Characterization of Ultrasound-Triggered Bulk Antibiotic Release from Novel Spinal Hardware

This study evaluated the efficacy of ultrasound-triggered drug delivery devices against bacterial infection in an ex vivo cadaveric rabbit spine model. Polylactic acid (PLA)-coated, vancomycin (VAN)-loaded polyether ether ketone (PEEK) devices (1 cm3) with a drug-loading reservoir (0.785 cm3) were 3D printed. Two device designs were evaluated: 1 large hole for drug release vs 2 smaller holes. Clips were implanted medial to the spinal midline in mature (~6 months, 3 kg) female White New Zealand cadaveric rabbits (n=4) under an IACUC-approved protocol. To simulate infection, 104 CFU of Staphylococcus aureus were added to 2 of the 4 sites; the other 2 sites were left clean. Two of the 4 sites (1 inoculated, 1 clean) were insonated for 20 minutes with a Logiq E9 ultrasound scanner (GE Healthcare, Waukesha, WI, USA) equipped with a C1-6 curvilinear probe, using power Doppler imaging (1.7 MHz frequency, 6.4 kHz PRF, 100% acoustic output power) to induce rupture of the PLA coating for VAN release. Infected sites showed marked reduction in bacterial colonization following ultrasound-triggered VAN release, while uninsonated sites exhibited little reduction in bacterial colonization. At 48 hours, there was significantly greater VAN release from the insonated clips compared to the uninsonated clips (p < 0.04). There was significantly greater ultrasound-triggered total VAN release from the 1-hole device design than from the 2-hole design (7420 ± 2992 µg vs. 3500 ± 954 µg, p < 0.0001). These levels are sufficient to prevent adhesion of S. aureus to implant materials. This study demonstrated the feasibility of an ultrasound-mediated antibiotic delivery device, which could become a potent weapon against spinal surgical site infections.