by decreased programmed cell death. Using cultured human microvascular endothelial cells, we show that vitronectin, in comparison to growth on alternative provisional matrix molecules (fibronectin, fibrinogen plus thrombin), collagen I, and basement membrane molecules (collagen IV), significantly reduces microvascular endothelial cell death in vitro. This reduction was observed using morphologic criteria, TdT-mediated dUTP nick end labeling (TUNEL) assay, histone release into the cytoplasm, and thymidine release into the supernatant. Though our data confirm that vitronectin may bind to more than one integrin receptor to reduce MEC apoptosis, binding to the alpha(v) component appears to be the critical integrin subcomponent for reducing apoptosis.

The cost to care for a patient with pressure sores can be exorbitant. One part of this expense results from the prolonged postoperative immobilization. Published protocols report 3 to 8 weeks of immobilization; however, there are no studies that establish longer periods to be superior. To justify our 3-week protocol, we conducted a prospective, randomized trial of 2 versus 3 weeks of postoperative immobilization. Each patient was randomized preoperatively to either 2 or 3 weeks of postoperative immobilization. A total of 42 patients with a diagnosis of paraplegia, tetraplegia, or multiple sclerosis and a solitary pressure sore were enrolled over 5 years. The complication rates in the two groups (9 of 23, or 39 percent, for the 2-week group and 9 of 19, or 47 percent, for the 3-week group) were not statistically different (p < 0.493). However, the time to mobilization was significantly reduced in the 2-week group (16.1 +/- 6.1 versus 22.9 +/- 4.9 days, p < 0.0003), as was the time to achieve sitting (21.2 versus 28.9 days, p < 0.0026). In summary, 2 weeks of postoperative immobilization following surgery is adequate for uncomplicated solitary pressure sores.

The inflammatory response following a thermal insult begins with the skin itself. Langerhan's cells, tissue macrophages, keratinocytes, fibroblasts, and endothelial cells contribute to the initial events of wound healing with active and passive release of cell mediators. One of the mediators potentially important to the repair process is monocyte chemoattractant protein-1 (MCP-1). Macrophages, fibroblasts, endothelial cells, and keratinocytes can produce MCP-1 in response to inflammatory stimuli. Therefore, we evaluated 10 human burn wound specimens for MCP-1 mRNA using in situ hybridization. Selected specimens of different ages were examined using combined in situ hybridization and immunocytochemistry to identify cell types that expressed MCP-1 mRNA. Antibodies to HAM56 for macrophages, CD45 for bone marrow-derived cells, Factor VIII for endothelial cells, and Factor XIIIa for dermal antigen-presenting cells were included in these experiments. By Postburn Day 2, basal layer keratinocytes at the edges of the wound had upregulated MCP-1 message; the increased signal persisted in the rate pegs deep in the dermal wound bed through 49 days postinjury. Occasional FXIIIa+ immunostained dermal cells expressed MCP-1 mRNA. Islands of granulation tissue throughout the wound bed were positive for increased expression of MCP-1; endothelial cells and inflammatory cells both contributed to this upregulated signal. Our data support the theory that the skin itself is a component of the immune system and that noninflammatory cells contribute to the initiation and maintenance of the inflammation at a wound site. Failure to produce MCP-1 or other related mediators by indigenous cutaneous cells may delay the inflammatory response to injury and potentially disrupt other essential phases of wound repair.

Objectives: The purpose of this review was to determine the incidence of pressure sores in acute trauma patients and to identify the causes. DESIGN AND MATERIALS AND METHODS: This study is a retrospective chart review of all acute trauma patients admitted to Harborview Medical Center between January of 1991 and December of 1993 who were discharged with an ICD-9 diagnosis of acute pressure sore., Results: A total of 7,492 trauma patients were admitted. Thirty-two patients developed 44 pressure sores for an incidence of 0.4%. All of the patients were severely injured (mean Injury Severity Score, 21). Eighteen (41%) of the pressure sores developed as a result of unrelieved positional pressure. Thirteen (30%) of the pressure sores were equipment induced., Conclusion: Our incidence of pressure sores in acute trauma victims is very low and occurs in those most seriously injured. The majority of the wounds are due to unrelieved pressure from body positioning or equipment failure.

Hemangiomas are vascular tumors that appear at or shortly after birth and undergo a rapid growth before involuting. During the proliferative phase, hemangiomas are infiltrated by macrophages, cells that are capable of initiating angiogenesis. Vascular malformations grow slowly, commensurate with the child, and do not regress or become infiltrated by macrophages. We demonstrate by in situ hybridization increased monocyte chemoattractant protein-1 (MCP-1) mRNA expression during hemangioma and vascular malformation growth. We found markedly upregulated expression of MCP-1 mRNA in all proliferative hemangioma specimens, expressed by alpha-actin perivascular smooth muscle cells and interstitial HAM 56+ macrophages. In contrast, 9 of 10 clinically involuting hemangiomas displayed no expression of MCP-1 mRNA. We found no expression of MCP-1 mRNA in vascular malformations, which correlates with the minimal monocytic infiltration of these lesions. We also showed that dexamethasone and interferon-alpha downregulate MCP-1 mRNA in cultured human vascular smooth muscle cells. Glucocorticoids can be efficacious in 30-50% of cases when given in the proliferative phase of hemangioma growth, but have no beneficial effect on vascular malformations. Interferon-alpha has been used to dramatically induce regression of steroid-refractory hemangiomas. Both of these agents' beneficial action on proliferative hemangiomas may, in part, result from reduced MCP-1 production and reduced influx of angiogenic macrophages.

Transplant arteriosclerosis (TA) is an immune mediated vascular injury that results in rapid intimal thickening and parenchymal ischemia. The monocytes that infiltrate the transplanted organ's vessels may mediate this effect. One of the cell signals that may initiate monocyte infiltration in rat models of acute and chronic vessel wall rejection is the product of the JE gene, a potent and selective chemoattractant for monocytes. We used combined in situ hybridization and immunocytochemistry to localize JE gene expression in specific cell types using a rat aortic transplant model of acute arterial graft rejection. Within 3 days of transplantation, there was JE mRNA expression in adventitial mesenchymal cells, probably fibroblasts, and this signal increased until 10 days post-transplantation. During this period, the number of adventitial monocytes, identified by immunocytochemistry, increased around the mesenchymal cells expressing JE. Intimal JE mRNA expression between 7 and 20 days localized to undefined mesenchymal intimal cells and occasionally blood-borne monocytes. There was no JE gene expression detected in the intima after 20 days. JE mRNA expression at 20 days was limited to alpha-actin-positive vascular smooth muscle cells in the outer aspect of the media. At 40 and 60 days, there was no JE hybridization signal in the media or adventitia despite increasing medial monocyte infiltration. The temporal sequence of JE mRNA expression, starting in the adventitia, intima, and finally the media, preceded or coincided with monocyte/macrophage infiltration. These results support our hypothesis that early JE gene expression may lead to the initial monocyte recruitment in acute vessel wall rejection. Because JE gene expression is downregulated by glucocorticoids, their immunosuppressive effect may be partly due to decreased JE gene expression by transplanted vessels.

Elastin is an important component of normal blood vessels and the extracellular matrix of atherosclerotic plaques, but its role in intimal thickening in the arteries of transplanted organs has not been defined. We have looked at elastin gene expression (by in situ mRNA hybridization) in an animal model using an abdominal aortic transplant between 2 strains of rats disparate for MHC class I antigens. The normal aortic wall of adult rats lacks elastin mRNA. Aortic allografts at 7 days after transplantation exhibit increased elastin mRNA in the medial vascular smooth muscle cells. This medial elastin mRNA expression is present only until 20 days after transplantation, and at later times, only the juxtaluminal cells of the neointima express elastin mRNA. Stainable elastin is detectable only in regions that previously demonstrated high levels of elastin mRNA. Combined in situ hybridization and immunocytochemistry reveals that most elastin mRNA-expressing cells in the media are alpha-actin-positive smooth muscle cells. In the neointima, elastin mRNA-expressing cells do not stain with antibodies to either smooth muscle alpha-actin or macrophage proteins. This cell population may represent a "synthetic" phenotype of vascular smooth muscle cell lacking alpha-actin protein. We presume there is immune cell-mediated injury leading to a vascular smooth muscle cell response and part of the vascular smooth muscle cell response may be increased elastin mRNA expression and elastin deposition in the allografts.

The role of endogenous growth factors in normal wound healing is not clear. Most of the data on growth factors in healing wounds have been obtained from the application of recombinant exogenous growth factors to animal and human wounds. We describe the immunolocalization of basic fibroblast growth factor (bFGF) in the injured dermis of skin from patients with partial and full-thickness burns. Three antibodies demonstrate an extracellular staining pattern of bFGF corresponding to areas of tissue injury that was most intense in specimens collected between 4 and 11 days post-burn injury. In contrast, bFGF staining appeared markedly decreased by Postburn Day 17 and was more consistent with uninjured tissue in a 30-day-old burn that had virtually reepithelialized. Basic FGF staining in the non-burned skin from the same patients was restricted to the dermal capillary basement membranes and the sweat glands, which is consistent with other reports of immunoreactive bFGF localization in normal adult skin. The immunohistochemical results were confirmed with Western immunoblots of the same tissue. The major band at 16.5 kDa, which is within the recognized range of the bFGF molecule's several forms, was detected in both burned and unburned tissue from the same patient. These findings support the hypothesis that bFGF is a presynthesized mediator that is stored in either the cells or extracellular matrix, is released locally from sites of direct injury, and may be important in early wound healing.

Antibodies to the proliferating cell nuclear antigen allow identification of proliferating cells in fresh tissue specimens using routine immunocytochemical methods. However, the use of such proliferation markers has not been verified for autopsy-derived tissue specimens, in which there is often a significant delay between the time of death and tissue specimen fixation. To assess the reliability of anti-proliferating cell nuclear antigen antibodies to identify proliferating cells in autopsy tissue specimens, an autopsy simulation was performed using fresh monkey and rat tissue specimens. These tissue specimens were kept at room temperature for predetermined numbers of hours before fixation. The proliferation specific staining was most reliable for tissue specimens obtained within 6 hours of death. There was reliable staining of proliferating regions up to 12 hours, although sensitivity was decreased. The only exception was skin, which was able to withstand much longer periods. Quantitative data from monkey spleen white-pulp regions showed 63% of the cells to stain for proliferating cell nuclear antigen when fixed immediately; this decreased to 29% of the cells after 12 hours and only 19% by 18 hours of postmortem simulation. Representative tissue specimens obtained from human autopsy material revealed similar postmortem staining patterns. Rapid procurement and fixation of tissue specimens and the use of control tissue specimens derived from the same autopsy material (eg, lymph node tissue) are recommended. These studies do suggest that anti-proliferating cell nuclear antigen antibodies can be used to identify proliferating cells in human autopsy tissue specimens obtained within approximately 12 hours of death, with some compromise in overall sensitivity.

Transplant arteriosclerosis (TA) has emerged as an obstacle to the long-term survival of transplanted organs, especially cardiac transplants. The animal models that have been used to study TA have not been fully characterized with regard to features such as the time course of cell proliferation and the sequence of cell types arriving in the developing intimal lesion. We present a model of TA based on a transplanted segment of abdominal aorta that helps address these questions. Two strains of rats (PVG x DA) underwent orthotopic aortic transplantation without immunosuppression and were killed at 14, 20, 40, and 60 days after transplantation. The within-strain control group displayed minimal evidence of cellular rejection with minimal to absent intimal lesions. In contrast, the allograft group showed a linearly increasing intimal lesion, up through 60 days after transplantation. The mechanism of intimal thickening was by an increase in cell number at the earlier time points with the later deposition of extracellular matrix. The early intimal lesion consisted mostly of mononuclear inflammatory cells (45%) with gradually increasing presence of smooth muscle cells (SMC) in the intima between 20 and 60 days. Conversely, the media showed gradual infiltration by macrophage-type cells with virtual loss of all SMC from the media by 40 days. The proliferative index showed a peak of 6% and 8% at 20 days in both the intima and media, respectively, and was preceded by the presence of macrophages. In fact, most of the proliferating cells at the earlier time points were either monocytes/macrophages, or were immediately adjacent to monocyte-/macrophage-rich regions. This straight artery segment model of transplant arteriosclerosis provides an easily quantifiable system in which the effects of different interventions (e.g., immunosuppressive regimens) can be tested.

Most injuries to the lung can be managed nonoperatively and rarely require resection. A case of bilateral hilar injuries requiring bilateral pulmonary lower lobectomies is presented. The resulting pulmonary hypertension, right ventricular failure, and cardiogenic shock explain the high mortality following extensive pulmonary resection in thoracic trauma and support the concept of a conservative approach.

Arterioportal fistulae can develop from a variety of causes, either congenital, iatrogenic, or acquired. They can have a varied clinical presentation including acute upper gastrointestinal bleeding, ischemic colitis, abdominal pain, ascites, and abdominal bruit. In the past the treatment has been ligation and surgical excision of the fistula with repair of the artery and vein or hepatic lobectomy. We report two patients with arterioportal fistulae between the hepatic artery and portal vein as a result of liver biopsy and transhepatic portography. Both patients were treated successfully by nonoperative radiologic intervention.