Mathieu van Steenberghe

Streptozotocin-induced diabetes in large animals (pigs/primates): role of GLUT2 transporter and beta-cell plasticity.

Background. To induce irreversible diabetes in large animals, the efficiency of streptozotocin (STZ) was evaluated in pigs, primates and compared to the gold standard model in rats. Methods. Low (50 mg/kg) and high (150 mg/kg) doses of STZ were tested. Hepatic/renal function, glucose metabolism (intravenous glucose tolerance tests, fasting blood glucose) and histomorphometry were evaluated prior to, 1, and 4 weeks after STZ treatment. Results. In rats and primates, expressing a high level of GLUT2 expression on &bgr; cells, a dose of 50 mg/kg STZ induced irreversible diabetes (due to the 97% destruction of beta cell mass) without provoking liver or renal failure. In pigs, despite the use of high STZ dose, partial correction of hyperglycaemia was observed four weeks after STZ injection (decreased fasting blood glucose and intravenous glucose tolerance tests; increased insulin production). The correction of hyperglycaemia was associated with significant hypertrophy of immature pig &bgr;-cell clusters (+30%, P<0.05), whereas no hypertrophy was observed in rats/primates. Conclusion. These results demonstrated that STZ might be used to induce irreversible diabetes in rats and primates. In contrast, the low STZ sensitivity in pigs related to a low expression of GLUT2, higher number of immature &bgr; cells and compensatory &bgr;-cell hypertrophy, renders STZ-induced diabetes inappropriate for studying islet allografts in swine.

Single coronary artery and neonatal arterial switch operation: early and long-term outcomes†

OBJECTIVES The presence of single coronary artery (CA) in the arterial switch operation (ASO) for neonatal treatment with transposition of the great arteries (TGA) has been reported to be an independent risk factor for early death after surgical repair and late reintervention. The study objective was to evaluate the mortality and the CA stenosis risk at early and long term in neonatal ASO for TGA and single CA. METHODS Between January 1987 and January 2010, 979 neonates underwent an ASO, of which 73 had a single CA (7.5% of all cohort): right ostium with posterior left CA loop was the most frequent pattern (63%), followed by left ostium with an anterior right CA loop (26%). Mean age at operation was 9.3 ± 5.7 days. Mean follow-up was 9.8 years (range: 1-20 years). RESULTS Eight patients (11%) died, 6 of coronary-related death. Overall, survival was 90.1 ± 1.9% at 1 year and 88.6% ± 3.8% at 2, 5, 10 and 15 years, respectively. Independent risk factor for mortality was associated surgery before 2001. Freedom from coronary events was 91.6 ± 3.3% at 1 year and 88.7 ± 3.8% at 5, 10 and 15 years respectively. No patients required late reintervention for CA surgery or angioplasty. CONCLUSIONS All coronary-related death occurred within the first 6 months after ASO, and all patients but 1 were operated before 2001. In our experience, it appears that a single CA is not any more a risk factor for early and late mortality after ASO for TGA. Mortality has drastically reduced since 2001 and is now close to that found in TGA with standard coronary patterns. The acquired experience shared between the surgeons and the institution offsets the undeniable surgical difficulty.

Feasibility of transcatheter techniques for intracardiac and extracardiac cavocaval connection in principle for Fontan completion in chronic animal models

OBJECTIVES We report the safety and feasibility of various transcatheter techniques of cavocaval connection in principle for the completion of Fontan circulation in viable, chronic and ovine heart models. Surgically simulated preparations of both intracardiac and extracardiac cavocaval connections were studied. METHODS Sixteen sheep were divided into two groups per the type of surgical preparation. All animals underwent standard right thoracotomy with interposition of a 20-mm Gore-tex(®) conduit between the superior vena cava (SVC) and the right atrium (RA). Nitinol rings were placed around the SVC and the inferior vena cava (IVC). In Group I (intracardiac, n = 10), the SVC-RA junction was closed using a polytetrafluoroethylene (PTFE) membrane 1 cm below the SVC-Gore-tex(®) anastomosis. In Group II (extracardiac, n = 6), a 20-mm Gore-tex conduit de-aired and filled with heparinized saline was anastomosed to connect the SVC and the IVC. The IVC end was anastomosed in a termino-lateral fashion and the SVC end in a termino-terminal fashion; both the ends were occluded with a PTFE membrane. Animals were scheduled for transcatheter cavocaval connection after a variable healing period. RESULTS Four animals in Group I died; three early and one late after surgical preparation. After a median interim period of 1 month (0-9 months), five sheep from Group I and six from Group II underwent successful transcatheter cavocaval connection. Perforation of the PTFE membrane was successful in all animals. Covered stents were deployed precisely and with good stability ensured by the nitinol rings. All animals survived transcatheter completion and were sacrificed after a median follow-up of 4 months (0-8 months) per protocol. No stent migration, thromboembolic events, residual shunts or paraprosthetic leak was noticed on angiographic evaluation or at autopsy in any animal. CONCLUSIONS Transcatheter techniques for completion of cavocaval connection in surgically simulated, chronic animal models is safe and feasible. Both techniques were equally successful with no failures or short-term complications. Such techniques should work in principle for completion of intracardiac and extracardiac Fontan circulation.

Feasibility of creating a novel animal heart model to test transcatheter techniques for a cavocaval connection that mimics a Fontan completion

OBJECTIVE The objective of this study was to create a novel animal model to foster the future development of interventional techniques for a cavocaval connection that mimics a Fontan completion. METHODS Ten sheep were studied. All had the superior vena cava-right atrium junction closed using a polytetrafluoroethylene membrane. A valveless Gore-Tex conduit connecting the terminal portion of the superior vena cava to the right atrium was used to bypass the polytetrafluoroethylene occlusion and to allow normal venous drainage through the right atrium. Radio-opaque nitinol rings were placed around the inferior vena cava near its entry in the right atrium to allow better fluoroscopic visualization and to enhance stent stability during transcatheter cavocaval connection. RESULTS The first 3 animals died during the learning curve as a result of technical issues. The subsequent 7 surviving sheep showed good flow dynamics on cardiac catheterization. Transcatheter cavocaval connection was performed successfully performed in 6 sheep at 0 to 9 months after the initial surgery. Sacrifice was done electively in all animals at 1 to 9 months per protocol. One animal was euthanized early because of an untreatable infection. One sheep was observed as a control without a transcatheter cavocaval connection and was sacrificed at 12 months. CONCLUSIONS A novel, chronic ovine model to foster development of transcatheter techniques for cavocaval connection to mimic a Fontan completion was created. The successful model is easily reproducible after a short learning curve and shows good survival.

Enhanced Vascular Biocompatibility and Remodeling of Decellularized and Secured Xenogeneic/Allogeneic Matrices in a Porcine Model

Background/Purpose: Calcifications and absence of growth potential are the major drawbacks of glutaraldehyde-treated prosthesis. Decellularized and secured xeno-/allogeneic matrices were assessed in a preclinical porcine model for biocompatibility and vascular remodeling in comparison to glutaraldehyde-fixed bovine pericardium (GBP; control). Methods: Native human (fascia lata, pericardium) and porcine tissues (peritoneum) were used and treated. In vitro, biopsies were performed before and after treatment to assess decellularization (hematoxylin and eosin/DAPI). In vivo, each decellularized and control tissue sample was implanted subcutaneously in 4 mini-pigs. In addition, 9 mini-pigs received a patch or a tubularized prosthesis interposition on the carotid artery or abdominal aorta of decellularized (D) human fascia lata (DHFL; n = 4), human pericardium (DHP; n = 9), porcine peritoneum (DPPt; n = 7), and control tissue (GBP: n = 3). Arteries were harvested after 1 month and subcutaneous samples after 15–30 days. Tissues were processed for hematoxylin and eosin/von Kossa staining and immunohistochemistry for CD31, alpha-smooth muscle actin, CD3, and CD68. Histomorphometry was achieved by point counting. Results: A 95% decellularization was confirmed for DHP and DPPt, and to a lower degree for DHFL. In the subcutaneous protocol, CD3 infiltration was significantly higher at day 30 in GBP and DHFL, and CD68 infiltration was significantly higher for GBP (p < 0.05). In intravascular study, no deaths, aneurysms, or pseudoaneurysms were observed. Inflammatory reaction was significantly higher for DHFL and GBP (p < 0.05), while it was lower and comparable for DHP/DPPt. DHP and DPPt showed deeper recellularization, and a new arterial wall was characterized. Conclusions: In a preclinical model, DPPt and DHP offered better results than conventional commercialized GBP for biocompatibility and vascular remodeling.