Feasibility of Neovessel Embolization in a Large Animal Model of Tendinopathy: Safety and Efficacy of Various Embolization Agents

Abstract

Targeting neovessels in chronic tendinopathies has emerged as a new therapeutic approach and several embolization agents have been reported. The aim of this study was to investigate the feasibility of embolization with different agents in a porcine model of patellar tendinopathy and evaluate their safety and efficacy. Eight 3-month-old male piglets underwent percutaneous injection of collagenase type I to induce patellar tendinopathies (n = 16 tendons). They were divided into four groups (2 piglets, 4 tendons/group): the control group, 50-100 µm microspheres group, 100-300 µm microspheres group, and the Imipenem/Cilastatin (IMP/CS) group. Angiography and embolization were performed for each patellar tendon on day 7 (D7). The neovessels were evaluated visually with an angiography on day 14. The pathological analysis assessed the efficacy (Bonar score, number of neovessels/mm²) and safety (off-target persistent cutaneous ischemic modifications and presence of off-target embolization agents). The technical success was 92%, with a failed embolization for one tendon due to an arterial dissection. Neoangiogenesis was significantly less important in the embolized groups compared to the control group angiographies (p = 0.04) but not with respect to histology (Bonar score p = 0.15, neovessels p = 0.07). Off-target cutaneous embolization was more frequently depicted in the histology of the 50-100 µm microspheres group (p = 0.02). Embolization of this animal model with induced patellar tendinopathy was technically feasible with different agents and allowed assessing the safety and efficacy of neovessel destruction. Particles smaller than 100 µm seemed to be associated with more complications.

Keywords: embolization; imipenem/cilastatin; neovessels; tendinopathy.

Figure 1

Flowchart of the study. US: ultrasound; D: day; DSA: digital subtraction angiography; *: one tendon was not treated due to femoral artery dissection on D7 angiography.

Figure 2

Embolization procedure at D7 using various embolization agents: microspheres (A–C) and imipenem/cilastatin (D–F). (A) Initial digital subtraction angiography (DSA) of a right geniculate artery showing neovascularization-related blush (white circle). (B) After embolization using 50–100 µm microspheres according to the “pruning” technique, DSA showed a “dead tree” aspect of the embolized area with a filling area by collaterals (white arrow). (C) DSA control at D14 showed no recurrence of neovascularization at the site of tendinopathy (white arrowhead). (D) The initial DSA of a right geniculate artery showing a blush related to neovascularization (black circle). (E) After embolization with an IMP/CS emulsion in an iodinated contrast medium, opacification revealed a truncated appearance of the geniculate artery (black arrow). (F) The DSA control at D14 showed repermeabilization of the native artery with no recurrence of neovessels (black arrowhead).

Figure 3

Histological evaluation of the safety of embolization. Using a standard stain (Hemathein Eosin Saffron) at ×40 (A), ×100 (B,C), and ×200 (D) magnification on tendon and skin samples taken 7 days after embolization. (A,B) Tendon treated by 50–100 µm microspheres. The peritendinous artery obliterated by the embolization material (arrows) with a granulomatous reaction. (C,D) Skin samples in the area of a transient livedo episode during embolization with Imipenem/Cilastatin (IMP/CS) emulsion. Dermohypodermal arteriole with granulomatous reaction with giant cells (black arrowheads) resorbing exogenous debris (IMP/CS crystals).

Figure 4

In vitro exploration. (A) Size distribution of 50 mg of imipenem/cilastatin (IMP/CS) particles in 1 mL of iodinated contrast dispersed during 10 s. (B) ×400 magnification of 50 mg/mL IMP/CS in iodinated contrast emulsion.