Diaphragm Repair with a Novel Cross-Linked Collagen Biomaterial in a Growing Rabbit Model

Neonates with congenital diaphragmatic hernia and large defects often require patch closure. Acellular collagen matrices (ACM) have been suggested as an alternative to synthetic durable patches as they are remodeled by the host or could also be used for tissue engineering purposes.

2.0x1.0 cm diaphragmatic defects were created in 6-weeks old New-Zealand white rabbits. We compared reconstruction with a purpose-designed cross-linked ACM (Matricel) to 4-layer non-cross-linked small intestinal submucosa (SIS) and a 1-layer synthetic Dual Mesh (Gore-Tex). Unoperated animals or animals undergoing primary closure (4/0 polyglecaprone) served as age-matched controls. 60 (n = 25) resp. 90 (n = 17) days later, animals underwent chest x-ray and obduction for gross examination of explants, scoring of adhesion and inflammatory response. Also, uniaxial tensiometry was done, comparing explants to contralateral native diaphragmatic tissue.

Overall weight nearly doubled from 1,554±242 g at surgery to 2,837±265 g at obduction (+84%). X-rays did show rare elevation of the left diaphragm (SIS = 1, Gore-Tex = 1, unoperated control = 1), but no herniation of abdominal organs. 56% of SIS and 10% of Matricel patches degraded with visceral bulging in four (SIS = 3, Matricel = 1). Adhesion scores were limited: 0.5 (Matricel) to 1 (SIS, Gore-Tex) to the left lung (p = 0.008) and 2.5 (Gore-Tex), 3 (SIS) and 4 (Matricel) to the liver (p<0.0001). Tensiometry revealed a reduced bursting strength but normal compliance for SIS. Compliance was reduced in Matricel and Gore-Tex (p<0.01). Inflammatory response was characterized by a more polymorphonuclear cell (SIS) resp. macrophage (Matricel) type of infiltrate (p<0.05). Fibrosis was similar for all groups, except there was less mature collagen deposited to Gore-Tex implants (p<0.05).

Matricel induced a macrophage-dominated inflammatory response, more adhesions, had appropriate strength but a lesser compliance compared to native tissue. The herein investigated ACM is not a viable option for CDH repair.