Experimental study of tendon sheath repair via decellularized amnion to prevent tendon adhesion

Abstract

The adhesion of tendon and surrounding tissue is the most common complication after repairing an injured tendon. The injured flexor tendons in zone II are frequently accompanied by tendon sheath defects, which lead to poor recovery. A variety of biological and non-biological materials have been recently used for repair or as substitute for tendon sheaths to prevent tendon adhesion. However, non-biological materials, such as polyethylene films, have been used to prevent tendon adhesions by mechanical isolation. The possibility of tendon necrosis and permanent foreign body remains due to the lack of permeability and the obstruction of nutrient infiltration. The natural macromolecule amniotic membrane derived from organisms is a semi-permeable membrane with the following characteristics: smooth; without vascular, nerve, and lymphatic; and rich in matrix, cytokines, enzymes, and other active ingredients. The unique structure of this membrane makes it an ideal biomaterial. In the experiment in Henry chicken, the model of tendon sheath defect and the flexor digitorum tendon in zone II was established and randomly divided into control group, medical membrane group, and decellularized amniotic membrane group. Samples were obtained at the 2nd, 4th, 8th, and 12th week after operation. General, histological, and biomechanical tests were performed to investigate the preventive effect of repaired tendon sheath by decallularized amniotic membrane. Experimental results showed the following: the amniotic membrane group and the medical membrane group had mild inflammatory reaction and tissue edema, and nearly no adhesion was observed in the surrounding tissue; the fibroblast-like cells were distributed in layers under the light microscope; the amniotic membrane group was denser than the medical membrane group cells, and numerous fibroblasts were disorganized in the control group. Biomechanical measurements showed that the sliding distance of tendon, the total flexion angle of the toes, and the tendon maximum tensile breaking strength at the early postoperative were significantly better than in the control group. Through this experiment, the amniotic membrane, as a natural biological substitute material in the construction of tendon sheath, can effectively inhibit exogenous healing and promote endogenous healing to prevent tendon adhesion.

Fig 1. Similarity of chicken toe tendons to human digital flexor tendons.

Flexor digitorum superficialis(FDS), flexor digitorum profundus (FDP). A and B quoted and modified from Xu Y, Tang JB. J Hand Surg. 2003;28A:944–1001[11]. A. The red box is the range of tendon sheath around the tendon. B. The incision and suture of the tendon after stripping part of the tendon sheath. C. Schematic diagram of acellular amniotic membrane repairing tendon sheath.

Fig 2. Biomechanical testing of tendons.

A. Inside the red box is the tendon repaired after injury. B. The traction force gradually increased and the tendon began to break partially.

Fig 3. General observation of each group at the 2,4,8 week after operation.

(A,B,C) No complete tendon sheath was observed in the control group. The tendon anastomosis was connected with fibrous connective tissue, which was firm and difficult to remove around the tendon tissue. The red circle refers to an adhesion formed between the tendon and the surrounding tissue. In the medical membrane group (D,E,F) and amniotic membrane group (G,H,I), the tendon sheath was reconstructed, and the tendon anastomosis healed well without adhesion. The medical membrane was vaguely visible while the amniotic membrane was unrecognizable.

Fig 4. Histological observation of the tendon sheath defection in each group at the 2nd, 4th, and 8th week after operation (hematoxylin–eosin staining, x 400).

At the 2nd week, each group was filled with hyperemia, edema, and inflammatory cell infiltration, but the amniotic membrane group (G) was the lightest. At the 4th week, inflammatory cells were significantly reduced than before. In the medical membrane group (E) and the amniotic membrane group (H), fibroblast cells were layered, and the amniotic membrane group was dense. In the control group (B), numerous fibroblasts were disorderly distributed. At the 8th week, the fibroblasts in the medical membrane group (F) and the amniotic membrane group (I) were arranged neatly, and the structure was dense. In the control group (C), the structure of tendon sheath was loose, and the distribution of fibroblasts was disorganized.

Fig 5. Transmission electron microscopy observations of the defection area of tendon sheath at the 4th and 8th week after operation.

The number of synovial cells in the control group (A–D) was significantly lower than that of the two other groups. Moreover, the content of the rough endoplasmic reticulum in the cell matrix was low, and the extension degree was not evident. The medical membrane group (E–H) was similar to the amnion group, but the number of synovial cells was lower than that of the amnion group. Numerous synovial cells were observed in the amniotic group (I–L). Furthermore, numerous dilated rough endoplasmic reticulum and secretions in the endoplasmic reticulum were found in the matrix.

Fig 6. Comparison of biomechanical results at different time points after operation.

The SNK method was used to compare the two groups, and the P < 0.05 indicated that the difference was statistically significant. (A) No significant difference was found in the slipping distance between the amniotic membrane group and the medical membrane group at each time point, which was higher than that of the control group (P < 0.05). (B) No significant difference was found in the angle of total flexion between the amniotic membrane group and the medical membrane group at each time point after the operation, which was higher than that of the control group (P < 0.05). (C) No significant difference was found in the maximum tensile breaking strength between the 2nd and 12th week after operation (P > 0.05). At the 4th and 8th week, the maximum tensile breaking strength of the amniotic membrane group and medical membrane group was higher than that of the control group (P < 0.05), whereas the amniotic membrane group was higher than that of the medical membrane group (P < 0.05).