Summary Comparative genomic hybridization (CGH) was used to detect copy number changes of DNA sequences in the Ewing family of tumours (ET). We analysed 20 samples from 17 patients. Fifteen tumours (75%) showed copy number changes. Gains of DNA sequences were much more frequent than losses, the majority of the gains affecting whole chromosomes or whole chromosome arms. Recurrent findings included copy number increases for chromosomes 8 (seven out of 20 samples; 35%), 1q (five samples; 25%) and 12 (five samples; 25%). The minimal common regions of these gains were the whole chromosomes 8 and 12, and 1q21-22. High-level amplifications affected 8q13-24, 1q and 1q21-22, each once. Southern blot analysis of the specimen with high-level amplification at 1q21-22 showed an amplification of FLG and SPRR3, both mapped to this region. All cases with a gain of chromosome 12 simultaneously showed a gain of chromosome 8. Comparison of CGH findings with cytogenetic analysis of the same tumours and previous cytogenetic reports of ET showed, in general, concordant results. In conclusion, our findings confirm that secondary changes, which may have prognostic significance in ET, are trisomy 8, trisomy 12 and a gain of DNA sequences in 1 q.

Ezrin is a member of the ezrin-radixin-moesin family of proteins that link the actin-containing cytoskeleton to the plasma membrane. Ezrin is also connected to signaling molecules involved in the regulation of cell survival, proliferation and migration. Here, we examined the expression of ezrin in 95 primary cutaneous melanomas and correlated ezrin expression with conventional prognostic factors and biomarkers. From 12 patients metastatic tissue samples were also examined. In addition to ezrin staining, Mib-1 proliferation antigen, p53 and Bcl-2 were evaluated. Ezrin immunoreactivity was seen in most tumors; only 19 (20%) melanomas were negative. A total of 48 (51%) tumors had weak immunoreactivity and 28 (29%) strong immunoreactivity. The intensity of ezrin immunoreactivity was associated with tumor thickness (Breslow, P ¼ 0.0008) and with tumor invasion level (Clark, P ¼ 0.004), thicker tumors having stronger immunoreactivity. Also, there was a correlation between higher Mib-1 index in tumors and strong ezrin expression. All metastatic samples (n ¼ 12) showed positive ezrin immunoreactivity. In univariate analysis of survival, patients (n ¼ 76) with positive ezrin immunoreactivity had worse clinical disease behavior than those (n ¼ 19) without ezrin immunoreactivity, but the difference was not significant (P ¼ 0.19). In multivariate analysis of survival, the ezrin immunoreactivity was not a significant marker. The results indicate that ezrin is expressed in most primary melanomas of the skin and in all metastatic tumors. Ezrin expression correlates with tumor thickness and level of invasion suggesting an association between ezrin expression and tumor progression.

A retrospective analysis of 75 consecutive free flap patients, operated on during 1989-1990, was performed to find out more about factors associated with free flap failure or immediate vascular complications. The overall failure rate was 9.3% (7/75) and the immediate vascular complication rate 22.7% (17/75). Sixteen patients required explorative surgery during the first postoperative day. The results were statistically analysed to find factors promoting either failure or vascular complications. Pre-operative infection of the recipient site or prolonged operation time correlated with flap failure. The use of a vein graft or long per-operative ischaemia correlated with immediate vascular complications. It is interesting that many factors often blamed for failure (age, body mass, history of cardiovascular disease, smoking, or previous irradiation of the recipient site) were not significant in this study.

Edema and insufficient blood perfusion are common problems in reconstructive surgery. The blood vasculature is reconstructed in microvascular flaps, whereas lymphatic vessel function is lost after surgical incision. Here, we demonstrate that vascular endothelial growth factor C (VEGF-C) gene transfer can be used to reconstruct a lymphatic vessel network severed by incision of skin flaps. We used adenoviral VEGF-C gene transfer at the edges of epigastric skin flaps in mice. Our results show that VEGF-C gene expression results in the formation of anastomoses between the lymphatic vessels of the skin flap and the surrounding lymphatic vasculature. Some spontaneous lymphangiogenesis also took place in the control mice, but the lymphatic vessels generated remained nonfunctional even 2 months postoperatively. In contrast, the VEGF-C treated mice demonstrated persistent lymphatic vessel function during the 2 month follow-up despite the transient nature of the adenoviral VEGF-C gene expression. The restoration of lymphatic function by VEGF-C in skin flaps provides new tools to promote vascular perfusion and to reduce tissue edema in skin and muscle flaps. These results have important implications for the prevention and treatment of surgically induced secondary lymphedema.

Free flap surgery is routine today, yet little is known of its pathophysiology. In this study, the authors evaluated the hemodynamics in different types of free microvascular flaps, by measuring intraoperative transit-time flow. Eighty-six free transplants--21 free TRAM flaps for breast reconstruction, 18 radial forearm flaps for head and neck reconstructions, and 47 muscle flaps for head and neck, trunk and lower extremity reconstructions--were studied. Donor artery flow was highest in the radial artery (mean: 57.5 +/- 50 (SD) ml/min) but dropped (p < 0.001) to one tenth (6.1 +/- 2 ml/min) after anastomosis. The flow was lowest (4.9 +/- 3 ml/min) in the recipient artery of the TRAM flap but, after anastomosis, increased significantly (13.7 +/- 5 ml/min) to the level of the flow in the donor artery. The donor-artery flow in muscle flaps had a mean of 15.9 +/- 11 ml/min, and it significantly increased after anastomosing (23.9 +/- 12 ml/min). Weight-related intake of blood was highest in the radial forearm flap (18.5 +/- 6 ml/ min/100g) and lowest in the TRAM flap (2.5 +/- 1 ml/min/100g). The study showed that blood flow through a free microvascular flap does not depend on recipient artery flow. Even low-flow arteries can be used as recipients, because the flow increases according to free-flap requirements. The blood flow through a free microvascular flap depends on the specific tissue components of the flap.