Multi-layer electrospun membrane mimicking tendon sheath for prevention of tendon adhesions

Abstract

Defect of the tendon sheath after tendon injury is a main reason for tendon adhesions, but it is a daunting challenge for the biomimetic substitute of the tendon sheath after injury due to its multi-layer membrane-like structure and complex biologic functions. In this study, a multi-layer membrane with celecoxib-loaded poly(l-lactic acid)-polyethylene glycol (PELA) electrospun fibrous membrane as the outer layer, hyaluronic acid (HA) gel as middle layer, and PELA electrospun fibrous membrane as the inner layer was designed. The anti-adhesion efficacy of this multi-layer membrane was compared with a single-layer use in rabbit flexor digitorum profundus tendon model. The surface morphology showed that both PELA fibers and celecoxib-loaded PELA fibers in multi-layer membrane were uniform in size, randomly arrayed, very porous, and smooth without beads. Multi-layer membrane group had fewer peritendinous adhesions and better gliding than the PELA membrane group and control group in gross and histological observation. The similar mechanical characteristic and collagen expression of tendon repair site in the three groups indicated that the multi-layer membrane did not impair tendon healing. Taken together, our results demonstrated that such a biomimetic multi-layer sheath could be used as a potential strategy in clinics for promoting tendon gliding and preventing adhesion without poor tendon healing.

Figure 1

Scanning electron microscopy (SEM) observation for cross-sectional features of the multi-layer membrane (A) and surface morphological features of the poly(l-lactic acid)-polyethylene glycol (PELA) electrospun fibers (B) and celecoxib-loaded PELA electrospun fibers (C).

Figure 2

(A) Gross evaluation of the rabbit flexor digitorum profundus (FDP) tendon model in the untreated control group, PELA membrane group, and multi-layer membrane group. Tendon (T) is indicated in the figures and adhesion tissue is pointed to using black arrows. Tendon repair and peritendinous adhesions are evaluated by determining macroscopic evaluation of tendon adhesions (B), maximum tensile strength (C), and work of flexion (D). * p < 0.05 compared with control group; ^† p < 0.05 compared with PELA membrane group. Data are expressed as mean ± SEM for six tendons/group.

Figure 3

HE (A) and Masson (B) staining of untreated repair site, repair sites wrapped with unloaded PELA fibrous membrane and wrapped with a multi-layer fibrous membrane. White arrowheads indicate the interface without peritendinous adhesions, while black arrowheads indicate peritendinous adhesions between the membrane (M) and tendon (T) (Scale bar = 1 mm). Peritendinous adhesions and tendon repair are evaluated by determining histological adhesion grade (C) and histological healing grade (D). * p < 0.05 compared with control group; ^† p < 0.05 compared with PELA membrane group. Data are expressed as mean ± SEM for six tendons/groups.

Figure 4

Western blotting assay for collagen I and collagen III expression in repair sites with the untreated group, PELA fibrous membrane group, and multi-layer fibrous membrane group for three weeks (A). Densitometry of collagen I (B) and collagen III (C) from Western blot experiment. Data are expressed as mean ± SEM for six tendons/group.

Figure 5

Schematic illustration of multi-layer membrane as physical barrier for preventing tendon adhesions after tendon injury.