[Analysis of outcome of free jejunal-autograft for head and neck reconstruction--postoperative complications and functional results of swallowing in 49 cases].

We evaluated postoperative results in 49 patients (39 men, 10 women) who underwent pharyngoesophageal reconstruction with free-jejunal autograft following total pharyngolaryngoesophagectomy in the Department of Otolaryngology, Hokkaido University School of Medicine from 1989 to 1997. Evaluation was performed regarding the following points: 1) postoperative complications, 2) factors that determine the functional results of swallowing, 3) relation between forms of jejunal anastomosis and swallowing. The primary malignancy site was hypopharynx (39), cervical esophagus (4), larynx (3), thyroid (2) and trachea (1), Reconstructions were made with free jejunal autograft alone in 45 cases and with free jejunum in combination with gastric pull-up in 4. In patients who underwent reconstruction with jejunum alone, the anastomosis of the jejunum to the pharynx was performed in side-to-end fashion in 22, end-to-end in 18 and rho-shaped in 4. In the 1 remaining case, we used jejunal-patch graft. Postoperative complications including minor or nongraft related, occurred in 24 of 49 (49.0%) patients. Among these, graft-related complications were graft failure in 1 (2.0%), fistula formation in 3 (6.1%) and graft stricture in 2 (4.0%). Re-operations were required in one case of graft failure and 2 of fistula formation. Consequently, the overall graft-survival rate was 98.0% (48/49). Therefore, we considered the method of reconstruction to be a reliable procedure with a high-success rate. The swallowing function after reconstructive surgery was studied in 35 patients who underwent side-to-end (18) and end-to-end (17) anastomosis of the jejunum to the pharynx. We indicated that appropriate tension in the jejunum was the most important factor for adequate swallowing function. The end-to-end group had a higher rate of taking normal diet compared with the side-to-end group. The rate of swallowing dysfunction was only 5.9% (1/17) in the end-to-end group. On the other hand, 4 of 18 (22.2%) cases in the side-to-end group were regarded as having poor swallowing function. As a result we considered the end-to-end proximal jejunal anastomosis to be the more desirable form.

Production of matrix metalloproteinases at the bone-implant interface in loose total hip replacements.

BACKGROUND: Incidence of aseptic loosening of hip prostheses is increasing in recent years. Previous studies suggested involvement of proteinases and cytokines in the accelerated bone lysis associated with loosening. EXPERIMENTAL DESIGN: To investigate the role of matrix metalloproteinases (MMPs) in the loosening we immunolocalized MMP-1 (tissue collagenase), MMP-2 (gelatinase A), MMP-3 (stromelysin-1), MMP-9 (gelatinase B) and their common inhibitors, tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2), in the bone-prosthesis interface membranes. In situ hybridization was performed for the detection of MMP-9 mRNA in the membranes. The amounts of these MMPs and TIMPs in the tissue were measured by the sandwich enzyme immunoassays and enzyme activities assayed using radiolabeled collagen, gelatin, and carboxymethylated transferrin substrates. We also examined the ability of the cells from interface membranes to resorb mouse calvaria bone. RESULTS: The membranes obtained from the loose bone-implant interface were composed of fibrous granulation tissue containing numerous multinucleated giant cells with high density polyethylene debris. Immunohistochemical examination revealed that the giant cells were strongly positive for MMP-9 and weakly for MMP-1. Expression of MMP-9 mRNA in the cells was demonstrated by in situ hybridization. MMP-2 and TIMP-2 were immunolocalized mainly in the fibroblasts. TIMP-1 was localized in the endothelial cells of the blood vessels and weakly in fibroblasts. However, MMP-3 was almost negative in the membrane tissue. Sandwich enzyme immunoassays showed that MMP-9 levels are significantly higher in both homogenates and culture media of the cup and stem interface membranes than the control pseudocapsule. Gelatinolytic activity was also remarkably higher in the membrane samples than the control. The cells isolated from the membranes had no ability to resorb calvaria bone. CONCLUSIONS: These data demonstrate that MMP-9 is produced by the multinucleated giant cells appeared by the reaction to polyethylene debris in the interface membranes. This proteinase may play a role in degradation of the extracellular matrix macromolecules present around and on the surface of the bone trabeculae, facilitating the osteoclastic bone resorption.

Matrix metalloproteinase 7 (matrilysin) from human rectal carcinoma cells. Activation of the precursor, interaction with other matrix metalloproteinases and enzymic properties.

Matrix metalloproteinase 7 (MMP-7) has been purified as an inactive zymogen of M(r) 28,000 (proMMP-7) from the culture medium of CaR-1 human rectal carcinoma cells. The NH2-terminal sequence of proMMP-7 is Lys-Pro-Lys-Pro-Gln-Glu, which is identical to that of matrilysin. The zymogen is activated by 4-aminophenylmercuric acetate (APMA), yielding an intermediate form of M(r) 21,000 and an active species of M(r) 19,000 which shows the new NH2-terminal sequence of Tyr78-Ser-Leu-Phe-Pro-Asn-Ser. Although trypsin fully activates the zymogen, the activation rate by plasmin or leukocyte elastase is confined to approximately 50%. ProMMP-7 can be activated by MMP-3 (stromelysin 1) to its full activity in a single-step mechanism and generates the same NH2 terminus obtained by APMA activation, whereas MMP-1 (tissue collagenase), MMP-2 (gelatinase A), and MMP-9 (gelatinase B) do not have such an effect. On the other hand, proMMP-1 is activated by MMP-7 to an activity similar to that obtained by APMA and the activation by MMP-7 is enhanced up to approximately 6.5 fold in the presence of APMA. This enhanced activity is donated by specific cleavage at the Gln80-Phe81 bond of proMMP-1. MMP-7 can also activate proMMP-9 up to approximately 50% of the full activity with a new NH2 terminus of Leu16-Arg-Thr-(Asn)-Leu. Incubation of proMMP-2 or proMMP-3 with MMP-7 results in no activation of these proMMPs. MMP-7 degrades type IV collagen, laminin-1, fibronectin, proteoglycan, type I gelatin, and insoluble elastin. These results suggest that in vivo MMP-7 may play a role in degradation of extracellular matrix macromolecules in concert with MMP-1, -3, and -9 under pathological conditions.